Copyright is owned by the Author of the thesis.  Permission is given for 
a copy to be downloaded by an individual for the purpose of research and 
private study only.  The thesis may not be reproduced elsewhere without 
the permission of the Author. 
 
Phase change, flowering and 
postharvest characteristics 
of Metrosideros excelsa 
(Myrtaceae) 
A thesis presented in  partial fulfilment of the 
requirements for the degree of 
Doctor of Philosophy 
m 
Plant Biology 
at the 
Institute of Molecular B ioSciences 
Massey University, Palmerston North, 
New Zealand 
Robert E. Henriod 
200 1 
Abstract 
The development of Metrosideros excelsa (pohutukawa) as an ornamental crop has been 
limited by a lack of knowledge on the cultural requirements and underlying 
physiological processes associated with: (a) vegetative phase change (maturation) 
following micropropagation, (b) the environmental control of flowering, and (c) the 
postharvest characteristics of the cut-flower. These three concerns were addressed in 
this thesis. 
First, plantlets of M. excelsa that had undergone rejuvenation following 
micropropagation, were subjected to shoot and root restriction treatments to accelerate 
vegetative phase change. Leaves of shoot-restricted, single-stemmed plants became 
progressively more adult with increasing node position, whereas root restriction reduced 
root growth but did not accelerate vegetative phase change. In single-stemmed plants, 
light saturated maximum rate of photosynthesis and leaf carbon isotope discrimination 
decreased within increasing node position. However, carbon isotope composition in 
leaves of these plants diverged away from those exhibited by leaves of adult plants, 
possibly reflecting physiological changes resulting from altered source/sink relations. 
Second, the effects of photoperiod, temperature and irradiance on floral initiation and 
development were examined in M. excelsa by manipulating these parameters in 
controlled and greenhouse environments .  M. excelsa responded as a facultative short?
day plant with maximum flowering occurring following a 1 5  weeks cool (mean 1 5?C) 
short-day ( 10  h) inductive treatment. An irradiance of 567 IJ.mol m?2 s? 1 during 
induction provided the optimal conditions for floral primordial growth and subsequent 
flower development. Buds initially 2 .0-3 .0 mm in diameter had the highest probability 
of becoming floral, whilst those less than 2.0 mm in diameter were more likely to 
remain vegetative or to not break. 
Finally, the postharvest characteristics of M. excelsa as a cut flower were assessed. 
Generally, holding solution treatments containing sucrose extended vase life, whereas 
those containing HQC (applied alone or as a pulse) were detrimental. Cut flowers were 
sensitive to exogenous ethylene and pre-treatment with inhibitors of ethylene action 
(STS and 1 -MCP) conferred significant protection. 
11 
This thesis has contributed significantly to furthering our understanding and knowledge 
of cultural and physiological factors that underlie vegetative phase change, flowering 
and vase life characteristics in flowers of M. excelsa. 
lli 
Acknowledgements 
My sincerest gratitude goes to my supervisors for providing not only the opportunity to 
undertake this project but also for the guidance provided throughout. Specifically, I 
would like to thank my main supervisor Dr. John Clemens for the countless discussions, 
encouragement, expertise and most of all his friendship. I am also especially thankful 
to my eo-supervisor Professor Paula Jameson whose advice, encouragement and 
expertise throughout my thesis has been extremely appreciated. A special thanks also 
goes to my eo-supervisor Mr. Garry Burge (The Crop and Food Research Institute) for 
providing a friendly and supportive role, including access to laboratory space and 
equipment. 
Thank you to Dr. Dennis Greer (The Food and Horticultural Research Institute) for 
providing technical support with the use of photosynthetic equipment, laboratory space 
and interesting discussions. 
Thank you to members of the Institute of Molecular BioSciences who have contributed 
to this project. This includes those who have provided technical support, helpful advice 
and assistance. Thank you to my lab colleagues Lehka, Suzanne, Ivan, Greg (and many 
others) who have provided wonderful company over the years. 
I would also like to thank the expert statistical advice provided by Duncan Hedderley 
and Dr. Steve Haslett from Institute of Social Sciences,  and to Ray Johnstone and 
assistants, Lindsey and Lesley, at the Plant Growth Unit, Massey University. 
I am extremely appreciative of the moral and loving support provided by my parents, 
Emesto and Jill .  Thank you to all my friends including lab colleagues, Atawhai 
flatmates and close friends away from home for providing the support and confidence. 
Last but not least, I gratefully acknowledge funding provided by the Public Good 
Science Fund, the Institute of Molecular B ioSciences, the New Zealand Society of 
Horticultural Science and the Royal Society of New Zealand for contributing financially 
to either the research, conference travels or living expenses incurred throughout my 
studies. 
IV 
Abstract 
Acknowledgements 
Table of Contents 
List of Figures 
List of Tables 
List of Plates 
List of Abbreviations 
Chapter 1. Introduction 
Table of Contents 
1 . 1  Vegetative Phase Change in Woody Plants 
1 . 1 . 1  Introduction 
1 . 1 .2 Applications 
1 . 1 .3 Ontogenetic development and terminology 
1 . 1 .4 Models of phase change 
1 . 1 .5 Apical Meristems: Location of Phase-related Expression 
1 . 1 .5 . 1 Role of apical meristem 
1 . 1 .5 .2 Apical meristem: time keeper for phase change? 
1 . 1 .6 Timing of phase change 
1 . 1 .7 Morphological and anatomical features 
1 . 1 .8 Plant size in relation to phase change 
1 . 1 .9 Environmental factors 
1 . 1 .9. 1 Light and temperature 
1 . 1 .9.2 Water stress 
1 . 1 . 1  0 Carbon isotope discrimination studies 
1 . 1 . 1 1 Photosynthetic characteristics and chlorophyll concentration 
1 . 1 . 1 2 Role of carbohydrates 
1 . 1 . 1 3 Plant hormones 
1 . 1 . 1 3 . 1  Gibberellins and abscissic acid 
1 . 1 . 1 3 .2 Cytokinins 
1 . 1 . 14 Acceleration of phase change 
1 . 1 . 1 5 Markers of phase change 
1 .2 Floral Induction and Development in Ornamental Woody Plants 
1 .2 . 1 Introduction 
1 .2 .2 Terminology 
1 .2 .3 Flower induction models 
1 .2.4 The molecular frontier 
1 .2 .5 Floral and vegetative growth relationships 
1 .2 .6 Environmental factors 
1 .2.6. 1 Light: duration 
1 .2.6.2 Light: spectral quality 
1 .2.6.3 Light: intensity 
1 .2.6.4 Temperature 
V 
11 
IV 
V 
X 
XIV 
XVI 
XVll 
1 
3 
3 
4 
5 
7 
1 1  
1 1  
1 3  
1 3  
1 4  
1 6  
1 8  
1 8  
1 8  
19  
20 
2 1  
22 
22 
24 
25 
26 
28 
28 
29 
29 
30 
34 
34 
35 
38 
39 
4 1  
1 .2 .7 Carbohydrates 45 
1 .2 .8  Hormonal regulation 48 
1 .2 .8 . 1 Gibberellins 48 
1 .2 .8 .2 Cytokinins 50 
1 .2.9 Effect of cultural factors 5 1  
1 .3 Postharvest Physiology of Cut Flowers 52 
1 .3 . 1  Introduction 52 
1 . 3 .2 Flower senescence 53 
1 . 3 .2. 1 Petals 53 
1 . 3 .2 .2 Sepals 54 
1 . 3 .2 .3 Stamens 54 
1 .3 .2.4 Style 55 
1 .3 .3  Water relations 55 
1 .3 .4 Water uptake 56 
1 . 3 .5 Transpiration 56 
1 . 3 .6  Water temperature 57 
1 . 3 .7 Vascular occlusions 58 
1 .3 .8  Biocides 59 
1 .3 .9 pH of vase solution 59 
1 .3 . 1 0  Carbohydrates 60 
1 .3 . 1 1  Ethylene 60 
1 .3 . 1 2  Background on ethylene 6 1  
1 .3 . 1 3  Ethylene biosynthetic pathway 6 1  
1 .3 . 1 3  Signal transduction 63 
1 .3 . 1 4  Physiology of abscission zones and ethylene 63 
1 .3 . 1 5  Ethylene and flower abscission 64 
1 .3 . 1 6  Ethylene production 65 
1 .3 . 1 7 Ethylene sensitivity 66 
1 .3 . 1 8  Hormone interaction: Auxins 68 
1 .3 . 1 9  Ethylene inhibitors 69 
1 . 3 .20 Inhibitors of ethylene biosynthesis 69 
1 .3 . 2 1  Silver thiosulfate (STS) 70 
1 .3 .22 1 -Methylcycopropene ( 1 -MCP) 72 
1 .4 Summary and Thesis Objectives 73 
Chapter 2. Effect of shoot and root restriction on phase change in Metrosideros 
excelsa 77 
2 . 1 Introduction 77 
2.2 Materials and Methods 79 
2.2 . 1 Plant material 79 
2.2 .2 Experimental layout 80 
2.2.3 Growth measurements 80 
2 .2.4 Image analysis 8 1  
2 .2 .5 Image analysis protocol 8 1  
VI 
2.2.6 Plant biomass and leaf mineral determination 
2.2.7 Carbon isotope analysis 
2.2.8 Statistical analyses 
2.3 Results 
2.3.1 Growth parameters 
2.3.2 Image analysis :  optical parameters 
2.3.3 Image analysis :  dimensional parameters 
2.3.4 Dry weight accumulation 
2.3.5 Leaf mineral concentrations 
2.3.6 Carbon isotope discrimination 
2.4 Discussion 
Chapter 3. Gas exchange and carbon isotope discrimination characteristics during 
82 
82 
83 
83 
83 
83 
88 
91 
93 
93 
96 
phase change in Metrosideros excelsa. 100 
3.1 Introduction 100 
3.2 Materials and Methods 103 
3.2.1 Plant materials 103 
3.2.2 Greenhouse environment 103 
3.2.3 Shoot treatments applied to greenhouse-grown plantlets 104 
3.2.4 Measurements made on greenhouse plants 104 
3.2.5 Controlled environment 105 
3.2.6 Gas exchange measurements in the controlled environment 106 
3.2.6.1 Light response curves 106 
3.2.6.2 C02 assimilation I intercellular C02 (A/Ci) curves 107 
3.2.7 Leaf image analysis 107 
3.2.8 Carbon isotope discrimination 108 
3.2.9 Gas exchange measurements under greenhouse conditions 109 
3.2.1 0 Leaf carbohydrate analyses 109 
3.2.1 0.1 Soluble sugars extraction and determination 109 
3 .2.10.2 Starch extraction and determination 111 
3 .2.11 Statistical analyses 111 
3.3 Results 112 
3.3.1 Characterisation of shoot growth 112 
3.3.2 Leaf development in adult plants 114 
3.3.3 Image analysis of leaf dimensional and optical properties 117 
3.3.4 Leaf morphological attributes 120 
3.3.5 Carbon isotope discrimination 120 
3.3.6 Relationship of carbon isotope discrimination and leaf characters 123 
3.3.7 Photosynthetic response and relationship to carbon isotope 
discrimination in plants grown under controlled conditions 125 
3.3.8 Photosynthetic response of plants under greenhouse conditions 129 
3.3.9 Carbohydrate analyses 133 
3.4 Discussion 134 
Vll 
Chapter 4. Effects of temperature, photoperiod and bud size on flowering in 
Metrosideros excelsa. 141 
4.1 Introduction 141 
4.2 Materials and Methods 142 
4.2.1 Plant materials 142 
4.2.2 Controlled environment experiment 143 
4.2.3 Greenhouse experiment 144 
4.2.4 Statistical analyses 146 
4.3 Results 146 
4.3.1 Controlled environment experiment 146 
4.3.2 Greenhouse experiment 148 
4.4 Discussion 153 
Chapter 5. Effect of irradiance on floral induction and development in 
Metrosideros excelsa. 157 
5.1 Introduction 157 
5.2 Materials and Methods 158 
5.2.1 Plant material 158 
5.2.2 Experimental environments 159 
5.2.3 Bud measurements 160 
5.2.4 Bud histology 161 
5 .2.5 Inflorescence measurements 161 
5.2.6 Vegetative measurements 162 
5.2.7 Chlorophyll determination 162 
5.2.8 Carbohydrate extraction and determination 163 
5.2.9 Statistical analyses 163 
5.3 Results 163 
5.3.1 Histological examination of the buds 163 
5.3.2 Proportion of plants flowering 168 
5.3.3. Flowering time 171 
5.3 .4 Inflorescence morphology 171 
5.3.5 Leaf chlorophyll concentrations 172 
5.3.6 Leaf carbohydrate concentrations 172 
5.3.7 Vegetative growth 176 
5 .4 Discussion 176 
Chapter 6. The Postharvest Characteristics of Metrosideros excelsa as a cut flower 182 
6.1 Introduction 182 
6.2 Materials and Methods 184 
6.2.1 Plant material 184 
6.2.2 Harvest and experimental preparations 184 
6.2.3 Vase life room 185 
6.2.4 Scoring flower condition 185 
6.2.5 Experiments 1 A-C: Holding solutions : effects on flower quality 185 
Ylll 
6.2.6 Experiment 2: Effect of humidity on flower quality and endogenous 
ethylene production. 1 86 
6.2.7 Experiment 3 :  Effect of applied exogenous ethylene on flower quality 
and endogenous ethylene production. 1 87 
6.2 .8 Experiment 4: Efficacy of 1 -methylcyclopropene ( 1 -MCP) and silver 
thiosulfate (STS) on ethylene-induced responses 1 88 
6 .2 .8 . 1  Protection treatment 1 88 
6 .2 .8 .2 Ethylene treatment 1 88 
6.2.9 Statistical analyses 1 88 
6.3 Results 1 89 
6.3 . 1  Experiment 1 A: Holding solutions: effects on flower quality in 
'Lighthouse' .  1 89 
6.3 . 1 . 1  Floral development 1 89 
6.3 . 1 .2 Water relations 1 89 
6.3 . 1 .3 Stamen wilting 1 9 1  
6 .3  . 1 .4 Floral Abscission 1 95 
6 .3 .2  Experiment 1 B :  Holding solutions: effects on flower quality in 
'Vibrance' 1 95 
6.3 .2. 1 Floral development 1 95 
6.3 .2.2 Water relations 1 95 
6.3 .2.3 Stamen wilting 1 97 
6.3 .2.4 Floral Abscission 1 99 
6 .3 .3 Experiment 1 C: Holding solutions: effect of HQC and pH on flower 
quality in 'Lighthouse' . 200 
6 .3 .4 Experiment 2: Effect of humidity on flower quality and endogenous 
ethylene production. 200 
6.3 .4 . 1  Floral development and abscissions 200 
6 .3 .4.2 Water relations 203 
6 .3 .4.3 Endogenous ethylene production 205 
6 .3 .5  Experiment 3 :  Effect of exogenous ethylene on flower quality and 
endogenous ethylene production. 205 
6 .3 .5 . 1 Floral development 205 
6 .3 .5 .2 Abscission 207 
6 .3 .5 .3 Wilting 207 
6 .3 .5 .4 Endogenous production of ethylene after exposure to exogenous 
ethylene 208 
6 .3 .6 Experiment 4: Effect of 1 -MCP and STS on ethylene-induced responses 208 
6.3 .6 . 1 Water relations 208 
6 .3 .6.2 Stamen wilting 2 1 0  
6 .3 .6.3 Stamen abscission 2 1 2  
6 .3 .6.4 Flower and petal abscission 2 1 2  
6.4 Discussion 2 1 5  
Chapter 7 .  General Discussion 22 1 
References 
Appendix I Wax infiltration procedure 
Appendix 11 Staining schedule: safranin I fast green 
IX 
225 
255 
256 
List of Figures 
Figure 1 . 1  Figure 1.1. Models of phase change described by Poethig and Kester. 8 
Figure 1.2 Alternate models of phase change or maturation. 9 
Figure 1.3 General models of the multiple flowering pathways in Arabidopsis 
thaliana. 32 
Figure 1.4 Schematic diagram model illustrating the regulatory loop involvement of 
sucrose and cytokinins in the flowering process in Sinapis Alba. 46 
Figure 1.5 Models of ethylene biosynthesis and ethylene signal transduction pathway. 62 
Figure 2.1 Increase in mean plant height and mean total number of nodes in single?
stemmed plants of Metrosideros excelsa 'Scarlet Pimpernel ' over the 
experimental period. 84 
Figure 2.2 Changes in optical properties with increasing node position for the abaxial 
leaf surface in single-stemmed plants and branched plants of Metrosideros 
excelsa. 86 
Figure 2.3 Dimensional changes with increasing node position for the abaxial leaf 
surface in single-stemmed plants and branched plants of Metrosideros 
excelsa. 90 
Figure 2.4 Effects of container size on growth in single-stemmed and branched plants 
of Metrosideros excelsa 'Scarlet Pimpernel' .  92 
Figure 2.5 Mean carbon isotope discrimination values for leaves collected at different 
nodes for single-stemmed, branched and adult plants of Metrosideros 
excelsa 'Scarlet Pimpernel ' .  95 
Figure 3.1 Changes in mean plant height, node number and internode length of single?
stemmed, branched and adult plants of Metrosideros excelsa 'Vibrance' 
over the experimental period. 113 
Figure 3.2 Difference in carbon isotope discrimination levels from bud and leaf tissue 
in mature plants of Metrosideros excelsa 'Vibrance' collected in July -
August 2000. 116 
Figure 3.3 Changes in dimensional properties during successive sampling dates for 
leaf abaxial surfaces in single-stemmed, branched and adult plants of 
Metrosideros excelsa 'Vibrance'. 118 
Figure 3.4 Changes in optical properties between successive sampling dates for leaf 
abaxial surfaces in single-stemmed, branched and mature plants of 
Metrosideros excelsa 'Vibrance'. 119 
X 
Figure 3 . 5  Comparison of leaf carbon isotope composition in  single-stemmed, 
branched and adult plants of Metrosideros excelsa 'Vibrance'. 
Figure 3 .6  Comparison of  carbon isotope composition from leaves harvested at 
different node positions for single-stemmed and branched plants with the 
mean value for adult plants of Metrosideros excelsa 'Vibrance' collected at 
1 22 
the end of the experimental period. 1 22 
Figure 3 .7 Comparison of variables from light response and C02 assimilation curves 
for leaves in single-stemmed, branched and adult plants of Metrosideros 
excelsa 'Vibrance ' .  1 26 
Figure 3 .8  Mean photosynthetic light response curves for branched, single-stemmed 
and mature plants of Metrosideros excelsa collected in a controlled 
environment. 1 27 
Figure 3 .9 Representational C02 assimilation to intercellular C02 response curves for 
a leaf measured on single-stemmed, branched and adult plants of 
Metrosideros excelsa. 1 28 
Figure 3 . 10 Correlation of water use efficiency with carbon isotope discrimination for 
leaves from single-stemmed, branched and adult plants of Metrosideros 
excelsa. 1 3 1  
Figure 3 . 1 1  Diurnal comparison of leaf gas exchange parameters in single-stemmed, 
branched and adult plants of Metrosideros excelsa 'Vibrance'. 1 32 
Figure 4. 1 Effects of temperature and photoperiod on flowering in plants of 
Metrosideros excelsa 'Scarlet Pimpernel'  and 'Vibrance' 147 
Figure 4.2 Rate of floral development in plants of Metrosideros excelsa 'Scarlet 
Pimpernel' treated for 10  or 1 5  weeks in controlled environments before 
being transferred to a forcing greenhouse. 149 
Figure 4.3 Percentage of Metrosideros excelsa 'Scarlet Pimpernel '  buds in each of 
seven bud size classes that were floral , remained vegetative, or did not 
break in plants transferred to the forcing greenhouse after 1 5  weeks 
treatment with cool ( 1 5?C), short days ( l Oh) .  1 50 
Figure 4.4 Rate of floral development in plants of Metrosideros excelsa 'Scarlet 
Pimpernel' and 'Vibrance' grown continuously in greenhouses maintained 
at either ambient temperature with either ambient daylengths or a 
photoperiod of 1 6  h, and at a day/night temperature of 24! 1 7?C with 
ambient daylength. 152  
Figure 5 . 1  Effect of irradiance and ambient conditions on the cumulative proportion of 
floral meristems per plant at 20 and 23 weeks from the start of the 
experiment in three bud size classes of Metrosideros excelsa 'Lighthouse' .  1 66 
XI 
Figure 5.2 Correlations between bud size at time of harvest with maximum length and 
width of floral meristems in Metrosideros excelsa Lighthouse'. 167 
Figure 5.3 Effect of irradiance treatments and ambient conditions during floral 
induction on the size of floral meristems at 23 weeks after the start of the 
experiment. 
Figure 5.4 Effect of inductive irradiance and ambient conditions on the mean number 
of inflorescences per plant in Metrosideros excelsa Lighthouse' and 'Scarlet 
169 
Pimpernel'. 170 
Figure 5.5 Effect of irradiance and ambient conditions on the cumulative proportion of 
inflorescences per three bud size classes per plant of Metrosideros excelsa 
'Scarlet Pimpernel' at 11 weeks after transference to the forcing greenhouse 
conditions. 170 
Figure 5.6 Effect of inductive environment on the percentage of terminal buds of 
inflorescences that had broken and from which vegetative shoots were 
elongating, that had aborted, or that had not broken in Metrosideros excelsa 
Lighthouse' and 'Scarlet Pimpernel assessed at Stage 1 of inflorescence 
development. 173 
Figure 5.7 Effect of irradiance and ambient conditions during floral induction on 
vegetative shoot length in two cultivars of Metrosideros excelsa determined 
11 weeks after transference to forcing conditions. 175 
Figure 5.8 Relationship between vegetative shoot length and leaf total chlorophyll 
concentration in Metrosideros excelsa Lighthouse' grown under different 
irradiance treatments. 175 
Figure 6.1 Effect of holding solution on the stage of floral development in cut cymules 
of Metrosideros excelsa Lighthouse'. 190 
Figure 6.2 Changes in mean water flux (uptake and transpiration) pooled for all 
holding solution treatments and mean cymule mass for individual 
treatments in cut cymules of Metrosideros excelsa Lighthouse'. 192 
Figure 6.3 Effect of holding solution on stamen wilting and abscission in cut cymules 
of Metrosideros excelsa Lighthouse'. 193 
Figure 6.4 Changes in mean water flux (uptake and transpiration) pooled for all 
holding solution treatments and mean cymule mass for individual 
treatments in cut cymules of Metrosideros excelsa 'Vibrance'. 196 
Figure 6.5 Effect of holding solution on stamen wilting and abscission in cut cymules 
of Metrosideros excelsa 'Vibrance'. 198 
Xll 
Figure 6.6 Effect of HQC and pH holding solutions on stamen abscission and wilting 
in cut cymules of Metrosideros excelsa Lighthouse'. 20 1 
Figure 6.7 Effect of relative humidity on stage of floral development in cut cymules of 
Metrosideros excelsa 'Lighthouse' . 202 
Figure 6.8 Effect of relative humidity on water flux (water uptake and transpiration) 
and cymule mass in cut cymules of Metrosideros excelsa 'Lighthouse'. 204 
Figure 6.9 Headspace ethylene concentration above cut cymules of Metrosideros 
excelsa Lighthouse' at 24 and 48 h after incubation per enclosure day. 206 
Figure 6. 10 Effect of preventative treatment on water flux (water uptake and 
transpiration) for cut cymules treated with either 0 or 5 f.ll r' exogenous 
ethylene. 209 
Figure 6. 1 1  Change in cymule mass following exposure to a preventative treatment and 
exogenous ethylene (0 and 5 f.ll r' ethylene data pooled). 2 1 1 
Figure 6. 1 2  Effect of preventative treatment on stamen wilting. 2 1 1 
Figure 6. 1 3  Change in stamen abscission in cut cymules after a preventative pre?
treatment of 1 -MCP and STS before application of either 0 and 5 f.ll r' of 
exogenous ethylene. 2 1 3  
Figure 6. 14  Change in flower abscission with time following exposure to 5 f.ll r' 
exogenous ethylene in pre-treated cymules of Metrosideros excelsa 
'Lighthouse' with 1 -MCP and STS . 2 14  
Figure 6. 1 5  Effect of petal in-rolling on Day 9 following exposure to 5 f.ll r' exogenous 
ethylene in pre-treated cymules of Metrosideros excelsa 'Lighthouse' with 
1 -MCP and STS. 2 14 
Xlll 
List of Tables 
Table 1 . 1  Features that distinguish juvenile and adult phases of English ivy (Hedera 
helix) and maize (Zea mays). 1 7  
Table 2 . 1 Comparison of single-stemmed and branched plants of Metrosideros 
excelsa 'Scarlet Pimpernel'  with respect to leaf optical (abaxial and adaxial 
surfaces) and dimensional (adaxial surface) parameters at node positions 
16- 1 7 and 26-27 . 87 
Table 2.2 Effect of node position within single-stemmed plants of Metrosideros 
excelsa 'Scarlet Pimpernel ' on optical (abaxial and adaxial) and 
dimensional (adaxial) leaf parameters , and means for corresponding 
parameters in mature leaves .  89 
Table 2 .3 Mineral nutrient concentrations in the leaves of single-stemmed and 
branched plants of Metrosideros excelsa 'Scarlet Pimpernel' grown in 0.82 
and 7 .2 1 containers. 94 
Table 3 . 1 Comparison of leaf characteristics in single-stemmed, branched and adult 
plants of Metrosideros excelsa 'Vibrance' .  1 2 1  
Table 3 .2  Correlation of leaf dimension, colour, mass and water content variables 
with leaf carbon isotope discrimination values in single-stemmed, branched 
and adult plants of Metrosideros excelsa 'Vibrance' .  1 24 
Table 3 .3  Comparison of  leaf gas-exchange parameters in  single-stemmed, branched 
and adult plants of Metrosideros excelsa 'Vibrance' . 1 30 
Table 3 .4 Comparison of carbohydrate concentrations in fully expanded leaves 
collected from nodes 10  and 40 in single-stemmed plants, node 30 in 
branched plants and from the canopy top in adult plants of Metrosideros 
excelsa 'Vibrance' on 25 June 200 1 ,  between 0600-0630 h. 1 34 
Table 4. 1 Classification of stages of floral development in Metrosideros excelsa. 145 
Table 5 . 1 Effect of inductive environment and cultivar on inflorescence 
morphological characteristics. 1 72 
Table 5 .2  Effect of  irradiance and ambient conditions during floral induction on leaf 
chlorophyll and carbohydrate concentrations in Metrosideros excelsa 
'Lighthouse' after 20 weeks in induction treatments . 1 74 
Table 6. 1 Effect of holding solution on mean water uptake and transpiration in cut 
cymules of Metrosideros excelsa 'Lighthouse' over the experimental 
period. 1 9 1  
xiv 
Table 6.2 Correlation of mean stamen wilting and cymule fresh weight between Days 
8- 1 2  for each holding solution treatment for cut cymules of M etrosideros 
excelsa 'Lighthouse' .  1 94 
Table 6.3 Correlation of mean stamen wilting and cymule fresh weight between Days 
4- 1 6  for each holding solution treatment for cut cymules of Metrosideros 
excelsa 'Vibrance' .  1 99 
Table 6.4 Effect on humidity on percentage of stamen and petal abscission in cut 
cymules of Metrosideros excelsa 'Lighthouse' . 203 
Table 6.5 Effect of exogenously applied ethylene on percentage of abscised flowers . 207 
Table 6.6 Effect of exogenously applied ethylene on percentage of mean stamen 
wilting. 208 
Table 6.7 Correlation of stamen wilting and cymule fresh weight of cut cymules of 
Metrosideros excelsa pre-treated with preventative treatments of either 
1 -MCP or STS before application of exogenous ethylene (data pooled for 0 
and 5 Ill r' ethylene) . 2 1 0  
XV 
List of Plates 
Face plate Metrosideros excelsa 'Vibrance' 
Plate 3 . 1 Cross-section of vegetative bud illustrating the continuous formation of 
XVlll 
new leaf primordia. 1 1 5 
Plate 5 . 1 Microscopic study of vegetative and floral initiation and development in 
buds collected after the start of the experiment with 20 weeks treatment to 
different inductive irradiances. 1 65 
xvi 
ABA 
ACC 
A/Ci 
AOA 
AVG 
<l'app 
? 
CK  
DL 
DNA 
DMF 
GA 
IAA 
FAA 
GENMOD 
HQC 
HQS 
lmax 
LD 
1 -MCP 
NBD 
PPF 
Pmax 
PPFsat 
PP333 
RH 
Rubisco 
SD 
STS 
Ycmax 
WUE 
List of Abbreviations 
abscisic acid 
1 -aminocyclopropane- l carboxylic acid  
C02 assimilation over intercellular C02 concentration 
aminooxyacetic acid 
arninoethoxyvinylgycine 
apparent (C02-limited) photon yield 
carbon isotope discrimination 
cytokinin 
day length 
deoxyribonucleic acid 
dimethylforrnamide 
gibberellin 
indole-3-acetic acid 
90% formalin, 5% acidic acid and 5% alcohol fixative solution 
generalised linear model using maximised likelihood estimations 
8-hydroxyquinoline citrate 
8-hydroxyquinoline sulfate 
maximum electron transport rate 
long-day 
1 -methylcyclopropene 
2,5-norbornadiene 
photosynthetic photon flux 
maximum photosynthetic photon flux at 99% light saturation 
l ight saturated maximum rate of photosynthesis 
paclobutrazol 
relative humidity 
ribulose-! ,5-bisphosphate carboxylase/oxygenase 
short-day 
silver thiosulfate 
maximum rubisco carboxylation rate 
water-use efficiency 
xvii 
Metrosideros excelsa 'Vibrance' 
XV111 
Chapter 1 
Introduction 
1 .0 Introduction 
The worldwide consumption for ornamental flowers is growing rapidly and the demand 
for novel flowering crops is increasing accordingly. In 1 995, the world flower use was 
estimated at around US $3 1 billion dollars, with the three most important markets for 
floral products being the European Union, North America and Japan (Anon. 200 1 a) .  
New Zealand, although only being a small player in this industry, has a relatively high 
level of horticultural technology and a well-established container and cut flower 
industry which is oriented towards the export market. Export earnings for flower 
products for the year ended March 200 1 were approximately NZ $46.2 million, and this 
estimate is anticipated to rise in the coming years (Anon. 2000, Anon. 200 1 b) .  
Currently, a number of native genera exported annually, often as liner plants, include 
live plants of Hebe, Leptospermum, Metrosideros, Pittosporum, Sophora and Uncinia, 
and smaller volumes of cut native material such as leaves of Leptospermum and 
Phormium. 
New Zealand has a unique and diverse range of flora with many indigenous shrubs and 
trees that are not well known worldwide. Thus, many woody plants native to New 
Zealand have excellent ornamental potential as either container plants or as cut flowers . 
However, their development has been limited by a lack of knowledge of their cultural 
requirements, particularly with regards to the hastening of flowering in plants that have 
lost the ability to flower following micropropagation, and a lack of understanding of 
environmental factors that control flowering. Moreover, the paucity of research into the 
postharvest characteristics of native cut flowers has been a limiting factor for further 
expansion of novel flowers into the international floricultural trade. In light of this ,  the 
Native Ornamental Plants Program funded by the Public Good Science Fund (PGSF) 
was established through the New Zealand Institute for Crop & Food Research (Crop & 
Food Research). This program sought to address these problems by conducting 
scientific research into selected native species with potentially high ornamental value. 
The results would enable the production and rapid technological development of native 
genera for expansion to both the New Zealand and export markets, while contributing 
invaluable scientific knowledge on the flowering biology of the native New Zealand 
flora. Massey University was sub-contracted by Crop & Food Research to achieve 
specific objectives in relation to Metrosideros excelsa. This formed the basis for the 
doctoral study described in this thesis. 
Metrosideros excelsa (Myrtaceae), also known as the pohutukawa or New Zealand 
Christmas tree (Hewett 1993) ,  is a native New Zealand tree species considered by the 
PGSF to have considerable potential as an ornamental flowering crop. In the wild, this 
species is typically found along the coastal regions of the North Island. Flowering 
displays that are exhibited by M. excelsa under natural conditions during the summer 
months from December through to January, have been described as a "brilliant 
spectacle", hence, the popularity of many cultivars among New Zealand growers 
(Salmon 1980).  The flowers in this genus have conspicuous long stamens,  and are often 
brightly coloured making it particularly attractive as an ornamental plant within the 
floricultural industry. 
Development of M. excelsa has been limited by several factors associated with 
flowering. Firstly, seedling-grown plants of M. excelsa typically reach an adult 
reproductive phase within five to ten years, whereas cuttings from adult foliage are 
capable of flowering within one year (Oliphant et al. 1992) .  However, it has been 
reported to be difficult to root cuttings from adult trees. The preferred method of 
propagation for obtaining mass clonal material has, therefore, been through the use of 
rnicropropagation (Oliphant 1990) .  However, during rnicropropagation, rejuvenation of 
parent material occurs, and can result in a delay of three or more years before plants are 
able to bear flowers . Thus, an effective method was required to accelerate vegetative 
phase change. Secondly, once plants of M. excelsa have become reproductively 
competent, there is a further need to understand what environmental factors are 
involved in the flowering process. This not only entails an understanding of factors that 
promote, inhibit or accelerate flowering, but also an understanding of factor(s) intrinsic 
to the plant that are indicative of flowering ability. Finally, production of an ornamental 
crop would not be complete without an assessment and evaluation of its postharvest 
characteristics. This would include research into methods for extending the vase-life 
and flower quality of cut material . 
2 
In this thesis, all three aspects discussed above for advancing M. excelsa as a flowering 
crop were investigated. The literature reviewed in Chapter 1 on phase change, 
environmental control of flowering and postharvest characteristics in woody plants 
provides a background from which appropriate hypotheses are advanced. In Chapter 2,  
several cultural methods for accelerating vegetative phase change in micropropagated 
plants were tested and evaluation of ontogenetic status was examined using image 
analysis methods. The physiological changes occurring during vegetative phase change 
are characterised in Chapter 3, with specific emphasis on leaf gas-exchange and carbon 
isotope discrimination properties. In Chapter 4, the effect of photoperiod, temperature 
and bud size on floral induction and subsequent development were investigated using 
reproductively competent adult plants of M. excelsa . The effect of irradiance on flower 
induction and subsequent development was examined in Chapter 5, including a 
microscopic investigation of the changes occurring during the initial stages of floral 
organogenesis, and their relation to irradiance, bud size and timing of development. In 
Chapter 6, the postharvest characteristics of M. excelsa as a cut flower were examined, 
including a quantitative and qualitative assessment of the effects of holding solution, 
endogenous ethylene emanation and exogenous ethylene applications on flower quality 
and duration of vase-life. The effectiveness of treatments designed to inhibit the effects 
of ethylene-mediated damage was also investigated. Finally, a general discussion and 
conclusion of the experimental results are provided in Chapter 7 of this thesis .  
1.1 Vegetative Phase Change in Woody Plants 
1.1 .1  Introduction 
The transition through a series of developmental phases during growth is inherent in all 
living organisms. In higher animals, developmental phases represent different episodes 
in the life of an organism which can encompass changes throughout the entire body 
plan. Higher plants also exhibit distinct developmental phases during growth, although 
these changes are manifested within the plant in the developing shoot apical meristem, 
which are subsequently expressed in parts of the plant derived from this region. The 
shoot apical meristem in higher plants passes through several, usually three, defined 
3 
post -embryonic states of development, which comprise a juvenile vegetative phase, an 
adult vegetative phase, and a reproductively competent phase (Poethig 1990) .  The 
transition between the developmental states in the life of a plant is commonly referred to 
as maturation or phase change (Hackett 1985, Poethig 1990, Greenwood 1995) .  An 
understanding of the mechanisms that underlie phase change is not only scientifically 
interesting in its own right but also important from a commercial perspective, given the 
need to produce crops (either flowers, seeds or fruiting bodies) from reproductively 
competent plants as rapidly as possible. In this section, the current literature on 
vegetative phase change in woody plants is reviewed. Particular emphasis is placed on 
the physiological, biochemical and molecular processes occurring in woody 
angiosperms during phase change, and during specific life-stages with the intent of 
defining and establishing future directions in this field of research. It should be noted 
that the study of phase change per se is not limited to woody angiosperms, and when 
appropriate relevant and important findings in other studies of model plants from 
different groups (e.g. herbaceous angiosperms and gymnosperms) are referred to and 
discussed. 
1 .1.2 Applications 
Methods for controlling or accelerating the timing of phase change would have 
significant economic benefits from both an agricultural and horticultural perspective. 
This is because the breeding efficiency of woody perennials and the selection of 
improved cultivars is greatly improved by reducing the length of the juvenile period 
(Hackett and Murray 1992) .  Similarly, a reduction in the length of the juvenile period 
can also advance the time to when plants are reproductively competent, hence 
shortening the time to flower or fruit production. This is in light of the fact that most 
plants exhibiting juvenile vegetative characteristics can rarely be induced to flower. 
The same philosophy also applies to the forestry industry given that the quantity and 
quality of productivity of a forest tree species is directly related to its degree of maturity 
(Hackett and Murray 1992) .  With regards to propagation technology, the ease of 
cutting propagation and in vitro organogenesis and somatic embryogenesis for all types 
of woody perennials is strongly affected by ontogenetic age (Hackett and Murray 1992) .  
4 
1 .1.3 Ontogenetic development and terminology 
The three primary phases of development following post-embryonic development that 
have been recognised include a juvenile vegetative phase, an adult vegetative phase and 
a reproductively competent phase. The juvenile phase of development starts with the 
initiation of a stem, true leaves and axillary buds and may last from a few days to 
several years depending on the species and conditions (Poethig 1990). This phase is  
often characterised by a number of distinguishable and unique vegetative traits (e.g. leaf 
morphology) and is, in almost all cases, exempt from producing reproductive organs.  
The juvenile phase is followed by the adult vegetative phase, which is characterised by 
a different suite of vegetative traits. The onset of the reproductively competent phase 
entai ls  the potential for the transformation of the vegetative shoot apex into a 
reproductive structure where either flowers, inflorescences or cones are formed. Rarely, 
vegetatively juvenile plants can be reproductively competent and bear flowers on 
branches bearing juvenile foliage, such as in certain Eucalyptus species (Wiltshire et al. 
1991 ). 
The transition between each of these different phases has been referred to as ontogenetic 
ageing by Fortanier and Jonkers (1976), meristem ageing (cyclophysis) by Oleson 
(1978), maturation by W areing (1959) and phase change by Brink ( 1962). This is 
different to physiological ageing as described by W areing ( 1959), which is often 
associated with a reduced growth rate or vigour as a result of increasing size and 
complexity of the plant. Phase change as currently referred to by several authors 
(Poethig 1990, Greenwood 1995, Hackett and Murray 1997), can be characterised by 
progressive changes in a number of species-specific morphological and developmental 
attributes such as leaf shape, phyllotaxy, thorniness, shoot orientation, shoot growth 
vigour, anthocyanin pigmentation and by the ability to form adventitious roots (Hackett 
1985, Hackett and Murray 1992). Changes in these characteristics during development, 
which are permanently recorded as variation along a shoot axis, have been referred to as 
heteroblasty (Kerstetter and Poethig 1998). This term was originally coined by Goebel 
( 1900) to describe the ontogenetic sequence of morphological features of shoots, 
although more recently, usage of this definition has been extended to include not only 
the changes in anatomical/morphological traits, but also physiological traits (e.g. 
photosynthetic rate) during ontogenetic development (Kerstetter and Poethig 1998). 
5 
The abruptness with which these differences are expressed between ontogenetic phases 
can be further expressed as either heteroblastic or homoblastic development. Plants that 
exhibit gradual changes in leaf morphologies between juvenile and adult vegetative 
phases have been described as undergo homoblastic development (Goebel 1900, Godley 
1985) .  Examples of homoblastic development are seen in New Zealand species of 
Avicennia, Corynocarpus, Griselinia, Macropiper, Meryta, Metrosideros, Myosotidium, 
and Planchonella (Godley 1985) .  This is in contrast to heteroblastic species that exhibit 
a dramatic transition between strikingly different juvenile and adult leaf morphologies 
(Goebel 1900), a feature typified by more than 200 New Zealand plant species, 
including Pseudopanax crassifolius (lancewood) Knightia excelsa (Cockayne 1928 ,  
Godley 1985) and Elaeocarpus hookerianus (Day e t  al. 1997). Several authors who 
have recently characterised phase change in Eucalyptus globulus assert that the 
progressive changes in leaf characters during phase change follows heteroblastic 
behaviour, although they describe the phenomenon as occurring in a "gradual" manner 
(James and Bell 200 1).  Based on the definition proposed by Goebel ( 1900) and by the 
subsequent use of the term (e .g. Godley 1985), E. globulus would be better described as 
a homoblastic rather than a heteroblastic species. 
According to Jones ( 1999), part of the problem associated with the confusion in the 
literature on phase change has been not only the lack of clearly defined terminology but 
also the inconsistent use of specific terms to describe a particular feature or 
phenomenon. The terms mature and adult have both been used widely to refer to the 
same phase(s) of development, although preference for a particular term has often been 
author(s) specific (e.g. Poethig 1990, Hackett and Murray 1997) .  In order to maintain 
consistency throughout this thesis, the term adult has been applied to the ontogenetic 
phases occurring after the juvenile vegetative phase. This term (as opposed to the term 
mature) allowed for a distinction to be made between either the adult vegetative phase 
or the adult reproductively competent phase. Similarly, the phrase phase change as 
defined by Poethig ( 1990) and applied in this thesis,  refers to the transition between all 
ontogenetic phases of development. This is unlike the term maturation which only 
implies a transition to when an organism may become reproductively competent. 
6 
1.1 .4 Models of phase change 
Several models have been proposed to account for the changes occurring during phase 
change. To account for intermediate leaf forms during phase change, Poethig ( 1990) 
proposed the combinatorial model (Figure 1 . 1  A) based on the model plant Zea mays. 
According to this model, shoot development can be described as a series of overlapping 
programs (juvenile, adult, reproductive), which are independently regulated and are 
modulated by expression of a common set of developmental processes rather than a 
series of mutually exclusive developmental programs. Therefore, during the transition 
from juvenile to adult leaves, different developmental programs can be expressed on the 
same leaf. A classic example of this is found in Acacia heterophylla where intermediate 
leaves express both the juvenile (pinnately compound leaf) and adult (phyllodes) leaf 
morphology on the same leaf, with the juvenile traits expressed at the distal part of the 
leaf (Taiz and Zeiger 1998) .  Similar patterns of growth are also exhibited in woody 
plants, such as Eucalyptus and Hedera helix (Hackett and Murray 1992), and 
herbaceous plants, such as maize (Lawson and Poethig 1995). Thus, cells at the base of 
the leaf are committed to an adult program whereas those at the tip are committed to the 
juvenile program (Poethig 1990) .  Kester ( 1976) had previously proposed a similar 
model based on the behaviour observed in woody plants (Figure 1 . 1  B) .  According to 
his model, juvenile and adult characteristics can coexist but are expressed at different 
locations on the plant during growth and development of the main shoot axis and 
axillary branches. Thus, juvenile traits are expressed at the lower portions of the stem. 
During developmental growth, intermediate structures are formed which are 
subsequently followed by the expression of adult traits that eventually reside on the 
upper portions and periphery of the plant body. 
More recently, Hackett and Murray ( 1997) proposed a set of alternate models for phase 
change (Figure 1 .2). According to their models, five proposed pathways in which phase 
change can be regulated include a set of phenotypic characters (A' -D') controlled by 
either a single-switch, a parallel set of switches or a series of switches. The latter two 
regulatory pathways can be further classified into switching programs that control 
groups of characters. Thus far, support for a single-switch event for controlling phase 
change appears to be lacking. Evidence for this has been based on observations of the 
timing of expression of particular traits during phase change. In Picea sitchensis (sitka 
7 
A. 
0 Juvenile 
0 Adult 
? Reproductive 
e Flower 
B .  
0 Mature form 
? Intermediate for:-r 
- Juvenile torm 
Vegetative 
Young adult 
plant 
Flowering 
adult plant 
Figure 1 . 1 Models of phase change described by Poethig (A) and Kester (B) .  A: 
Schematic representation of the expression of juvenile vegetative, adult vegetative, and 
reproductive traits in the shoot of an annual plant (Poethig 1990) .  B: Schematic 
representation of the localisation of different ontogenetic states on an adult (mature), 
woody, seedling-grown plant (Kester 1976) .  
8 
\0 
Juveni l e  
Phenotyplc  
Characters 
A 8 C D 
Juveni le 
Phenotypic  
Characters 
A B C D 
? 
T TTTT 
A' B '  C' D '  
Mature 
Phenotypic  
Characters 
Single  
Master 
Switch 
A '  B '  C' 
Mature 
Phenotypic 
Characters 
D '  
Pr imary Swi tch 
w i th Secondary 
Swi tches . 
A 
? 
A' 
A' 
Juveni l e  
Phenotypf c  
Characters 
B C 
? 
8 '  
? 
C'  
D 
t ? 
8'  C'  
Mature 
Phenotypic 
Characters 
Paral le l  
swi tches 
D' 
? 
D'  
Juveni l e  
Phenotypic 
Characters 
A 8 C D 
? 
A' 8 '  
A'  8 '  
? 
C' D" 
1 
C' 
Mature 
Phenotyplc 
Characters 
D' 
Separate Para l le l  
Swi tches for 
Groups of Characters 
Juveni le 
Phenotypi c 
Characters 
A 8 C D 
? 
A '  
L1 
B '  
L1 
C '  
L1 
D '  
? 
A '  B' C' D '  
Mature 
Phenotypfc 
Characters 
Ser ies of 
Swi tches 
Juveni l e  
Phenotyp i c  
Characters 
A B C D 
? 
LJ ? 
8 '  C'  
t L1 
A' B '  C '  
D '  
? 
Mature 
Phenotypfc 
Characters 
D '  
Separate Series 
of Swi tches for 
Groups of Characters 
Figure 1 .2 Alternate model s  of phase change or maturation (Hackett and Murray 1 997) .  Note: "mature" refers to adult  plants. 
spruce) ,  for example, various needle morphological features (e.g. leaf dimension and 
mass characteristics) were characterised from tree stands of different ages (Steele et al. 
1989) .  Consequently, most needle characters changed at different rates with age, 
suggesting that a multiple set of regulatory processes were involved during phase 
change. Similarly, the transition from the juvenile to adult leaf form of Eucalyptus 
globulus is also accompanied by different temporal expression in leaf characters (J ames 
and Bell 200 1) .  Moreover, further evidence of a multiple regulatory system comes from 
rejuvenation experiments . Application of varying concentrations of GA3 to adult plants 
of Hedera helix caused rejuvenation and resulted in differential sensitivity of various 
physiological and morphological phase-related characteristics (Rogler and Hackett 
1975) .  Low doses of GA3 only suppressed flowering ability, whereas higher doses 
inhibited not only flowering ability but also expression of various other adult characters, 
such as rooting ability or absence of anthocyanin. Saturating GA3 concentrations 
appeared to repress expression of all characters associated with the adult form (Rogler 
and Hackett 1975).  The results of these and other studies certainly suggest the role of a 
multiple switch mechanism and may indicate that a change in one character is required 
before others are able to be switched (Hackett and Murray 1992).  Differential and 
independent expression of phase-specific traits is also supported by genetic studies. 
Law son and Poethig ( 1995) contend that pathways involved in phase change are likely 
to be complex with many branching and interacting pathways. 
Based on developmental studies of woody species, Greenwood ( 1995) proposed several 
mechanisms that ought to be considered in models that account for phase change. First, 
the onset of the adult phase is usually gradual but in certain cases (e.g. branching 
frequency) can occur abruptly. Second, a wide variety of morphological, physiological 
and biochemical traits are involved in phase change and occur independently of one 
another. Thirdly, maturational traits are often persistent, and their maintenance is not 
always a function of plant size or root proximity. Next, cells in the apical meristem in 
some woody plants appear to become determined. Finally, the process of rejuvenation 
results in a gradual reversion to the juvenile condition. Based on these criteria, 
Greenwood ( 1995) contended that a simple model involving only one switch would 
account for the complex phenomenon of phase change. Greenwood proposes that the 
switch would reside in individual cells within the apical meristem. Upon activation of 
the switch, juvenile cells in the meristem would become adult so that over time the 
10  
apical meristem would resemble a "mosaic" with an ever-increasing proportion of adult 
cells. Thus, the number of adult characters at any given time would be a function of the 
ratio of juvenile to adult cells in the meristem. This hypothesis may explain why fully 
expanded leaves at intermediate stages of phase change possess phenotypic characters 
(or cell types) typical of both juvenile and adult phases as exhibited in many woody 
species (James and Bell 200 1) .  With regards to rejuvenation, Greenwood ( 1995) 
hypothesised that this process could be explained by exposure to conditions that favour 
rapid multiplication of juvenile cells in the apex (Greenwood 1995) .  
1 . 1 .5 Apical meristems: location of phase-related expression. 
1 .1 .5.1 Role of apical meristem 
Research conducted on phase change has often focused on the role of the apical 
meristem during this process (Poethig 1990, Greenwood 1995, Hackett and Murray 
1997) .  There is strong evidence to suggest that the apical meristem is the site or source 
of phase change activity. However, it is yet unclear whether the apical meristem 
behaves autonomously, meaning it acts independently of the rest of the plant, or 
whether extrinsic factors in other parts of the plant are involved (Sussex 1976,  Hackett 
1985).  This may involve chemical signalling between the roots and/or leaves to the 
apical meristem, subsequently initiating a developmental phase change response. 
Evidence to suggest that the apical meristem behaves as an autonomous unit as opposed 
to an apical system reliant of extrinsic signals has been obtained through grafting 
experiments. In order to determine whether the development of grafted juvenile shoots 
is influenced by an adult rootstock, a number of workers have grafted seedling scions 
onto reproductively competent adult trees. The formation of reproductive structures has 
usually been a reliable indicator as to whether phase change has occurred (Poethig 
1990) .  Thus, many of the studies on woody species examining the effect of grafting on 
phase change have used flowering or cone formation as this indicator. Robinson and 
W areing ( 1969) grafted juvenile scions of Larix leptolepis and L. decidua onto 
reproductively competent trees and found that less than two percent of the scions 
produced cones after one year. The plants from which scions were derived did not form 
cones during that time period. Conversely, grafting adult scions onto juvenile 
rootstocks has resulted in a rejuvenation response in some species. For example, 
1 1  
experiments where adult apices of Citrus or Sequoiadendron with one to three leaf 
primordia were grafted onto juvenile rootstock resulted in plants that grew out to have 
adult vegetative characteristics (Navarro et al. 1975 , Monteuuis 199 1) .  It would appear 
that upon reaching the adult phase, the rootstock has little effect on the scion status ,  
such as  in causing rejuvenation. It is possible that cells in  the apical meristem are 
comprised predominantly of adult rather than juvenile cells (Greenwood 1995) and/or 
have become determined and remain ontogenetically stable (Hackett and Murray 1997) .  
In contrast to  the above findings, there is  also evidence to suggest that factors extrinsic 
to the meristem can influence the ontogenetic status of the plant. Visser ( 1973) found 
that apple seedlings grafted onto dwarfing rootstock precociously flowered compared to 
those grafted on an "invigorating" rootstock. With regards to vegetative phase change, 
Brand and Lineberger ( 1992) grafted adult scions of Betula onto six-month old seedling 
rootstocks. They found partial rejuvenation evidenced by the occurrence of leaves with 
juvenile characteristics, although this occurred for a transitory period. There has also 
been some anecdotal evidence that adult characteristics in cuttings of adult plants of 
Populus and Eucalyptus disappear rapidly after being grafted or rooted (Greenwood 
1995) .  These studies suggest that other plant organs, such as the roots, can affect some 
phase-dependent characteristics within the plant. Grafting studies, however, have 
l imitations since the process of grafting itself affects the scion by initiating stress when 
cutting, providing contact with endogenous substances in the adult tree, and by 
subjecting the scion and rootstock to competition for assimilates by other branches 
(Snowball 1989) .  
The transition from the juvenile through to the adult vegetative phase may require some 
level of cellular determination (Hackett 1985 , Greenwood and Hutchinson 1993).  
Poethig ( 1990) argues that both intrinsic and extrinsic factors regulate phase change at 
the cellular level in the apical meristem per se, but that the onset of adult features is  
initiated by factors extrinsic to the apical meristem. Thus, changes in the apical 
meristem and changed inputs into that meristem (e .g. carbohydrates, hormones and 
water) function in determining its expression, and these factors are likely to be 
modulated by increasing size and complexity of the plant (Hackett 1985, Poethig 1990) .  
12  
1.1.5.2 Apical meristem: time keeper for phase change? 
The timing of phase change transition may, at least in part, be intrinsic to the 
meristematic cellular tissue. The shoot meristem undergoes a number of morphological 
and physiological changes during phase change (Poethig 1990, 1997) .  For example, the 
apical meristems of adult plants in comparison to those in juvenile plants, are generally 
larger, cells are smaller and have slower rates of subapical division (Stein and Foskett 
1969) .  Early studies by Brink et al. ( 1968) and Robinson and Wareing ( 1969) suggest 
that cells in the apical meristem can act as a time-measuring system whereby 
information can be maintained on the number of cell divisions. Clearly, phase change 
can be accelerated by placing plants in conditions that promote rapid growth (Brink et 
al. 1968, Robinson and Wareing 1969, Hackett 1985, Snowball l989, Snowball et al. 
1994) . Under these favourable growth conditions,  Sussex ( 197 6) proposed that cells at 
the apical meristem would be undergoing a rapid cyclic turnover, which also coincides 
with plants reaching a minimum size, and thus, attainment of a minimum number of 
cells divisions would be achieved before phase change occurs. 
1.1.6 Timing of phase change 
Recent quantitative genetic studies on Eucalyptus have focused particular attention on 
the timing or rate of developmental processses (heterochrony) (Wiltshire et al. 1998, 
Jordan et al. 1999).  Jordan et al. ( 1999), using the onset of leaf morphological changes 
and time to first flowering as markers of phase change in E. globulus trees, 
demonstrated in open-pollinated and controlled-cross field experiments that the onset of 
these events was independently regulated and were under strong genetic control . E. 
globulus occuring in numerous populations throughout south-eastern Australia is known 
to have a high level of genetic differentiation between populations with regards to the 
time to phase change (Dutkowski and Potts 1999) .  Thus, by the comparing ontogenetic 
development of E. globulus grown under a common environment, using progeny from 
individual populations and crosses from different populations, Jordan et al. ( 1999) were 
able to establish that the onset of phase change in this taxon was under moderate to 
strong additive genetic control. However, there was no quantitative genetic association 
between the timing of phase change and first flowering. In other words, vegetative 
phase change was not conditional for flowering, since flowers can coincide in plant 
1 3  
parts having both adult and/or juvenile foliage (Potts and Wiltshire 1997) .  Their 
findings support previous studies suggesting that many Eucalyptus species may have 
undergone rapid heterochronic evolution of morphologically and ecologically distinct 
groups (Potts and Wiltshire 1997) .  The timing of phase change, at least in Eucalyptus, 
is therefore suggested to be under strong genetic control. 
1 . 1.7 Morphological and anatomical features 
Characterisation of developmental phases in plants has been achieved through 
quantitative examination of numerous morphological and anatomical features (Day et 
al. 1997) .  Leaf shape and size present some of the most recognisable and conspicuous 
morphological traits that have been used as markers of phase change (Frydman and 
Wareing 1973a, Hackett 1976, Day et al. 1997) .  Juvenile leaves produced immediately 
after germination are usually smaller, simpler and anatomically different from leaves 
produced during the adult phase (Poethig 1997) .  Juvenile leaves of English ivy (Hedera 
helix), for example, typically have three-to five-lobed palmate leaves arranged in 
alternate phyllotaxy, while adult leaves are entire and ovate, developing in a spiral 
arrangement (Frydman and Wareing 1973a). Under natural conditions, the length of 
this juvenile period can last between five to ten years (Hackett 1985) .  This is not 
atypical ,  considering that the duration of the juvenile period can be as short as 20 to 30 
days in perennial Rosa, and up to 30 to 40 years in F agus sylvatica (Hackett 1985) .  
Many New Zealand woody species which include around 200 species from 37 families 
(Cockayne 19 1 1) exhibit heteroblastic leaf characteristics .  Clearwater and Gould 
( 1993) characterised leaf development in Pseudopanax crassifolius. Juvenile P. 
crassifolius produce long and narrow (25- 100 cm long, 0 .5- 1 .5 cm wide) sharply 
toothed leaves in contrast to the short and broad, ( 10 -20 cm long, 1 .5 -3 .0 cm wide) 
dentate leaves found in the adult phase (Clearwater and Gould 1993) .  One New 
Zealand species, Elaeocarpus hookerianus, is known to exhibit three distinct 
ontogenetic leaf morphological and anatomical phases, comprising a juvenile, an 
adolescent and an adult phase (Day et al. 1997) .  Each of these phases is characterised 
by having distinct leaf shapes, being larger in terms of length, width and area in the 
adult form in comparison to leaves from the adolescent and j uvenile phases. Also, adult 
leaves are typically obcordate, oval or elliptic in shape and are longer (lanceolate) and 
1 4  
wider compared to the non-adult phases (Day et al. 1997) .  The variation in leaf shape 
or shoot architecture characteristics between ontogenetic phases has been hypothesised 
to be an environmental adaptation response, whereby these phase-specific modifications 
provide efficient light capture under forest canopies (Day and Gould 1997 , Day et al. 
1997) and/or tolerance mechanisms against environmental stresses (James and Bell 
200 1 ) . 
Differences in other leaf properties have also been described for a number of model 
woody plants. In H. helix and Solanum aviculare leaf thickness can vary between 
ontogenetic phases, being significantly larger and having a greater leaf mass during the 
adult phase (Bauer and Bauer 1980, James and Mantell 1994). These ontogenetic 
differences have been attributed to a more weakly differentiated palisade mesophyll 
layer in juvenile tissue, and not to differences in the adaxial or abaxial epidermal layers 
between phases (Day et al. 1997, James and Bell 200 1) .  Additionally, environmental 
factors can also influence leaf anatomy (Poethig 1997).  Day et al. ( 1997) showed that 
Elaeocarpus hookerianus plants, irrespective of their growth phase, produced fewer leaf 
cell layers and palisade was less well developed when grown under shadier conditions. 
Within the family Myrtaceae, several studies have characterised the changes in leaf 
morphology between ontogenetic phases (Ito and Suzaki 1990, Chalmers 1992) and 
during phase change (James and Bell 200 1) .  Most workers have focused particular 
attention on Eucalyptus, considering that it is a large genus comprising around 700 
species (Williams and Brooker 1997) with many species that exhibit homoblastic leaf 
development. Leaves of juvenile Eucalyptus are typically broader and thinner than their 
adult counterparts. Juvenile leaves can also undergo a transition in stomatal placement, 
developing on both adaxial and abaxial surfaces (amphistomy) during the adult phase 
(James and Bell 200 1 ) .  Many leaf characteristics in this genus, however, may not 
always be reliable markers of reproductive phase change, since some species are 
capable of producing flowers deriving from juvenile shoots (Potts and Wiltshire 1997) .  
S tudies of phase-related changes in leaf morphology in other genera within the 
Myrtaceae would provide interesting developmental comparisons. 
S tudies conducted on several plant species have suggested that the transition between 
j uvenile and adult morphological characters is also correlated with an increase in the 
15  
size of the shoot apical meristem (Hackett 1985 and references within) .  This is in 
consideration of the fact that the apical and subapical meristem of the shoot apex is  the 
site where different paths of cellular, tissue and organ determination and differentiation 
originate (Hackett and Murray 1992) .  Therefore, it was logical to conclude that these 
are the regions that are involved in the origin of ontogenetically-related changes in 
development (Poethig 1990, Hackett and Murray 1992) .  In juvenile plants of Hedera 
helix, for instance, meristematic regions in the apices are smaller, although they contain 
larger apical cells than those found in adult plants. Cell division rates differ between 
these ontogenetic phases, and a decrease in cell division in the subapical region of adult 
plants is correlated with a shorter internode distance (Stein and Fosket 1969) .  Many of 
these and other differing phase-dependent features described above have also been 
reported for a various other woody plant genera such as Betula (Brand and Lineberger 
1992), Solanum (J ames and Man tell 1994 ), Eucalyptus (J ames and Bell 200 1 )  and in 
various crop plants including blueberry (Lyre ne 198 1 )  and elderberry (Hrib et al.  1980).  
An example of various features that distinguish the juvenile and adult phases in model 
plants of the woody English ivy and the herbaceous maize can be seen in Table 1 . 1 
1.1.8 Plant size in relation to phase change 
The size of the developing plant may be important in affecting the ontogenetic status .  
Various studies have shown that attainment of a minimum size i s  correlated with the 
transition to an adult condition, as indicated by the attainment of reproductive structures 
(Hackett 1985) .  Generally, ' size' refers to the distance from the apical mersitem to the 
base of the plant (distance to the roots) but it can also be attributed to the mean stem 
diameter (Visser 1964) or number of nodes (Snowball et al. 1994) .  Visser ( 1964) for 
example, showed that the stem diameter of eight- to twelve-year old apple and pear 
seedlings was inversely correlated with the length of the juvenile period. Similarly, the 
attainment of a minimum size (height) in Ribes nigrum seedlings was strongly 
correlated with the capability of the plant to respond to short-day inductive cues for 
flowering (Robinson and W areing 1969). Further, conditions that promoted rapid 
growth were negatively associated with the length of the juvenile period (Visser 1964, 
Robinson and W areing 1969) .  How plant size affects the phase change process remains 
unclear, although several hypotheses have been proposed. 
16  
Table 1 . 1 . Features that distinguish juvenile and adult phases of English ivy (Hedera 
helix) and maize (Zea mays) (from Poethig 1990) .  
Traits 
Leaf shape 
Leaf thickness 
Phyllotaxy 
Plastochron 
Growth habit 
Anthocyanin 
Aerial roots 
Rooting ability 
Flowers 
Cuticle thickness 
Epidermal cell shape 
Epicuticular wax 
Aerial roots 
Epidermal hairs 
Bulliform cells 
Lateral buds 
Anthracnose resistance 
Juvenile 
Hedera helix 
Entire 
230 Jlffi 
Alternate 
1 week 
Plagiotropic 
Present 
Present 
Good 
Absent 
Zea mays 
1 ).lm 
Circular 
Present 
Present 
Absent 
Absent 
Tiller-like 
Poor 
Adult 
Lobed 
330 Jlm 
Spiral 
2 weeks 
Orthotropic 
Absent 
Absent 
Poor 
Present 
3 J.Lm 
Rectangular 
Absent 
Absent 
Present 
Present 
Ears or absent 
Good 
Attainment of a particular size may not, in itself, cause phase change but rather serve as 
an indicator of the ability of the plant to undergo phase change. Hackett ( 1 985) outlined 
three hypotheses to account for this, suggesting firstly that size is likely to act as an 
indicator of the number of cell divisions undertaken in the apical meristem. Secondly, 
he hypothesised that size can also reflect the position of apical meristems in relation to 
the other parts of the plant (e.g. roots) and with the corresponding distances between 
these parts, particularly with respect to translocation of assimilates, water and 
hormones. Assimilate accumulation and redistribution to shoot meristems might, 
therefore, be dependent on the complexity of the plant (tree).  Thirdly, size can serve as 
an indicator of relative cumulative sizes and activities of various meristems, both in the 
shoot and root system. The requirements for growth of these meristems on the plant 
system as a whole would also influence the competition for assimilates such as 
photosynthates, water and hormones (Hackett 1985) .  
17  
1 .1.9 Environmental factors 
1.1.9.1 Light and temperature 
Environments that favour rapid or vigorous growth are also associated with accelerated 
phase change (Hackett 1 985,  Snowball et al. 1988,  1 994) . These environmental 
conditions are usually 'warm' temperature and ' long' day-lengths. This was 
demonstrated in early studies by J onkers ( 1 97 1 )  who showed that young apple seedlings 
planted in fields at various latitudes across Europe (Finland to Italy) responded to 
relatively warmer and longer daylengths in the southern locations by growing more 
rapidly and reaching reproductive maturity sooner. More recently, several studies 
growing plants under controlled conditions have been able to quantify the effects of 
different environments on phase change. James and Bell (2000a) grew juvenile plants 
of Eucalyptus under varying degrees of shade ( 100, 50 and 1 0% sunlight) . Plants 
exposed to low light conditions not only showed reduced growth, biomass and leaf area, 
but also an increased retention of the juvenile phase. Thus, the time taken for initiation 
of vegetative phase change increased with decreasing l ight availability despite the fact 
that the position of phase change on the main shoot remained the same (lames and Bell 
2000a). S imilarly, other experiments on Citrus seedlings have also shown that 
environments that enhance growth and promote attainment of a critical size (critical 
node number) resulted in a shortening of the juvenile phase (Snowball et al. 1 988 ,  
1 994 ) . The prolonging of the juvenile phase or rejuvenation caused by low light 
intensity and high temperature are likely due to conditions that reduce the availability of 
carbohydrates necessary for growth and development towards phase change (Hackett 
1 985, Tsai et al. 1 997). 
1 .1.9.2 Water stress 
Little information is available on the effects of various environmental stresses on phase 
change. Most reports, however, have emphasised the role of water stress in affecting 
the transition from adult vegetative plants to those that become competent to flower and 
few on the transition between juvenile and adult vegetative characteristics .  There is  
some evidence, although mostly anecdotal, that water stress in particular can play an 
important role in the timing into the reproductive ontogenetic phase. R.P. Pharis ,  in  
personal communication with Hackett ( 1 985), observed that Pseudotsuga menziesii 
1 8  
plants growing on South Vancouver Island in a low rainfall area flowered within six to 
ten years . In comparison, the same species growing 50 miles away in a cool wet region 
did not flower until 15 to 20 years in age. These observations were clearly confounded 
by environment, and as noted by Hackett ( 1985), it would, therefore, be difficult to 
distinguish the effect of water stress on phase change from those conditions that 
promote induction of flowering. 
1.1 .10 Carbon isotope discrimination studies 
Assessment of plant physiological performance has been evaluated using stable isotopes 
obtained from plant tissue (Farquhar et al. 1989, Donovan and Ehleringer 199 1 ,  1994; 
Poss et al. 2000) .  Discrimination against the naturally occurring stable 1 3C occurs along 
the C02 diffusion pathway and during photosynthetic C02 fixation in plants. The extent 
to which discrimination occurs and the composition of the carbon isotopes in leaf tissue 
of C3 plants are primarily determined by the ratio of intercellular to atmospheric partial 
pressure of C02 during assimilation and by the isotopic composition of the source 
(Farquhar et al. 1982). Both C02 assimilation and stomatal conductance directly affect 
the ratio of intercellular to atmospheric partial pressure of C02. The ratio of C02 
assimilation to transpiration also provides a long-term measure of the intrinsic water-use 
efficiency (carbon gained per unit of water lost) of the plant, a variable that is highly 
and consistently correlated with carbon isotope discrimination itself (Farquhar and 
Richards 1984, Farquhar et al. 1989) .  Thus, plants that are highly water-use efficient 
tend to have tissue with relatively lower carbon isotope discrimination (?) values. 
Carbon isotope discrimination has also been correlated with a diversity of gas exchange 
components, including leaf conductance, transpiration, and photosynthetic capacity 
(Farquhar et al. 1989, Ehleringer et al. 1993, Sparks and Ehleringer 1997) including 
properties intrinsic to the leaf, such as leaf mass per unit area (Osorio and Pereira 1994 ), 
leaf thickness (Hanba et al. 1999) and mineral status (Sparks and Ehleringer 1997 , Poss 
et al. 2000) .  
Inter- and intraspecific variation in 1 3C discrimination in woody plant species has been 
reported in numerous plant ecological and physiological studies (e.g. Donovan and 
Ehleringer 199 1 ,  Sparks and Ehleringer 1997 , Bonal et al. 2000).  Considerable 
attention has been given to the level of ?' and its relationship to photosynthetic 
1 9  
pathways (Farquhar et al .  1989), water use and gas exchange at a s ingle leaf (Robinson 
et al. 1993) and whole plant canopy scale (Donovan and Ehleringer 199 1, Stew art et al. 
1995, Bonal et al. 2000) .  Reports on the intraspecific variation in 6. between woody 
plants at different ontogenetic states has been limited primarily to ecophysiological 
studies of plants from desert assemblages (Sandquist et al. 1993, Donovan and 
Ehleringer 1994) and Acacia and Metrosideros populations found on the Hawaiian 
islands (Hansen 1996, Cordell et al. 1998) .  Discrimination is often reported to be 
higher for juvenile plants than for adult plants, and is associated with higher 
photosynthetic rates and decreased water-use efficiency during the juvenile phase. This 
phenomenon is often reported for successful establishment of juvenile plants in 
environments where water is often a limiting factor (Donovan and Ehleringer 199 1, 
Bond 2000) .  Therefore, differences in physiological responses between plants of 
different ontogenetic states may be confounded when examined under variable field 
conditions. As noted by Donovan and Ehleringer ( 1994 ),  juvenile and adult plants are 
potentially under different selective pressures, and variation in ecophysiological 
characters (e.g. water access as function of root depth) may affect growth and survival. 
In this thesis,  woody plants of Metrosideros excelsa were grown under controlled 
conditions and in the absence of water-stress. This would provide a more accurate 
evaluation of plant ontogenetic state during the transition between phases. 
Consequently, these results would make an interesting comparison with the results of 
many of the ecophysiological studies, where findings may have been confounded by the 
prevailing environmental conditions. 
1.1 .11 Photosynthetic characteristics and chlorophyll concentration 
Differences in photosynthetic characteristic have been attributed to difference in 
ontogenetic states. A variety of studies generally indicates that both photosynthesis and 
stomatal conductance decrease with increasing age or ontogenetic state, although this 
pattern is not consistent for all species (Bond 2000). For example, decreased net 
photosynthetic rates with increasing ontogenetic development have been reported in 
Picea abies (Kull and Koppel 1987) ,  Sequoiadendron giganteum (Grulke and Miller 
1994) and Pinus ponderosa (Kolb and Stone 2000), whereas other woody species such 
as Quercus rubra (Boardman 1977) show the reverse trend. 
20 
With regards to chlorophyll content, juvenile leaves of a number of species, such as 
Solanum aviculare (J arnes and Man tell 1994) have been reported to have higher total 
chlorophyll concentrations than those of adult leaves, whereas adult leaves of Larix 
have significantly higher levels of total chlorophyll (Greenwood et al. 1989) .  
Variations i n  the different types of chlorophyll can be confounded by season. In 
Solanum aviculare, the chlorophyll a and b ratio can be higher in adult leaves during the 
summer, and reversed in the winter. Considering that differences in chlorophyll content 
or ratio can vary spatially within a plant and between different seasons or environments 
(J ames and Man tel l  1994, Bond 2000), these parameters are generally not considered 
reliable markers of ontogenetic status. 
1.1.12 Role of Carbohydrates 
The role of carbohydrates in affecting shoot development has often been neglected in 
studies involving phase change, particularly during vegetative phase change. Most 
studies focusing on reproductive phase change have shown strong correlations between 
accumulation of carbohydrate levels and the tendency to flower (reviewed in Bemier et 
al. 198 1 a). A reduction in carbohydrates can be directly affected by exogenous factors . 
Environmental treatments that reduce carbohydrate levels such as extreme temperature 
conditions and low light intensities subsequently delay vegetative phase change (Njoku 
1958, Hackett 1976 , Schwabe 1976, Hackett 1985) .  Low light conditions, for example, 
can delay changes in juvenile and adult leaf shapes in Ipomoea caerulea (Njoku 1958) .  
Treatments that promote carbohydrate accumulation in  various parts of  the plant can 
also affect the transport and accumulation of other substances (e .g. hormones), and thus, 
complicate the interpretation of the effects of carbohydrates on development. With the 
recent advance of molecular studies, however, researchers have been able to clearly 
elucidate the role of photosynthates in vegetative phase change (Tsai et al. 1997) .  Tsai 
et al. ( 1997) were interested in determining what effect source strength (carbohydrate 
production) would have on shoot development. Using an antisense mutant of Nicotiana 
tabacum that had reduced amounts of Rubisco in comparison with a wild-type tobacco 
plant, Tsai et al. ( 1997) were able to make comparisons of differential shoot 
development in juvenile plants. Both types of plants displayed distinct phases of shoot 
development as evidenced by differences in shoot elongation rates, internode distances, 
plastochron indices, leaf sizes and leaf morphologies. The juvenile phase of the 
2 1  
antisense plants showed a delay in transition to an adult phase. This was suggested to 
occur in response to the attainment of a threshold source strength, which was delayed in 
the mutant plants (Tsai et al. 1997) .  Their study clearly emphasised the important role 
of source strength in the timing of phase change. However, since the delay in phase 
change in antisense plants was only temporary, this strongly supports the hypothesis 
that several factors, other than carbohydrate levels, are involved in phase change. 
1 .1 .13 Plant hormones 
In addition to nutritional factors, there is convincing evidence to suggest that 
phytohormones are also intricately involved in phase change. In particular, the ratios of 
different hormones such as abscisic acid (ABA), gibberellins (GAs), auxin (IAA) and 
cytokinins (CKs), rather than their absolute concentrations, are suggested to play an 
important role in promoting phase change (Greenwood 1995) .  It i s  well established that 
hormones such as cytokinins, auxins and gibberellins create metabolic s inks which 
mobilise assimilates (Sachs 1977) .  These hormones may serve to create competition 
between sinks for assimilates (Hackett 1976, Hackett 1985) and are likely to be 
involved in the partitioning of assimilates in the apical region where phase change 
occurs (Hackett 1985) .  
1 . 1. 13. 1 Gibberellins and abscisic acid. 
Numerous classical studies have closely linked the role of GAs and ABA with phase 
change (Robbins 1957, Frydman and Wareing 1973a, 1973b, Frydman and Wareing 
1974) . The suggestion of Robbins ( 1957) that GAs are important in phase change was 
first prompted by the finding that exogenously applied GA3 caused rejuvenation in 
Hedera helix. Subsequent studies showed that "gibberellin-like substances", which can 
be synthesised in both roots and immature leaves and buds of H. helix, could act in the 
regulation of phase change in the apical region (Frydman and W areing 1973a, 1973b ) .  
High levels of  endogenous GAs appear to  be associated with juvenility (W areing and 
Frydman 1976), and the application of exogenous GA3 to adult plants induced the 
development of juvenile characteristics (Rogler and Hackett 1975) ,  a feature described 
as causing "complete rejuvenation" (Frydman and Wareing 1974). The application of 
suggested antagonists (e.g. chlormequat, ancymidol and darninozide) or inhibitors (e.g. 
22 
ABA) of endogenous gibberellins appear to stabilise the adult phase in H. helix. Plants 
treated with 5 nM of GA3 and 5 mM of ABA prevented rejuvenation, but when GA3 
was increased to five-fold with the same levels of ABA, reversion did occur. They 
suggested that the relative rather than the absolute amount of each hormone used was 
important in controlling the growth form. Although stabilisation of the adult phase may 
be maintained with these treatments, the application of similar treatments (chlormequat 
and ABA) has not led to the acceleration of phase change (Frydman and Wareing 1974).  
More recent work by Horrell et al. ( 1990) showed that exogenous applications of GA3 
to adult apices of H. helix resulted in rejuvenation, but only partially as opposed to 
complete reversion as previously suggested (Frydman and Wareing 1974). This 
included the incomplete reversion of characteristics such as leaf lobing and anthocyanin 
pigmentation that typify the juvenile form. Application of paclobutrazol (PP333) (an 
inhibitor of endogenous gibberellin biosynthesis) and ABA treatments were ineffective 
in promoting phase change in juvenile H. helix (Horrell et al. 1989b, 1990) .  Contrary to 
expectations, PP333 promoted shoot extension in juvenile plants, a response 
hypothesised to be the result of gibberellin production in aerial roots (Horrell et al. 
1989b ) .  Considering the atypical response of this species to PP333 compared with 
other heteroblastic species, Horrell et al. ( 1990) proposed that H. helix was not an ideal 
model for studies of phase change. 
The effects of plant growth regulators have been reported for various New Zealand 
plant species. Similar to H. helix, application of GA3 to adult plants of Carpodetus 
serratus, Pennantia corymbosa and Parsonsia heterophylla caused only partial 
reversion to a juvenile phase (Horrell et al. 1989a). Applications of PP333 or ABA to 
juvenile plants of these New Zealand species and others (e.g. Pseudopanax crassifolius) 
did not promote phase change but, unlike in H. helix, shoot elongation was inhibited or 
otherwise plants died (J ameson and Horrell 1988, Horrell et al. 1989a) . Endogenous 
GAs are, therefore, implicated in the retention of the juvenile phase, although the role of 
GAs in phase change is, as yet, unclear. The results from these and other studies 
involving application of anti-gibberellin treatments suggest that plant responses to 
factors other than a decrease in GAs are likely to be involved in phase change from the 
juvenile to adult state. 
23 
1.1 .13.2 Cytokinins 
Most physiological studies examining phase change have focused on GAs, although 
there is some evidence suggesting a role played by cytokinins (Ross and Pharis 1976 , 
Irnbault et al. 1988, Day et al. 1995 , Diekrnan 1997) .  Cytokinins (CKs), which are 
predominately root-produced hormones, have been implicated in a host of plant 
hormone interactions and developmental processes (Voesenek and Blom 1996) .  Among 
these has been their role in vegetative to reproductive development. Zaerr and Bonnet?
Masimbert ( 1987), for instance, reported that a reduction in endogenous CKs in the 
shoot of Douglas fir was associated with treatments that promoted this transition. 
Similarly, exogenous applications of CKs have also been demonstrated to depress 
gibberellin-induced reproductive bud production in Picea mariana (black spruce) 
(Smith and Greenwood 1995) .  It has been hypothesised that CKs may regulate but not 
promote flowering, considering that juvenile plants do not flower even in the presence 
of exogenous cytokinin (lrnbault et al. 1988) .  
A few studies have compared the relative levels of cytokinins between different phases 
of vegetative development (Carswell et al. 1996, Mercier and Endres 1999) .  Seedlings 
of the woody angiosperms Sophora microphylla and S. prostrata were treated with 6N?
benzyladenine (BA) (Carswell et al. 1996) .  In S. prostrata, BA appeared to have no 
effect on growth, whereas in S. microphylla the divaricating shoot growth typical of 
juvenile plants was enhanced. This was exhibited by an increase in the number of 
outgrowing branches and by a widening of branch angles in comparison with water?
treated control plants. Examination of assays suggested the ratio of active to storage 
CKs was greater in divaricating than non-divaricating forms.  S imilar reports suggesting 
that higher concentrations of CKs occur in juvenile tissue have been supported by 
investigations in Elaeocarpus hookerianus (Day et al. 1995), Tillandsia recurvata, an 
epiphytic bromeliad (Mercier and Endres 1999) and in the rubber-tree Hevea 
brasiliensis (Perrin et al. 1997) .  In E. hookerianus, treatment of seedlings, j uvenile and 
adult plants with exogenous CK resulted in differential responses (Day et al. 1998) .  
Juvenile plants showed greater variation in  leaf morphology (leaf length, proportion of 
linear leaves and leaf angle) and growth habit (height and internode length) than 
seedling or adult plants. However, application of exogenous CK did not induce any 
substantial ontogenetic development of juvenile plants, nor did it cause rejuvenation in 
24 
adult plants. A recent examination of endogenous CKs in j uvenile and adult Pinus 
radiata has revealed several novel forms of CK glucosides that are differentially 
expressed, in terms of concentration, in juvenile and adult buds (Zhang 1 998,  Zhang et 
al. 200 1 ) .  
Recent genetic studies are advancing our understanding of the role played by CKs in 
plant development and phase change, although much progress is still needed. The 
Arabidopsis mutant, altered meristem program I (amp I) ,  that produces elevated levels 
of CK, has provided key insights into the role of CKs during phase change (Diekman 
1 997). Observations of phenotypes in amp] in comparison with the wild-type indicate 
that heightened levels of CK prolong the juvenile phase. This may be because of the 
increased number of cotyledons and greater number of rosette leaves expressed in the 
amp I .  Another mutant, cyr I ,  produces under-developed cotyledons and true leaves, 
and is insensitive to CKs (Diekman and Ulrich 1995), suggestive of a disruption in CK 
perception or  signal transduction (Diekman 1 997). This may explain why cyr I has a 
relatively short juvenile period. These studies imply that CK is required for the 
expression of the juvenile phase. The repression of genes involved in flower formation 
by CK would also support this conclusion (Venglat and Sawhney 1 996) . 
1.1 .14 Acceleration of phase change 
A number of cultural practices have been employed in order to accelerate phase change. 
The most successful method of shortening the juvenile phase in woody species has been 
to grow plants under environmental conditions that promote rapid or vigorous growth 
(see Section 1 . 1 .9) (Poethig 1 990) . This usually equates to relatively warm and high 
l ight environments . Moreover, this method is often coupled with various cultural 
practices that physically alter the growth pattern of a plant and accelerate phase change. 
For example, Snowball et al. ( 1 994) grew various citrus species trained to a single-stem 
(lateral buds removed) in either a warm greenhouse (min/max: 1 5-30?C) ,  in an outdoor 
shade-house (50% shade) or under ambient year-round outdoor conditions. Growth 
rates (number of nodes attained) were significantly higher for citrus plants grown under 
greenhouse conditions, and generally those that flowered were obtained only from this 
environment. The practice of growing plants as a single-stem has also been successful 
in accelerating phase change (reduction to the time to flower) in kiwifruit (Davis 1 99 1 )  
25 
and various other woody plants (reviewed in Snowball 1 989). This practice of growing 
plants as a single-stem, as opposed to allowing plants to branch freely, is suggested to 
encourage growth to a minimum size required for phase change to occur, since this 
feature is often inversely related to the length of the juvenile period (Hackett 1 976, 
1 985) .  In contrast, most practices that tend to reduce growth or promote flowering in 
adult plants, such as girdling, scoring, bark inversions, grafting onto dwarfing stocks 
and application of plant growth regulators, generally have little to no effect on the 
length of the juvenile period (Zimmerman 1 972, Hackett 1976) . 
The timing of phase change, at least in some species, can be under strong genetic 
control (Wiltshire et al. 1998).  Selective breeding of desired genotypes can be a method 
for obtaining plants with short juvenile phases. For example, by selecting offspring of 
Betula pendula with precocious flowering, V on Stern ( 1 96 1 )  was able to reduce the 
juvenile period from five to 1 0  years down to two years . This has also been conducted 
in various other woody crops including Pinus species (Heimburger and Fowler 1 969), 
apple (Way 1 97 1 )  and pear (Zimmerman 1 977) . 
There has been some speculation that root restriction (small pot size) can accelerate 
phase change (Zirnrnerman 1 972, Oliphant et al. 1 992). Various reports have claimed 
that potted trees reach reproductive maturity sooner when grown in environments that 
promote vigorous growth (reviewed in Zirnrnerman 1972) . However, in many of these 
and other studies, attainment of the adult phase has been assessed using flowering as an 
indicator of maturity and, therefore, no quantitative distinction has been made of the 
progressive changes in vegetative characters and the onset of flowering. 
1 .1.15 Markers of phase change 
The search for reliable markers specific to ontogenetic phases of development has been 
an ongoing pursuit in studies of phase change. An outcome of such studies would allow 
assessment of the progression of phase change or the degree of stability in phase-related 
characteristics (Hackett and Murray 1 997) .  This would provide a basis for 
understanding the mechanisms associated with stability, as well as devising methods or 
protocols for manipulating phase change. Several promising biochemical markers 
indicative of a developmental phase have been described. Anthocyanin pigments, 
26 
which are flavonoid glycosides, have been found to accumulate in stems and leaf 
petioles of juvenile Hedera helix, although not in adult tissue (Hackett and Murray 
1 997). Murray and Hackett ( 1 99 1 )  found that the lack of activity in dihydroflavonol 
reductase (DFR), an enzyme involved in the anthocyanin biosynthetic pathway, limited 
the accumulation of anthocyanin in adult tissues . Thus, they concluded that DFR 
activity could be a definitive marker for the juvenile phase in H. helix. Various other 
proteins have been also been identified that are either unique to juvenile shoots or show 
a higher activity in juvenile than adult shoots. For example, catalase isozymes in 
Cupressus sempervirens show phase-specific differences that may be suitable as 
biochemical markers for rooting ability (Racchi et al. 1996). Other potential 
biochemical markers have also been described for Sequoia sempervirens (Berthon et al. 
1 987), Prunus avium (Hand et al. 1996) and citrus crops (Snowball et al. 1 99 1 ) .  
However, the relative differential expression (between species) and species-specificity 
of a number of endogenous substances, such as proteins, appears to be the primary 
disadvantage for the use of these biochemical markers. 
Morphologically, the presence of floral structures has readily been used as a marker to 
distinguish between the vegetative adult and reproductively competent adult phase, 
and/or signifying the end of the juvenile phase (Hackett 1 985, Poethig 1 990) . However, 
this method is not always a consistent indicator given that some species can produce 
reproductive structures arising from branches containing distinctly juvenile foliage (e.g. 
some Eucalyptus species) (Wiltshire et al. 1 99 1 ) . An approach to characterise different 
phases of vegetative growth has been the use of leaf morphological markers . This has 
been particularly applicable for heteroblastic species where differences between 
different phases of ontogenetic development are visibly apparent and often represented 
by striking differences in morphological features (Day et al. 1997, J ames and Bell 
200 1 ) .  Characterisation of homoblastic species such as Metrosideros excelsa presents 
more of a challenge given the subtleties in expression of phenotypic characters during 
development. However, once the attainment of leaves characteristic of an adult 
vegetative phase has been reached, further inductive treatments are usually required to 
confirm or dismiss reproductive competence. 
27 
1.2 Floral Induction and Development in Ornamental Woody Plants 
1 .2.1 Introduction 
An intimate understanding of processes that govern flowering has posed a formidable 
challenge for plant biologists, despite the large volume of literature investigating this 
phenomenon (References within: Kinet 1 993, Meilan 1997, Reeves and Coupland 
2000) . This is because floral development, per se, is a multistage process entailing 
often a complex series of events that are spatially and temporally related (Bernier et al. 
1 993) .  First, plants must generally undergo a period of reproductive incompetence 
generally during the juvenile and adult vegetative states (Hackett 1 985) .  Second, 
following a period of suitable environmental conditions for induction can the flowering 
process commence. Each developmental event, starting from floral induction through to 
flower senescence and subsequent fruit and seed development, is dependent upon 
specific requirements and is affected differently by environmental and chemical 
parameters (Kinet 1 993). In many species, environmental factors act as cues for the 
timing of flowering, by providing communication on the time of year and conditions 
suitable for sexual reproductive growth (King et al. 1 995). 
Numerous fundamental studies have provided us with a broad understanding of the 
physiological and molecular mechanisms involved in flowering (Bernier et al. 1 993,  
Levy and Dean 1998). Additionally, applied studies have sought to determine the 
optimum combinations of environmental and/or chemical factors affecting the flowering 
process (Jackson and Sweet 1 972, Moncur 1992, Myster 1 999) . A large range of 
reviews has been published on flowering, including a six-volume Handbook of 
Flowering edited by Halevy ( 1985, 1986, 1989) and a three-volume monograph on the 
physiology of flowering (Bernier et al. 1 98 1 a, 198 1 b, Kinet et al. 1 985) .  Some reviews 
have focused on the molecular (Amasino 1 996, Levy and Dean 1 998, Reeves and 
Coupland 2000) and physiological (Bernier et al. 1 993, Srivastava and Iqbal 1 994) 
mechanisms involved in the flowering process, whereas other have directed their 
attention to specific plant groups, e .g. coniferous trees (Bonnet-Masimbert 1 987),  
woody angiosperm trees (Meilan 1 997), and particular ornamental crops (Criley 1 998) .  
There have been no published reviews, at least to the author' s knowledge, on the factors 
28 
affecting flowering specifically in woody angiosperms that have floricultural 
significance. In this review, both the applied and fundamental studies addressing the 
role of environmental and chemical factors regulating the flowering process in woody 
ornamental flowering plants are investigated, although when relevant, reference to other 
crop and/or model plant systems will be discussed. 
1.2.2 Terminology 
The termflowering, which is often broadly or ill-defined, is referred to in this thesis as 
the reproductive period including flower opening (King et al. 1 992). Specific 
terminology describing the early stages preceding the development of a floral 
reproductive structure follow those specified by Evans ( 1 969). Evans ( 1 969) made the 
distinction between induction which refers to the events in the leaf which commit plants 
to the flowering process, and evocation which refers to the events that occur at the shoot 
apex following the arrival there of a floral stimulus, but which precedes the onset of 
differentiation of the floral organs. 
Subsequent stages in the flowering process are defined according to K.inet ( 1 993) who 
subdivided it into three stages :  ( 1 )  floral initiation, which refers to the reactions 
required for the initiation of flowers and, given the appropriate conditions, leads to the 
"irreversible commitment" of the shoot apical meristem towards inflorescence and/or 
flower primordia production; (2) floral morphogenesis, which pertains to the formation 
of recognisable features of the meristem, including floral primordia and other floral 
organs;  and (3) floral development, which refers to the development of reproductive 
structures through to anthesis .  
1 .2.3 Flower induction models 
Over the past decades, scientists have sought to explain by way of models the 
mechanism by which flowering is induced and maintained. One of the first 
physiological hypotheses suggesting the role of a stimulus (florigen) for inducing 
flowering was proposed in the 1 930' s by Mikhail Chailakhyan (reviewed in Salisbury 
and Ross 1 992). This model proposed that a key substance (hormone) signalled the 
onset of flowering. The concept of florigen was based on the transmissibility of 
29 
substances or signals across graft unions, as was evident when a reproductively 
competent shoot was grafted with a vegetative recipient (Levy and Dean 1 998) .  It was 
suggested that florigen was synthesised in leaves exposed to favourable photoperiods, 
and was subsequently transported to the shoot apex via the phloem, where it acted as a 
floral-promoter. A contrasting view sought to explain the onset of flowering based on 
an antiflorigen concept, which was graft-transmissible, yet acted as a floral inhibitor. 
Many unsuccessful research years, however, have been spent searching for the illusive 
key florigen or antiflorigen substance within the phloem. 
Due to the inability to identify a single florigenic compound, a second model, the 
nutrient diversion hypothesis, was introduced by Sachs and Hackett ( 1 983) .  The 
nutrient diversion hypothesis proposed that inductive conditions modified the 
source/sink relationship in the plant, subsequently leading to an increase in the level of 
assimilates directed to the shoot apical meristem for reproductive growth. However, 
this model did not account for either the diversity of flowering responses observed in 
different species, or the influence of different environmental cues, or the concept of 
inhibitive or promotive role of endogenous substances. 
With the concept that flowering was not solely induced by assimilates, Bemier ( 1 988)  
proposed the multifactorial control model which not only accounted for assimilates but 
other factors such as hormones, which acted as promoters and inhibitors controlling the 
timing of flowering. Thus, this model maintained that flowering could occur only if the 
limiting factors present at the shoot apical meristem were present at the correct times 
and in the appropriate concentrations. This multifactorial control model is currently the 
most accepted model for explaining the role of environmental and chemical factors 
regulating flowering and this hypothesis is supported by findings in various model 
plants such as pea, cereals and Arabidopsis. 
1 .2.4 The molecular frontier 
Whatever factors are involved in the flower transition, ultimately control will be exerted 
at the genetic level. Molecular research has generally been undertaken using 
herbaceous model plants, in particular Arabidopsis, Antirrhinum and Pisum. This i s  
primarily because woody species are not ideal model systems due to  maintenance of a 
30 
long juvenile period, the accumulation of secondary metabolites and subsequent 
difficulty with transformations and regeneration (Meilan 1 997). However, much 
progress has been made using these herbaceous models .  
The rosette plant Arabidopsis has been an important model in investigations in the 
timing of flowering and floral development. Recent research into the role of light in 
controlling flowering has been the focus of photoreceptors, being light-sensitive 
molecules involved in light perception located in the leaf organs (Thomas and Vince?
Prue 1997). Photoreceptors such as phytochrome (PHY A) and cryptochrome (CR Y2) 
initiate signals that interact with the circadian clock and entrain the circadian rhythm 
(Levy and Dean 1998). The signalling mechanism possibly provides information on 
light/dark transitions to the circadian clock, since both PHY A and CRY2 protein levels 
have been found to drop rapidly with exposure to light (Thomas and Vince-Prue 1 997) .  
Once daylength is  perceived and somehow measured by these photoreceptors (or the 
length of the dark period has decreased below a critical threshold), 'flowering time' 
genes such as CONSTANS (CO) and GIGANTEA (GI) in Arabidopsis are activated, 
leading to the upregulation of floral meristem identity genes (Amasino 1 996, Weigel 
1 998, Reeves and Coupland 2000) (Figure 1 .3 A). Products of the FLOWERING 
LOCUS C (FLC) gene inhibit the transition from vegetative growth to flowering, 
although low temperature can depress the abundance of FLC mRNA suggesting an 
important role of this gene in the vernalisation pathway (Figure 1 .3 B) .  The FRIGIDA 
(FRI) gene is involved in the autonomous pathway and promotes FLC expression, 
whereas other genes such as LD, FCA and VRN2, play an inhibitory role in its 
expression (Reeves and Coupland 2000) . 
Two classes of genes have been identified that regulate floral development, which 
comprise meristem identity genes and organ identity genes .  The first class of genes 
contains positive regulators of the floral meristem identity and includes several 
important genes described in Arabidopsis, such as LEAFY (LFY) and APETALAI (AP 1)  
(Weigel et  al. 1 992 , Bowman et  al. 1993) .  Both genes act as developmental switches 
activating the entire genetic program for the formation of the floral meristem (Weigel 
and Nilsson 1 995).  These genes are also positive regulators of a set of genes known as 
the floral organ identity genes. In Arabidopsis, APETALA2 (AP2), APETALA3 (AP3) ,  
PISTILLATA (PI) and AGAMOUS (AG) are floral organ identity genes that are involved 
3 1  
w 
N 
A. l'hotoprDI "".jt"" 
CO/Cii FLC 
rT/FWA ? ? 
? I ') 
? ? -
?D 
AP? I ?_!_] 
.--/\ P-1 /'-A P 2 r t\ eT] 
[ . Floral pattcrning 
B. 
Vegetat ive 
development 
1\ utononmus pathway 
promoters 
( ! . D. FCA,  J:V L : . e tc . ) 
F ? <:----1 FRIG IOJ\ 
? sr 
V <.:r l  Ht l  i sui  ion 
r : lom.;r ing 
Figure 1 .3 General models of the multiple flowering pathways in  Arabidopsis thaliana. A: Molecular and genetic in teractions in  the 
photoperiodic pathway and vernal isation. Long-day photoperiod leads to upregulation of circadian clock and l ight receptor mediated genes (CO, 
GI) i n  addition to the upregulation of other photoperiodic responsive genes (FT, FW A). The autonomous and photoperiodic pathways act 
addit ively to promote flowering and lead the upregulation of floral meristem identity genes (e.g. AP I ,  LFY), although the stage at which 
conversion of these pathways (?)  occur is unknown (modified from Reeves and Coupland 2000). B: Molecular and genetic interactions i nvolved 
i n  the autonomous pathway. The FRIGID  A (FRI) gene upregulates FLC despite the inhibitory effect of the autonomous pathway genes. 
in specifying the identity of the organ in each whorl (Weigel and Meyerowitz 1 994 ) .  
Coen and Meyerowitz ( 1 99 1 )  proposed the ABC model, i n  which three unique 
combinations of activities (A, B and C) of these homeotic genes determine the organ 
identity within each of the whorls. In this model, they proposed that the activity of type 
A genes controls organ identity in the first and second whorls. Type B genes determine 
floral organs in the second and third whorls .  And finally, type C activity is involved in 
organ identity in the third and fourth whorls of the flower (Coen and Meyerowitz 1 99 1 ,  
Weigel and Meyerowitz 1 994). The pattern of floral organ formation in the Arabidopsis 
wild type and most of the mutant phenotypes are supported by this model, although 
extensions to the model have been proposed (e.g. SEPALLATA) (Pelaz et al. 2000). 
To date, many of the floral genes isolated from model species are highly conserved and 
function in heterologous systems. Genes cloned from the herbaceous model systems 
can, therefore, be used as probes to find orthologues in woody plant species. 
Orthologues of several meristem identity genes, such as LEAFY have already been 
reported in Eucalyptus (Southerton et al. 1 998), Populus (poplar) (Rottmann et al. 
2000) and in Malus pumila (apple) (Kotoda et al. 2000), including APETALAJ in the 
latter species. Recently, partial orthologues of LEAFY (McKenzie et al. 1 997) and 
APETALAJ genes have been cloned from Metrosideros excelsa, in addition to the 
isolation of the partial orthologue of the TERMINAL FLOWER 1 gene (L. Sreekantan, 
pers. comm.).  
Thus far, several candidate woody species look promising as models for molecular 
research. Eucalyptus, cranberry, and poplar are capable of being transformed and 
regenerated with ease and are capable of floral induction through chemical applications 
(Serres and McCown 1 993 , Griffin et al. 1993, Meilan 1 997). Moreover, flowering can 
be induced in transgenic poplar in vitro (Meilan 1 997). The New Zealand woody 
species, Metrosideros excelsa, also looks to be an attractive model system for 
investigating the molecular and physiological processes associated with flowering. 
Further studies examining the role of environmental factors in promoting and regulating 
flowering in this species are investigated in this thesis, and complement existing 
research to date (Sreekantan et al. 2001 ) . 
33 
1 .2.5 Floral and vegetative growth relationships 
Woody angiosperm plants often undergo cyclic and repetitive patterns of vegetative 
growth with intermittent periods of flowering in response to changing seasons 
(Srivastava and Iqbal 1 994) .  During these periods, reductions in vegetative growth 
coincide with the production of flowers (Jackson and Sweet 1 972).  Treatments that 
manipulate environmental conditions and/or applications of growth retardants, have 
been shown to both reduce vegetative growth and subsequently enhance flower number 
(Jackson and Sweet 1 972, Helier et al. 1 994). These treatments probably cause the 
diversion of nutrients and carbohydrates from vegetative shoots to floral meristems 
(Sachs and Hackett 1983) .  In Lolium temulentum, for example, an increase in sucrose 
concentrations at the apex occurs after evocation and enhances inflorescence 
development (King and Evans 1 99 1  ). However, this phenomenon of antagonism 
between vegetative growth and flowering is not always observed. In many cases 
growth inhibition is often a secondary event that can occur at advanced stages in the 
floral transition. In the case of Bougainvillea (Hackett and Sachs 1 966), Boronia 
(Roberts and Menary 1 989) and Hypocalymma (Day et al. 1994a), both cycles are not 
mutually antagonistic and a reduction in vegetative growth is not necessary for 
flowering to occur. Therefore, a reduction in vegetative growth may enhance flowering 
although it does not appear to be part of the same mechanism controlling floral 
evocation. 
1.2.6 Environmental factors 
For many plant species, environmental factors play an important role for signalling the 
initiation of reproductive structures (Kinet 1 993). Manipulation of photoperiod, light 
quality, l ight intensity, temperature, and the availability of nutrients and water can serve 
as signals for the timing and regulation of flowering (Levy and Dean 1 998) .  The effects 
of each of these factors on the flowering process can function independently or in 
combination with other factors, and promote flowering based on ecologically optimal 
limits (King et al. 1 995).  These environmental factors are, however, discussed 
independently in order to ascertain the effect each factor predominantly has on the 
flowering process. 
34 
1.2.6.1 Light: duration 
Light is one of the major environmental factors regulating floral development. 
Photoperiodism, defined as the ability of an organism to detect the length of a day, was 
first observed in plants in classic experiments conducted by Garner and Allard ( 1 920) . 
They demonstrated the effect of daylength on the flowering response using a mutant 
tobacco plant, Maryland Mammoth, which they noticed flowered in the greenhouse only 
during winter. Garner and Allard ( 1 920) artificially shortened the daylength during the 
long summer daylengths by enclosing the tobacco plants in light-tight tents, simulating 
winter day-length conditions. The tobacco plants subsequently flowered, prompting the 
hypothesis that plants had the ability to measure varying daylengths throughout the 
seasons, which subsequently signals the time to flower under optimal daylengths .  Their 
hypothesis was confirmed using various other plant species (including soybean and 
spinach) and under various conditions. This work laid the foundation for subsequent 
and extensive work on photoperiodic responses .  To date, it is known that photoperiodic 
responses in plants are not limited to just flower initiation but to numerous other in vivo 
responses such as asexual reproduction, storage organ formation and the onset of 
dormancy (Vince-Prue 1985). 
Classification of plants based on their preference for an optimum day length for 
flowering can vary widely between species. Flowering in short-day (SD) plants is 
promoted only when the daylength is less than (or the night length exceeds) a critical 
threshold in every 24 h cycle. A number of temperate woody species, such as 
Chamelaucium uncinatum (Shillo et al. 1 985,  Dawson and King 1993) ,  Coleonema 
aspalathoides (Helier et al. 1994) ,  and Leptospermum scoparium (Zieslin 1 985,  Zieslin 
and Gottesman 1 986) have been classified as SD plants. Opposite to SD plants, 
flowering in long-day (LD) plants is promoted only when the day length exceeds (or the 
night length is less than) a critical duration. Eucalyptus occidentalis (Bolotin 1 975) and 
Hamelia patens (Armitage 1 995) are but a few species reported to be LD plants. 
In contrast to both SD and LD plants, intermediate-day plants, flower only when 
exposed to a narrow margin of a day length (e.g. 1 2 - 14 h photoperiods). 
Contrastingly, in day-neutral or autonomous flowering plants, a specified daylength or 
low temperature is not an inductive requirement for undergoing reproductive 
35 
development, and flowering is facilitated with time and under favourable growth 
conditions (Kinet 1 993). Several species such as Eucalytpus landsdowneana (Moncur 
1 992), Pimelea rosea (King et al. 1992), Boronia megastigma and Hypocalymma 
angustifolium (Day et al. 1 994a) have been described as day-neutral plants. In 
summary, plants can therefore respond (or not respond) to a specific day length 
requirement for flowering, and only by examining a range of photoperiods can a 
suitable photoperiodic classification be established. 
In many horticultural systems, manipulation of daylength has been accomplished within 
growth rooms, greenhouses or under outdoor ambient conditions through the use of 
artificially supplied lighting systems. These can include incandescent light-extensions 
that provide additional daylight hours (Bredmose 1993) ,  night-break interruptions 
(Langton 1977), and/or through the utilisation of light-covers preventing light 
penetration at required times (Shillo et al. 1 997). For a few woody crops of 
horticultural importance,  flowering can be controlled solely by the manipulation of 
photoperiod, although other factors such as temperature can have an interacting effect. 
In the case of the omamentally-valued Chamelaucium uncinatum (family Myrtaceae),  
an evergreen shrub native to the Mediterranean climate of Western Australia, plants 
were treated to two photoperiod treatments at various inductive intervals in order to 
determine their flowering response (Shillo et al. 1 98 1 ,  1 985) .  A minimum of four 
weeks of SD (8 h) before transference to LD ( 1 3- 14 h) conditions was sufficient for all 
C. uncinatum plants to produce flowers (Shillo et al. 1 985) .  Moreover, applying SD 
treatments for longer than four weeks served only to enhance flowering by further 
increasing the number of flowers and flowering shoots per branch (Shillo et al. 1985) .  
In some woody species, the inductive daylengths can affect flower initiation and/or 
development. Arrnitage ( 1 995) reported that both floral initiation and development 
occurred in Hamelia patens in response to LD treatments. While flower bud initiation 
occurred in approximately two-thirds of the plants treated to 8 and 1 2  h photoperiods, 
flower development was totally arrested in the 8 h photoperiod and only 20 percent of 
plants under the 12 h photoperiod reached anthesis. Under 16  h photoperiods, all H. 
patens plants reached anthesis. Arrnitage ( 1 995), therefore, suggested that floral 
development in this species was more sensitive to unfavourable photoperiods than was 
floral initiation. 
36 
Flowering responses at different stages in development can also be mediated by factors 
other than photoperiod. In Hardenbergia violacea, King ( 1 998) reported that there was 
an obligate requirement for both short days ( < 1 2.5 h) and sustained cool temperatures 
( 1 5 - 1 8?C) for inflorescence initiation. Plants rarely initiated inflorescences under long 
days, particularly under warm conditions. However, maintenance of inflorescences 
beyond an intermediate stage in floral development (carpel and anther primordia 
microscopically evident) required a narrow range of temperatures ( 1 5- 1 6?C) in order to 
prevent the rapid ( <30 days) and complete abortion of all inflorescences . Therefore, the 
flowering response in H. violacea was more sensitive to daylength early during floral 
initiation, and to temperature during floral development (King 1 998) .  
Photoperiodic responsive plants, other than Hardenbergia violacea, can also be 
sensitive to temperature during floral development. Photoperiod appears to be the main 
factor affecting floral initiation in Chamelaucium uncinatum, while temperature was the 
primary factor regulating flower development (Shillo et al. 1985, Dawson and King 
1 993) .  By experimenting with both temperature and photoperiod treatments applied in 
factorial combination for up to nine weeks, Shillo et al. ( 1985) showed that plants 
formed flowers after exposure to low day/night temperatures (20/ 14?C),  in both SD (8 
h) and LD ( 1 3- 14h).  However, at high temperatures (24/1 8?C), only plants exposed to 
SD flowered, whereas none flowered under LD conditions. The number of flowers per 
shoot that formed was also positively related to the number of days that plants were 
placed under inductive conditions. Shillo et al. ( 1 985) subsequently classified C. 
uncinatum as a facultative SD plant. 
Although flowering in C. uncinatum occurred after exposure to SD with both high and 
low temperatures, and LD at only low temperatures, the number of flowers produced 
between treatments was significantly different. Shillo et al. ( 1985) observed that 
approximately twice as many flowers per branch were formed in those plants held under 
SD rather than LD conditions. Thus, the application of moderately cool temperatures 
can overcome the effects of a LD treatment in C. uncinatum, despite a reduction in 
flower number. Evidence that decreasing temperatures can progressively nullify the SD 
or LD photoperiod requirements for flowering has been well documented, as in the case 
37 
with Bougainvillea (Hackett and Sachs 1966) and in various herbaceous species (Kinet 
et al. 1 985) .  Therefore, daylength can interact with other environmental variables to 
significantly affect the final number of flowers produced. 
Unfavourable photoperiods during floral initiation or development can cause a decline 
in flower quality (decreased flower number or flower malformations), an issue 
particularly important in horticultural crop production. Shillo et al. ( 1 984) reported that 
flowering in C. uncinatum plants ceased after the first flowering flush when plants were 
maintained continuously under inductive SD conditions .  They attributed this outcome 
to the formation of developing flowers that inhibited subsequent flower initiation. Only 
removal of the flower buds at an early stage enabled the continuous formation of 
flowers under SD conditions (Shillo et al. 1 984) . Unfavourable photoperiods during 
flower formation are also associated with malformations in the inflorescences/flower 
(Kinet et al. 1985). These abnormalities may affect bract, flower or floral primordium 
development indicating that any given floral primordium during development is not 
committed during initiation and their developmental path may remain optional (Kinet et 
al. 1 985) .  In Impatiens balsamina, for example, vegetative growth occurs in 
undifferentiated floral organs when exposed to unfavourable photoperiods 
(Krishnamoorthy and Nanda 1 968). Floral buds have also been reported to abort early 
in development in response to unfavourable photoperiods, as in the case of Boronia 
megastigma when treated to constant LD conditions (Roberts and Menary 1 989).  In 
summary, exposure to inappropriate photoperiods can clearly have deleterious effects 
on flowering. However, the responses exhibited can be species-specific . 
1 .2.6.2 Light: spectral quality 
The quality of light is an important aspect in horticultural crop production, particularly 
under controlled environmental conditions. Relatively few studies, however, have been 
conducted on flowering responses in woody species to different wavelength ratios .  In 
Pinus radiata, Morgan et al. ( 1 983)  described an increase in shoot elongation rates, 
internode length and stem weight in adult and juvenile plants as the proportion of far?
red (FR) to red (R) light increased. In kiwifruit and grape, exposure to light with a high 
FR:R ratio coincided with an increase in petiole elongation or internode length, 
respectively, and with a declining trend in the number of flowering nodes/clusters per 
38 
shoot (Morgan et al. 1 985) .  Perception and translation of the light quality are linked to 
the involvement of phytochrome photoreceptors for controlling vegetative and 
flowering growth (Thomas and Vince-Prue 1 997) . 
The effects of different light qualities (spectral distribution) can also affect growth. 
Maas and Bakx ( 1 997) examined the effect of different spectral qualities applied 
continuously at a PAR of 1 30 ?mol m?2 s? 1 on vegetative growth in the rose hybrid 
'Mercedes' .  After six weeks, they found that shoots in plants grown under amber (85% 
blue light filtered out) and orange-light ( 100% blue light filtered out) were considerably 
longer (increase of 10- 1 6  cm) than those grown under control white light (containing 
400-500 nm blue light) . Plants held in amber and orange light also partitioned relatively 
more dry weight into stem than into leaves in comparison with white-light treated 
plants .  None of the different spectral treatments, however, had an effect on the number 
of flowering shoots per plant. M or and Halevy ( 1 984) suggested that other factors such 
as the ratio of R to FR affected the number of flower buds sprouting and that irradiance 
affected subsequent flower development. 
1 .2.6.3 Light: intensity 
Although the role of light intensity on flower formation has been less studied than that 
of photoperiod, its involvement in the flowering process is equally important for many 
species. Irradiance can play an important role in the flowering process for many 
autonomously flowering species. For instance, floral initiation in Leucospennum, an 
ornamentally-valued day-neutral plant, is induced under high light intensities (0 - 30 
percent shading), although this requirement occurs only in conjunction with a cessation 
of shoot growth and through the release of apical dominance (e.g. removal of the 
primary inflorescence) (Jacobs 1 985) .  Other day-neutral species such as various species 
of Rosa and grape (e.g. Vitis vinifera) are also sensitive to irradiance. For example, 
transference of V. vinifera plants from a low (220 11mol m-2 s? 1 ) to a high irradiance (600 
?mol m?2 s? 1 ) environment, or continuous maintenance (>50 days) under a high 
irradiance, produced the highest yield of flowers per shoot (Morgan et al. 1 985) .  The 
enhancement effect of high irradiance levels on the number of flowers or floral buds 
produced has been reported in numerous other woody species, including ornamentals 
39 
such as Leucospermum (Zieslin 1 985),  Begonia x hiemalis (Myster 1 999), Acacia 
pycnantha (Sedgley 1 985), Pimelea ciliata (Slater et al. 1 994) and Boronia megastigma 
(Roberts and Menary 1989). 
The intensity of light is often important for flower formation considering that relatively 
low or high levels can have deleterious effects (Sedgley 1 985). Low light during 
induction can result in abnormalities in reproductive structures. For instance, a decrease 
in light intensity after flower initiation in Kalanchoe blossfeldiana has been shown to 
cause the formation of inflorescences with vegetative parts (Kinet et al. 1 985) .  Low 
light conditions can also lead to abortion of reproductive structures .  In Boronia and 
Rosa, shady conditions are correlated with a decline in the number of flowers and this is 
suggested to be due, at least in part, to an increase in flower bud abortion (Zieslin and 
Mor 1990, Plummer et al. 1998). Finally, flowering time can also be affected by 
irradiance. For example, low light significantly delayed the time to anthesis in 
Coleonema aspalathoides (Heller et al. 1 994) . 
Relatively few studies have attempted to document the critical stages during floral 
development that are most affected by irradiance (Sedgley 1985, Roberts and Menary 
1 989). Of the many studies investigating the effects of irradiance on flowering in 
woody species, Sedgley ( 1 985)  is one among the few investigators who has employed 
the use of microscopic examinations to detail the independent effects of irradiance (and 
temperature) on floral primordium development. Sedgley ( 1985) found that floral 
initiation in Acacia pycnantha ceased at a very early stage of development when plants 
were grown under low light conditions. Differentiation did not progress beyond the 
initiation of the individual primordia. Temperature also had an effect on floral 
development but at a later stage in development. Floral development in plants grown 
continuously in a heated greenhouse as opposed to those grown outside proceeded 
normally up to the stage of megasporogenesis and microsporogensis before floral buds 
aborted (Sedgley 1 985) .  Thus, both optimum irradiance and temperature levels were 
important in floral development in A .  pycnantha, and extremes in both factors blocked 
different stages in floral development. 
In some circumstances, light intensity may have a deleterious effect on inflorescences 
late in development. Transferring Pimelea ferruginea plants to low light conditions 
40 
($ 1 1 3  J.Lmol m-2 s-1 ) late in development (4 days before anthesis) resulted in 30% of 
inflorescences failing to open (King et al. 1992), whilst in two cultivars of Azalea low 
light conditions delayed anthesis (Bodson 1983). In Leucospermum, reductions in light 
intensity of up to 50% had no effect on the rate of flower development (J acobs and 
Minnaar 1 980). However, flower quality decreased significantly as determined by a 
decline in the number of styles per flower head, decreased receptacle length and 
diameter, and decreased inflorescence dry weight (Jacobs and Minnaar 1 980). 
The effect of different irradiance levels on flowering can also be affected by other 
factors. In particular, irradiance can interact strongly with daylength in 
photoperiodically responsive species. Applications of a high light intensity during LD' s  
can override the photoperiodic signal i n  the SD Bougainvillea plant, resulting in 
flowering (Kinet 1 993). In other instances, the effect of irradiance on flower 
development can be modified by other environmental factors . Dawson and King ( 1 993) 
tested the effects of different levels of irradiance, C02 and daylength on flowering in 
Chamelaucium uncinatum. Their findings showed that the number of flowers present 
was positively correlated with irradiance level with the highest yields occurring at 800 
J.Lmol m-2 s- 1 . A further increase in flower number was observed when C02 was doubled 
at moderate and high irradiances. The high light intensity and C02 effects, however, 
were clearly secondary to a need for an exposure to short days ( 1 0  h). 
Photosynthetic assimilates appear to play a predominant role in controlling flower 
development, and a decrease in their production within the plant appears to inhibit 
inflorescence development and can induce bud abortion (Kinet et a/. 1 985,  Halevy 
1 987). Therefore, treatments that reduce photosynthetic output such as reduced 
irradiance or low ambient C02 concentrations, photosynthetic inhibitor treatments, 
and/or increases in plant density, can subsequently have an impact on floral 
development (Kinet 1 993). 
1 .2.6.4 Temperature 
Maintenance of low temperatures,  varying from a few weeks to several months, plays 
an important role for many species in order to achieve reproductive development. The 
promotion of flowering by exposure to this period of low temperature is known as 
41 
thermoinduction or vemalisation (Kinet 1 993) .  The site of perception of 
thermoinduction can be localised to the shoot apical meristem (Thomas and Vince-Prue 
1 984). Evidence of this has been on excised shoot tips that have been florall y induced 
with low temperatures, an effect that appears to be largely independent of the 
temperature experienced in the rest of a plant (Taiz and Zeiger 1 998) .  However, in 
some species, such as mango or lychee, leaves are the sites of induction from which an 
unidentified floral signal(s) is transferred to the apex (Shu and Sheen 1 987, Nunez?
Elisea et al. 1 993). The age of the leaf is also positively related to its competence for 
floral induction (Nunez-Elisea and Davenport 1995). The effect of low temperature for 
promoting flowering is possibly related to changes in the pattern of DNA methylation 
which can alter gene expression. For example, the late-flowering ecotypes of 
Arabidopsis can flower early if vemalised (Bum et al. 1 993). However, application of a 
demethylating agent (5-azacytidine) to non-vemalised plants caused them to flower 
earlier than untreated control plants. Therefore, genes required for early flowering can 
be inhibited by DNA methylation, at least in late flowering Arabidopsis ecotypes (Bum 
et al. 1993) .  
The response to low temperatures and its duration for floral induction varies according 
to species. For a number of plants species, cool temperatures are obligatory for a 
flowering response. Moncur ( 1 992) showed that Eucalyptus lansdowneana seedlings 
were capable of floral induction only when plants were transferred from a greenhouse 
heated to 24/ 1 9?C (day/night) to a cold regime ( 1 5/ 10?C) for five to 1 0  weeks before 
returning to the former environment. Several other Australian species also show a 
similar response to cool temperatures .  Both Boronia megastigma and Hypocalymma 
angustifolium initiated flowers irrespective of the daylength after transfer to warm 
conditions (25!17?C) (day/night), following a 10 to 1 5  week exposure to a cool 
temperature regime ( 1 7  /9?C) (Day et al. 1 994a). Under constant 25/ 1 7?C, however, no 
flowers reached anthesis in either species. An obligate low temperature response has 
also been described for several other Australian woody species including Ixodia 
achillaeoides (Weiss and Ohana 1 996), and the day-neutral Pimelea ciliata (Slater et al. 
1 994) and P. rosea (King et al. 1 992).  
42 
Several Australasian species also display a facultative response to temperature. For 
instance, flowering in Leptospermum scoparium occurred under both moderate 
(20! 10?C) (day/night) and high temperature (26/20?C) regimes under 8 h short days, 
although flower number was enhanced in the cooler conditions (Zieslin 1 985) .  
Additionally, flowering in  the SD plant Chamelaucium uncinatum 'Purple Pride' 
responded favourably in terms of number of flowers when treated to high temperatures 
(20-25?C), whereas lower temperatures (::::: 1 6?C) resulted in low flower number 
(Dawson and King 1 993) .  In contrast, a C. uncinatum x C. floriferum hybrid produced 
a greater number flowers per plant under the cooler ( 12- 1 5?C) SD regime than did C. 
uncinatum 'Purple Pride' ,  although Dawson and King ( 1 993) did not test this cultivar 
using temperatures above 15?C (daily mean).  
Experimentation on differences in day versus night temperatures (DIF) influence both 
vegetative and floral developmental characteristics. Generally, increasing positive DIF 
(higher day than night temperatures) is associated with an increase in internode length, 
leaf expansion, while negative DIF can decrease plant height (Myster and Moe 1995) .  
DIF effects on internode length and leaf expansion are primarily due to an increase in 
cellular elongation rather than division (Myster and Moe 1995) .  With regards to the 
flowering process, high night temperatures have been reported to be particularly 
detrimental to flower development compared to high day temperatures (Moe 1990). 
Increasing positive DIF has been shown to promote flower stalk elongation in some 
species (Myster and Moe 1995), whereas negative DIF can delay flower development 
(Mortensen 1 994). In the case of Kalanchoe blossfeldiana, negative DIF delayed 
flower opening by two to four days at low C02 levels, although the process could be 
hastened with C02 enrichment (Mortensen 1 994 ). Mortensen ( 1994) attributed the 
delayed flowering response to a depletion in carbohydrates during times of both high 
night temperature and high dark respiration rates, whilst C02 enrichment was suggested 
to counteract this effect .  High C02 concentrations have also been shown to decrease 
dark respiration rates (e.g. in Rumex crispus) which could subsequently improve 
endogenous carbohydrate levels (Amthor et al. 199 1 ) .  
Application of  high temperatures can modify the timing of an thesis by  hastening the 
rate of morphological development of reproductive structures (K.inet et al. 1 985) .  
43 
Temperature response curves on growth can delineate the times at which flowers may 
be expected to reach anthesis. Armitage ( 1 995), for example, demonstrated that 
growing Hamelia patens under a 30?C rather than a 20?C temperature regime 
accelerated the time to visible flower bud by 30 days. Studies describing the hastening 
of anthesis have also been reported in a number of other woody ornamental species 
including Chamelaucium uncinatum (Dawson and King 1993), Heliotropium 
arborescens (Park and Pearson 2000), Pimelea ciliata (Slater et al. 1 994) and several 
cultivars of Hebe (Noack et al. 1996) . 
Extremes in temperature can often have unfavourable effects on the flowering process. 
For many species dependent on low temperatures for flowering, brief or prolonged 
exposure to high temperatures at any stage in development can induce abortion of 
reproductive structures. At an early stage in floral development (during meiosis), 
transferring Acacia plants to warm conditions (max/min: 28/1 6?C) resulted in a 
cessation of floral development before bud abortion occurred (Sedgley 1 985) .  
S imilarly, floral buds of Boronia megastigma underwent flower reversion or otherwise 
aborted but at the stage where carpels had initiated (Day et al. 1994a). In young buds, 
exposure to high temperatures resulted in floral reversion evident by the development of 
flowers with leaf-like sepals at the time of anthesis (Day et al. 1994a). Pimelea 
ferruginea also requires cool temperatures for flowering (King et al. 1 992). However, 
in this species no flower buds aborted as temperature was increased at any stages of 
floral development. In summary, high temperatures can have deleterious effects on 
flower development, although the 'window' of sensitivity, if present, can be species?
specific. 
The ability to induce flowering at low temperature may be limited by factors intrinsic to 
the plant. Eryngium planum, a perennial herbaceous plant from central Europe, has 
been used as a cut flower in Israel for over 20 years . However, efforts to hasten and 
synchronise flowering time have been unsuccessful (Ohana and Weiss 1 998) .  
Treatments such as application of growth retardants showed little to no effect. 
However, an investigation into the effects of low temperatures on the size of the 
initiated bud on root cuttings was related to flowering time and quantity. Large buds 
(diameter > 4 mm) on vemalised roots reached anthesis faster and at a higher 
percentage than smaller buds. Ohana and Weiss ( 1 998) suggested that the ability to 
44 
respond to low inductive temperatures related to the size of the apical meristem, which 
was correlated with bud size. A similar response with regard to bud size was reported in 
cranberry (Vaccinium macrocarpon),  showing that small buds (<1 mm) had a lower 
probability of becoming floral the following summer than did buds > 1 mm in diameter 
(Patten and W ang 1994 ) .  There is strong evidence, therefore, to suggest that the size of 
the bud may be a limiting factor controlling flowering. However, whether this is  a 
common pattern exhibited among other woody plant remains uncertain given the 
paucity of studies investigating this relationship. 
1.2.7 Carbohydrates 
Recently, increasing attention has been given to the role of carbohydrates in flower 
induction (Bernier et al. 1993, Jiao and Grodzinski 1 998, King and Ben-Tal 200 1 ) .  
From the beginning o f  the nineteenth century, the role o f  carbohydrates i n  controlling 
flowering had been debated, resulting in the forging of several hypotheses for 
describi1 ?, ?.,.,.r induction models; Section 1 .2.3) .  Following the 
florigen 
carboh? 
in ducti 
along 
al. 1 9  
LD o 
growth substances, the role of 
iding mechanism once floral 
1ence suggests that carbohydrates 
ative to floral transition (Bernier et 
;! shown that exposure to either one 
at th, 
the ; 
,s to increase rapidly and transiently 
The increased supply of sucrose to 
gy-consuming processes in early 
staf 
Be1 
on 
(a? 
tr 
m) (Bernier et al. 1 993) .  However, 
200 1 )  does not regard sucrose as the 
.he involvement of cytokinins as well 
1 P. Jameson at IPGSA 200 1 )  contends 
.ce role in the floral induction process .  
Further c "  ? --- - - 1crose in the flowering process comes 
from a floral induction study conducteo uu L .A.chsia hybrida (King and Ben-Tal 200 1 ) .  
Their study found a strong positive correlation between flower number and sucrose 
concentration in shoot apices with the level of irradiance applied during induction. 
45 
?Apimi ? 
I Tr l 
: 2 ? : : \+ : Mtbn leaf 
Figure 1 .4 Schematic diagram model illustrating the regulatory loop involvement of 
sucrose and cytokinins in the flowering process in Sinapis Alba. LD perception by 
leaves ( 1 )  leads to mobilisation of starch (2) to stems and leaves and sucrose (2) to root 
and shoot apical regions through the phloem (solid arrow). Zeatin riboside ([9R]Z) and 
isopentenyladenine riboside ( [9R]iP) are transported (dashed arrow) in the xylem from 
the roots to leaves (3) and is followed by transportation (dotted arrow) of 
isopentenyladenine (iP) in leaves via the phloem to the apical meristem (4). These 
hypotheses are supported by experimental manipulation of vascular pathways through 
bark ringing and reduction in transpiration (RH 1 00%) and by chemical analysis of sap 
(Bemier et al. 1 993). 
46 
However, when plants were treated to inductive (2-4 LD) but with low light conditions, 
flowering was induced although no concomitant increase in shoot apex sucrose content 
was observed. S imilar findings have also been reported in the LD plant Lolium 
temulentum whereby flowering occurred following inductive low light conditions 
despite there being no changes in apical sucrose content (King and Evans 199 1  ) . King 
and Ben-Tal (200 1)  conclude that sucrose plays a florgenic role in F. hybrida, although 
it does not act as a stimulus for flowering specific to LD photoperiodic exposures. 
The mobilisation of starch early during the inductive treatment has been suggested to 
play an important role in flower induction. The increased supply of sucrose to the shoot 
apical meristem of induced plants seems to arise from the storage of carbohydrates 
(possibly starch) from stems and leaves, and not necessarily from increased 
photosynthesis (Bernier et al. 1993).  Studies using an Arabidopsis starchless mutant 
(pgm TC75)  and a starch overproducer mutant (sop TC26T) have shown that the 
transport of starch to the apex can affect flowering (Bernier et al. 1 993).  When grown 
in continuous lighted conditions, the mutants responded similarly in both growth and 
flowering time to that of the wild-type. However, as the daylength was decreased, 
growth was reduced and flowering was delayed in the mutants . This response was 
suggested to be the result of an inability to mobilise starch, a feature exhibited by both 
mutants (Bemier et al. 1 993). 
Investigations into the role of carbohydrates in flowering in woody species have centred 
on the changes occurring between shoot/leaf and shoot meristems under various 
environmental conditions (Jiao and Grodzinski 1 998). In Rosa plants, leaves on the 
flowering shoot act as source leaves even before the presence of a visible bud (Mor and 
Halevy 1 979) and subsequently act in transporting assimilates to the developing flower 
bud (Mor and Halevy 1 979, Jiao et al. 1 989). Using 1 4C radioisotope experiments, Jiao 
and Grodzinski ( 1 998) demonstrated that in Rosa various environmental conditions 
could affect photosynthesis and concurrent export rates from the source leaves to the 
developing flower. These authors showed that export rates of carbon (sucrose) were 
greatest at late stages of floral development (e.g. petal colour visible) and this occurred 
under both ambient and enriched C02 levels, being 35 and 90 Pa, respectively (Jiao and 
Grodzinski 1 998) .  However, at higher temperatures (25-40?C), photosynthesis 
decreased by 40 percent and carbon (sucrose) export rates dropped by 80 percent. 
47 
Irradiance also affected sucrose export. Export rates increased significantly under 
higher irradiance levels (> 1 000 ?-tmol m-2 s- I ) due to heightened stimulation of 
photosynthesis which caused the mass flow of carbon into sucrose synthesis (Jiao and 
Grodzinski 1 998) .  Low irradiance levels during induction are generally unfavourable 
for continued flower development, and can subsequently cause flower bud abortion 
(Zieslin and Mor 1 990, Plummer et al. 1 998). Plummer et al. ( 1 998),  for instance ,  
showed that the number of Boronia heterophylla flowers that reached anthesis 
decreased significantly under shaded conditions (less than 39% shade) and this was 
attributed to early abortion due to a limited assimilate supply. Their findings are also 
supported by Richards ( 1 985) who suggested that flower abortion in B. heterophylla can 
occur in response to a diversion of photoassimilates away from the flower bud and 
towards vegetative shoots. In summary, various environmental factors can act 
independently or in combination to affect flowering through changes in carbohydrate 
levels and/or fluxes. 
1.2.8 Hormonal regulation 
In higher plants, plant hormones are intimately involved in the regulation of 
metabolism, growth and morphogenesis and serve as chemical signals within the plant 
(Taiz and Zeiger 1 998) .  Plant development is primarily regulated by the five classic 
types of hormones, being auxins, gibberellins, cytokinins, ethylene and abscisic acid. 
There is compelling evidence for the addition of a sixth group, these being a steroid 
family of plant hormones - the brassinosteroids . However, the vast majority of research 
into the flowering process has specifically focused on the role played by gibberellins, 
and to a lesser extent by cytokinins. 
1.2.8.1 Gibberellins 
The effects of applied or endogenous gibberellins (GAs) on flowering responses have 
been inconsistent across plant groups and between species. Application of exogenous 
GAs (e.g. G?n) has been successful in promoting flowering in conifers (Pharis et al. 
1 986), but gibberellins generally appear ineffective in many SD or woody angiosperrn 
species (Shillo et al. 1 985) .  In many woody angiosperrn plants, responses to growth 
regulators have generally been two-fold, including reports on their effects on timing and 
48 
on the number of flowers initiated. For instance, Day et al. ( 1994b) reported on the 
effects of PP333 on flowering in Hypocalymma angustifolium and Boronia megastigma. 
In comparison with untreated plants (control), PP333 increased the proportion of 
flowers produced in H. angustifolium, but not in B. megastigma. However, with regards 
to days to anthesis, only B. megastigma differed significantly from control plants, 
reaching anthesis five weeks earlier than controls (averaging 3 1  weeks) (Day et al. 
1994b ). In contrast, applications of PP333 to Acacia elegans and A. vestita produced 
significantly fewer inflorescences per plant than the untreated control plants (Parletta 
and Sedgley 1996). However, plant height and width were decreased in all three species 
by this growth retardant, presumably through the effect of PP333 on branching 
frequency and/or a reduction in shoot internode length (Day et al. 1 994b ). Moreover, 
application of G? or 2,2-dimethyl G? stimulated shoot elongation in micropropagated 
plants of Metrosideros collina, whereas high doses of G? reduced flowering on treated 
shoots (Clemens et al. 1 995). 
Reductions in plant stature with growth retardant (e.g. PP333,  chlormequat chloride) 
applications are not surprising given that GAs are associated with promoting internode 
elongation (Oren-Shamir and Nissim-Levi 1 999). However, responses that include (but 
do not exclude) a decline in flower number or differences in flowering time with 
application of growth retardants implicate gibberellins in the flowering process, at least 
in the above species and in various other woody angiosperms, such as Chamelaucium 
(Shillo et al. 1 985) and Eucalyptus (Moncur and Hasan 1 994) . 
In summary, the precise role of GAs in the transition to flowering remains unclear 
although recent molecular approaches to this problem are beginning to unravel the role 
of gibberellin in flowering. In Arabidopsis, GA activates, at least in part, the expression 
of LEAFY (LFY), a floral meristem identity gene, and the activity of the LFY promoter 
can be maximised when GA3 is present with sucrose (Blazquez et al. 1 998).  Therefore, 
this suggests that in some species GAs stimulate flowering through a pathway 
regulating the transcription of the LFY gene. 
49 
1 .2.8.2 Cytokinins 
Another plant hormone group implicated in the flowering process is that of the 
cytokinins. Studies on Sinapis alba suggest that cytokinins (CKs) are involved in the 
flowering transition pathway in response to inductive LD photoperiods (Bernier et al. 
1 993) (see Figure 1 .4). Experiments in which a ring of living tissue (including phloem) 
was excised between the root and lowest stems portion (bark-ringing) of induced S. alba 
plants inhibited flowering when applied during hour eight of the LD, but not during or 
after hour 12 .  This is suggestive of a flowering signal that is translocated from mature 
leaves to the root system (Bernier et al. 1993).  The chemical nature of this signal is not 
known although sucrose is suggested to be a candidate since sucrose levels increase 
rapidly (within 1 h) in the roots at the onset of the LD photoperiod (see Figure 1 .4). In 
addition to sucrose, CKs are also suggested to play a signalling role since a bark-ringing 
treatment at hour eight which inhibited flowering could be reversed by application of 
CKs to the apical meristem at hour 1 6  (Bernier et al. 1 993). From these and other 
experiments, Bernier et al. ( 1993) hypothesised that the transportation of sucrose to the 
roots lead to the export of mostly the zeatin riboside form of CK to shoots and leaves, 
possibly via the xylem. Within the leaves, isopentenyladenine riboside (a  CK) was 
subsequently exported to the shoot apical meristem where increasing levels were 
detected within 1 6  h of the inductive treatment (Bernier et al. 1 993). CK, applied in low 
doses to apical meristems when under inductive conditions, are known to increase the 
rate and synchronisation of cell divisions, vacuole splitting and decrease the size of 
DNA replication units in half (Bernier et al. 1 993).  Bernier (pers. comm. P. Jameson at 
IPGSA 200 1 )  contends that CKs as well as sucrose play an essential role in the signal 
transduction process, although these factors alone can not replace a LD signal . 
Application of exogenous synthetic CKs (e.g. 6N-benzyladenine) has been trialed on 
several ornamental woody plants . In Boronia megastigma, 6N-benzyladinine (BA) was 
applied to plants during transference to an environment with marginally inductive 
temperatures (night/day: 191 1 1 ?C). BA treatments reduced the number of days to 
anthesis to less than half the time ( 1 3  weeks) required in untreated plants (control), 
although comparatively, the number of florally induced buds was significantly 
depressed by the BA treatment (Day et al. 1 994b). Day et al. ( 1994b) acknowledged 
that above-optimal temperatures probably inhibited the effectiveness of B A  for 
50 
flowering, since Boronia grown in cooler optimum conditions showed no inhibitory 
responses with BA applications (Richards 1 985,  Day et al. 1 994b ) . 
Successful floral bud initiation and development has been achieved in B. megastigma in 
vitro under optimal environmental conditions. However, flower bud reversion was 
observed in media containing high levels of CK whilst media without added CK 
prevented anthesis from occurring (Roberts et al. 1993). Synthetic CKs have been used 
on the premise that a stimulation of branching, either by pruning and/or with BA 
applications, can increase the number of growing points for floral initiation (Parletta and 
Sedgley 1 996) . In many cases, an increase in branching is usually achieved, although 
this does not always appear to be associated with an enhancement in final flower 
number (e.g. King et al. 1 992, Dawson and King 1 993). Therefore, evidence exists that 
CKs may play a role in the flowering process in woody angiosperm species. However, 
like GAs, plant responses to exogenous applications have been inconsistent between 
species and can be modified by other environmental treatments. 
1 .2.9 Effect of cultural factors 
A number of studies have utilised physical methods for controlling flowering. 
Treatments involving scoring or applying various forms of girdling (e.g. overlapping 
girdles, wire girdles, scoring) to shoots have been implicated in promoting flowering 
(reviewed by Meilan 1997) . These treatments restrict vascular flow and cause the 
accumulation of shoot produced metabolites (e.g. ABA, auxin and carbohydrates) above 
the restriction zone and below that zone for root-produced metabolites (e.g. gibberellins 
and nitrogen) (Meilan 1 997). Those techniques that manipulate possible carbohydrate 
and/or hormone levels within the plant have been successful in promoting higher crop 
yields in apple (Veinbrants 1 972) and olive (Eris and Barut 1 993), cone formation in 
conifers (Pharis et al. 1 986) and flowering in lychee (Young 1 977) and mango 
(Reboucas and Jose 1 997) . 
5 1  
1.3 Postharvest Physiology of Cut Flowers 
1.3.1 Introduction 
Extensive research has been conducted on the postharvest features of numerous types of 
cut flowers. In particular, the vast majority of these investigations have focused on the 
suitability of herbaceous rather than woody perennial crop species for ornamental 
purposes . Introduction of novel cut flower species with attractive floral displays 
varying in either colour, shape, and/or prominence of flower parts may have commercial 
appeal. Many new woody genera on the market meet these standards and have been 
successfully introduced into the floricultural trade, including Leptospermum (Burge et 
al. 1 996), Chamelaucium (Joyce 1 988) and Verticordia (Joyce and Poole 1 993). The 
commercial success of any new woody crop, however, has been dependent on an 
understanding of the flowering biology of the species in question, with particular 
emphasis on knowledge of postharvest technologies for delaying senescence-related 
symptoms and abscission of flowers or floral organs. In an exhaustive study of the 
effects of ethylene on petal senescence conducted on 93 species (23 families), 
Woltering and van Doom ( 1 988) showed that most plant families, not sensitive to the 
ethylene, were characterised by flowers that underwent wilting as the primary 
symptoms of senescence. Flowers, however, that were sensitive to low exogenous 
concentrations of ethylene displayed abscission of petals as the initial symptoms of 
senescence. Thus, the basis for much of the research on postharvest flowering has been 
on the role of ethylene as well as water relations on senescence and abscission of floral 
organs. 
In this review, the biology of flower senescence and associated floral organs is 
discussed. Secondly, a description is provided of the effect of various components of 
vase solutions on vase-life and flower quality with particular reference to woody 
species. Finally, treatments employed to extend vase-life by delaying senescence via 
inhibition of ethylene biosynthesis or ethylene action are reviewed. The focus of this 
review is primarily concerned with the postharvest physiology of cut flowers of woody 
ornamental plants, although this would be incomplete without reference to the 
52 
numerous fundamental and applied studies conducted on herbaceous plant species that 
have contributed to our understanding in this field. 
1 .3.2 Flower senescence 
Broadly defined, senescence refers to the combination of events that leads to the death 
of cells, tissues or organs (Reid and Wu 1992). The senescence process involves the 
occurrence of a series of highly coordinated physiological, biochemical and ultra?
structural changes in cells, typified by increases in hydrolytic enyzmatic activity, 
degradation in starch and chlorophyll ,  disintegration of cellular compartments, and 
climacteric surges in respiration (Van Altvorst and Bovy 1 995) .  These events are 
associated with changes in gene expression and protein synthesis (Van Altvorst and 
Bovy 1 995). In many plant species, ethylene regulates these processes during flower 
senescence. Generally, flower senescence is independent of flower morphology so that 
closely related species may exhibit similar senescent symptoms despite their differing 
floral morphology (Woltering and V an Doom 1988) .  
1 .3.2.1 Petals 
Characteristic changes in petal structure have been described for countless species 
during the senescence process. Depending on the species and conditions, these 
responses can include symptoms of petal wilting, inrolling or folding, withering, fading, 
discoloration (pigment breakdown) and/or organ abscission (Reid and Wu 1992, Van 
Doom 1 997, Van Doom and Stead 1 997) . Symptoms of abscission-related senescence 
are generally  restricted to dicotyledonous plants, whereas petal senescence in 
monocotyledons are typically displayed by symptoms of wilting or withering 
(McKenzie and Lovell 1 992). Moreover, the similarity in the types of petal senescence 
responses can be expressed at a family level. For example, flowers featuring the 
abscission of turgid petals have been reported for many woody species from families 
such as Fagaceae, Malvaceae, Myrtaceae, Proteaeae, Rosaceae, Rutaceae and 
Solananceae (Woltering and Van Doom 1 988,  Van Doom and Stead 1 997). In some 
families (e.g. Ericaceae and Primulaceae), turgid petal abscission occurs in some species 
and petal wilting, either preceding abscission or not at all ,  in others (Van Doom and 
Stead 1 997). Abscission of petals can occur following various physiological events . In 
53 
some species, petal shedding can occur in response to cell wall dissolution situated at 
abscission zones, whereas in others, this action can occur as a result of fruit growth 
(Van Doom and Stead 1 997) . Pollination or fertilization can also hasten petal 
abscission, as observed in Pelargonium x hortorum (Hilioti et al. 2000). 
1 .3.2.2 Sepals 
While petals are typically regarded as the most prominent part of cut flowers, other 
floral structures such as sepals have ornamental value. This include species in the 
genera Ceratopetalum (Johnson and Ronowicz 2000) and Eucalyptus (Sun et al. 2001 ) . 
In some Eucalyptus species, the sepals remain united at the top of the flower forming a 
cap (calyptra), derived from either the sepals, petals or both, which abscises upon flower 
opening (Moncur and Boland 1 989). In most plant families, sepals do not undergo 
abscission but rather dessicate while remaining attached to the plant, as in the case of 
species within the Rutaceae and Rosaceae families (Van Doom and Stead 1 997). In a 
few exceptions, such as species within Scrophulariaceae, sepals grow after fertilisation 
and eventually cover the ovary (Endress 1 994). Sepals may also form part of the fruit 
during development, as in the case of species within Malvaceae (Endress 1 994 ) .  
1 .3.2.3 Stamens 
Relatively little attention has been given to stamen senescence, probably because other 
floral structures, specifically petals or flowers with fused petals and sepals, are typically 
the focus of postharvest flower senescence studies. Despite this, several woody plant 
families (e.g. Myrtaceae and Proteacae) contain species with flowers that contain 
relatively prominent and attractive stamens with commercial appeal . Specifically, these 
include several woody myrtaceous plants within the genus Metrosideros which are 
characterised by having flowers (borne in cymules subtended by a pedicel, on a 
compound inflorescence) that contain numerous (- 25), unfused stamens (Orlovich et 
al. 1 996,  Sun et al. 2000). Other ornamental genera with prominent stamens include 
Verticordia (Joyce and Poole 1 993) and Syzygium (Payne 1 997). According to Van 
Doom and Stead ( 1997), there are various dicotyledonous families that feature 
particular forms of stamen senescence. These include: (a) abscission of turgid stamens 
before petal abscission (e.g. in Balsarninaceae ) , (b) simultaneous abscission of both 
54 
wilted stamens and petals (e.g. in Malvaceae and Solanaceae), (c) abscission of 
desiccated stamens with or after petal fall (e.g. in Ranunculaeae), and (d) abscission of 
turgid or partly wilted stamens with or after petal drop (e.g. in Rutaceae and 
Scrophulariaceae) . Metrosideros flowers typically follow the latter pattern of abscission 
with turgid stamens shedding within three to four days of becoming fully expanded 
(Sun et al. 2000). 
1 .3.2.4 Style 
Depending on the species or family, various patterns in senescence of the style can 
occur. In families such as Rutaceae or Rosaceae, abscission typically occurs when 
styles are fully turgid. In contrast, style abscission usually occurs after wilting such as 
in Eucalyptus species (Myrtaeae) (Moncur and Boland 1989). In both the Boraginaceae 
and Onagraceae family, styles desiccate before abscising. It is unclear, however, as to 
whether style shedding in both these families is due to abscission or to mechanical 
tearing from the developing fruit (Van Doom and Stead 1997). 
1.3.3 Water relations 
The limiting factor regulating the vase-life in many cut flowers is water stress.  Intact 
flowers undergo senescence either by colour changes (Jones et al. 1 993a) ,  flower 
closure, petal wilting/in-rolling (Zieslin and Gottesman 1983, J ones et al. 1 993a) or 
abscission of floral parts (Joyce 1 988,  Joyce and Poole 1993) . However, when flowers 
are cut and placed in water, symptoms of senescence, as exhibited in intact flowers, are 
not always observed but rather those of water stress, such as premature wilting of flower 
parts or leaves (Burge et al. 1996, Wirthensohn et al. 1996). At a cellular level, growth 
of floral organs (e.g. petals) is due to an increase in cell number and cell size. A 
balanced water status plays a role in maintaining cell turgor, even when petals have 
fully expanded. However, once petals are fully expanded, the duration that cells remain 
turgid can, depending on the species, vary from a few hours to several months (Stead 
and V an Doom 1 994 ). In flowers where turgid petal abscission is not the first symptom 
of senescence, partial or full wilting or withering can occur. This is usually preceded by 
a loss of solutes (e.g. inorganic ions, organic acids, reducing sugars, amino acids and 
anthocyanins) that result in cell leakiness (Stead and Moore 1 983,  Van Doom 1997). 
55 
The direct cause of this leakiness is not known but is suggested to be related to a loss of 
semi-permaneability of the tonoplast and the plasma membrane (Van Doom 1 997). 
Thus, adverse water relations may inhibit growth and maintenance of turgidity in floral 
organs and subsequently accelerate premature senescence. 
1 .3.4 Water uptake 
The pattern in water uptake following cutting of flowers can vary depending on the 
taxa. In many species, the rate of water uptake may initially be high due to a low water 
potential at cutting (Van Doom 1997) . Subsequently, water uptake may either stabilise 
or decline rapidly, as in the case of many cut woody species . This includes genera such 
as Rosa, Anigozanthos (Kangaroo paw), Banksia, Grevillea, Thryptomene, 
Leptospermum, Chamelaucium and Telopea (Mayak et al. 1 974, Faragher 1989). In 
contrast, water uptake in other species such as Heliconia is low even after the stem is 
freshly cut and placed in water (Ka-Ipo et al. 1 989). The rate of water uptake can also 
depend on other factors, other than species type, such as on the transpirational pull of 
the flower, the temperature and composition of the holding solution, and factors innate 
to the flower stem that may prevent water uptake. 
1.3.5 Transpiration 
The extent of water stress within a cut flower is dependent upon water balance, and a 
water deficit can occur when the rate of water uptake is less than the rate of 
transpiration. Therefore, water stress can be alleviated or delayed by reducing the rate 
of transpiration. Stomatal transpiration can be decreased through the removal of leaves 
since cut (flower and foliage) stems with relatively small leaf areas lose less water per 
unit stem area and time than those with relatively large leaf areas . Stomata per se are 
usually present in all green epidermal tissues, such as leaves and can also be present in 
stem and in epidermal tissue of non-green flower parts (e.g. petals) .  Stomata have also 
been reported to be present on stamens and nectaries, although in the latter case, they 
are suggested to function in nectar exudation rather than in gas exchange (Zer and Fahn 
1 992). 
56 
The addition of plant hormones to vase solutions can subsequently affect transpiration. 
The addition of exogenous abscisic acid (ABA), at concentrations of 1 0  ppm or more, 
effectively preserved the water balance in Chamelaucium uncinatum and extended vase?
life up to 1 0.6  days compared to 6.4 days for control stems (Joyce and Jones 1 992). 
Similar results have also been obtained in cut roses but only when applied in solution 
(Halevy and M a yak 1 98 1 ) and not as a foliage spray (Muller et al. 1 999) . The 
effectiveness of ABA in extending vase-life is attributed to its role in inducing stomatal 
closure, thus ,  inhibiting transpirational water loss (Aspinall 1980). The relatively high 
cost of ABA, however, may not be conducive for commercial use (Joyce and Jones 
1 992). Transpiration has been reported to decrease when solutes are added to the vase 
solution. V an Doom ( 1 997) suggested that sugars can promote bacterial growth in 
vascular vessels, which can lead to a water stress response causing stomatal closure. 
1.3.6 Water temperature 
Water temperature has been of considerable importance particularly for dry-stored 
stems. In some species, the rate of rehydration is correlated with water temperature, 
hence, warm water (35-40?C) has been recommended for use with many commercial 
cut flowers, such as Freesia, Gladiolus, Lilium and Protea and Syringa (Sacalis 1 993) .  
The mode of action is  not fully understood but may relate to the decreased viscosity of 
water with increasing temperature. Cut roses, on the other hand, rehydrate more rapidly 
in cold (2?C) water in comparison to water at 23?C (Durkin 1 979) . The increased 
solubility of gases in cooler water may also contribute to improved water uptake (V an 
Doom 1 997).  
In freshly cut flowers, dipping of stem ends in hot water has been implicated for 
improving water relations (R. Lill pers. comm. with Burge et al. 1 996).  However, 
treatments, in which stem ends of Leptospermum scoparium were dipped in 60?C water 
for 1 minute, resulted in little to no effect on water uptake or vase life (Burge et al. 
1 996) . 
57 
1 .3.7 Vascular occlusions 
A decrease in water uptake and subsequent turgor may be attributed to vascular 
occlusions. These occlusions may develop via several means: through ( 1 )  gas (air) 
uptake, (2) a plant response mechanism or (3) microbial growth. The formation of air 
bubbles in cut stems may impede water uptake. This usually occurs from air that has 
aspired directly after cutting or from vascular cavitation (Williamson and Milbum 
1 995). Re-cutting of stems under water is an effective method for preventing air 
vascular blockages and promoting water uptake (V an Doom 1997), and is  currently a 
standard practice used in the floricultural industry. 
Plant-specific responses to cutting may affect water uptake. Development of vascular 
occlusions as a result of wounding (cutting) or as a defensive mechanism to wounding 
can lead to the deposition of material in xylem conduits, such as suberin, lignin, tannin 
or various gums (Van Doom 1 997) . Additionally, cutting may lead to exudation-related 
blockages (e.g. resin or latex) or to the formation of tyloses (cell outgrowths) that 
inhibit water flow (Zimmerman 1 983) .  The addition of ethanol ( 1  %) in holding 
solutions containing detached flowering peach (Prunus spp.) stems increased xylem 
hydraulic conductivity by reducing the number of plugged vessels and delayed a surge 
in ethylene production (Munoz et al. 1 982). This may be an effective treatment in plant 
families where exudate deposition occurs following cutting (e.g. Proteaceae and 
Rutaceae) (Chattaway 1 948). 
There is strong evidence that the accumulation of microbes in cut stems can lead to 
vascular occlusions. Microbial organisms typically found in vase solutions and at the 
base of stem cuttings include yeast, filamentous fungi, bacteria and their degradation 
products (Put and Clerkx 1988 ,  V an Doom 1997) .  Unless vase water changes are 
conducted regularly, microbial growth can increase rapidly over time. For example, 
bacterial counts (cfu rnl-1 )  on vase water containing cut stems of Thryptomene calycina 
increased 3x 104 fold over a 72 hour period (Jones et al. 1993a). Further, microscopic 
examinations (SEM) of xylem vessels in cut flowers of Gerbera and Rosa (at 24 hours 
vase life) showed the adhesion of microbes primarily on the cut stem surfaces causing 
blockages of xylem vessels (Put and Clerkx 1 988) .  The extent of microbial infiltration 
58 
is also dependent on the number, shape and size of microbes found in the vase water 
and on the width of xylem vessels .  Eventual microbial blockages in the xylem can lead 
to water stress and to premature wilting of flowers (Put and Clerkx 1 988) .  
1.3.8 Biocides 
Various compounds have been developed to suppress microbial growth in vase 
solutions. Most compounds when used at concentrations that adequately control 
microbial growth are toxic to cut flowers (Van Doom 1 997) . With exception, a 
considerable number of compounds which have been shown to be non-toxic to cut 
flowers and have been associated with improved vase-life include salts of various 
elements, chlorine/chlorinated aromatics and ammonium or quinoline based compounds 
(Van Doom 1 997). Quinoline based compounds such as 8-hydroxyquinoline citrate 
(HQC) or 8-hydroxyquinoline sulfate (HQS) have been incorporated into a number of 
studies examining vase-life of Australian and Pan-Pacific ornamental plants. These 
include studies on cut foliage from Eucalyptus spp. (Joyce et al. 1 993) and cut flowers 
from Leptospermum scoparium (Burge et al. 1 996), Chamelaucium uncinatum (Joyce 
and Jones 1 992), Metrosideros collina (Sun et al. 2000) and Eucalyptus ficifolia (Sun et 
al. 200 1 ) . In cut flowers of E.ficifolia, HQC concentration (maximum 400 ppm) was 
correlated with decreased flower mass, as recorded on day five after harvest (Sun et al. 
200 1 ) . In cut flowers of L. scoparium, slightly lower concentrations of quinoline were 
equally effective. The use of 200 ppm of HQS effectively delayed the gradual decline 
in water uptake and leaf moisture content and extended the vase life by two days (Burge 
et al. 1 996). Similar concentrations have also been reported to be effective in other 
myrtaceous species, for example, M. collina (Sun et al. 2000) and C. uncinatum (Joyce 
and Jones 1 992). 
1 .3.9 pH of vase solution 
The effectiveness of pH on water uptake has been evaluated in only a few cut flower 
species. Solutions with a pH below 7 clearly improved water uptake in Eucalyptus 
ficifolia, since flower mass was highest at Day 5 when solutions were maintained at pH 
4.3  (Sun et al. 200 1 ). In Metrosideros collina, the effects of a factorial combination of 
sucrose and pH (2.7-7 .0) of solution were tested on the stamen quality. A sucrose (30-
59 
40 g r 1 ) and HQC solution (pH 5)  produced the least amount of stamen wilting, whereas 
in treatments with the highest stamen wilting (pH 3), stamen abscission was the lowest 
(Sun et al. 2000). Increased solution acidity is  associated with a decline in microbial 
growth, which may improve water uptake via a reduction in stem blockages (Sacalis 
1 993, Jones et al.  1 993a) . 
1 .3.10 Carbohydrates 
The carbohydrate content of a vase solution is an important factor for controlling the 
vase-life of many cut species. Sugars, such as sucrose, provide a source of carbon by 
substituting for the natural depletion of carbohydrates during the life of a cut flower 
(Marousky 1972, V an Doom 1 997) . The addition of sucrose to vase solutions applied 
continuously or as a pulse have been effective in extending the vase-life of a number of 
woody species, such as Rosa spp. (Ichimura et al. 1 999), Chamelaucium uncinatum 
(Joyce 1 988), Leptospermum scoparium (Burge et al. 1 996), Metrosideros collina (Sun 
et al. 2000), Grevillea (Joyce and Beal 1999) and Eucalyptus foliage crops (Jones et al. 
1 993b, Wirthensohn et al. 1 996). In some other species, however, sucrose treatments 
(0.5-5%) did not enhance the quality or longevity of cut flowers, such as in Banksia 
(Sedgley 1 998) or Eucalyptus tetragona and E. youngiana (Delaporte et al. 2000). 
Additionally, high sucrose concentrations (? 1 0%) can damage tissue as exhibited by the 
browning of leaf margins in cut eucalypt foliage (Jones and Sedgley 1 993) .  Generally, 
sucrose concentrations of 1 -2% in vase solutions were effective for extending the vase?
life in several other myrtaceous cut flowers, for example in C. uncinatum (Joyce 1 988),  
Thryptomene calycina (J ones et al .  1 993a), M. collina (Sun et al. 2000) and E. ficifolia 
(Sun et al. 200 1 ). Typically, biocides are added with sucrose in vase solutions in order 
to prevent the rapid accumulation of micro-organisms that lead to occlusions and 
subsequent premature flower senescence (Van Doom 1997). 
1 .3. 1 1  Ethylene 
Plant growth substances play a critical role in regulating different events in the life cycle 
of plants. Ethylene is renowned for its role in initiating and regulating the processes 
that eventually lead to programmed cell death in whole plants or specific organs or 
tissues (Bleecker and Kende 2000).  Increased endogenous production of this gas can 
60 
occur in response to physical wounding or physiological stress (e.g. flooding, chilling 
disease, temperature and drought) and during normal developmental processes, such as 
in leaf or flower abscission and senescence (Reid and Wu 1 992). The role of ethylene 
in affecting flower abscission or senescence processes has for many years been of 
considerable interest to postharvest physiologists researching ornamental crops . This i s  
because ethylene, which is not only a potential hazard as  a pollutant, is involved in 
causing damage and premature senescence in  both leaf and floral organs (Nell 1 994). 
1 .3.12 Background on ethylene 
To date, a considerable amount of research has been undertaken on ethylene gas and its 
role in plant growth and development. This has included numerous studies that have 
investigated the processes incurred during in vivo ethylene evolution, production and 
perception, through to its deleterious effects on plant organs (e.g. flower abscission). 
From a postharvest perspective, a thorough understanding of ethylene biosynthesis, 
mode of action and consequential effects on specific ornamental crops are prerequisites 
for understanding the factors that either inhibit ethylene production and/or action. 
1.3.13 Ethylene biosynthetic pathway 
Knowledge of the ethylene biosynthetic pathway has been reviewed in various reports 
(Yang and Dong 1 993 , Van Altvorst and Bovy 1 995, Bleecker and Kende 2000) . Here, 
a summary is presented of the processes that occur during ethylene biosynthesis 
(Figure 1 .5 A). In the first step in the pathway, the enzyme S-adenosylmethionine 
synthetase catalyses the conversion of methionine and A TP to S-adenosylmethionine 
(AdoMet). AdoMet is subsequently converted by 1 -arnino cyclopropane- 1 -carboxylic 
acid synthase (ACC synthase) to form 5' -methylthioadenosine (MT A) and 1 -arnino 
cyclopropane- 1 -carboxylic acid (ACC), the immediate precursor to ethylene. Recycling 
of MT A to methionine occurs, thus allowing high production rates of ethylene even if 
methionine concentrations are low. Oxidation of ACC by the enzyme ACC oxidase 
results in the production of C02, HCN and ethylene (Bleecker and Kende 2000). It is 
suggested that hydrogen cyanide detoxification occurs through conversion of /3-
cyanoalanine by the enzyme {3-cyanoalanine synthase (Van Altvorst and Bovy 1 995) .  
6 1  
0\ 
N 
A. 
Methionine 
JJ:doMct synthetase 
ADO MET 
Methionine cycle r<::::. n /\CC S!Ji t / lwse -().Se ?V  (1.5Je' 
{ 4?(1. t;_1t o,J l,?t.?)'/ ('f-. -1(' ?Jsr? 0 
M-ACC ? 
MTA cC \(0-\ f'.' ?  + ? o.'i\Y ? c?w-? 
 I
J ?G-ACC 
/\CC m.idase 
Ethylene 
B.  
E thylene  ( C? I I t ) 
cl1 
i ?:TR I I :TR2 1 ?: 1ZS I I ?: I<S1 UN4 -=;::---_r;:;:n;=-famil?? ---, !\H'I t'n! ll)lt\! !Ct 
j} r?-
cTR l -?-------' 
r ___ n _ _  :] ?,;,:: ;?.?::?.] 
1: 1 N2""'-'---===== 
j} 
l<q:.ulakr,. 
F IN3 E I L I I - J L2?--
?
l 
- n ? I ' J{ I - 1 ? j} 
? 
D 
Responses 
Figure 1 .5 Models of ethylene biosynthesis (A) and ethylene signal transduction pathway (B). A: Schematic representation of the major pathway 
of ethylene biosynthesis in higher plants and the enzymes involved. Shown also are the paths to the methionine cycle and conjugation reactions 
that inhibit ACC from being directly available for ethylene production (Fiuhr and Mattoo 1 996) . B :  Schematic representation of the genetic 
interactions and biochemical identities of components in the ethylene signal transduction pathway. Negative regu lation of the ETR-family of 
membrane-associated receptors by ethylene interacts with the Raf-like kinase CTR l regulatory domain, which in turn negatively regulates the 
EIN2 membrane protein. The cytoplasmic EIN2 domain positively signals down-stream to the EIN3 family (nucleus) which subsequently 
promotes the ERF l gene followed by a down-stream activation of a subset of ethylene responses (modified from B leecker and Kende 2000) . 
1.3.13 S ignal transduction 
An ethylene response is suggested to occur following the binding of ethylene to a 
specific ethylene receptor. An activation signal is subsequently transcribed along a 
signal transduction pathway. Most of the recent research has been based on studies 
using etiolated mutants of Arabidopsis thaliana that do not display the 'triple response' 
characteristics .  Thus far, three types of mutants have been identified that ( 1 )  are 
insensitive to ethylene, (2) display an ethylene response in the absence of the hormone 
or (3) are affected only in specific tissues (Van Altvorst and Bovy 1 995).  
To date, a thorough understanding of ethylene signalling in Arabidopsis is  still 
incomplete, although a conceptual model of the interactions is starting to emerge 
(Bleecker and Kende 2000, Hall et al. 2000, Hirayama and Alonso 2000) (Figure 1 .5 
B) .  Based on these recent publications, a summary of the events occurring along the 
ethylene signal transduction pathway is described. The binding of ethylene occurs at a 
receptor site known as a two-component-receptor that is membrane-localised. In the 
two-component-receptor, the N-terrninal domain is responsible for binding with 
ethylene (sensor) while the intracellular portion of the protein receptor is a protein 
kinase (response regulator) which is activated upon the binding of ethylene. This is  
includes a family of five membrane associated receptors (ETR 1 ,  ETR2, ERS 1 ,  ERS2 
and EIN4) . The protein kinase transmitter domains interact with the regulatory domain 
of Raf-like kinase CTR 1 which, in turn, negatively regulates EIN2, a membrane protein. 
Downstream positive signalling by the cytoplasmic C-terrninal domain of EIN2 occurs 
on the nucleus bound family of transcription factors (e.g. EIN3) . A member of a second 
family of transcription factors, ERF1 ,  can subsequently activate a subset of ethylene 
responses. This can entail an increase in rnRNA transcript levels of numerous genes 
including those that encode for cellulase, chitase, ?- 1  ,3-glucanase, peroxidase, chalcone 
synthase, pathogenesis-related protein, in addition to ripening-related genes and 
ethylene biosynthesis genes (Bleecker and Kende 2000). 
1.3.14 Physiology of abscission zones and ethylene 
Prior to abscission of floral organs, abscission zones are reported to undergo various 
anatomical and physiological changes. Anatomically, variations in the size of cells 
63 
located at abscission zones have been observed. For instance, Van Doom and Stead 
( 1 997) classified the size of abscission-zone cells as either those that were ( 1 )  smaller 
than adjacent cells but mostly isodiametric ,  (2) smaller than adjacent cells but mostly 
oblong, or otherwise, (3) of similar size to adjacent cells. The role of changing cell size 
with respect to the abscission process,  however, remains unclear, although it has been 
suggested to assist in breaking of vascular strands (Sexton and Roberts 1 982). 
Additionally, the process of cell swelling at the abscission zone is apparently not 
present in some species, such as in tobacco, tomato and Phaseolus flowers (Jensen and 
Valdovinos 1 968, Webster and Chiu 1 975).  
More consistent reports on the abscission process refer to an increase in intracellular 
activities of cells at abscission zones (Oberholster et al. 1 99 1 ,  Evensen et al. 1993) .  
Ultrastructural data suggest an increase in protein synthesis and high secretory activity 
of material towards cell walls in abscission-zone cells, indicative of increased enzymatic 
activity required for cell wall separation (Van Doom and Stead 1997) .  This is 
evidenced by the appearance of numerous rough endoplasmatic reticulum profiles and 
ribosomes occurring at an early stage of the abscission process. In Pelargonium x 
hortorum, for example, protein levels doubled before the occurrence of petal abscission 
in conjunction with newly synthesised proteins (Evensen et al. 1993).  The enzymes 
most commonly associated with the abscission process include cellulases and 
pectinases,  and their regulation, at least in part, is suggested to be the result of ethylene 
(Brown 1 997). Ethylene regulation is through the transcription of genes that encode 
these enzymes. Subsequently, abscission zone cells undergo enzymatic solubilisation of 
their middle lamella and degradation of the primary cell wall ,  which is often followed 
by the partial or complete autolysis of cell contents (Van Doom and Stead 1 997) . 
1.3.15 Ethylene and flower abscission 
The role of ethylene in plant organ abscission is well recognised. Exposure to 
exogenous ethylene has been shown to hasten inflorescence or floral organ abscission in 
a number of woody ornamental plants. These include a number of species within the 
family Myrtaceae, such as Verticordia nitens (Joyce and Poole 1 993), Chamelaucium 
uncinatum (Joyce 1 989), Metrosideros collina (Sun et al. 2000), Leptospermum 
petersonii and various other Australian cut flowers (Macnish et al. 2000). Other woody 
64 
ornamental genera also include Grevillea '(I oyce et al. 1995) and Camellia (Woolf et al. 
1 995, 1 999) .  
The abscission of  floral organs via exogenous ethylene applications appears to be 
regulated by various factors, including the concentration and duration of ethylene 
exposure, the timing of endogenous production, the sensitivity at various developmental 
stages (Zieslin and Gottesman 1983,  Mor 1987, Wu et al. 1 99 1 )  and species lineage 
(e.g. family level) (Woltering and Van Doom 1988) .  Metrosideros collina exemplifies 
that the abscission response of ethylene-sensitive flowers can depend on the ethylene 
concentration used and on the duration of the exposure. Petal and stamen abscission in 
this species occurred over a seven-day period following exposure to low levels (0. 1 
?1 r1 ) of exogenously applied ethylene. Concentrations ?0.5 !J.l r1 of ethylene, 
however, resulted in abscission of greater than 80 percent of floral buds (receptacles) by 
day five (Sun et al. 2000). In another myrtaceous species, Leptospermum scoparium, 
the concentration of exogenous ethylene treatments was positively related to the 
proportion of flower abscissions (Zieslin and Gottesman 1 983). A 30 h exposure to 1 00 
ppm ethylene was required to cause half of all floral abscissions on shoots of L. 
scoparium, whereas a further 1 8  hour exposure increased the proportion of abscissions 
to 80 percent (Zieslin and Gottesman 1983).  
1.3.16 Ethylene production 
The rate of endogenous ethylene production usually increases before flower abscission 
(Reid and Wu 1992). The time course for ethylene production as observed in carnation, 
rose and other flowers typically follows a profile composed of three distinct phases, 
including ( 1 )  a low steady state, (2) an increased rise to a maximum and (3) a decline in 
ethylene production (Mor 1 987). Each phase can be associated with various events 
during senescence. The onset of the second phase, for instance, is usually associated 
with terminal stage of senescence while visual symptoms of the effects of ethylene (e.g. 
flower abscission) appear at the end of that phase (Mor 1 987, Van Doom and Stead 
1 997) . For example, Mayak et al. ( 1 972) reported that endogenous ethylene production 
by rose flowers peaked two days before the onset of petal abscission. 
65 
The start of the second phase in ethylene production can also be associated with a rise in 
respiration (known as climacteric) and subsequent C02 evolution (Taiz and Zeiger 
1 998).  The climacteric pattern of ethylene production during senescence has been 
particularly well described for carnation (Wu et al. 199 1 )  and Petunia (Borochov et al. 
1 997). In the carnation cultivar 'Crowley Pink' , the increasing ethylene production 
during the senescence process also serves in accelerating the rate of senescence itself 
(Wu et al. 1 99 1 ) .  This can be expressed as an autocatalytic response, whereby the 
triggering of tissue to produce ethylene is catalysed by previously produced ethylene. 
In non-climacteric species, such as in excised Grevillea flowers, ethylene production 
increases over time, peaking at day five although respiration levels remained relatively 
constant. Other non-climacteric flowers include Geraldton waxflower (Joyce et al. 
1 993, Olley et al. 1 996) and ethylene-insensitive species, such as Sandersonia (Eason 
and DeVre 1 995) and daylily (Bieleski and Reid 1992) . 
The source of ethylene production within a flower appears to be initiated in specific 
floral organs. Pollination in Lathyrus odoratus (sweet pea) flowers serves as a catalyst 
for ethylene production, although the level of this production varies between floral 
organs. Combined, both the style and stigma produce more ethylene than the petals ,  
ovary and receptacle collectively, thus indicating that ethylene production first increases 
in the stigma and style before extending to various other flower parts (Singh and Moore 
1 994) . More recent findings support this hypothesis. Hilioti et al. (2000) proposed that 
rapid petal abscission occurred in Pelargonium x hortorum during increased post?
pollination ethylene production via the activation of ACC synthase primarily in the 
stigma and style. Applications of aminoethoxyvinylglycine (A VG), an inhibitor of 
ACC synthase activity, to the stigma effectively prevented pollination-induced ethylene 
production and petal abscission (Hilioti et al. 2000) . Post-transcriptional regulation of 
ACC synthase is suggested to occur in response to other factors besides pollination, 
including various types of stress (Felix et al. 1 99 1 ,  Mathooko et al. 1 998) and after 
cytokinin treatment (Vogel et al. 1998) .  
1.3.17 Ethylene sensitivity 
Species-specific differences in responses to ethylene have been broadly examined. 
Weltering ( 1 987), for instance, conducted a large study to compare the relative effects 
66 
of exogenous ethylene on intact flowers in various ornamental pot plants. Leaves on 
plants marketed as foliage crops exhibited less sensitivity to exogenous ethylene (0- 1 5  
!1-1 r' for 24-72 hours) than did floral organs on flowering plants. Flower and floral bud 
abscissions were frequently observed at relatively low ethylene concentrations or during 
the short exposure time. For example, Hibiscus plants, classified as 'extremely 
sensitive' to ethylene, incurred flower and floral bud abscission following a short (24 
hour) exposure to low ( <4 !1-1 r ' ) concentrations of exogenous ethylene. The 
significance of this study was highlighted by the establishment of a classification 
scheme for a number of popular omamentals, thus, enabling comparison of plant 
responses from other studies (e.g. H?yer 1 996a, 1 996b; Sun et al. 2000). This study 
also broadly described within-plant differential responses of leaf and floral organs to 
various ethylene treatments . Woolf et al. ( 1 999) suggested that the terminology 
defining the relative effects of ethylene on abscission rates (e.g. time to 50%) required 
clarification. They proposed that the term "responsiveness" be defined as differences in 
abscission response to a high ethylene concentration (e .g. 10 1-11 r' ) ,  whereas differences 
in abscission response to low ethylene concentrations (< 1 !-!1 1" 1 ) be defined as 
"sensi ti vi ty" differences. 
Not only does the concentration of ethylene or exposure time cause a response in 
flowers, but rather the combination of both these factors and the sensitivity of the plant 
(or plant part) to ethylene have been considered in a number of reports (Moe and Srnith?
Eriksen 1 986, Evensen 199 1 ,  H?yer 1 996b) . For flowering plants in general, there is a 
gradual increase in ethylene sensitivity with increasing age, and thus, older flowers 
usually require relatively lower concentrations of ethylene to induce flower abscission 
and senescence (Zieslin and Gottesman 1 983, H?yer 1996b). This response has been 
observed in flowers of carnation (Camprubi and Nichols 1978), Hibiscus (Woodson et 
al. 1 985),  Begonia x cheimantha (Moe and Srnith-Eriksen 1986), Pelargonium x 
domesticum (Evensen 1 99 1 ) and Camellia (Woolf et al. 1995, 1 999). Woolf et al. 
( 1 995), for example, demonstrated that different organs in Camellia (leaves, floral and 
vegetative buds) displayed differential sensitivities to exogenous ethylene, and this 
effect could be mediated by temperature. Generally, floral buds rather than vegetative 
tissue were more sensitive to ethylene. S imilar responses were also observed following 
treatment with ethephon, an ethylene-releasing compound (Woolf et al. 1 992) . The 
67 
effect of exogenous ethylene on plant organs of Camellia was also dependent on the 
physiological age of that tissue, showing increased sensitivity and responsiveness with 
age (Woolf et al. 1 999). 
Increased floral sensitivity has been reported following pollination or fertilisation. 
Porat et al. ( 1 994) showed that increased flower sensitivity to ethylene (occurring after 
four hours) appeared to be independent of endogenous ethylene production, at least 
following pollination in Phalaenopsis hybrids (orchids). Ethylene production was 
detected only 1 2  hours after pollination and reached its peak after 30 hours. Similarly, 
petunia flowers also increased in ethylene sensitivity by at least seven hours after 
pollination, while ethylene production started a few hours later (Halevy and Whitehead 
1989). 
The biological explanation to account for differences in ethylene sensitivity in the life?
cycle of a flower remains, as yet, unclear. Brown et al. ( 1 986), however, suggested that 
ethylene sensitivity was not related to an increased ethylene binding capacity but rather 
to synthesis of an ethylene sensitivity factor (Van Altvorst and Bovy 1 995). Increases 
in tissue sensitivity to ethylene have been associated with the accumulation of short?
chain saturated fatty acids produced in carnation petals during early stages in 
senescence. The pollination of carnation styles also coincides with the formation of 
fatty acid synthesis (Whitehead and Vasiljevic 1 993). The increased fatty acid 
concentrations are hypothesised to increase the ethylene binding affinity by altering the 
properties of certain membranes (V an Altvorst and Bovy 1995) .  
1.3.18 Hormone interaction: Auxins 
Auxin has been implicated in playing a role as an inhibitor of floral abscission. A 
number of studies successfully delayed or inhibited the abscission of flowers or floral 
parts in a number of woody crop species by using exogenously applied synthetic auxins, 
such as a-naphthaleneacetic acid (NAA) or indole-3-acetic acid (IAA) (Einset 1 986, 
Joyce 1 989, Joyce and Beal 1999) . Auxin treatments can also inhibit abscission in 
specific floral parts, such as anthers in Cleome hassleriana (Koevenig and Sillix 1 973) 
and styles in Citrus (Einset 1 986). Experimental work on cut sterns of Rosa, showed 
that pedicel abscission following treatments with exogenous auxin was significantly 
68 
reduced, even after cut sterns were treated with ACC, the precursor to ethylene. 
Goszczynska and Zieslin ( 1 993) subsequently hypothesised that auxin played a role in 
decreasing the sensitivity of abscission zone cells to ethylene. The application of auxins 
to either prevent or delay floral abscission has not always been successful, as recently 
highlighted in a study where dip applications of IAA ineffectively prevented flower 
abscission in Grevillea (Joyce and Beal 1999). Moreover, excessive quantities of auxin 
can also induce organ abscission by promoting ethylene production (Abeles et al. 1 992). 
1 .3.19 Ethylene inhibitors 
Various chemical applications have been successfully employed for preventing 
ethylene-related damage (e.g. abscission of flowers or floral parts) in flowering 
ornamental crops . Specifically, two approaches to this problem have been addressed; 
incorporation of treatments either targeted at interfering with ethylene biosynthesis or 
aimed at reducing the effects of ethylene by inhibiting or depressing its action. 
Substances commonly used as inhibitors of ethylene biosynthesis include chemicals 
such as arnino-ethoxyvinylgycine and arninooxyacetic acid, whereas popular inhibitors 
of ethylene action have primarily been those with silver-based compounds (e.g. silver 
thiosulfate) (Macnish et al. 1 999) and/or with the use of the novel gaseous compound, 
1 -rnethylcyclopropene (Serek et al. 1 994, Macnish et al. 2000). 
1.3.20 Inhibitors of ethylene biosynthesis 
Several effective chemical inhibitors of ethylene biosynthesis include arnino?
ethoxyvinylgycine (A VG) and aminooxyacetic acid (AOA). Specifically, both A VG 
and AOA are suggested to inhibit the action of the enzyme A CC-synthase that converts 
S-adenosylrnethionine to 1 -arninocyclopropane- 1 carboxylic acid (ACC) (see Figure 1 .5 
A) . In carnation, for example, increased ethylene biosynthesis associated with petal 
senescence was accompanied by an increase in ACC synthase activity and ACC 
content. Treatment of climacteric carnation petals with 2,5-norbornadiene (NBD), 
another ethylene biosynthesis inhibitor, resulted in the subsequent inhibition of ACC 
synthase and ACC oxidase activities (Woodson 1 99 1 ) . These and other ethylene 
biosynthesis inhibitors have been effective in delaying ethylene-mediated senescence or 
abscission in Strongylodon macrobotrys (ornamental blue jade vine) (Furutani et al. 
69 
1 989), carnation (Staby et al. 1 993), Phalaenopsis (orchid) (Porat et al. 1 994) and 
various other species (Staby et al. 1 993). However, deterrent factors associated with 
their use have been attributed to the cost of the products themselves, the possible 
toxicity or side-effects of the chemicals (Staby et al. 1993) and to their lack of 
effectiveness in delaying or preventing senescence/abscission in some species (e.g. 
Digitalis flowers) (Van Doom and Woltering 1 99 1 ) . Furthermore, inhibitors of 
ethylene biosynthesis do not protect flowers and plants from the damaging effects of 
exogenous ethylene (Van Doom and Woltering 1 99 1  ) .  
1 .3.21 Silver thiosulfate (STS) 
Extensive research has been directed towards extending the post-harvest life of 
ornamental crops via treatments that inhibit ethylene action. In particular, the efficacy 
of silver thiosulfate, coined as the 'the original anti-ethylene compound' (Nell 1994), 
has been extensively researched on numerous flowering crops since the discovery of its 
effects in the late 1970's .  Originally, Beyer ( 1 976) showed that silver ions, in the form 
of silver nitrate, effectively inhibited ethylene-mediated processes. However, silver 
nitrate produced phytotoxic effects when applied as a spray, and was ineffectively 
translocated in cut flower stems when added to vase solutions. Silver in the form of a 
silver thiosulfate (STS) complex was subsequently found to be more suitable for 
protecting flowers from the damaging effects of ethylene. Not only was STS readily 
translocated in cut flowers,  but it was also deemed non-phytotoxic when used at 
effective concentrations (Staby et al. 1 993) .  
The effectiveness of STS lies in its ability to bind at ethylene receptor sites, thus 
preventing the subsequent binding and action of ethylene molecules (Beyer 1 976, Van 
Meeteren and Proft 1982). Numerous reports have described the effectiveness of STS 
in preventing floral abscission or delaying the onset of senescence-related symptoms in 
flowers of woody ornamentals. For instance, the postharvest potential of Grevillea spp. 
as cut flowers is limited by its short vase-life ( < 10  days) due to abscission of perianth 
and perianth segments (tepal) and/or whole flowers (Joyce and Beal 1 999). Short (::;; 1 
hour) pulse treatments of STS (4 mM r1 Ag+) applied in vase solutions successfully 
prevented flower abscission and extended vase life by inhibiting the action of 
endogenously produced ethylene (Joyce and Haynes 1 989, Joyce and Beal 1 999). 
70 
Similar studies on the efficacy of STS in preventing flower and flower bud abscission 
has been reported in numerous other ornamental woody crops, for example in Rosa 
(Serek 1 993), Boronia heterophylla (Macnish et al. 1999) and Bougainvillea glabra 
(Cameron and Reid 1 983) .  
Limited research has shown that ethylene i s  an important factor in the postharvest life of 
a number of species within the family Myrtaceae. Moreover, the efficacy of STS in 
either preventing ethylene-related abscission or shrivelling/wilting of floral parts has 
been investigated in Leptospermum scoparium (Zieslin and Gottesman 1 983) ,  
Chamelaucium uncinatum (Joyce 1988),  Verticordia spp. (Joyce and Poole 1 993) and 
Metrosideros collina (Sun et al. 2000) . Pulse pre-treatments of 4 mM of STS increased 
flower fresh mass and successfully prevented floral abscission in C. uncinatum and V. 
nitens against exogenous ethylene (8.6-9.4 J.Ll r1 ) treatments applied for 20 to 24 h, 
although these effects did not differ from controls (no STS or ethylene treatments) 
(Joyce 1 989, Joyce and Poole 1 993). Similar findings were observed in L. scoparium, 
although Zieslin and Gottesman ( 1 983) also reported on the level of ethylene emanation 
and petal abscission in young (at full petal opening) and mature ( 1 -2 weeks after petal 
opening) flowers exposed to conditions with high (> 70%) relative humidity (RH) . 
Under these conditions, petal abscission occurred only in mature flowers, irrespective of 
STS treatment, while both young and mature flowers emanated significantly higher 
rates of ethylene (compared to controls) only following STS treatment (Zieslin and 
Gottesman 1 983) .  Therefore, STS prevented ethylene-induced, but not natural 
abscission (senescence) of mature flowers. Under 70 percent RH, ethylene evolution 
occurred in a climacteric pattern, peaking at seven days after harvest (Zieslin and 
Gottesman 1 983) .  Therefore, applications of STS have been successful in preventing 
exogenous ethylene-induced damage on perianth flower parts, at least in these species. 
While the majority of investigations of STS-related effects in myrtaceous species have 
focused on perianth abscission, only a few studies has investigated its role in preventing 
both stamen and petal abscission. Sun et al. (2000) described the effects of pre?
treatments of STS (dip of 2.0 mM) on cut cymules of Metrosideros collina followed by 
24 h exogenous ethylene (0-5 J.Ll r1)  treatments. STS treatments conferred considerable 
protection against exogenous ethylene, since between Day 5-8, the level of abscission 
from petals and stamens was reduced by 32.4 and 1 2  percent per floral part, 
7 1  
respectively. In contrast, similar applications of STS on flowers of Eucalytpus ficifolia 
provided no protection from ethephon-induced (an aqueous ethylene-releasing 
compound) stamen wilting, symptoms that were probably associated with water loss 
rather than an ethylene-mediated response (Sun et al. 200 1 ) .  STS , therefore, appears to 
be an effective treatment for preventing exogenous ethylene-related damage in some 
species within Myrtaceae. However, the protection conferred has not been associated 
with an improvement in vase-life above that found in ethylene-untreated (control) 
plants. 
1 .3.22 1-Methylcyclopropene (1 -MCP) 
Concern over the toxic effects of silver (STS) on human health and the environment has 
prompted research into alternative protocols for preventing ethylene-induced damage in 
cut flowers (Halevy 1 994, Nell 1 994) . Interest in a relatively new gaseous compound, 
1 -methylcyclopropene ( 1 -MCP), has developed due to its odourless, non-toxic 
properties and relative ease of use (Serek et al. 1 994, Newman et al. 1 998) .  In a large 
screening study conducted by Macnish et al. (2000), the efficacy of 1 -MCP was 
examined on 14 different native Australian cut flowers treated with exogenously applied 
ethylene ( 10 111 r 1 )  for 1 3  h at 20?C. Pre-treatments of 1 -MCP successfully prevented 
floral abscission and associated loss in vase life in Ceratopetalum gummiferum, 
Chamelaucium uncinatum, two Grevillea hybrids, Leptospermum petersonii and 
Verticordia nitens. These results were in accordance with studies using STS-treated cut 
flowers from C. uncinatum (Joyce 1989, Macnish et al. 2000), V. nitens (Joyce and 
Poole 1 993) and Grevillea hybrids (Joyce et al. 1 993). In particular, 1 -MCP improved 
the vase-life of only one species, C. gummiferum (data not shown), while in another 
species, Telopea speciosissima, no significant protection from ethylene was provided. 
Overall ,  with the exception of C. gummiferum, the vase-life of all 1 3  species tested was 
not significantly extended in the absence of exogenous ethylene exposure (Macnish et 
al. 2000). 
S imilar findings with regards to the level of protection afforded by 1 -MCP have been 
reported in other woody plants. Muller et al. ( 1999) demonstrated that 1 -MCP was 
effective in preventing petal fading in intact flowers of two Rosa hybrid varieties after 
undergoing a simulated transportation experiment designed to induce a stress-related 
72 
ethylene production response. S imilar to the effects of STS, 1 -MCP effectively 
protected against the deleterious effects of exogenous ethylene but did not extend shelf?
life of these Rosa hybrids in the absence of ethylene. In some cases, pre-treatments of 
1 -MCP were more effective than STS in preventing exogenous ethylene-mediated 
effects. For instance, in Boronia heterophylla, 1 -MCP further reduced the level of 
flower wilting, stem fresh weight and extended vase-life (Macnish et al. 1 999) from 
STS-treated cut flowers. As in the case with most other studies, the efficacy of 1 -MCP 
and STS was not greater than that found in control plants. 
To date, only one study has examined the efficacy 1 -MCP in protecting ethylene?
induced stamen abscission. As previously discussed, pre-treatments of STS followed by 
exogenous ethylene exposures effectively reduced stamen and petal abscission in cut 
cymules of Metrosideros collina (Sun et al. 2000) . The results obtained from STS, 
however, did not parallel those found for 1 -MCP. Initially, stamen abscission was 
inhibited with 1 -MCP ( 1 50 nl r 1 applied over 6 hours) for a maximum of 24 hours after 
the ethylene (0. 1 J..ll r 1 ) treatment. Over the subsequent five days, stamen abscission in 
control and 1 -MCP treatments increased significantly (>87 percent by Day 7) in 
comparison with the STS treatment ( 1 3% by Day 7) .  Not only was 1 -MCP found to be 
ineffective in preventing ethylene-induced abscission in cut cymules, but it also 
enhanced endogenous ethylene production, thus corresponding with the increased 
stamen fall (Sun et al. 2000). The short-term protection afforded by 1 -MCP was 
attributed to the incomplete blocking of receptors or the subsequent synthesis of new 
ethylene receptors (Sisler and Serek 1 999). Whether the response exhibited by M. 
collina to 1 -MCP is unique within this genus is, as of yet, unknown and thus merits 
further investigation. 
1.4 Summary and Thesis Objectives 
The continuing efforts to understand the physiological, biochemical and genetic changes 
during ontogenetic development show promise. This is in spite of decades spent 
investigating this phenomenon, given that first studies examining this phenomenon date 
to the early part of the last century (Goebel 1 900). The use of mutants in genetic studies 
show promising results and are helping to establish the role of different genes and/or 
73 
hormones and their interactions during different phases in development. However, 
current methods to accelerate phase change are not always consistent and/or require 
high maintenance and extended periods of observation, for example, selection and 
breeding of genotypes with short juvenile periods. Until such a time that genetic 
engineering becomes the preferred method for accelerating phase change, reliance on 
cultural methods and environmental conditions that favour vigorous growth appear 
favourable. Moreover, a further understanding of vegetative phase change would 
benefit from studies that examine the physiological changes that occur during this 
transition as opposed to morphological comparisons between different states of 
ontogenetic development. These observations should also be considered in light of the 
currently proposed hypotheses and models so that a conceptual understanding of phase 
change can be advanced. 
Numerous investigations into the environmental and chemical factors that regulate 
flowering in woody angiosperm species have been reported. These findings are 
supported by basic research, particularly from model herbaceous plants have 
contributed to our understanding and knowledge of factors involved in the flowering 
process. Clearly, environmental and chemical factors underpinned by genetic 
mechanisms function to determine the timing and subsequent development of 
reproductive organs. However, even with an enhanced knowledge at the molecular 
level, there will still be a need to induce flowering, either through environmental 
manipulation and/or chemical applications. This is particularly important within the 
floricultural industry. Given the proper knowledge of the flowering requirements of a 
target crop, the timing, quantity and quality of flowers should then be able to be 
culturally manipulated. 
The postharvest life of cut flowers has been investigated in numerous studies. The role 
of ethylene in decreasing the vase-life of many woody species is well noted, although 
treatment applications that either inhibit ethylene biosynthesis or its action on flowers 
have not been consistent across species. In particular, further studies are needed to 
examine the role of ethylene and chemical inhibitors in preventing its deleterious effects 
on flower parts other than sepal/petals, such as stamens. With regards to water 
relations, most studies concede that treatments with biocides and/or sucrose are 
important in extending vase-life, at least in a number of woody species. Whether these 
74 
plant responses are exhibited in most species that have stamens as their ornamental 
feature, i s  not well documented. Additionally, an extensive and comprehensive 
postharvest knowledge of both water relations and ethylene-related research is lacking 
for numerous woody genera. Studies that encompass both of these factors would be 
particularly important in producing a successful flowering crop in the floricultural 
industr;. 
A proper evaluation of plant performance would not be complete without experiments 
conducted on various cultivars of a given species. In this thesis, several cultivars of 
Metrosideros excelsa were examined in order to assess their relative responses to 
different treatments. However, in certain instances, determination of responses from 
several cultivars was not possible due to the unavailability of particular cultivars at the 
start time of a given experiment(s) .  
In light of this review, several hypotheses with regards to phase change, environmental 
control of flowering and the postharvest treatment of cut flowers of Metrosideros 
excelsa are advanced and tested in this thesis. It is hypothesised that: 
1 )  The restriction of shoots to a single-stem axis and/or application of a root restriction 
regime accelerates vegetative phase change in M. excelsa, as has been documented 
for several other woody species (Zimmerman 1972, Davis 1 99 1 ,  Snowball et al. 
1 994) (Chapter 2). 
2) Juvenile plants of M. excelsa undergoing vegetative phase change exhibit decreasing 
carbon isotope discrimination (L1) because of changes in leaf morphology consistent 
with greater water use efficiency, notably more tomentose and rounder leaves. This 
may also be associated with a reduction in photosynthetic capacity with increasing 
node position. This hypothesis would be consistent with the bulk of published 
ecophysiological reports, in which leaves from adult plants exhibit lower L1 and 
associated differences in photosynthetic capacities than leaves from juveniles 
growing in the field (Donovan and Ehleringer 1 992, 1 994; Hansen 1 996, Cordell et 
al. 1 998) (Chapter 3). 
75 
3) Changes in ? are not a function of leaf age, but are the result of changes in  leaf 
conductance relative to photosynthetic capacity at the time of leaf expansion 
(Chapter 3) .  
4)  Both low temperature and/or short day-length are important in promoting floral 
induction in M. excelsa, as reported for a number of closely related genera (Shillo et 
al. 1 985, Zieslin 1 985, Moncur and Hasan 1 994) (Chapter 4). 
5) The size of the developing bud at the time of application of an inductive treatment 
will determine whether or not floral initiation will take place, with larger buds 
becoming floral more readily than smaller buds (Chapter 4). 
6) Irradiance is positively related to the number of flowers produced in M. excelsa 
when applied in conjunction with treatments inductive for flowering (Chapter 5) .  
7) The vase life of M. excelsa flowers is limited by adverse water relations and by the 
ethylene-related abscission of whole flower and floral organs .  Appropriate holding 
solutions and inhibitors of ethylene action may, therefore, provide effective 
treatments for extending the vase life and minimising ethylene-related damage, as 
demonstrated for closely related taxa (Burge et al. 1996, Macnish et al. 2000, Sun et 
al. 2000, 200 1 )  (Chapter 6). 
76 
Chapter 2 
Effect of shoot and root restriction on vegetative phase change in 
Metrosideros excelsa 
2.1 Introduction 
Early developmental phase change in higher plants is characterised by a transition from 
a juvenile through to an adult vegetative phase. During this transition there are a 
number of accompanying changes in vegetative characteristics such as in leaf 
morphology, phyllotaxy, thorniness or leaf 'hairiness '  (Godley 1985, Poethig 1990, 
James and Bell 2001 ) . A considerable number of New Zealand plants (greater than 200 
species) exhibit heteroblastic development, showing distinct and dramatic changes in 
leaf morphology between the juvenile and adult stages (Godley 1 985,  Day et al. 1 997).  
This is in contrast to homoblastic species such as those in the genera A vicennia, 
Macropiper and Metrosideros where a gradual transition between these life stages 
occur. For example, Metrosideros excelsa has characteristic phase-dependent leaf 
morphologies; leaves in adult compared with juvenile plants are typically more rounded 
and possess a characteristic downy tomentum on abaxial surfaces (Cockayne 1 928,  
Dawson 1968) .  
In higher plants, the competence to undergo reproductive development is usually 
dependent upon attainment of an adult vegetative state (Hackett 1985) .  However, 
attainment of this adult vegetative state can be strongly influenced by propagation 
methodology. For example, seedlings of Metrosideros typically reach an adult 
reproductive phase within five to ten years, whereas cuttings from adult foliage are 
capable of flowering within one year (Oliphant et al. 1 992). In light of the difficulties 
in rooting cuttings, such as in Metrosideros, an alternate and preferred method of 
propagation has been through micropropagation, where multiplication of clonal material 
can be obtained rapidly. 
Data published from this chapter: Clemens, J. ,  Henriod, R.E., Bailey, D.G.,  Jameson, P.E. ( 1 999) 
Vegetative phase change in Metrosideros: Shoot and root restriction. Plant Growth Regulation 28:  
207-214. 
77 
During micropropagation, rejuvenation of parent material is a common occurrence in 
many woody plant species (Hackett and Murray 1 992). Rejuvenation entails the 
ontogenetic reversion of adult material to the juvenile phase and results in plantlets 
bearing leaves that are morphologically and physiologically distinct from adult leaves 
(Oliphant 1 988) .  This process is different to the reversal of physiological ageing, which 
can arise from cutting propagation, grafting or hedging and is often associated with 
renewed shoot vigour (Fortanier and Jonkers 1976, Wareing and Frydman 1 976, 
Zimrnerman et al. 1985) .  Rejuvenation following micropropagation is reported to occur 
in Metrosideros and the time until which plants can bear flowers can exceed three years 
(Oliphant et al. 1 992) . Moreover, floral development in Metrosideros occurs only on 
branches bearing adult morphology. This is in contrast to certain Eucalyptus species 
(also in the family Myrtaceae), where flowers have been reported to arise on branches 
bearing juvenile foliage (Wiltshire et al. 199 1 ) .  Therefore, treatments applied to 
rejuvenated plants of Metrosideros that might accelerate the transition between the 
juvenile to adult phase might be expected to shorten the time to flower. 
Application of various cultural practices can affect the timing of vegetative or 
reproductive phase change (Oliphant et al. 1992, Snowball et al. 1 994). Acceleration of 
phase change as a result of manipulation of the growing environment (e.g. high light) or 
plant material (e.g. shoot restriction) has been attributed to conditions that promote 
vigorous growth (Wareing and Frydman 1976, Poethig 1990) . Subsequently, treatments 
that tend to retard growth, such as water stress, low light and temperatures, and poor 
carbohydrate and mineral nutrition, have been associated with prolongation of the 
juvenile phase (Poethig 1990). In several studies, limiting growth to a single-stem by 
removal of axillary branches has been shown to hasten the onset of developmental 
phase change (Davis 199 1 ) .  Additionally, root restriction has also been reported to 
accelerate phase change (Zimrnerman 1 972). However, in both these and other studies 
attainment of the adult phase has been assessed using flowering as an indicator of 
ontogenetic status and, therefore, no quantitative distinction was made of the 
progressive changes in vegetative characters and the onset of flowering. 
In this study, vegetative phase change was characterised in plantlets of Metrosideros 
excelsa derived by micropropagation. The experiment was designed to assess the 
78 
relative effects of shoot and root restriction, applied in factorial combination, on 
accelerating vegetative phase change. Shoot restriction was imposed through removal 
of lateral shoots resulting in a single-stemmed plant, whereas a root restriction regime 
was afforded by using pots of different sizes. By applying a shoot restriction regime, 
the level of ontogenetic development in leaves from complementary height and node 
positions was assessed in branched (control) and single-stemmed plants through 
evaluation of leaf morphological characters . 
Typically, phases of shoot development are permanently recorded as variation in the 
character of structures along the axis of the shoot (Poethig 1 990). Thus, the capturing 
of these changes in morphological and optical properties of leaves was undertaken using 
an image analysis procedure. This technique was particularly suitable for examination 
of ontogenetic development of leaves in the homoblastic M. excelsa, where the 
transition between juvenile and adult leaf forms occurs gradually. Specifically, the use 
of optical parameters enabled quantification of the gradual accumulation of tomentum 
(increasing whiteness) on the abaxial surfaces of leaves undergoing ontogenetic 
development. 
In this experiment, the following hypotheses regarding the effect of restriction 
treatments on phase change were tested. It is hypothesised that: 
1 )  The restriction of shoots to a single-stem axis and/or application o f  a root restriction 
regime accelerates vegetative phase change in M. excelsa, as has been documented 
for several other woody species . 
2.2 Material and Methods 
2.2.1 Plant material 
Micropropagated plants of Metrosideros excelsa 'Scarlet Pimpernel ' growing in 50 mm 
square plastic were supplied by Lyndale Nurseries Auckland Ltd. Initial plant height, and 
total leaf, stem and root dry weights of a sample of five plants were 1 1 .6 cm, 1 .46 g, 0.63 
g and 0.85 g, respectively. Plants were highly-branched, each bearing 8-9 terminal 
shoots. Experimental containers were made from 100 mm sections of PVC water pipe of 
79 
five different internal diameters. A synthetic, non-woven fabric (Syntex, Permathene 
Plastics Ltd. ,  Auckland) was glued to the base of each section to give rooting volumes of 
0.82, 1 .7, 2.7, 4. 1 and 7 .2 1 .  Plants were transplanted on 29 August 1997 into the pots 
using a growing medium composed of split, expanded clay granules ( 1 -4 mm diameter 
range, Hydroton, NZ Hydroponics International Ltd. ,  Tauranga) and vermiculite (4 mm 
grade, Revertex Industries (NZ) Ltd.)  (9: 1 v/v). Each plant (regardless of pot size) 
received the same amounts of controlled release fertilisers ( 1 .0 g 3-4 month release, and 
3 .0 g 8-9 month release Osmocote Plus, Grace Sierra, Heerlen, The Netherlands) placed 
adjacent to the root ball of the transplant. These application rates were comparable with 
those commonly used for woody plants grown for nine months in 2-3 L containers . 
2.2.2 Experimental layout 
In the first week after planting, all shoots with the exception of the tallest shoot on each 
plant were removed. On half of the plants grown in each of the five container sizes, the 
tallest shoot was allowed to grow unchecked and was trained in an upright position. Any 
branch development in these plants (referred to as single-stemmed plants) was prevented 
by removal of axillary buds as these swelled before bud break. On the remaining plants 
(referred to as branched plants), the single shoot was pinched once to encourage a 
branched growth habit. There were five blocks and two internal replicates (plants) of 
each of the 10 treatments, which were completely randomised within blocks. The plants 
were placed on capillary irrigation tables wetted four times daily for 1 5  minutes to ensure 
an uninterrupted supply of water to the plants. The experiment was conducted over nine 
months in a temperature controlled greenhouse set to vent at 23?C, and warmed when the 
temperature fell below 1 5?C. 
2.2.3 Growth measurements 
Plant height to the tip of the expanding leaves (shoot length) and number of nodes 
bearing leaves longer than 1 5  mm were recorded every 14 days on single-stemmed 
plants. Shoot length (the linear distance from the medium surface to the shoot tip of the 
longest shoot) and node number in the branched plants was recorded immediately 
before destructive harvest on 2 June 1 998 .  
80 
2.2.4 Image analysis 
For image analysis, shoots from each single-stemmed plant were removed at the surface 
of the growing medium, and the longest shoot in each branched plant was removed at a 
point below which no green leaves were attached. Detached shoot bases were placed in 
vials of distilled water to maintain leaf turgor. Images were captured of both adaxial 
and abaxial surfaces of the four leaves at pairs of contiguous nodes at up to six locations 
within each shoot (node pairs 6-7, 1 6- 1 7, 26-27, 36-37, 46-47, and 5 1 -52). For shoots 
not bearing fully-expanded leaves to nodes 5 1 -52, measurements were made at the same 
five locations up to nodes 46-47, and at the highest nodes bearing fully expanded leaves .  
In general, branched plants bore leaves on 35-40 nodes, and leaves on the lowest nodes, 
which had been heavily shaded by surrounding leaves,  had abscised. Typically, image 
analysis was possible for leaves at nodes 1 6- 1 7, 26-27 and 36-37 in these plants. In 
addition, images were captured for the lower surface of one leaf at every node for all 
plants grown in the intermediate container size (2.7 1) .  
2.2.5 Image analysis protocol 
Leaf dimensional and optical properties were captured on a Sony DX3000P colour (3  
chip) video camera mounted on a Polaroid MP4 copy stand containing four 75 W 
incandescent bulbs. The capturing and processing of leaf parameters was through PC 
installed hardware (lmaging Technology Colour Frame Grabber hardware) and software 
(Vision Image Processing System), respectively. Prior to any leaf measurements, the 
system was calibrated for aspect ratio and pixel size (Bailey 1995) .  The red, green and 
blue colour channels were normalised against a pre-calibrated white background. 
Leaves were placed at the base of the copy stand below the camera lens and covered 
with a glass pane to prevent folding. Optical measurement were made of leaf length, 
width, length/width ratio, area, perimeter and roundness (leaf perimeter)2/leaf area), as 
described by Bailey and Hodgson ( 1 988) and Bailey ( 1 995). To make leaf colour 
measurements, the segmented leaf image was eroded by two pixels to avoid pixels at the 
leaf edge that may have included background. This was then used to mask the original 
colour image, which was transformed into HLS (Hue, Lightness and Saturation) colour 
8 1  
space (Foley and V an Dam 1982). The mean and standard deviation of leaf hue (scale 0-
256), lightness (percentage scale) and saturation or strength of colour (percentage scale) 
were calculated. 
2.2.6 Plant biomass and leaf mineral determination 
Shoots were dissected into leaf and stem fractions.  Leaf area per plant was measured, 
and the total number of terminal shoots > 1 5  mm in length was recorded for branched 
plants. Root systems were washed clean of the growing medium, and fine roots 
recovered by sieving under running water. Leaves, stems and roots were dehydrated at 
80?C for 24 h and weighed. Complete mineral analyses were obtained (Hill 
Laboratories, Hamilton, NZ) for four replicates of the most recently fully expanded 
leaves for single-stemmed and branched plants grown in 0.82 and 7.2 1 containers. 
Fully expanded leaf samples were collected at nodes 17 ,  27, 37, 4 7 and 52 for single?
stemmed and branched (on available nodes only) plants and from the canopy top of 
adult plants ( -3 year-old plants) for determination of carbon isotope discrimination. 
2.2. 7 Carbon isotope analysis 
Leaf samples were dried at 80?C for 48-72 h and were analysed for determination of 
1 3C/1 2C isotope ratio at the Stable Isotope Unit, University of Waikato, New Zealand. 
The dry tissue samples were finely ground ( <200 11m) and combusted, and the resulting 
C02 was analysed for Op using a Dumas Elemental Analyser (Europa Scientific 
Tracermass) .  A precalibrated C4 sucrose reference was used that had been standardised 
relative to Pee Dee Belemnite by CSIRO, Canberra, Australia. Carbon isotope 
discrimination (?) was calculated from the sample (op) and ambient carbon isotope 
ratios relative PDP (oa) (equated to -0.0078 %o) using the formula (Farquhar et al. 
1989): 
82 
2.2.8 Statistical analyses 
Data were analysed by analysis of variance using the SAS statistical package (SAS 
Institute, Cary, N .C . ,  USA). Time was included as a factor for repeated measures made 
of node number and plant height. Where appropriate, node position on single-stemmed 
plants was treated as a within-plant split plot for analyses of shape, size and colour 
parameters . Statistical comparison of leaves on single-stemmed and branched plants 
grown in 2. 7 1 containers were also made at a shoot length measured 280 mm from the 
surface of the growing medium. This was the maximum height at which fully expanded 
leaves were borne on harvested shoots of all plants, and corresponded with node 
positions 1 2  and 23 in single-stemmed and branched plants, respectively. 
Transformation of data to normalise variance was not required except for analysis of 
some mineral concentrations. A log1 0(x+ 1 )  transformation before analysis of variance 
was used in these cases. Means were compared for significant differences at the 5%  
level. 
2.3 Results 
2.3.1 Growth parameters 
Single-stemmed plants grew to ten times their initial height over the thirty-five week 
experimental period, each adding approximately 50 nodes (Figure 2. 1 ). A key feature 
of the shoot restriction treatment was the greater number of nodes and greater shoot 
length that the single-stemmed plants attained compared to the branched plants. The 
longest shoot in branched plants was both shorter and bore fewer nodes ( -40 cm and 32 
nodes, respectively) than that in s ingle-stemmed plants ( - 100 cm and >50 nodes) 
(Figure 2. 1 ) . Branched plants bore an average of 1 6-2 1 terminal shoots > 1 5  mm long. 
Container size had no significant effect (P>0.05) on any of these growth parameters . 
2.3.2 Image analysis: optical parameters 
Measurements made at every node in both single-stemmed and branched plants grown 
in an intermediate container size (2.7 1) revealed contrasting trends in leaf development 
between these groups of plants as node number increased. Thus, while hue of the 
83 
1 20 r? 
1 00 + 
i 
80 1 
_......_ ! E 
u 
---?
..... 
..c 
r
o 
40 
.... 
(!) 
..0 
E 
.? 60 (!) 30 ? 
..c 
20 + 
I 
I 
0 4 8 1 2  1 6  
Time (weeks) 
------- height 
........._ nodes 
20 24 
- 20 
1 0  
28 
Figure 2 . 1 Increase i n  mean plant height and mean total number of 
nodes in  s ingle-stemmed plants of Metrosideros excelsa 'Scarlet 
Pimpernel '  over the experimental period. 
(!) -o 0 
z 
84 
abaxial leaf surface decreased steadily in branched plants from approximately 60-65 
units (yellow-green) at the basal nodes to 50-55 units (yellow) at nodes in the apical 
region of the shoot, that in the single-stemmed plants initially decreased slightly before 
increasing to above 70 units (approaching green) at node 50 (Figure 2.2 A).  This trend 
of divergence between single-stemmed and branched plants above intermediate node 
positions was also observed for colour saturation of abaxial leaf surfaces. Leaves in the 
upper portion of single-stemmed plants became less saturated (had less colour) than 
those in branched plants (Figure 2.2 B).  Leaf hue and saturation in single-stemmed 
plants attained or approached the corresponding values for leaves on adult plants. 
However, leaf hue and saturation in branched plants diverged from or did not approach 
these values (Figure 2.2 A-B) .  
Lightness (white { 100% } as opposed to black { 0% } )  of abaxial leaf surfaces tended to 
increase in both single-stemmed and branched plants . Lightness in the basal portion of 
branched plants rose sharply so that leaves at nodes 30-35 in this group of plants were 
as light as those at nodes 50-55 in single-stemmed plants (Figure 2.2 C) .  Mean leaf 
lightness, which was comparable in the more apical leaves of both sets of plants at 
harvest, was approximately 95% that of leaves on adult plants (Figure 2.2 C). The 5 %  
saturation attained for the abaxial surface of single-stemmed plants (approximately 25% 
that of branched plant leaves), coupled with a comparable lightness in  the two sets of  
plants, gave the most recently expanded leaves of single-stemmed plants a greater 
"mealiness" or whiteness than those of branched plants. 
Statistical analysis of leaf colour parameters for single-stemmed and branched plants 
were made across all root restriction treatments at the two node positions for which 
comparisons could be made (nodes 1 6- 17  and nodes 26-27) .  There were highly 
significant differences (P<0.000 1 )  for abaxial leaf surface hue between single-stemmed 
and branched plants at the two node positions (Table 2 . 1 ) .  Similarly for the saturation 
and lightness of abaxial leaf surfaces, there were highly significant differences 
(P<0.000 1 )  between single-stemmed and branched plants at the two node positions 
tested (Table 2 . 1 ) .  Hue, saturation and lightness of adaxial leaf surfaces in single?
stemmed plants were also significantly different (P<0.000 1 )  from those in branched 
plants. Leaves on single-stemmed plants were greener, less saturated and lighter both at 
nodes 1 6- 1 7  and 26-27 (Table 2. 1 ). Container size did not have a significant effect on 
85 
75 r--
70 
<!) 
::l 
...c 
4-
ro 
<!) 
.....l 
55 
50 
25 -?- ------
B 
20 
? 
c 
1 5  0 ...., ro ..... 
2 
ro 1 0  VJ 
4-ro <!) 
.....l 
5 
0 
55 T c 
50 t . 
? I l 
Vl 45 -Vl 
C!.J I 5 
...c 
Ol) 
40 ? "-"' 
C!.J ......l 
30  +----,---,,---?---,----,----,--------,-------? 
5 10 1 5  20 25 30 35 40 45 50 55 
Node position 
Figure 2 .2  Changes in mean optical properties w ith i ncreas ing node pos i tion for the 
abax ial l eaf surface in s i ngle-stemmed p lants (red l ine) and branched plants 
(b lue l i ne) of Metrosideros excelsa 'Scarlet P impernel '  grown in  2 . 7  I containers. 
A: hue. B: saturation. C: l i ghtness .  Green dotted l i ne show the correspond ing means 
for adul t  leaves. 
86 
Table 2. 1 Comparison of single-stemmed and branched plants of Metrosideros excelsa 'Scarlet Pimpernel' with respect to leaf optical (abaxial and 
adaxial surfaces) and dimensional (adaxial surface) parameters at node positions 1 6- 1 7  and 26-27. Means shown for single-stemmed and branched 
plants with each node position and leaf surface column were highly significantly different (P<O.OOO l ) . Data (mean ? s.e .)  are for all container sizes. 
Parameter 
Hue (units) 
Saturation (%) 
Lightness (%) 
Width (mm) 
Length/width ratio 
Length (mm) 
Area (mm2) 
Perimeter (mm) 
Roundness 
00 
-....) 
Leaves at nodes 1 6- 1 7  
Abaxial surfaces Adaxial surfaces 
Single- Branched Single- Branched 
stemmed stemmed 
63 . 1 7 ? 0.34 58. 1 6  ? 0. 3 1  1 02 . 1 8 ? 1 .46 74.56 ? 1 . 1 5 
15 .52 ? 0.22 17 .53 ? 0.22 7.09 ? 0. 1 1  12 .57 ? 0.36 
34.84 ? 0. 1 9  40. 12  ? 0.26 27 . 1 5 ? 0. 1 9  32.29 ? 0.3 1 
27.27 ? 0.53 25.30 ? 0.54 
n .s .  n .s .  
54.52 ? 1 . 1 8  48.99 ? 0.97 
1058 ? 37 874 ? 29 
1 32.4 ? 2.6 120. 1 ? 2.2 
n.s .  n .s .  
Leaves at nodes 26-27 
Abaxial surfaces Adaxial surfaces 
Single- Branched 
stemmed 
63.37 ? 0.44 57 .34 ? 0.4 1 
15 .85 ? 0.26 1 6.76 ? 0.24 
35 .62 ? 0. 1 8  42.86 ? 0.44 
Single-
stemmed 
99.07 ? 1 .92 
7.07 ? 0. 10 
26.28 ? 0.2 1 
33 .64 ? 0.54 
1 .58 ? 0.03 
5 1 .82 ? 0.65 
122 1  ? 22 
1 39.2 ? 1 .2 
1 5 .95 ? 0.08 
B ranched 
65.70 ? 1 .02 
17 . 10 ? 0.57 
35 .32 ? 0.57 
23.72 ? 0.6 1 
1 . 83  ? 0.02 
42.85 ? 0.9 1 
730 ? 28 
107 .8  ? 2.3 
16 .35 ? 0.09 
any of the three leaf colour parameters, nor was there a significant interaction between 
container size and shoot restriction. 
For statistical comparisons of the six node positions within single-stemmed plants (from 
nodes 6-7 to nodes 46-47) ,  there were highly significant effects (P<0.000 1 )  of node 
position on abaxial leaf hue, saturation and lightness (Table 2.2) .  Node position also 
had a significant effect (P<0.0001 )  on adaxial leaf saturation and lightness (although not 
hue) in single-stemmed plants, the means generally converging on those for adult leaf 
adaxial surfaces (Table 2.2). There were no significant main effects of root restriction 
on the three leaf colour parameters, and generally no interactions between root 
restriction and node position. 
2.3.3 Image analysis: dimensional parameters 
With respect to shape and size parameters, leaves of single-stemmed plants tended to 
become more similar to those of adult plants with increasing node position than did 
those of branched plants grown in the intermediate container size (2.7 L). For example, 
leaf width and leaf length/width ratio in single-stemmed plants were comparable with 
those in adult plants above 30-40 nodes (Figure 2 .3A and B) .  Leaf length, area and 
perimeter were greatest at nodes 1 0-40 in single-stemmed plants, and at 10-25 nodes in 
branched plants. Values for these parameters in single-stemmed plants became smaller 
at higher node numbers, and approached corresponding values for leaves of adult plants . 
In branched plants these parameters tended not to coincide with those for adult plants 
(Figure 2.3 C-E). Leaves became more rounded with increasing node number at a 
similar rate in single-stemmed and branched plants. However, leaves of branched plants 
had not attained the compactness of adult leaves by the end of the experiment (Figure 
2 .3  F) .  
Means for four of the six leaf size and shape parameters in single-stemmed and 
branched plants were highly significantly different in comparisons made at nodes 1 6- 1 7  
and 26-27 (Table 2 . 1 ) . There was no significant difference between single-stemmed 
and branched plants at the lower node position for leaf length/width ratio and leaf 
compactness (Table 2. 1 ), which was consistent with whole-shoot trends (Figure 2 .3  B 
and F). There was no effect of container size or interaction of this factor with shoot 
88 
Table 2.2 Effect of node position within single-stemmed plants of Metrosideros excelsa 'Scarlet Pimpernel' on optical (abaxial and adaxial) and 
dimensional (adaxial) leaf parameters, and means for corresponding parameters in mature leaves. Means bearing the same superscript are not 
significantly different (p<0.05). 
Parameter p 
Abaxial hue (units) 0.000 1 
Adaxial hue (units) n.s. 
Abaxial saturation (%) 0.000 1 
Adaxial saturation (%) 0.000 1 
Abaxial lightness (%) 0.000 1 
Adaxial lightness (%) 0.02 
Width (mm) 0.000 1 
Length/width ratio 0.000 1 
Length (mm) 0.000 1 
Area (mm2) 0.000 1 
Perimeter (mm) 0.000 1 
Roundness 0.000 1 
00 
\0 
Nodes 6-7 
63.62a 
l 3 .02b 
6.92a 
35 .9 1 c 
30.09c 
1 3 .97a 
2 .2 1  d 
30.63a 
323a 
73.57a 
1 7.98e 
Nodes 16- 1 7  
63. 1 9a 
1 5 .49c 
7 . 10a 
34.90a 
27 .2b 
27.23b 
2.0 1 c 
54.36c 
1 055b 
1 32. 1 bd 
l 6.93d 
nodes 26-27 Nodes 36-37 Nodes 46-47 Matures leaves 
63.37a 65.53b 68 .95c 74.23 
1 5 .83c 1 2 .44b 9. 1 1  a 6.2 1 
7.07a 7 . 1 9a 8 .36b 1 4. 1 
35 .69ab 38 .08c 39.58d 50. 1 8  
26.35a 25.93a 27.46b 28.77 
33 .62c 34.94c 34.72c 34.65 
1 . 57b 1 .3 1  a 1 .25a 1 . 1 5 
5 1 .65c 45.03b 43.07b 39.62 
l 2 1 9d 1 142bcd 1 1  08bc 1 045 
I 38 .8dc 1 3 1 .7bc l 27.6b 1 24.6 
1 5 .93c l 5 .26b 1 4.72a 1 5 .06 
\0 
0 
50 60 
A 
40 s 50 ,.--.. s E 530 '--' ..s:: -5 bb 40 -"'0 '!i 2o t:: ? ...... <...,. ell ell Q) Q) 30 ? 1 0  .....:I 
0 20 
1 400 -
I c 
1 60 -?-----? -? 1 E 1 40 
? 20 
..... Q) ?1 00 E 
?c:  & 80 -
1a 
? 60 
40+--r-,--r-??--????? 
2.5 I 
.9 
I A
B I 1 200 -I D 1 9 y- - F - ? --- ------ -? 1 8  -
I 
E 
? 2 
'!i 
--.. 
..s:: bD 
_
@ .5 
...... 
ell 
Q) ? 
5 
,.-..! 000 -
N E 
E 800 '-' 
ell 
? 600 
ell 
...... ? 400 ? 
JJtw.. ' 200 
0 
1 0  1 5  20 25 30 35  40 45 50 55 5 
Node position 
1 0  1 5  20 25 30 35 40 45 50 55 
Node position 
17 
I 
0 
' 
"' 
- -..::s - - - -
?1 6  
"\A--??. 1 5  
1 4  +--,--r--,-,,-,--,--,--,--r-? 
5 I 0 1 5  20 25 30 35 40 45 50 55 
Node position 
F igure 2.3 Mean dimensional changes with increasing node position for the abaxial leaf surface in single-stemmed plants (red l ine) and 
branched plants (blue l ines) of Metrosideros excelsa 'Scarlet Pimpernel' grown in 2.7 I containers. A: with . B : length/width ratio. 
C: length. 0:  area. E: perimeter. F :  roundness. Green dotted l ines show the corresonding means for adult leaves. 
restriction for shape and size parameters at the two node positions compared 
statistically. 
There were highly significant effects (P<0.000 1 )  of node position on shape and size 
parameters in single-stemmed plants when means at six locations ranging from nodes 6-
7 to nodes 46-47 were compared (Table 2.2) .  As with the colour parameters, the means 
were consistent with the whole shoot trends shown by plants grown in the 2 .7 L 
containers (Figure 2 .3 A-F) . There were no significant effects of root restriction or 
interaction between root restriction and node position for these parameters . 
Comparing development at a linear shoot distance of 280 mm above the growing 
medium indicated no significant difference between leaves on single-stemmed and 
branched plants for any dimensional parameter. However, leaves of single-stemmed 
plants had higher hue, and lower saturation and lightness than branched plants (62. 1 and 
58 .2, 14 .7% and 1 7 . 1 %, 35 .5% and 4 1 .5%, respectively). 
2.3.4 Dry weight accumulation 
Single-stemmed plants were significantly smaller than branched plants (P<0.000 1 )  with 
respect to root dry weight (Figure 2.4 A), total leaf area (7.00 and 34.00 dm\ leaf dry 
weight ( 1 6 .3  and 6 1 .0 g), stem dry weight ( 1 2.2 and 22.0 g) and total shoot dry weight 
(28.5 and 83 .0 g). The most marked effect of container size was on the growth of roots 
in branched plants. Restricting roots caused a significant decline in root dry weight in 
these plants (Figure 2.4 A).  However, an effect of container size on root dry weight was 
not seen in the single-stemmed plants (Figure 2.4 A). Container size did not have a 
significant effect on leaf area, leaf dry weight, stem dry weight or total shoot dry weight 
in branched or single-stemmed plants. 
There was a significant interaction (P<0.0 1 )  between container size and shoot restriction 
treatments for shoot/root dry weight ratio. In both branched and single-stemmed plants 
shoot/root dry weight ratio increased as container size decreased, although the response 
to decreasing container size was more marked for the branched plants (Figure 2.4 B) .  
Whereas the increase in shoot/root ratio in branched plants was due to the reduction in 
root dry weight with decreasing container size (Figure 2 .4 A), that for the single-
9 1  
6 0  r ?? -? - - ---- -- - -- ? - -
50 I 
::?9 40 
? _c: on 
Q) 
? 30 1 
>-, 1-. 
-o 
0 
? 20 ? 
0 +,-----,-----.----?-----,----?----?----?----? 
3 .5 -- - ---- -- - --- -- - - -- -
B 
3 
0 
? 
"' 
1-. 
? 2.5 a 
0 
0 
1-. 
? 2 a 
0 
0 
_c: 
Vl 
1 .5 
1 
0 2 4 6 
Pot s ize ( l itres) 
Figure 2.4 Effects of conta iner s ize on growth of s ingle-stemmed 
(red l ines) and b ranched plants (b l ue l i nes) of Metrosideros 
excelsa 'Scarlet Pimpernel ' .  A: f inal  root dry weight (mean ? SE) . 
B:  shoot dry weight/root dry weight rat io (mean ? SE) . 
8 
92 
stemmed plants was a result largely of a consistent (but non-significant) increasing trend 
in stem dry weight as container size decreased (P>0.05) .  
2.3.5 Leaf mineral concentrations 
Leaves of branched plants contained significantly lower concentrations of most major 
nutrients (N, P, K, Ca and Mg) than single-stemmed plants (20-50% less) ;  
microelements Fe, Mn, Zn and B were also present at lower concentrations by 20-55% 
(Table 2.3).  There was no  interaction between shoot and root restriction treatments for 
these minerals, and no main effect of root restriction. The only main effect of root 
restriction was to significantly increase leaf Ca concentration (P<O.OOO 1 )  in plants 
grown in the smallest pot size compared to those in the largest pot size ( 1 . 12% and 
0.79%, respectively) . There were also significant interactions between root and shoot 
restriction treatments for S and Na. 
2.3.6 Carbon isotope discrimination 
Carbon isotope discrimination (?) values differed between branched and single?
stemmed plants, being significantly lower in single-stemmed than branched plants at 
nodes 27 and 37 (P<0.0 1 )  (Figure 2.5) .  Comparisons between plant types were not 
possible above node 37 since shoots in branched plants did not exceed this nodal 
position. There was also a significant main effect of node position (P<0.000 1 )  but no 
interaction of plant type and node position on leaf ? levels. In both single-stemmed and 
branched plants, mean leaf ? values showed a diverging pattern away from those 
exhibited by adult plants, although values within branched plants did not differ 
significantly. In single-stemmed plants, a comparison of mean ? levels at node position 
7 with those at position 27 were significant (P<0.05) ,  whereas comparisons with higher 
node positions showed an increase in significance (above node position 37, P<0.0001 ) .  
93 
Table 2 .3  Mineral nutrient concentrations in the leaves of single-stemmed and branched plants of Metrosideros excelsa 'Scarlet Pimpernel' grown in 
0.82 and 7.2 1 containers . 
\0 
? 
Shoot restriction 
Treatment 
Branched 
Single-stemmed 
P-value 
N p K Ca Mg 
(%) (%) (%) (%) (%) 
0.7 1? 0.05 0.07? 0.00 0.86? 0.04 0.79? 0.05 0.23? 0.0 1 
1 .39? 0.06 0. 1 1? 0.00 0.96? 0.02 1 . 12? 0.09 0.33? 0.0 I 
0.000 1 0.000 1 0.046 0.000 1 0.000 1 
Fe Mn Zn Cu B 
mg kg- 1 mg kg- 1 mg kg- 1 mg kg- 1 mg kg- 1 
25.5? 2.3 59. 1? 2. 1 7 .0 ?0.5 3 .75 1 8 .8? 0.9 
35 .0? 4.7 1 3 1 . 1? 6.4 10.5? 0.3 3 .38  24.3? 0.5 
0.04 0.000 1 0.00 1 n.s .  0.00 1 
Cl 
? 
c 
0 
....... 
C';l c 
E 
'-u <Zl 
-o 
<J.) 
CL 
3 0 <Zl 
c 
0 
..0 '-
C';l 
u 
2 1  - . .  -- .. 
20 
1 9  - D Branched 
0 Single-stemmed 
1 8  Adult 
1 7  
1 6  
7 1 7  27 37 47 52 
Node posit ion 
Figure 2 .5 Mean carbon isotope d iscr iminat ion values for leaves 
col lected at d ifferent nodes for s i ngle-stemmed and branched plants 
and from the canopy top of adult p lants (dashed l i ne) of Metrosideros 
excelsa 'Scarlet P impernel '  at the end of the experimental period . 
95 
2.4 Discussion 
The hypothesis that restriction of shoots to a single-stem accelerated vegetative phase 
change was supported by this study, whilst root restriction appeared to have no 
significant effect. The progressive transition in single-stemmed plants of Metrosideros 
excels a from juvenile to adult phenology for several optical and dimensional leaf 
parameters in typified homoblastic development (Goebel 1900) . It was remarkable that 
leaves at a particular node position in branched plants did not display the same level of 
ontogenetic development as those in single-stemmed plants. Most notable of these 
features, was the progressive acquirement of a downy tomentum in single-stemmed 
plants evident by the increasing whiteness (less shade) or 'mealiness' in the abaxial leaf 
surfaces. In contrast, leaf optical parameters in branched plants above nodes 1 5-20 
revealed a divergence in values for leaf saturation and hue away from those expressed in 
adult leaves. 
The use of an image analysis procedure provided an effective and novel method for 
quantifying the level of ontogenetic development with ascending node position. Prior to 
this study, there had been no reports in which homoblastic behaviour had been fully 
characterised. Therefore, this protocol was particularly suitable for this homoblastic 
species where a gradual and subtle transition between juvenile and adult leaf forms 
occur. This is in contrast to a considerable number of native New Zealand species or 
certain closely related genera (e.g. Eucalyptus) where heteroblastic leaf development is 
exhibited and phenotypic expression of leaf morphological features from different life?
stages are discernible (Day et al. 1 997 , James and Be11 200 1 ) .  
The employment of a shoot restriction regime to accelerate phase change was consistent 
with findings in other woody species (Davis 1 99 1 ,  Snowball et al. 1 994 ) .  Davis ( 1 99 1 ) ,  
for example, reported a decrease in  the length of the juvenile period in seed grown 
plants of Actinidia deliciosa (kiwifruit) following application of a shoot restriction 
regime. S imilarly, phase change occurred with increasing node position in single?
stemmed plants of Citrus (Snowball et al. 1994). In both these studies, however, phase 
change was assessed based on the ability of plants to flower and not on a quantification 
of the progressive change in vegetative characters exhibited along the shoot axis. 
Nonetheless, the increase in shoot length and accumulation of nodes with increasing 
96 
distance from the medium surface has also been reported to be a determinant of phase 
change in other woody species such as Solanum aviculare (J ames and Man tell 1997) 
and Elaeocarpus hookerianus (Day et al. 1 997). This is in contrast to M etrosideros 
where changes in leaf parameters are less abrupt between node positions .  
A common report with regards to juvenility in woody plants is that a minimum size 
(number of nodes) must be achieved before the juvenile phase is complete (Davis 1 99 1 ,  
Snowball e t  al. 1994) . In single-stemmed plants of M. excelsa the acquisition of adult 
leaf characteristics based on values for optical parameters were evident in leaves above 
node 50 where values were similar to those of adult leaves .  Similar node heights were 
also required for stabilisation of most morphological parameters on par with values 
recorded in adult leaves .  Thus, the requirement of vigorous growth for attainment of 
adult characteristics does not appear to be a requirement for ontogenetic development in 
M. excelsa. Interestingly, the size of the plant (linear distance from roots to shoot apex) 
also appeared to have little effect on vegetative phase change between single-stemmed 
and branched plants. Although leaf dimensional parameters did not differ between both 
sets of plants when measured at 280 mm, there were significant differences in colour 
parameters . Differences in morphological parameters, however, were further evident at 
higher node positions (or greater distance from roots) indicating a divergence in the two 
sets of plants rather than parallel development. 
In contrast to single-stemmed plants, most leaf morphological parameters in branched 
plants displayed a divergence away from values expressed in adult leaves, despite 
having a greater leaf, stem and root mass than that of single-stemmed plants . While 
single-stemmed plants showed an increasing similarity in leaf properties with adult 
plants as node position increased, branched plant appeared to show the opposite pattern. 
This pattern of development was typified by a reversion in colour (decrease in hue: 
yellowing) and in most size and shape morphological parameters above nodes 1 5-20 
towards values exhibited at lower nodes (nodes 5- 10) .  Although shading in this 
experiment was kept to a minimum, it is possible that the reversion in these qualities 
was attributed to the environment in which the developing leaves were being formed. 
Increasing lateral growth and the successive formation of developing leaves along these 
lateral branches might cause a self-shading response on more plastic juvenile leaves 
(Poethig 1 997, Day et al. 1 998). Tsai et al. ( 1 997) showed that the level of source-
97 
strength in leaves of tobacco was a determinant of the level ontogenetic development 
achieved. Therefore, a relative increase in the source-strength in the less shaded leaves 
of single-stemmed plants may have been adequate for developmental phase change to 
occur. 
Interestingly, container size did not appear to have an effect on leaf parameters in 
single-stemmed plants. Since root dry mass was significantly lower in single-stemmed 
plants compared to branched plants, an effect of container size would have been more 
pronounced in the latter plant type. Thus, i f  root restriction is associated with 
acceleration in vegetative phase change, then this phenomenon would have been 
expected to be marked in branched plants. This treatment, therefore, did not advance 
vegetative phase change in M. excelsa, although it may have an effect on the 
reproductive behaviour of adult plants (Zimmerman 1 972). 
Confinement of roots to small rooting volumes can affect plant performance (Webster et 
al.  1 997). Moisture and nutrient stress are often associated with a reduction in rooting 
volume (Webster et al. 1 997), although it is unlikely that either factor had an effect on 
plant growth in this study. This is in light of the fact that plants were irrigated with a 
more than sufficient level of water (four times daily for 1 5  minutes) and supplied at the 
start of the experiment with a standard nine-month slow-release fertiliser. Moreover, 
although leaves analysed from branched plants at the end of the experiment indicated a 
lower leaf nutrient concentration compared with single-stemmed plants, the absence of 
an effect of container size on shoot growth parameters and an associated reduction in 
shoot/root biomass and root dry weight confirmed that changes in root biomass were a 
result of root restriction, and not of moisture or nutrient stress associated with a small 
container size. Furthermore, although leaf mineral concentrations required for growth 
have not been reported in Metrosideros, comparable values regarded as adequate for 
growth have been reported in the closely related genera Leptospermum and Psidium 
(Reuter and Robinson 1 986). 
An examination of !1 values in leaves of single-stemmed plant showed a strong 
divergent pattern with increasing node position away from values expressed at lower 
node positions and away from those of adult plant leaves. This is in spite of the fact that 
leaves in single-stemmed plants were becoming increasingly more adult with increasing 
98 
node position. The use of stable isotopes from plant tissue can provide a long-term 
indication of plant performance (Farquhar et al. 1989), and has been strongly correlated 
with leaf gas exchange components (e.g. stomatal conductance) and features limiting 
carbon acquisition (Farquhar et al. 1 989, Ehleringer et al. 1993, Sparks and Ehleringer 
1 997). Intra-specific differences in ? values can vary between juvenile and adult plants, 
as previously acknowledged to occur in M. polymorpha (Cordell et al. 1 998) and 
various other woody species (Donovan and Ehleringer 199 1 ,  1 994) .  However, the 
decline in leaf ? values in single-stemmed plants away from those exhibited in adult 
plants appear anomalous, particularly since these leaves approach leaf characters 
exhibited by adult plants with increasing node position. It is possible that this 
phenomenon occurs only during phase change, and once a critical number of nodes have 
been attained, ? values may return to similar values exhibited in leaves of adult plants. 
If this is the case, then ? may be a reliable indicator of the level of phase change in 
Metrosideros. This supposition is evaluated further in Chapter 3, and discussed in 
relation to leaf gas-exchange processes during ontogenetic development in single?
stemmed plants of M. excelsa. 
In conclusion, this study supports the hypothesis that a shoot restriction regime 
effectively accelerates vegetative phase change in micropropagated plants of 
Metrosideros excelsa. As acknowledged by several authors (Zimmerman et al. 1 985),  
the attainment of foliage in single-stemmed that phenotypically resembles that of adult 
leaves does not imply a competence to flower. However, exposure to inductive 
treatments which might include the addition of plant growth regulators (e.g .  inhibitors 
of gibberellin biosynthesis) (Clemens et al. 1 995), could be expected to be successful 
once a suite of vegetative characteristics that comprise the adult phase has been attained. 
99 
Chapter 3 
Gas-exchange and carbon isotope discrimination characteristics 
during phase change in Metrosideros excelsa 
3.1 Introduction 
In woody perennials, developmental phase change from a juvenile to an adult vegetative 
state is frequently accompanied by changes in morphological and physiological 
properties of leaves (Hackett 1985, Poethig 1 990, Greenwood 1995, Day et al. 1 997) .  
Morphologically, this can encompass variations in leaf dimensions between ontogenetic 
phases (J ames and Bell 1 996). Anatomical features such as variation in leaf thickness, 
and trichome and stomatal properties can also vary between juvenile and adults leaves 
(Day et al. 1 997, James et al. 1 999, James and Bell 2000a). These factors, in turn, can 
influence gas exchange parameters within a leaf, such as limiting the diffusivity of 
external C02 into leaf tissue and affecting the loss of water vapour into the atmosphere 
(Choinski and Wise 1 999, Niinemets et al. 1 999, James and Bell 2000b) .  Thus, gas 
exchange relations are likely to be influenced by alterations to surface (tomentum), 
stomata and the intercellular properties associated with different phase-dependent states . 
Numerous plant ecology and physiology studies have employed the use of stable 
isotopes for assessing plant performance (Farquhar et al. 1 989, Donovan and Ehleringer 
1 99 1 ,  1 994; Poss et al. 2000) . Isotopic discrimination of 1 3C relative to 1 2C in plant 
material reflects the differential diffusivity and internal fractionation of C02 and, thus, 
has been correlated with a diversity of gas exchange components, including leaf 
conductance, transpiration, photosynthetic capacity and water-use efficiency (Farquhar 
et al. 1 989, Ehleringer et al. 1993, Sparks and Ehleringer 1 997) .  In many woody plant 
species, intra-specific differences in carbon isotope discrimination (?) values can vary 
between juvenile and adult plants (Donovan and Ehleringer 199 1 ,  1 994; Cordell et al. 
1 998).  In these and other related studies, ? is often reported to be higher in juvenile 
than in adult leaf tissue. Often associated with the higher leaf ? levels in j uvenile plants 
is an increase in several photosynthetic parameters such stomatal conductance, 
photosynthetic rate and stomatal transpiration, and an overall decrease in plant water-
100 
use efficiency (WUE) (Hansen 1 996, Cordell et al. 1 998). However, comparisons of 
plants at different ontogenetic phases have not been conducted under unstressed 
conditions and controlled environments, but rather under field and/or common garden 
conditions, where interpretation of findings can be confounded by the prevailing 
environmental conditions (Donovan and Ehleringer 1 99 1 ,  Hansen 1 996, Cordell et al. 
1 998). 
The homoblastic species, Metrosideros excelsa, exhibits a gradual transition between 
juvenile and adult vegetative characteristics during phase change (Cockayne 1 928, 
Dawson 1 968). The most notable characteristic of the adult phase is the acquisition of 
rounder leaves with a downy (hairy) tomentum on leaf abaxial surfaces, which is 
otherwise lacking in juvenile plants (Dawson 1 968). Previously, vegetative phase 
change in M. excelsa ' Scarlet Pimpernel' was characterised using leaf shape/dimension 
and optical parameters to define ontogenetic ageing (Chapter 2). Additionally, 
significant differences in .0. were found between ontogenetic states in M. excelsa as 
evidenced by leaf .0. values in juvenile plants undergoing vegetative phase change (along 
single-stemmed plants) diverging away from values exhibited by adult plants bearing 
leaves with adult morphology (Chapter 2) .  A similar response was observed for 
seedlings and plantlets of M. excelsa ' Scarlet Pimpernel' ,  which exhibited adult leaf 
morphology at the end of four months when grown under a controlled temperature 
regime of 24/ 16?C (day/night) (M. Sismilich, P.E. Jameson and J. Clemens, 
unpublished data) . A decline in leaf .0. values in juvenile plants undergoing phase 
change suggests an overall decrease in leaf stomatal conductance and an increase in 
WUE with increasing node position (Donovan and Ehleringer 1994, Hansen 1 996, 
Cordell et al. 1 998,  Rundel et al. 1 999). However, characterisation of physiological 
markers that may typify a phase-dependent state, e.g. photosynthesis, stomatal 
conductance and transpiration, have not been reported in M. excelsa. Moreover, 
whether or not the divergent pattern in .0. can serve as a reliable physiological marker of 
phase change merits further investigation. 
1 0 1  
To account for these observations it is hypothesised that: 
1) Juvenile plants of M. excelsa undergoing vegetative phase change exhibit decreasing 
D. because of changes in leaf morphology consistent with greater water use 
efficiency, notably more tomentose and rounder leaves. This may also be associated 
with a reduction in photosynthetic capacity with increasing node position. 
In addition, with regards to leaf age, interpretation of leaf D. data collected at the end of 
an experimental period would be limited, considering that leaves harvested along the 
axis of a single-stemmed plant at one point in time would be of different ages. Thus, an 
effect of age would be confounded with that of phase change. Therefore, an analysis of 
D. for leaves of the same age collected from given node positions at different times 
during an experimental period would provide evidence complementary to analysis of 
leaves collected at one point in time. Therefore, it was further hypothesised that: 
2) Changes in D. are not a function of leaf age, but are the result of changes in leaf 
conductance relative to photosynthetic capacity at the time of leaf expansion. 
To test these hypotheses, juvenile rnicropropagated plantlets of one cultivar of M. 
excelsa were grown as single-stemmed plants in order to accelerate vegetative phase 
change. Morphological and physiological changes in single-stemmed plants undergoing 
vegetative phase change were assessed using a complementary image analyses 
technique, and instantaneous (leaf gas exchange) and long-term integrated 
measurements (leaf D.) of photosynthetic capacity conducted under controlled conditions 
on three sampling times. Comparative results were assessed against leaves from 
rnicropropagated plantlets that were allowed to branch freely (same genotypic 
population as single-stemmed plantlets), and adult leaves of >3 year-old plants of the 
same cultivar. 
102 
3.2 Material and Methods 
3.2.1 Plant materials 
Plants of Metrosideros excelsa 'Vibrance' were obtained at two developmental stages. 
Adult plants (approximately 1 .5 m in height) supplied by Joy Plants, Hamilton, in 6 1 
bags in August 1 998 were transplanted into 30 I containers. S ixty micropropagated 
plantlets were obtained from Lyndale Nurseries Ltd. ,  Auckland, in 50 mm square plastic 
pots in April 2000 in a rejuvenated state, six months from exflasking. Plantlets were 
transplanted into 7 1 containers in early June, 2000. The potting medium consisted of a 
mixture of peat and pumice (80:20 v/v) supplemented with a control release fertiliser 
(4 .0 g of 8-9 month release Osmocote Plus Grace Sierra, Heerlen, The Netherlands). 
Plants were supplied twice daily with overhead irrigation and maintained under ambient 
outdoor conditions at the Plant Growth Unit (PGU), Massey University, Palmerston 
North. 
3.2.2 Greenhouse environment 
Adult and micropropagated plants were transferred in mid-June 2000, from outdoor 
(mid-winter) conditions into a heated greenhouse for the experimental duration. During 
this time, the greenhouse was vented at 26?C and warmed at night, when required, to 
ensure the minimum temperature was no lower than 14?C. Temperatures were 
monitored regularly using a Hobo shuttle logger (model H8, Scott Technical 
Instruments, Hamilton, New Zealand) to ensure that temperatures remained within the 
respective range. Photoperiod was extended to 14 h during the winter months (June ?
September) using overhead 100 W incandescent lights ( - 10 11mol m?2 s? 1 ) .  Plants were 
watered twice daily to container capacity using an automatic irrigation system with one 
microtube hose per plant container stemming from a central hose line. Four irrigation 
microtube hoses were fitted to containers holding adult plants. 
1 03 
3.2.3 Shoot treatments applied to greenhouse-grown plantlets 
In the third week after transference to the greenhouse ( 1 8  June 2000), pruning 
treatments were applied to micropropagated plantlets. In half of the plantlets, all side 
branches were removed and the tallest shoot was trained in an upright position and 
allowed to grow unchecked. Branch development in these plants (referred to as single?
stemmed plants) was prevented by pinching of axillary buds as these swelled before bud 
break. In the remaining plantlets (branched plants), the main stem was pinched once at 
the apical meristem region at approximately node 6-8 ,  and commencement of axillary 
branching was not restricted in any way. Plants were randomly assigned to one of five 
blocks within the greenhouse based on a completely randomised block design. Within 
each block, there were six replicates each of branched and single-stemmed plants and 
two replicate adult plants. The experiment was conducted over a 10 month period. On 
three occasions, a random subset of plants per block was transferred to a controlled 
environment room for approximately 9 days for leaf gas-exchange measurements 
(described in Section 3 .2.6) before transference back to the greenhouse. Diurnal leaf 
gas exchange measurements were also conducted at the end of the experimental period 
under ambient greenhouse conditions. 
3.2.4 Measurements made on greenhouse plants 
The height and the number of accumulated nodes was measured on micropropagated 
plants every 14 d. Shoot length was measured based on the linear distance from the 
medium surface to the shoot tip of the longest shoot. Measurements were also taken on 
the diameter of expanding buds in adult plants ( 10 replicate buds per plant) through to 
bud break (parting of the fourth pair of scaleslbracts) and during elongation of the 
subsequent shoot (shoot length and node number) . 
Before bud break (approximately 6 weeks after the start of the experiment), an 
additional ten buds from four bud size classes (<4.0, 4-5 .0, 5 . 1 -6.0 and 6 . 1 -7 .0 mm) 
were randomly harvested from adult plants. Buds were excised with a razor blade and 
individually transferred to 1 ml Eppendorf tubes containing a 90% formalin, 5%  acetic 
acid and 5% ethyl alcohol (FAA) fixative solution. The protocol for histological 
preparation of plant tissue follow Johansen ( 1 940) . Bud samples were placed under 
104 
vacuum for 48 h before being washed three times in 70% ethanol solution over 2 days. 
Samples were transferred to a wax infiltration procedure (described in Appendix I) 
before being finely sectioned ( 1 2  J..Lm) on a microtome (Leica RM2145,  Germany) . 
After sections were transferred onto glass slides and allowed to dry ( 48 h), they were 
transferred through a safranin and fastgreen staining series (Appendix II) before 
viewing under a light microscope. 
3.2.5 Controlled environment 
Gas exchange measurements were conducted in a controlled environment room at the 
New Zealand Controlled Environment Laboratory, The Horticulture and Food Research 
Institute of New Zealand (HortResearch), Palmerston North. The environmental 
conditions within the controlled environment room were kept constant and were the 
same during each experimental period. Lighting in the growth room was provided by a 
water-screened array of four high intensity main 1 kW Metalarc lamps and four 1 kW 
tungsten halogen lamps ( 1 2  h) providing a photosynthetic photon flux (PPF) of 700 
J..Lmol m?2 s? 1 . A photoperiod extension of 1 h was provided by four low intensity 1 50 W 
Tungsten auxiliary lamps with a PPF of 7 J..Lmol m?2 s? 1 . Thermoperiod coincided with 
photoperiod. Day/night air temperature and relative humidity within the environment 
room were maintained at 24/ 1 9?C and 87/86%, respectively. This corresponded to a 
vapour pressure deficit of 0.4/0.3 kPa. C02 concentrations were maintained at 
approximately 350 ppm. 
Within the room, plants were placed on one of two trolleys holding either three adult 
plants or 1 2  branched and single-stemmed plantlets (6 replicates per pruning type) .  
Micropropagated plantlets were elevated to 1 .5 m in  height corresponding to the canopy 
height of adult plants to ensure an even exposure to environmental conditions within the 
room. Each trolley and individual plants on a trolley were rotated daily to ensure an 
even light distribution over time and to reduce any positional effects created by the 
room. 
105 
3.2.6 Gas exchange measurements in the controlled environment 
Leaf gas exchange measurements were conducted within the controlled environment 
using an open photosynthetic system (LI6400, Li-Cor, Lincoln, Neb. USA). Gas 
exchange measurements were conducted on three separate occasions, occurring on 22-
30 September, 1-9 December 2000, and 23 February-3 March 200 1 ,  and referred to as 
'September' , 'December' and 'March' ,  respectively. During each sampling date, gas 
exchange measurements on all plants were conducted after an initial 48 h 
acclimatisation period. Leaf measurements were conducted on newly emerged ( <2 
months) but representative, fully expanded, whole leaves. In adult plants, this included 
leaves sampled from the top portion of the canopy. In plantlets, gas exchange 
measurements for respective sampling times, September, December and March, 
corresponded with new fully-expanded leaves for node position 10, 30 and 40 in single?
stemmed plants, and node position 10, 20 and 30 in branched plants. 
3.2.6.1  Light response curves 
Leaf photosynthetic light response curves were analysed according to Greer and 
Halligan (200 1 ) .  The selected leaf was placed in the Li-Cor Infra-Red Gas Analyser 
(IRGA) chamber. Chamber light intensity was provided by a Licor LED lamp. 
Chamber C02 was supplied via the LI 6400 C02 injection and scrubbing system and 
controlled at a concentration of 400 11mol C02 mor 1 at an air flow rate of 500 11mol s- 1 ? 
Photosynthesis was measured at 22?C starting at an initial PPF of 600 11mol m-2 s- 1 ? 
Once photosynthesis had stabilised, PPF was increased to 1 200 11mol m-2 s- 1 . The rate 
of photosynthesis (Ps) was then recorded in 10  steps of decreasing PPF until dark, when 
respiration was measured. Data were then analysed using a non-linear regression fitted 
to the following rectangular hyperbola equation: 
Ps = (Pmax * Tanh (PPF* <pappi Pmax.)) - Rs 
where, Tanh represents the hyperbolic tan function, <papp the apparent photon yield, Pmax 
the light saturated maximum rate of photosynthesis, and Rs the rate of respiration. To 
determine the maximum photosynthetic photon flux at 99% light saturation (PPFsar), the 
106 
following equation was used, where Atanh represented the inverse hyperbolic tan 
function: 
PPFsat = Atanh (0.99) * Pmax I <i>app 
Gas exchange measurements from the light response curves included stomatal 
conductance, transpiration rate and intercellular C02 concentration. From these 
measurements, a measure of instantaneous water use efficiency (WUE) within the plant 
was calculated based on the difference in the maximum rate of C02 uptake per unit of 
transpired water. 
3.2.6.2 C02 assimilation I intercellular C02 (A/Ci) curves 
Leaf photosynthetic measurements were conducted on the same equipment described 
above for obtaining the light response curves. The selected leaf was placed in the leaf 
chamber where light intensity was provided by the LED system and the PPF was 
maintained at a constant 600 !lmol m-2 s- 1 and an air flow rate held at 500 !lmol s- 1 . 
Photosynthesis was measured at 22?C starting at an initial C02 concentration of 400 
!liDO! C02 mor 1 . When photosynthesis had stabilised, the C02 concentration was 
decreased to 50 11mol C02 m or 1 and increased over nine steps to 900 !lmol C02 m or 1 
during which time photosynthesis was recorded.  Based on the NCi response curves, 
biochemically based equations describing the potential limits to photosynthetic capacity 
such as the maximum Rubisco (ribulose- 1 ,5-bisphosphate carboxylase/oxygenase) 
carboxylation rate (V cmax) and the maximum rate of RuBP (ribulose- I ,5-bisphosphate) 
regeneration mediated by electron transport (J max) were calculated, according to the 
method of W alcroft et al. ( 1 997). 
3.2.7 Leaf image analysis 
Image analyses were conducted on detached leaf samples harvested after each gas 
exchange sampling time, including additional samples taken from plants within the 
greenhouse. A total of ten leaf samples were collected from each plant type (one 
1 07 
representative replicate leaf per plant) . Leaf samples corresponded to the node position 
examined during a given gas exchange sampling date. 
Each leaf was cut from the plant at the base of the petiole and transported immediately 
in distilled water to the Image Analysis laboratory, Massey University. Petiole length 
was measured and excised before each leaf was imaged. Images were captured for 
determination of leaf shape, size and optical properties of abaxial and adaxial surfaces 
following the methodology described in Chapter 2. 
Morphological measurements were taken of the petiole length, basal angle and total leaf 
mass. Basal angle was measured with a protractor between the leaf margin and the mid?
vein at the base of the leaf blade. The specific leaf area was determined within 2 h of 
collection and dry mass was determined after drying at 80?C for 48-72 h. Leaf specific 
mass was determined by dividing the leaf dry weight by leaf area (obtained from the 
image analysis procedure). Leaf water content (%) was expressed as a function of the 
difference in fresh and dry mass, and as a function of total leaf area. 
3.2.8 Carbon isotope discrimination 
Collection of tissue for determination of carbon isotope discrimination (.6.) occurred at 
four separate times during the experimental period. In the first harvest, leaf tissue 
samples were collected from adult plants over an eight week period (July-August 2000) . 
Tissue samples included those of (a) fully-expanded old leaves (growth from previous 
season) which subtended the tight bud, (b) unexpanded leaves enclosed within a tight 
bud (mean 5 . 1  mm in diameter), followed by (c) partly and fully expanded young leaves 
harvested from the second node of a newly derived shoot. Mean length/width 
measurements of partly ( - 1 5  day old) and fully expanded (- 32 day old) leaves were 
45/20 mm and 62/29 mm, respectively. In the three subsequent harvests, dry leaf tissue 
for determination of 11 was obtained from all plant types. The leaf samples 
corresponded to the nodes examined following each gas exchange sampling date. The 
leaf tissues collected during all four harvests were dried at 80?C for 72 h. Subsequent 
processing and determination of 11 of the dried samples followed the protocol described 
in Section 2 .2.7 . 
1 08 
3.2.9 Gas exchange measurements under greenhouse conditions 
The daily photosynthetic response of leaves under greenhouse conditions was 
characterised over two consecutive sunny (< 10% cloud cover) days occurring on 20-2 1  
March 200 1 .  Gas exchange measurements were conducted on terminal leaves from 
three adult plants and on branched (node 30) and single-stem (node 40) 
rnicropropagated plants . Measurements were started at 0600 h and ended at 1 800 h with 
observations recorded at 2 h intervals .  Gas exchange measurements were conducted 
with a LI6400 JRGA attached to a leaf chamber equipped with a transparent leaf 
chamber cover. Measurements were conducted under ambient C02 temperature and 
irradiance conditions within the greenhouse. 
3.2.10 Leaf carbohydrate analyses 
Leaf samples were collected between 0600-0630 h on 25 June 200 1 ,  from the three sets 
of plants growing in the greenhouse for determination of soluble sugars (sucrose, 
fructose and glucose) and starch concentrations. Extraction and determination of soluble 
sugars and starch are described below. Leaves sampled included those that were fully?
expanded from nodes 1 0  and 40 in single-stemmed plants, node 30 in branched plants 
and from the canopy top of adult plants .  Leaf samples consisted of eight to 1 2  
replicates per node or plant type. Carbohydrate measurements were conducted on a 
Hitachi U-2000 spectrophotometer using 1 .5 rnl curvettes containing the sample 
solution. 
3.2.10.1 Soluble sugars extraction and determination 
Each leaf was cut from the plant and immediately snap-frozen in liquid nitrogen and 
stored on wet ice in tin foil sachets. Samples were transferred to the laboratory and 
freeze dried for 72 h before being finely ground to a powder ( <200 11rn). Samples ( 40 
rng dry weight) were extracted in 2 rn1 62.5% (v/v) methanol in a water bath at 55?C for 
1 h while being vortexed every 1 5  min .  Samples were centrifuged for 5 rnin at 3000 
rpm and the pellet was retained for starch analyses. Chlorophyll, polyphenolics and 
other interfering substances were precipitated out with the addition of 10 111 of saturated 
109 
lead acetate per 400 Jll of supematant (Haslemore and Roughan 1 976).  The solution 
was centrifuged ( 1 5  min at 3000 rpm) and the supematant containing the soluble sugars 
was decanted and used in the assay. Soluble sugar concentrations (mol/g dry weight) 
were determined using Sucrose/D-fructose/D-glucose enzyme kits (Boehringer 
Mannheim Biochemicals). 
a) Determination of D-glucose and D-fructose 
A sample of supematant (30 Jll) was pipetted into a 1 .5 ml curvette well containing 400 
Jll of rnilli-Q water and 1 52 Jll of solution 2 (triethanolamine buffer, pH 7 .6, NADP 2 .4 
mg/rnl, ATP 5 .8  mg/rnl),  lightly vortexed (5 sec) and incubated for 3 rnin at room 
temperature before obtaining an absorbance reading (A1 ) at 340 nm. After adding 20 Jll 
of a six-fold dilution of solution 3 (hexokinase and glucose-6-phosphate dehydrogenase) 
to the curvette, samples were lightly vortexed (5 sec) and kept at room temperature for 
15  min to complete reactions before a second absorbance reading (A2) was obtained at 
340 nm. A six-fold dilution of solution 4 (20 Jll) (phosphoglucose isomerase) was 
added and a final absorbance reading (A3) was taken at 340 nm after 1 5  rnin at room 
temperature. A standard curve was constructed from which sample amounts were 
calculated. Determination of D-glucose concentration was calculated by A2 - A1 and D?
fructose concentration from A3 - A2. 
b) Determination of sucrose 
A sample of supematant (30 Jll) was pipetted into a 1 .5 rnl curvette with 30 Jll of 
Solution 1 (?-fructosidase). Sample solution was lightly vortexed (5 sec) and incubated 
for 1 5  min at 25?C. Solution 2 ( 1 52 Jll) and 400 Jll rnilli-Q water were added and 
lightly vortexed (5 sec) .  Absorbance (B 1 ) was obtained at 340 nm after incubating for 3 
min at room temperature. A six-fold dilution of Solution 3 (20 Jll) was added, and after 
1 5  rnin, a second absorbance reading (B2) was taken at 340 nm. Sample contents were 
determined from a standard curve. Sucrose concentrations were then calculated by B2 -
B J .  
1 10 
3.2.10.2 Starch extraction and determination 
The sample pellet from the soluble sugar extraction procedure was placed in a 1 5  rnl 
falcon tube to which 0.5 ml 8M hydrochloric acid (HCL) and 2 ml dimethyl sulphoxide 
(DMSO) was added. The solution was vortexed until the pellet was resuspended. The 
sample solution was heated at 60?C for 1 h in a water bath while being vortexed every 
1 5  min .  After being cooled on ice for 5 min. ,  0.5 rnl NaOH was added to the sample 
solution and then vortexed for 20 sec . The solution was made up to 10rnl with sodium 
citrate buffer (pH 4) and the supematant was clarified by centrifuging at 2600 rpm for 5 
rnin before being transferred to a fresh 1 .5 rnl curvette in preparation for the assay. 
Starch concentration (mollg dry weight) was determined using a starch enzyme kit 
(Boehringer Mannheim Biochernicals). 
a) Determination of starch 
A sample (50 Ill) was pipetted into a 1 .5 rnl curvette and mixed with 40 !-ll of Solution 1 
(amyloglucosidase). The curvette was incubated for 20 rnin at 60?C for 1 5  rnin before 
200 111 of Solution 2 (triethanolamine buffer, pH 7 .6, NADP 75 mg, 190 mg ATP and 
300 111 of rnilli-Q water was added. The solution was lightly vortexed (5 sec) and after 
3 rnin absorbances of the solution (A1 ) were read at 340 nm. Solution 3 (20 Ill) 
(hexokinase, glucose-6-phosphate dehydrogenase) was added to the solution and after 
completion of the reaction ( 1 5  rnin), absorbances of the solution (A2) were read at 340 
nm. The sample contents were determined from a standard curve. Starch concentrations 
were then calculated by A2 - A 1 . 
3.2. 1 1  Statistical Analyses 
Data analysis was performed using the SAS statistical package (SAS Institute, Cary, 
N.C. ,  USA). A repeated measure Analysis of Variance (AN OVA) was used to compare 
differences in plant height, node position and internode length, with the repeated factor 
being time. Statistical comparisons of leaf shape, size, colour, carbon isotope 
discrimination, morphological and photosynthetic parameters were conducted using a 
two-way ANOVA for comparisons of the main effects of time (session date) and plant 
1 1 1  
type. Comparison of leaf gas exchange parameters from different plant types measured 
in the greenhouse over a 1 2  h diurnal period was conducted using a one-way ANOVA, 
with comparison of plant types analysed within each 2 h interval . Differences between 
carbohydrate samples were analysed using a one-way ANOV A. In all analyses, 
differences between treatments were assessed using a Least Square Means (LSM) 
multiple comparison test. Where appropriate, data were log-transformed to normalise 
variances and means were compared for significant differences at the 5% level. 
3.3 Results 
3.3.1 Characterisation of shoot growth 
The average bud diameter in adult plants at the start of the experiment ( 1 6  June 2000) 
was 5 . 1 mm. Mean bud size increased significantly with time (P<0.000 1 )  at a rate of 
0.5 mm per 14  d at mean greenhouse temperatures of 22/1 8?C (day/night) over the first 
eight weeks (P<0.000 1 ) .  By week 8, 96 % of buds had broken as emerging vegetative 
shoots elongated. Shoot length and node number continued to increase significantly 
over the duration of the experiment (P<0.0001 )  (Figure 3 . 1  A and B) .  
There was a significant effect of shoot restriction treatment on the length and number of 
nodes produced in micropropagated plants over the experimental period (P<0.000 1 ) .  
Shoot length and number of accumulated nodes in single-stemmed plants was 
significantly greater than in branched plants after week 1 6  (shoot length) and week 4 
(node number) (P<0.05) (Figure 3 . 1  A-B) .  By the end of the experimental period, 
shoots from single-stemmed plants were longer and bore a greater number of nodes 
( -98 cm and 42 nodes) than branched plants ( -48 cm and 30 nodes) .  
In adult plants, the mean internode length (averaged across the entire stem length for 
each record) showed a cyclic pattern in shoot growth (Figure 3 . 1  C) .  The highest rates 
of internode elongation occurred between weeks 1 2- 1 8  and 22-28,  with intermittent 
periods of 6-8 weeks during which time internode length did not increase significantly 
(weeks 1 8-22, 29-36) (P>0.05) .  Observations of terminal shoot apices during the first 
intermittent period where internode length did not increase (week 1 8-22) corresponded 
1 1 2 
1 20 - -- - - - - - - - - -- -- .... 
100 
- 80 E u '--' 
.:;:: 60 
Cl) 
Q) 
:::r:: 40 
20 
0 
50 1 
45 
40 
35 Q) .D 30 E 
:::l 25 ? c I Q) -a 20 j 0 
z 1 5  
1 0  ? 
5 
0 
2 .4 
E' 2.2 2 
-5 2.0 -Cl) c Q) 1 .8 
Q) -o 0 c .... Q) 
E 1 .4 ....... 
1 . 2 
A 
__._ Single-stem 
Branched 
Adult  
B 
c 
- v-
1 .  0 +--,---..,-----,---,c--,---.----,--r--,--.---;---...,..--,-?-,----..,----,----
0 1 0  20 30 40 
Weeks 
Figure 3 . 1  Changes in mean (?SE) plant height (A) , node number (B) 
and in ternode length (C) of s ingle-stemmed, branched and adul t  plants 
of Metrosideros excelsa 'V ibrance' over the experimental period. 
1 1 3  
to a time when a high percentage of terminal shoot apices had formed a 'resting bud' 
(77%) , and a further 2 1 %  had aborted. 
In micropropagated plants, there was a significant effect of time (week) on the mean 
internode length (P<0.00 1 )  (Figure 3 . 1  C). In branched plants, the overall mean 
internode length initially declined with increasing node position and remained relatively 
constant after week 8 (mean internode length - 1 .6 cm between weeks 1 0-30). In single?
stemmed plants, the overall mean internode length increased with each node attained 
after reaching node 20 (week 20), becoming increasingly longer than in branched plants 
after week 22 (P< 0.05) .  Additionally, internode growth in single-stemmed plants 
above node position 20 increased for the remainder of the experiment, with intermittent 
periods during which time internode length did not increase significantly (e.g. between 
weeks 22-26, 30-32) (Figure 3 . 1 C). A maximum mean internode length of 2 .4 cm 
was recorded upon reaching 42 nodes. 
3.3.2 Leaf development in adult plants 
A histological examination of buds indicated that leaf meristems in quiescent buds from 
adult plants were produced continuously within the apical meristem region (Plate 3 . 1 ) . 
There was a strong positive correlation of bud size (diameter) with the number of 
newly-formed pairs of leaves (R2 = 0.78 1 ,  P < 0.000 1) .  Buds from the smallest 
diameter size class ( <4.0 mm) averaged 0.8 pairs of leaves, whereas large buds ( 6. 1-7.0 
mm) averaged 8 pairs of leaves. 
There was a significant difference in ? of leaf tissue at different development stages of 
expansion (P<0.001 )  (Figure 3 .2). Discrimination in leaf tissue was significantly higher 
in partly and fully expanded leaves (pooled mean 20.3%o) compared with bud tissue 
( 1 8 .6%o) (P<0.05) .  Isotope composition in older subtending leaves (previous growing 
season) did not differ from that of recently expanded leaves (both 50% and 1 00% 
expanded leaves) or newly formed bud tissue (P>0.05). 
1 14 
Plate 3 . 1  Cross-section of a vegetat ive bud (diameter before harvest: 4 .2 mm) 
i l l us trat ing the cont inuous formation of new leaf primord ia  from the ap ical meristem. 
Bar = 500 flm. A =  Apical meristem. L = Leaf I bract. S = Scale.  
1 1 :;  
"""" C) 
22 
21 -+ 
I ? 
c 20 -l-
. 2  I 
? I E 1 9  _j_ '- I 
tJ 8 
0 U1 
c I 
? : : I 
Bud 50% expanded 1 00% expanded Subtend ing  l eaf 
Leaf s i ze 
Figure 3 .2  D ifference i n  carbon isotope d iscrim ination levels (mean ? SE) from bud and 
l eaf tissue in adul t  p l ants of Metrosideros excelsa 'Vibrance' col lected in Ju ly - August 
2000. Tissue samples : t ight bud, 50% and 1 00 % expanded leaves of current season's 
growth and subtendi ng leaf from the prev ious season's growth . 
1 16 
3.3.3 Image analysis of leaf dimensional and optical properties 
Imposition of a shoot restriction treatment on micropropagated plantlets of Metrosideros 
excelsa 'Vibrance' effectively accelerated vegetative phase change. Leaves in single?
stemmed plants initially showed no significant differences from those in branched and 
adult plants in shape or size characteristics (node position 1 0  during September). At 
intermediate node positions (node position 30 during December), they tended to diverge 
away from leaf qualities expressed in juvenile as well as those of adult plants (Figure 
3 .3 A-F) . This included a progression towards larger leaves, in terms of length, width, 
total leaf area and perimeter, with increasing node position (P<O.OOO l )  (Figure 3 .3  A?
D). By node position 40 (March), however, leaves in single-stemmed plants began to 
approach qualities expressed in adult plants, resembling both the characteristic 
roundness and size parameters exhibited by adult leaves. 
In contrast, branched plants in general did not differ significantly between sampling 
dates from adult plants with respect to leaf width and area (P>0.05) (Figure 3 .3  B-C) . 
However, with each subsequent sampling date (or increasing node number), leaf 
parameters such as length, length/width ratio, perimeter and leaf roundness in branched 
plants showed a divergent pattern away from that of adult leaves (significant interaction 
of time x plant type: P<O.OOO 1 ) . Leaves of branched plants therefore became less oval 
and more elongated with increasing node number (Figure 3 . 3  A, D-F). 
Relative to adult plants, an assessment of leaf colour parameters on the abaxial surface 
of leaves in micropropagated plants provided an indication of the degree to which 
vegetative phase change (acquisition of white-coloured tomentum) occurred. There was 
a significant interaction of sampling date and plant type on hue and saturation values 
(P<0.05),  and only a main effect of plant type on values for leaf lightness (P<O.OOO l )  
(Figure 3 .4 A-C). Generally, leaf saturation and hue values in single-stemmed plants 
approached those of adult plants by March, whereas those of branched plants either 
remained constant or diverged away from values exhibited by leaves of adult plants 
(Figure 3 .4 A-B) .  Leaf lightness in single-stemmed plants, however, did not differ from 
that in branched plants during the three sampling dates, although i t  did show a trend 
towards the values for adult plants by node 40 (March) (Figure 3.4 C). 
1 1 7 
1 00 
90 
----A- ?---- - - -+-!':??;::? ?--- B 
Adult  ' 
I I 
- - 50 
45 
E 
40 s 
..c: ..... 
-o 
35 ? 4-?l Q) 
......l 
30 
50 ?------------------------? ?--------?------?------? 25 
3000 - -- ------- - --- -???-?- --???--c 
2500 
' r?-
1 D 
- - 1 250 
? 200 ] N E 
s "' 2000 
Q) ..... "' ....... "' 
Q) ......l 1 500 
1 000 
3.0 
E 
0 
c.:; :.... 
-5 2.5 
"Cl 
? 
..c Oo 
t: 
Q) 2.0 -
..._ "' Q) ......l 
1 .5 
-- - - - - --- --- -1 
T I 
Sept. Dec. Mar. 
Date 
F 
1 00 
20 
1 8  
1 6  
r--------.--------.-------? 1 4  
Sept. Dec . Mar. 
Date 
Figure 3 .3 Changes in d imensional properties (mean ? SE) during success ive 
Vl Vl Q) t: -o t: :::l 0 ..... 
4-(<) Q) ......l 
samp l i ng dates for leaf abaxial surfaces in s ingle-stemmed, branched and adu l t  pl ants of 
Metrosideros excelsa 'V ibrance' . A : length. B : width. C: area. D :  perimeter. E :  length I 
width rat io. F: roundness. Sampl i ng dates for September (Sept) , December (Dec) and 
and March (Mar) correspond to nodes 1 0 , 20 and 30 in  branched p lants and nodes 10 ,  
30  and 40 i n  s i ngle-stemmed plants ,  respectively .  Leaves in adu l t  p lants harvested 
from canopy top. See Materials and Methods for specific sampl i ng days .  
1 1 8 
V 
::l 
...c 
'+-
C<l 
V 
.....l 
? 
c: 0 .... 
C<l 1-
::l ..... 
C<l (/) 
'+-
C<l 
V 
.....l 
? 
(/) (/) 
V 
E ...c Cl) 
'+-
C<l 
V 
.....l 
45 
40 
35 
30 
25 
20 
: ? l 
16  
14  
1 2  
1 0  
8 
A 
B 
_..._ Single-stem 
Branched 
- - Adu lt  
? 
-? --
-- -- - - - - -- - l 
?-?---- - 1. 
6 +------------?-------------.------------?
65 - - ???-------- ---? - - ?--- --? ? - - - -? - -- - -
60 
55 
50 
45 
40 
35 
30 
c 
September December 
Date 
--- - l 
March 
Figure 3 .4 Changes in optical properties (mean ? SE) between success ive 
sampl ing dates for leaf abax ia l  surfaces i n  s i ngle-stemmed, branched 
and adu l t  p lants of Metrosideros excelsa 'V ibrance' . A: hue. B :  saturation 
C:  l ightness .  Sampl i ng dates for September, December and March 
correspond to nodes 10 ,  20 and 30 in b ranched p lants and nodes 1 0 , 
30 and 40 i n  s ingle-stemmed p lants, respective ly .  Leaves i n  adu l t  p lants 
harvested from canopy top .  See Materia l  and Methods for specif ic 
samp l i ng days. 
1 1 9 
3.3.4 Leaf morphological attributes 
In micropropagated plants, leaf morphology varied depending on the sample date and 
plant type examined (Table 3 . 1 ) . In single-stemmed plants, leaf petiole length increased 
with increasing node position, reaching a similar length by node position 40 as recorded 
in leaves of adult plants (pooled mean 7 .3  mm in March). Leaf petiole length in 
branched plants also showed a similar trend to that of single-stemmed plants, although 
this was still significantly shorter than petioles in adult plants by March (node position 
30 in branched plants). Values for leaf angle in single-stemmed plants were generally 
smaller at higher nodes compared with leaves of adult plants, whereas the opposite 
relationship occurred for leaves in branched plants. The total fresh and dry mass per 
leaf was generally higher in single-stemmed plants at higher nodes (node position 30-
40) in comparison with leaves of branched and adult plants. Leaf specific mass in 
branched and adult plants increased between the first and last sampling date, whereas no 
trends were apparent in single-stemmed plants. Leaf water content (L WC) generally 
varied depending on plant type and sampling date. Percent LWC per given mass in 
micropropaged plants decreased with increasing node position, reaching similar values 
for adult leaves by March. L WC per given area in leaves of single-stemmed plants 
primarily was significantly higher than leaves of branched and adult plants during 
December and March. 
3.3.5 Carbon isotope discrimination 
Measurements of isotope composition taken during the three sampling dates revealed 
contrasting trends in the levels of leaf carbon isotope composition (Figure 3 .5) .  There 
were significant main effects of sampling date (P<O.OOO l )  and plant type (P<O.OOO l )  on 
the level of isotope discrimination but no significant interaction between these factors 
(P>0.05).  Mean carbon isotope discrimination levels in leaves of single-stemmed plants 
during the second and third sampling dates (December and March) did not differ, but 
were significantly lower than values recorded during the first date (September). The 
mean values for leaves in single-stemmed plants during the third sampling date were 
also lower than those recorded for branched and adult plants. In both adult and 
branched plants, mean carbon i sotope discrimination values did not differ within and 
1 20 
,..... 
10 ,..... 
Table 3 . 1  Comparison of leaf characteristics in single-stemmed, branched and adult plants of Metrosideros excelsa 'Vibrance' .  
Means shown for each plant type collected at the end of each sampling date (Sept = on 22 - 30  September, Dec = 1 - 9 December 2000, and 
Mar = 23 February - 3 March, 200 1 )  correspond to leaves on nodes 10, 30 and 40 (shown in parentheses) in single-stemmed plants and 1 0, 20 
and 30 in branched plants, respectively. Analysis for variables in each row conducted using a two-way ANOV A for assessing the effect of plant 
type (P), sampling time (T) and their interaction (PxT). Mean separation in rows by Least Square Means tests, 5% significance level. 
Single-stemmed 
Parameter P-value Date (node) 
SeQt ( 1 0) Dec (30) Mar ( 40) 
Petiole length (mm) PxT 4.i 5 .6dc 6.7bc 
<0.000 1 
Leaf basal angle (0) PxT 36.9b 23 .4e 22.8c 
<0.05 
Leaf fresh mass (mg) PxT 783 . 1  c 1 6 1 6.9a 1 234. 1 b 
<0.000 1 
Leaf dry mass (mg) PxT 236. 1 c 46 1 .9a 447.8a 
<0.00 1 
Leaf specific mass P, T<O.OOO I 0. 19ab 0 . 17a 0.25a 
(mg cm-2) 
64.7bcd Water content - mass PxT 69.9a 7 l .6a 
(%)  <0.000 1 
- area P<O.OOO I 42.9ab 43 . 1  a 44.9a 
(% )  
Branched 
Date (node) 
SeQt ( 1 0) Dec (20) Mar (30) 
4.9cl 4.9el 5 .9cd 
42.9a 36.2b 34.2bc 
53 1 .2d 795.5c 825.0c 
1 52 .2d 245 .3bc 295 . 1 be 
0. 1 3c 0. 16de 0.20bc 
7 1 .4a 69.4a 64.3c 
32.5b 36.9b 36.4b 
Adult 
Date 
SeQt Dec Mar 
7.6ab 8 .3a 7 .8ab 
35 . 1 b 30.0cd 28.6d 
7 1 6.8c 829 .8c 786.9c 
264.9bc 28 1 .4bc 305 .4b 
0.20bc 0. 1 8cd 0.24a 
63.0cd 66. 1 b 6 1 .2d 
34.9b 36. 1 b 37 .4b 
23 - -- -------- .. .. --.. -?------?  -- ----?- -- ?- -- - - -- ---- - ---- -- -. 
c:: 
0 2 1  ...., ? c:: 
Single-stemmed 
D Branched 
D Adult  
E 2 0  
1-.. u 
(/) 1 9  u 
Q) 
g- 1 8  ...., 
0 (/) 
c:: 1 7  
0 _o 
? 1 6  
u 
September December March 
Date 
Figure 3 . 5  Comparison of leaf carbon isotope compos ition (mean ? SE) in s i ngle?
stemmed, branched and adult  pl ants of Metrosideros excelsa 'Vibrance' . Means 
shown for each p lant type col lected immed iate ly after leaf gas exchange measurements 
conducted on three success ive sampl i ng dates. Specific dates described in Materia l  
and Methods. 
2 2  ,--- -??------?- - -
? 2 1  ? 
c:: 
20 J 0 ...., ? c:: E r 
1-.. I u I C/l 1 9  -1 "0 I Q) 0.. I 0 
0 1 8  C/l 
c:: ' 0 I 
..0 1 7  l 1-.. ? 
u I 
I 1 6  
0 5 1 0  1 5  2 0  2 5  3 0  3 5  
Node posi tion 
4 0  
- l I 
45 5 0  5 5  
Figure 3 . 6  Comparison of carbon isotope composit ion (mean ? SE) from l eaves 
harvested at d i fferent  node pos i tions for s ing le-stemmed (red l i ne) and branched (b lue) 
p lants w ith the mean value for adu l t  plants (co l l ected from canopy top) of Metrosideros 
excelsa 'Vibrance' col lected at the end of the experimental period (March - May 2 0 0 1 ) .  
1 22 
between plant types during the second and third sampling date, although values from the 
second date were lower in comparison with the first. 
During the third sampling date, carbon isotope composition was examined at different 
node positions in single-stemmed plants, and compared with values obtained from 
leaves of branched and adult plants. As node position increased in single-stemmed 
plants, carbon isotope values decreased, diverging away from values recorded in leaves 
for adult and branched plants (P<0.001 )  (Figure 3 .6). Carbon isotope discrimination 
values at node 40 in single-stemmed plants decreased by approximately 1 2% compared 
with earlier values recorded at node 10 .  Additional leaf samples collected at node 
position 50 in single-stemmed plants also showed a continuation in this pattern of 
decreased carbon isotope discrimination values with increasing node position. 
In single-stemmed plants, there was no significant effect of leaf age on the level of 
carbon isotope discrimination, based on an assessment of leaves collected from different 
nodes (ages) during the third sampling date with corresponding nodes collected during 
previous sampling dates (P>0.05) (Figure 3 .5  and 3.6) .  Therefore, the older leaf 
samples collected in March from nodes 10 and 30 did not differ from corresponding 
node positions harvested in September and December, respectively, when leaves were 
relatively younger. 
3.3.6 Relationship of carbon isotope discrimination and leaf characters 
The number of leaf characters that were correlated with ? values varied depending on 
the plant type (Table 3 .2) .  In single-stemmed plants, all pre-defined features describing 
the size and shape characteristics of leaves were strongly correlated with ? values. For 
instance, leaf ? values decreased as leaves became rounder and smaller in size 
resembling a size and shape characteristic of adult leaves. Similarly, ? values also 
decreased as the abaxial leaf surface became relatively lighter, indicative of an increase 
in the amount of tomentum accumulated. In branched plants, relatively few colour and 
size/shape parameters were correlated with t:. values. In adult plants, leaf morphological 
parameters were relatively constant with time and showed no relationships with ? 
values, whereas only one colour parameter (leaf saturation) showed a negative 
correlation. The most notable differences in leaf mass and water content characteristics 
123 
-
N 
-!'> 
Table 3 .2 Correlation of leaf dimension, colour, mass and water content variables with leaf carbon isotope discrimination values in single?
stemmed, branched and adult plants of Metrosideros excelsa 'Vibrance' . Pearson's correlation coefficient (r) and significance test (P) shown for 
each plant type were pooled for data collected at the end of three sampling dates (22 - 30 September, I - 9 December 2000, and 23 Febmary - 3 
March, 200 1) .  Mean at 5% significance level. Symbol :  n.s. not significant. 
Leaf characteristic Single-stem Branched Adult All plants 
Size and shape ??-?? _ ?-- r r p r p 
Basal angle (0) 0.546 0.002 0.444 0 .014 n.s. 0.599 0.000 
Petiole length (mm) -0.575 0.00 1 n.s. n.s. -0.235 0.026 
Length (mm) -0.386 0.035 -0.625 0.000 n.s. -0.38 1 0.000 
Width (mm) -0.6 19  0.000 n.s .  n.s. -0.5 1 8  0.000 
Length/Width Ratio 0.370 0.044 -0.580 0.00 1 n.s. n.s. 
Perimeter (mm) -0.540 0.002 -0.47 1 0.009 n.s. -0.487 0.000 
Area (mm) -0.646 0.000 n.s. n.s. -0.558 0.000 
Roundness 0.464 0.0 1 0  -0.637 0.000 n.s. n.s. 
Colour 
Hue 
Saturation (%) 
Lightness (%) 
Mass and Water Content 
Total leaf dry mass (mg) 
Specific mass (mg I cm-2) 
Water content - mass (%) 
- area (%) 
0.438 
0.464 
-0.500 
-0.4 1 9  
n.s .  
n.s. 
n.s. 
0.050 
0.039 
0.025 
0.02 1 
n.s. 
n.s .  
n.s .  
-0.676 
-0.654 
-0.626 
-0.647 
0.429 
-0.579 
0.00 1 
0.002 
0.000 
0.000 
0.0 1 8  
0.00 1 
n.s. 
-0.444 0.050 
n.s. 
n.s. 
n.s. 
-0.552 0.002 
n.s. 
-0 .5 16  
-0.236 
-0.380 
n.s. 
n.s. 
n.s. 
0.000 
0.025 
n.s. 
0.000 
were in branched plants, showing increases in leaf mass and water content per given 
area as values for ? decreased. 
In summary, leaves from single-stemmed plants displayed morphologically and 
anatomically similar characteristics to those of adult plants as node position increased, 
such as a transition to smaller and rounder leaves with tomentum on abaxial surfaces . 
This was also associated with a decrease in ? values with increasing node position. 
3.3. 7 Photosynthetic response and carbon isotope discrimination relationship in 
plants under controlled conditions 
Gas exchange measurements made on leaves on the three sampling dates revealed 
contrasting responses to changes in PPF. There was a significant interaction of plant 
type and sampling time on both Pmax and <papp (P<0.05) (Figure 3.7 A-B, Figure 3 .8) .  
Generally, in comparison with leaves of adult plants, leaf values for P max and <papp in 
micropropagated plants decreased at successive sampling dates, being significantly 
lower than values for adult leaves, especially at node 40. However, PPFsat levels were 
similar in leaves from all plants (mean 1 0 1 8?53 11mol m-2 s- 1 ) ,  irrespective of plant type 
and sampling date (P>0.05) (Figure 3.7 C). 
Based on analyses from the NCi curves (see Figure 3 .9), there was a declining trend 
over time in values for rnicropropagated versus adult plants. There were significant 
main effects only of plant type and time (P<0.05) on values of Rubisco carboxylation 
activity (Vcmax) and maximum electron transport capacity CJmax), being primarily lower 
in single-stemmed and branched plants compared to adult plants on the March sampling 
date (Figure 3.7 D and E). However, in both types of micropropagated plants, mean 
values for Ycmax and lmax showed a declining (although not significant) trend with time. 
The general decline in these values was also reflected in the lower C02 assimilation 
curves for branched and single-stenlined plants compared with adult plants in March 
(Figure 3 .9). Respiration rate increased with time in single-stemmed plants and 
remained relatively constant over time for branched and adult plants (significant time 
and plant type interaction, P<0.05) (Figure 3 .7 F). 
1 25 
20 
- l s  
' Cl)  
"' E 
0 10  E :i 
.._,. "' 
? 5 
A 
?- t ? 
---Single-stemmed 
--- Branched 
Adul t  
-
-:V) 0.06 
"' 
E 
0 
E 
-.::;. ? 0.04 
Q) 
>-. 
c 
3 
0 .g_ 0.02 -
...... c Q) I?CI:) 
B - ?-l 1 600 - 1 400 Cl) 
"' E 1 200 
N 
0 
u 1 000 -
0 
1 soo 
? 
Q 600 -
? 0... 
"'.-...?-- -- ? ?--......_-___ .., 
c 
0 r-----?----?----? 
0.. 0.. -< 0.00 400 +-----,-------,------, 
60 -
Cl) 
"' 50 -E 
0 E 3-
?40 E >u 
30 
D 
- ? -- 1 
1 80 
Cl) 
"' E 
Cl) 6 1 40 
1-
u Q) 
Q) 
0 E 1 00 
3-
>< 
Cl:l E 
-, 60 
E 
0.0 
Cl) 
"' E_z.o 
0 E :i '-' 
c 0 
-?-4.0 l?
a.. Cl) 
Q) 
0::: 
-6.0 
F 
-(,) 
September December March September December March September December March Date Date Date 
Figure 3 .7  Comparison of vari ab les from l ight response and C02 ass imi lat ion curves for leaves in s ing le-stemmed, branched and adu l t  plants 
of Metrosideros excelsa 'V ibrance ' .  Means (?SE) shown for each plant type col l ected during three sequential sampl ing dates in a contro l led 
13 env i ronment correspond to leaves on nodes 1 0, 30 and 40 in s ing le-stemmed plants and 1 0, 20 and 30 in branched plants, respective l y .  
0\ 
1 6  - --- -- -???-?-- --- ----??--?? ?  - -??? ------ - --????--
1 4  A 
Q) _...-._ 1 2  
? "7 Cll 1 0  '--u ":' 
....... E 8 Q) ...c N ...... 0 s:: 
>. u 6 Cll 
3 0 0 E 4 ...c: ::t 0.... ___.. 
2 
.: 
/ 
0 
-2 ? -----.-------.-------.------?-------,------? 
1 6  -r-- --
? :' 
? ; j\ B 
e . Cil 
u ":' 1 0  ?;::; E Q) 8 -5 N 
s:: 0 
>. u 6 Cll 
3 0 4 o E 
t: 2- 2 
1 6  
1 4  
Q) _...-._ 1 2  ;; "7 Cll '-- 1 0  u ":' Q) E 8 ...c N ....... 0 s:: u 6 >. 
Cll 
3 0 4 0 E ...c ::i. 0.... ___.. 2 
0 
-2 
0 
- --------------?-?-??? ---------? ?-- --- --- -----
----------------- ?-- - -
c 
200 400 600 
Single-stem 
0 Branched 
0 Adult 
800 1 000 
Photosynthetic photon flux (?-tmol m-2 s" 1 ) 
1 200 
Figure 3 .8 Mean (?SE) photosynthetic l ight response curves for branched, s ingle?
s temmed and adu l t  pl ants of Metrosideros excelsa col lected i n  a control l ed env i ronmen; 
on (A) 22 - 30 September, (B) 1 - 9 December 2000, (C) 23 February - 3 March 200 1 .  
Leaf node sampled for s i ngle-stemmed (A) 1 0, (B) 30 and (C) 40, and for branched 
p lants nodes (A) 1 0, (B) 20 and (C) 30. 
1 27 
(!.) ..., ?..... 
c 0 
..., ? 
E <Zl <Zl ? N 
0 
u 
25 
A 
20 
...-.... 
, (1)1 5  ":' E 
01 0  E 2-
5 
0 ' 
25 1 I B I 
(!.) 
20 I 
; ? J ? ":' E E <Zl <Zl ? N 
0 
u 
? ? ,_ 
c 
0 
...... ? 
E 
<Zl <Zl ? N 
0 
u 
? 1 0  
2-
5 ' 
0 I 
25 
l e 
20 -1 
N<Zl 1 5  I E I i 
? 1 0  ? 
:::1. I 
? I 
I 
5 J 
J 
0 1 0  
o Sing le-stem 
Branched 
Adul t  
-- ? ?---- ---1 
20 30 40 
Interce l l u lar  C02 (Pa) 
50 60 
Figure 3 .9 Mean representat ional C02 ass im i l ation to intercel l u l ar C02 response curve? 
for a leaf measured on s ingle-stemmed, branched and adul t  p lants of Metrosideros 
excelsa co l lected on (A) 22 - 30 September, (B) 1 -9 December 2000, and (C) 23 
February 3 - March 200 1 .  Leaf nodes sampled for s i ngle-stemmed plants were (A) 10 ,  
(B )  30  and (C) 40, and for branched p lants nodes (A) 10 ,  (B) 20 and  (C) 30 .  Leaves 
measured from the canopy top for adu l t  p lants .  
1 28 
An examination of photosynthetic performance at P max showed some patterns related to 
plant type in gas exchange properties (Table 3 .3) .  Leaf stomatal conductance and 
transpiration rates in single-stemmed plants decreased at higher node positions (nodes 
30 and 40 in December and March, respectively), although these trends were not 
apparent in leaves of either adult or branched plants. Generally, intercellular C02 
concentrations (Ci) were relatively constant in leaves of adult plants (mean 232? 1 2  
!-!IDOl C02 mor1 ) and to a lesser degree in single-stemmed plants (234?8 !-!IDOl C02 
mor1 ). At all sampling times, the mean leaf chamber C02 was also relatively constant 
400? 1 !-!IDOl C02 mor 1 ? In branched plants, however, Ci increased significantly with 
increasing node position, increasing by an initial 1 1 % to 275?6 ).lmol C02 mor 1 by 
node 30 (March) (P<0.0 1 ) . 
Relationships in leaf WUE varied between plant types and time (Table 3 .3) .  Generally, 
WUE in leaves of single-stemmed plants increased with node position to levels that 
were not significantly different from those in adult plants. Leaf WUE in single?
stemmed plants showed a strong negative correlation with L1 (R2 = - 0.62 1 ,  p<0.00 1 )  
(Figure 3 . 10). In contrast, leaves of branched plants became less water-use efficient at 
higher nodes (Table 3 .3 ) .  There were significant correlations for values of L1 with 
stomatal conductance in single-stemmed plant (R2 = 0.557, p<0.00 1 )  and branched 
plants (R2 = 0.352, p<0.05) but not in adult plants (R2 = 0.005, p>0.05) .  Based on data 
pooled across all plant and sampling time treatments, L1 was strongly and positively 
correlated with stomatal conductance (R2 = 0.4 1 6, p<0.000 1 )  and transpiration (R2 = 
0.436, p<0.000 1 ) , and negatively correlated with WUE (R2 = -0.394, P <0.000 1 )  (Figure 
3 . 10) .  
3.3.8 Photosynthetic response of plants under greenhouse conditions 
Rates of photosynthesis differed significantly between plant types depending on the 
time of day (Figure 3 . 1 1 A) . Rates of photosynthesis were significantly higher in adult 
plants between 1000 and 1 600 h, averaging a peak rate of 8 .7 ).lmol C02 m-
2 s- 1 at 1 000 
h (mean PPF 1 104?23 !-!mol m-2 s- 1 ) .  The highest level of PPF occurred at 1 200 h with 
an average flux of 1 678?14  !-!mol m-2 s- 1 . Photosynthetic responses of leaves in 
branched and single-stemmed plants varied in accordance with the diurnal changes in 
1 29 
....... 
w 
0 
Table 3 . 3  Comparison of leaf gas-exchange parameters in single-stemmed, branched and adult plants of Metrosideros excelsa 'Vibrance' . Gas?
exchange values represent means obtained at l ight saturated maximum rate of photosynthesis .  Mean leaf chamber C02 across all treatments was 
400? 1 !-lmol C02 mor 1 ? Means shown for each plant type collected during three sampling dates(Sept = on 22 - 30 September, Dec = 1 - 9 
December 2000, and Mar = 23 February - 3 March, 200 1 )  in a control room at the New Zealand Controlled Environment Laboratory, 
Horticulture and Food Research, Palmerston North, correspond to leaves on nodes 1 0, 30 and 40 (shown in parentheses) in single-stemmed plants 
and 10, 20 and 30 in branched plants, respectively. Analysis for variables in each row conducted using a two-way ANOV A for assessing the 
effect of plant type (P), sampling time (T) and their interaction (PxT). Mean separation in rows by Least Square Means tests, 5% significance 
level. 
Parameter P-value 
Stomatal conductance PxT <0.05 
(mol H20 m-2 s- 1 ) 
Transpiration PxT<O.OO I 
(mmol H20 m-2 s-
1 ) 
Water use efficiency PxT<O.OO I 
(!lmol mmor1 ) 
Single-stemmed Branched 
Date (node) Date (node) 
Sept ( 1 0) Dec (30) Mar (40) Sept (10) Dec (20) Mar (30) Sept 
0.20a 0. 1 2? ___ 0. l lb 0.23a 0.2 1a 0. 1 8a 0. 1 5ab 
2.7a 1 . 8c 1 .6c 3 .0a 2 .8a 2 .3ab 2.0bc 
5 .2bc 6 .3a 6. 1 ab 4.7c 5 . 3b 4. 1 d 6.3ba 
Adult 
--
Date 
Dec Mar 
0. 1 8a 0. 1 9a 
2.4ab 2.5ab 
s .sab S .6abc 
I -- - --- ---- - ----26 "i 
0 11 
? 
? 24 .J I 
E 
.... 22 ? U I en I 
U I 
Cl) 0. 
I 
? 20 _, 
en 
c 
0 .D 
? 1 8  ? u 
! e Sing le-stem I 
Branched 
Adult 
22 
21  
20 
1 9  
1 8  
1 7  
1 6  
Date 
lx sept 1 
I I t::,. Dec 
? ' o Mar ! ? 
4 
? 
r = -0 .788 
P<0.001 
5 
? ? 
? 
X 
6 7 
I r = -0.628 
I P<0.000 1 
1 6 -?L=======?--- --L_ ______ L_ ______ ?----? 
3 4 5 6 7 
Figure 3 . 1 0  Corre lation of water use eff iciency (WUE) w ith carbon  i sotope 
d iscrim ination for leaves from s i ngle-stemmed, branched and adu l t  p l ants of 
Metrosideros excelsa poo led from the three sampl i ng dates (22 - 30 September, 
1 - 9 December 2000, 23 February - 3 March 200 1 ) .  Sma l l  w indow :  Corre lation 
of water use effic iency w ith carbon isotope d iscrim ination for s i ngle-stemmed 
plants w i th the three success ive sampl i ng dates displayed i n  descend i ng order. 
8 
1 3 1  
w 
bJ 
1 4 I ---?;nnlufo? ? .. 
Q) 1 2  ...... 0 
A r 1 600 
0 
___.__ St .. 5.?-???? . .  
? 'Cil1 0 
U 'i'  
___.__ Branched 
__._Adult 1 200 'Cil 'i' 
?? 8 8 Q) N ? 
0 6 ;::.-, U 800 
-PPF 
Cll -0 0 4 b 8 ...t:l :::t 2 ? '-' 
- 400 
I I ? o .....-= 
--. 6 
5 
-""" 
d 'en 4 
0 ?  ?? E 
.? 0 3 Q.,N  
? ? 2 
!--- ? '--" 
6 8 1 0  1 2  1 4  
Time of day (hour) 
1 6  1 8 
0 
8 
0 
s 
:::t '-' 
? 
p.. 
p.. 
0.20 
ko. t s j 
B 
-T-- -
i to w i - /"
1 f'- ? 
tij ::I:  
? -
E 
o 
B S0.05 
[/J 
0.00 
3 
0 
D 
' 
0 
E 2 E 
0 
E 
i
t 
0 
6 8 1 0  1 2  
T 
1 4  
Ime of day (hour) 
1 6  1 8  
Figure 3 . 1 1 Diurnal comparison of leaf gas exchange parameters (mean ?SE) in single-stemmed, branched and adult plants of 
Metrosideros excelsa 'Vibrance' .  Changes in (A) photosynthetic rate with photosynthetic photon flux (PPF), (B) conductance, 
(C) transpiration and (D) water-use efficiency (WUE) measured on 20 - 2 1  March 200 1 ,  within a greenhouse located at the Plant Growth 
Unit, Massey University, Palmerston North. Mean diurnal ambient C02: 358?0.07 J..lmol C02 m-
2 
s- 1 ?  Diurnal air temperature range: 25-35?C. 
PPF. Leaves in branched plants displayed a bi-modal photosynthetic pattern, with rates 
temporally decreasing significantly at mid-day ( 1 200 h) before a subsequent resurgence 
to 7 . 1 j.lmol C02 m-2 s- 1 ( 1400 h). A similar but less pronounced response also occurred 
in single-stemmed plants. 
Several other photosynthetic parameters varied between plant types over the 1 2  h 
period. Generally, leaf stomatal conductance and transpiration rates were higher in 
adult plants during the early (0800-1200 h) and later ( 1 200-1 600) hours of the day, 
respectively, in comparison to values in leaves of rnicropropagated plants (P<0.05) 
(Figure 3 . 1 1 B and C). Water-use efficiency (WUE) in leaves of single-stemmed plants 
was significantly higher than in leaves of both branched and single-stemmed plants 
early in the day (0800 h) ,  and only higher than branched plants at 1 200 h (P<0.00 1 )  
(Figure 3 . 1 1  D). In comparison with leaves from branched plants, those in adult plants 
became more WUE during the middle of the day ( 1 200- 1400), although these values 
did not differ from leaves of single-stemmed plants. 
3.3.9 Carbohydrate analyses 
There was a significant difference in the level of leaf carbohydrates between plant types 
(Table 3 .4). Total leaf soluble sugar concentrations in single-stemmed (node 40) and 
branched plants (node 30) plants were almost twice as high compared with leaves 
harvested from adult leaves. This was attributed mainly to higher fructose and glucose 
but not sucrose concentrations in the micropropagated plants. Starch levels in leaves of 
single-stemmed plants and branched plants were similar between plant types, being 
approximately double the concentration recorded in leaves of adult plants. 
133  
Table 3 .4 Comparison of carbohydrate concentrations in fully expanded leaves collected 
from nodes 1 0  and 40 in single-stemmed plants, node 30 in branched plants and from 
the canopy top in adult plants of Metrosideros excelsa 'Vibrance' on 25 June 200 1 ,  
between 0600 - 0630 h .  Replication consisted of 8- 1 2  leaf samples per node or plant 
type collected within a greenhouse located at the Plant Growth Unit, Massey University, 
Palmerston North. Treatment comparisons were conducted with a one-way ANOVA 
with mean separation in rows by Least Square Means tests, 5% significance level . 
Symbol : n .s .  not significant. 
Carbohydrate Single-stemmed Branched Adult 
(mol g? 1 DW) p-value node 40 node 30 canopy top 
Sucrose n.s. 0.028 0 .029 0.035 
Fructose <0.0 1 0.026a 0 .027a 0.0 1 5b 
Glucose <0.00 1 0.063b 0.046a 0.028c 
Total soluble <0.01  0. 1 1 7a 0. 1 02a 0.078b 
Starch (mg g? ' DW) <0.00 1 14 .7a 1 7 .5a 8 .0b 
3.4 Discussion 
Imposition of a shoot restriction treatment on rnicropropagated plantlets of Metrosideros 
excelsa 'Vibrance' accelerated vegetative phase change. Not only did leaves in single?
stemmed plants become smaller and rounder with increasing node position, but they 
also approached leaf optical values (representative of increasing tomentum on abaxial 
surfaces) similar to those of adult plants. This pattern of an acceleration in vegetative 
phase change is consistent with findings reported earlier in the single-stemmed plants of 
the cultivar ' Scarlet Pimpernel'  (Chapter 2), and in similar experiments where phase 
change occurred with increasing node and height above the roots (Davis 1 99 1 ,  Snowball 
et al. 1 994, James and Mantell 1997, Sisrnilich, Jameson and Clemens unpublished 
data) . Unique to this study, however, was the characterisation of leaf gas exchange and 
6. components during vegetative phase change in a woody species. 
Variation in 6. values between adult and j uvenile plants has been reported in a number 
of ecophysiological studies (Donovan and Ehleringer 1 994, Hansen 1 996, Cordell et al. 
1 34 
1 998, Rundel et al. 1999) . Typically, intraspecific differences in L1 values for juvenile 
plants are often reported to be higher than those of adult plants. A common theme in 
these and other studies has been the significance of L1 as a reliable indicator of long-term 
WUE, given the strong negative correlation between these two factors (Farquhar et al. 
1 989). For example, based on L1 values obtained from plants of Chrysothamnus 
nauseosus growing in common plots in semi-arid shrubland, Donovan and Ehleringer 
( 1 992, 1 994) were able to assess the WUE between both sets of plants. Juvenile plants 
of C. nauseosus were found to be less WUE than adult plants, which coincided with 
higher rates of photosynthesis and increased stomatal conductance and transpiration in 
the former plant type. Similarly, comparable L1 and gas exchange responses have also 
been reported for different life stages in Metrosideros polymorpha (Cordell et al. 1 998) 
and Acacia (Hansen 1 996). The distinct ecophysiological characteristics expressed by 
each life stage may occur because juvenile plants in the wild are potentially under 
different selective pressures than adult plants (Donovan and Ehleringer 1 99 1 ) . 
Therefore, the relatively higher rates of photosynthesis expressed in many juvenile plant 
species, particularly those under severe conditions (e.g. arid conditions), would be 
consistent with the hypothesis of a 'go for broke' strategy (Bond 2000). This would 
provide the growth potential required for establishment, despite a greater risk of 
mortality (Bond 2000) . 
The results obtained from single-stemmed plants of M. excelsa that had been grown 
under well-watered conditions while undergoing vegetative phase change appeared to 
be consistent with the above studies .  Not only did these plants exhibit short-term 
changes in gas-exchange characteristics, such as reduced rates of photosynthesis, 
transpiration and stomatal conductance with increasing node position, but also a long?
term indicator of plant performance (i .e. L1) suggested that these plants became more 
WUE as their leaves became more adult. Moreover, in single-stemmed plants, L1 values 
were also strongly correlated with most of the leaf morphological attributes, 
discriminating less as leaves became increasingly adult. Based on these observations, it 
would appear that these results support the hypothesis that j uvenile plants exhibit 
decreasing L1 because of changes in leaf morphology consistent with greater water-use 
efficiency, notably the acquisition of more tomentose and round-shaped leaves similar 
to those born on adult plants. 
135  
The comparatively higher values of !::. in leaves of adult plants with those of single?
stemmed plants, however, appeared anomalous .  The contrasting responses in gas 
exchange parameters between single-stemmed and adult plants also highlighted this 
anomaly. This is in light of the fact that these plant-type specific differences in short?
term (gas exchange) and long-term (/1) responses were obtained consistently when 
obtained under both climate-controlled (light and C02 response curves) and ambient 
greenhouse conditions, respectively. Furthermore, !::. values did not differ between the 
three sets of plants at the beginning of the experiment, despite their differences in size, 
ontogenetic state and leaf dimensional qualities. These findings were also not 
confounded by age of the leaf samples given that harvested samples represented newly 
( < 2 months) expanded leaf tissue from all plants. Moreover, the hypothesis that the age 
of the leaf would not play a role in affecting the expression !::. was also upheld by this 
study, since leaves harvested from single-stemmed plants at similar nodes but at 
different times (leaf age) showed no significant differences. Therefore, this would 
suggest that !::. levels were established during a time of leaf expansion. 
What factors may account for the discrepancy in photosynthetic rates between single?
stemmed undergoing vegetative phase change and adult plants? It is possible that 
removal of axillary buds in single-stemmed plants induced a wound or stress response 
that limited stomatal conductance. However, this is unlikely because photosynthesis 
measurements were spatially and temporally separate from the times and positions at 
bud removal occurred along the stem axis. For instance, gas exchange measurements 
were conducted at least six days after buds had been excised. Furthermore, 
photosynthetic measurements were conducted on leaves that were below the position 
( -3-5 nodes) of the most recently excised axillary buds. 
An alternative explanation relates to variation in source and sink relations as a result of 
differences in plant architecture . Following the removal of axillary meristems, the 
structural architecture of a single-stemmed plant would be expected to limit sink?
demand within that plant to root and shoot apical regions, whereas in adult plants, this 
would be dispersed over numerous growing points, such as developing axillary 
buds/shoots within the canopy. Sink demand relative to source strength in the single-
1 36 
stemmed plants would be expected to decrease further as these plants continued to grow 
and attain an increasingly greater number of source leaves (increased source strength). 
It has frequently been reported that treatments that reduce the translocation of 
photosynthates to sink regions (e.g. removal or reduction in sink demand) can lead to an 
excess accumulation and partitioning of these carbohydrates into starch within source 
leaves (Harlevy and Sharkey 1 99 1 ,  Stitt and Schulze 1 994, K.rapp and Stitt 1 995, Myers 
et al. 1 999, Nakano et al. 2000). Such experiments designed to induce alterations in 
sink demand have included excision of the sink material, including removal of shoot 
apices (Mondal et al. 1 978),  leaves (Myers et al. 1 999), flower and fruiting bodies 
(Nautiyal et al. 1 998, Nakano et al. 2000, Wi.insche et al. 2000), and continuous 
removal of axillary buds (Gezelius et al. 1 98 1 ,  Sawada et al. 1999). 
An increase in carbohydrate levels in source leaves has frequently been associated with 
a reduction in photosynthetic capacity (Krapp and Stitt 1995, Myers et al. 1 999, Nakano 
et al. 2000). Nakano et al. (2000) and others have suggested that a decrease in 
photosynthetic activity is  a result of a feedback inhibition on that activity caused by an 
accumulation of an end product (possibly starch) within the source leaves . The findings 
in M. excelsa are also consistent with this hypothesis. The higher concentrations of 
starch (and several soluble sugars) and the depressed rates of photosynthesis recorded in 
leaves of single-stemmed plants compared with those of adult plants, are suggestive of 
an end-product (possibly starch) induced feedback inhibition response on 
photosynthesis. S imilar to M. excelsa, the accumulation of starch in source leaves in 
Pinus taeda has also been attributed to a reduction in carboxylation-related processes, 
such as in electron transport rates and Rubisco activity (Myers et al. 1 999). This has 
been attributed to an extreme enlargement in starch grains within the leaf that physically 
damage chloroplasts, subsequently reducing photosynthesis and hindering C02 diffusion 
in chloroplast (Pritchard et al. 1997, Nakano et al. 2000). S imilar reports of increased 
carbohydrate content and a down-regulation in photosynthesis in source leaves has also 
been observed under various sink-limited conditions following bud removal in seedlings 
of Pinus silvestris (Gezelius et al. 198 1 ), terminal leader excision and girdling in Pin us 
taeda (Myers et al. 1 999), seed pod removal in Phaseolus vulgaris (Nakano et al. 2000) 
and fruit removal in trees of Malus sylvestris (Wi.insche et al. 2000). 
1 37 
Branched plants also exhibited a reduction in gas exchange parameters at higher node 
positions in conjunction with high starch level, although this was not reflected in the 
long-term WUE of the plant (e.g. L1). This suggests that the actions or events that 
triggered a reduction in these parameters were of relatively recent origin and not of a 
long-term nature. The reduced photosynthetic rates in branched plants at high nodes 
may be attributed to the lack of axillary branches observed on long (ea. 30 nodes) 
individual shoots of branched plants from which samples were collected. It could be 
speculated that a similar sink-reduced effect in elongating shoots of branched plants 
might have occurred as node position increased. This would be consistent with the high 
levels of starch and reduced rates of photosynthesis at high node positions, as also 
observed in single-stemmed plants . Further evidence to support this hypothesis may 
have been attained if carbohydrate analyses of leaves in branched and single-stemmed 
plants were collected at different node positions and/or sampling times. 
The decline in carbon isotope composition of single-stemmed plants is, therefore, 
consistent with alterations in gas exchange parameters. Not only was stomatal 
conductance and transpiration reduced in single-stemmed plants at higher nodes, but 
they also became increasing more water-use efficient. Both stomatal conductance and 
WUE, particularly in single-stemmed plants, were also strongly correlated with L1. This 
response is  suggestive of either greater stomatal closure or more heterogeneity in 
stomatal apertures over the leaf surface during photosynthesis, ultimately limiting 
atmospheric and intercellular C02 gas exchange (Lauteri et al. 1993, Patterson and 
Rundel 1 993) .  Thus, the reduced supply of atmospheric C02 in a stomatal-limited 
photosynthetic system would, theoretically, account for the decline in L1 values observed 
in single-stemmed plants. Therefore, the decline in leaf L1 values in single-stemmed 
plants undergoing phase change does not support the hypothesis that it is associated 
with changes in leaf morphology, but rather appears to reflect the physiological 
responses associated with a disruption in source and sink relations. 
Characterisation of shoot and leaf growth was also described in this study. It was 
interesting to note that the pattern of vegetative growth in adult and micropropagated 
plants differed significantly. Both single-stemmed and branched plants exhibited 
continuous elongation of either a main or an axillary shoot, respectively, whereas in  
1 38 
adult plants, shoot elongation was episodic. This episodic pattern of vegetative growth 
is typical of adult plants in Metrosideros (Gerrish 1 989, Clemens et al. 1 995), whereby 
growth is interrupted by intermittent periods during which time resting buds are formed 
(Sreekantan et al. 2001 ) .  Histological observations of adult buds of M. excelsa 
suggested that leaves were preformed inside the bud. This is in contrast to 
micropropagated plants where de novo leaf organogenesis appeared to arise 
continuously from the apical meristem. 
Within the developing buds on adult plants, the number of preformed leaves was 
strongly correlated with the size of the bud itself. It is possible that the number of 
subsequent leaves and internodes laid down following bud break is related to the 
number of preformed leaves within the bud, as thought to occur in various temperate 
and tropical tree species (Steeves and Sussex 1989). In the event that only preformed 
leaves expand, the first primordia formed at the shoot apex may begin to differentiate as 
cataphylls that would surround the next terminal bud, subsequently containing the next 
set of pre-formed leaves (Steeves and Sussex 1989). Presumably, the size of the shoot, 
being the number of nodes formed, would be a function of the size of the bud at the time 
of bud-break. Therefore, bud-break of relatively small or large buds may result in 
shoots containing few or many nodes, respectively. 
Relatively little attention has been given to the evaluation of carbon isotope signatures 
conducted during bud and subsequent shoot and leaf ontogeny. The fact that carbon 
isotope composition in partly to fully-expanded leaves of M. excelsa were similar to 
those expressed in older leaves ( -6 month in age), suggests that discrimination levels 
were not affected by the age of the leaf. However, !1 values in buds containing pre?
formed leaves, although not differing from that of old subtending leaves, were 
significantly lower than that of the newly expanded leaves .  This difference is probably 
related to the origin of organic matter obtained during bud development. Damesin et al. 
( 1 998) suggested that the origin of fractionated leaf-external organic matter differed 
between deciduous and evergreen species. In deciduous oak species, for example, this 
organic matter comprises reserves (mostly starch) stored from previous growth seasons, 
whereas leaf organic matter in evergreen species is mostly comprised of recently 
assimilated sugars (Damesin et al. 1998).  Considering that M. excelsa is an evergreen 
species, it was surprising that !1 values were comparatively lower in bud tissue, 
1 39 
particularly since recently synthesised photoassirnilates deriving from subtending leaves 
would be expected to have a similar isotopic signature. It is possible that some of the 
carbohydrates acquired during bud growth were obtained from storage carbohydrates, 
given that complex carbohydrates,  such as starch,  can have a higher affinity for 13C than 
soluble sugars (Brugnoli et al. 1 988). This would explain the relatively low ? values 
recorded in the developing bud. Further studies, however, using 13C labelling would 
confirm this hypothesis. 
In conclusion, the role of carbohydrates during the phase change process has, as of yet, 
received very little scientific attention. It has been suggested that treatments that 
promote carbohydrate accumulation in various parts of the plant can also affect the 
transportation and accumulation of other substances (e.g. hormones), which would 
subsequently complicate interpretation of carbohydrate effects on plant developmental 
and physiological responses. Nevertheless, the evidence thus far suggests that 
carbohydrates do play a role during phase change (Tsai et al. 1 997), although further 
studies incorporating molecular technology would elaborate on their role during 
vegetative phase change. The results of this study suggest that shoot restriction 
accelerates vegetative phase change in M. excelsa. In future work, examination of these 
trends over a longer growing period would provide a useful comparison between 
branched and single-stemmed plants. The hypotheses that a reduction in photosynthetic 
and ? parameters in single-stemmed plants undergoing phase change were attributed to 
either the age of the leaf or to differences expressed between juvenile and adult plants 
were not supported by this study. Instead, the decline in these parameters, although 
coinciding with vegetative phase change, were consistent with studies l inking these 
phenomena to alterations in source and sink relations. Therefore, based on the evidence 
presented in this study, the decline in leaf ? values with increasing maturity in single?
stemmed plants would not serve as a reliable marker of phase change in M. excelsa. 
Future research in this area may profit from a better understanding of the role of 
carbohydrates during vegetative phase change in woody species. 
140 
Chapter 4 
Effects of photoperiod, temperature and bud size on flowering in 
Metrosideros excelsa 
4.1 Introduction 
Manipulation of flowering time has been used extensively within the horticultural 
industry (Atherton 1987), particularly in the scheduled production of ornamental pot?
plants. Most research in this area has focused on a limited number of herbaceous 
species or woody plants cultivated for many decades .  However, relatively few studies 
have considered development of practical methods for controlling flowering in woody 
angiosperm species new to cultivation. Of those that have, manipulation of temperature 
and/or photoperiod conditions have been successful in promoting flowering in 
Leptospermum (Zieslin 1 985), Chamelaucium (Shillo et al. 1 985), Eucalyptus (Moncur 
1992), Hypocalymma (Day et al. 1 994a), Pimelea (King et al. 1996) and Hardenbergia 
(King 1 998) .  
Specific temperature and photoperiodic treatments required to initiate flowers vary 
between species. In Leptospermum and Chamelaucium (family Myrtaceae), sustained 
cool temperatures (mean 1 5-20?C) and short days (:=::; 1 2  h) were conducive for floral 
initiation (Shillo et al. 1 985,  Zieslin and Gottesman 1 986, Dawson and King 1 993).  
However, flowering in Eucalyptus and Hypocalymma was achieved using short (5- 10  
week) exposure to cold ( 1 2- 1 3?C) temperatures, irrespective of  photoperiod (Moncur 
1 992, Moncur and Hasan 1 994, Day et al. 1994a). Floral induction in response to 
relatively cool temperatures has also been reported for Pimelea (King et al. 1 996). 
The bud contains the active site (apical meristem) where endogenous signals for 
flowering are received from the subtending leaf (Bernier et al. 1 993).  However, 
Data published from this chapter: Henriod, R.E., Jameson, P.E., Clemens, J. (2000) Effects of 
photoperiod, temperature and bud size on flowering in Metrosideros excelsa (Myrtaceae). Journal of 
Horticultural Science & Biotechnology 75 :  55-61 .  
14 1  
relatively little attention has been paid to the effect on floral initiation that bud size at 
the time inductive conditions are applied might have (Patten and Wang 1 994) . In 
cranberry (Vaccinium macrocarpon) buds > 1 mm in diameter in autumn had a higher 
probability of becoming floral the following summer than did buds < 1 mm in diameter 
(Patten and Wang 1 994) . Therefore, it is important to understand the influence bud 
size can have on developmental outcome of buds. 
The unpredictable timing and non-uniform patterns of flowering (Clemens et al. 1 995) 
have hindered the development of Metrosideros for the floricultural market. The aim of 
this study was, therefore, to determine the effect of photoperiod and temperature on 
floral initiation and development, and to examine the effect of bud size on flowering in 
two important cultivars of M. excelsa. The approach taken in this experiment was to 
study potentially inductive combinations of lowered temperature and short photoperiod 
applied over a number of weeks in controlled environments. In a complementary 
experiment, other plants were grown through winter either under naturally inductive 
conditions, or in greenhouses to determine the deleterious effects on flowering of 
elevated temperature and/or daylength extension. 
Based on the current literature on factors that promote floral induction in species within 
the Myrtaceae, it is hypothesised that: 
1 )  Both low temperature and/or short day length are important in promoting floral 
induction in M. excelsa. 
2) The size of the developing bud at the time of application of an inductive treatment 
will determine whether or not floral initiation will take place, with larger buds 
becoming floral more readily than smaller buds. 
4.2 Materials and Methods 
4.2.1 Plant materials 
Commercial growers supplied two year-old, cutting-grown plants of Metrosideros 
excelsa 'Scarlet Pimpernel' and 'Vibrance' that had been grown under ambient 
conditions .  In August 1 997, when the plants were approximately 1 .5 m high, 
1 42 
they were potted into 7 1 containers using a peat and pumice growing medium (80:20 v/v) 
and appropriate control-release fertilisers. The plants were maintained in a greenhouse 
from November 1 997 (early summer) until the start of experimentation in February 
1 998. During this time the greenhouse was vented at 24?C and warmed at night when 
required ensuring the minimum temperature was no lower than 17?C. Daylength was 
extended to 1 6  h using incandescent lighting ( 1 0 ?-tmol m?2 s- 1 ) .  Two experiments were 
conducted using a single population of each cultivar for both experiments. 
4.2.2 Controlled environment experiment 
In the first experiment, temperature and photoperiod treatments were applied in four 
controlled environment rooms located at the New Zealand Controlled Environment 
Laboratory, The Horticulture and Food Research Institute of New Zealand 
(HortResearch), Palmerston North. Plants were grown using a factorial combination of 
two temperatures and two photoperiods. Day/night temperatures were maintained at 
either 1 2/9?C or 17/ 14?C (mean temperatures 10?C and 1 5?C, referred to below as 
"cold" and "cool") in combination with 1 0  h (short day) or 1 6  h (long day) 
photoperiods. Lighting in the rooms was provided by four high intensity main 1 kW 
Metalarc lamps with four 1 kW tungsten halogen lamps (8 h) with a photosynthetic 
photon flux (PPF) of 680-700 !-!IDOl m-2 s- 1 . A photoperiod extension of either 2 h (short 
days) or 8 h (long days) was provided by low intensity 1 50 W Tungsten auxiliary lamps 
(PPF 7 ?-tmol m-2 s- 1 ) .  In growth rooms maintained under cool conditions, the relative 
humidity (RH) was maintained at (day/night) 79/8 1 %  with a dew point (DP) at 1 31 1 1 oc. 
In cold rooms, RH was maintained at 7 1174% with a DP at 7 /5?C. In all rooms, C02 
concentration was 350 ppm and vapour pressure deficit was 0.4/0.3 kPa (day/night) 
(Fulton 1 998) .  
Plants were assigned to one of four blocks and subsequently to one of the four 
treatments. Each block contained seven plants, three of which were ' Scarlet Pimpernel' 
and four of 'Vibrance' .  Due to the low number of ' Scarlet Pimpernel' plants, they were 
placed in blocks per treatment using a balanced incomplete block design. Plants were 
blocked on four trolleys and were rotated within the rooms every seven days to ensure 
an even light distribution over time and to reduce any positional effects created by the 
rooms. Subsets of plants were subjected to the four environments for 0, 5 ,  1 0  or 1 5  
weeks before being transferred to a greenhouse maintained at 24! 17?C (day/night) under 
143 
natural illumination with photoperiod extended to 16  h using 1 00 W incandescent lights 
( 10  1-1mol m-2 s- 1 ) -
At the outset of  the experiment, 10  pairs of  terminal axillary buds from each plant were 
selected. Selection of buds was designed to capture a proportional range of bud sizes 
from each plant so that their fate could be established at the completion of the 
experiment. Initial measurements were made of bud diameter perpendicular to the 
petiole using digital calipers. Upon completion of the experiment in January 1999, buds 
were scored as floral, aborted, vegetative, or not broken (fewer than two pairs of scales 
parted) .  Records were kept of the number of plants that flowered, the number of 
inflorescences per plant, the number of cymules per inflorescence and the timing of 
anthesis. Changes in floral development were tracked every seven days using seven 
recognisable stages (Stages 1 -7 ,  Table 4 . 1  ) .  
4.2.3 Greenhouse experiment 
Plants were maintained in four greenhouses for the duration of the experiment (February 
1998-J anuary 1999) using factorial combinations of temperature and day length. The 
two temperature regimes were those occurring under ambient conditions, and 
maintenance of a warm temperature regime (241 1 rq (day/night). Mean ambient 
day/night temperatures during mid-winter were 1 3/8?C, rising to 25/ 1 7?C in mid?
summer. Photoperiod treatments were ambient conditions (declining through mid?
winter to 9 h and rising to 1 3  h in mid-summer), and maintenance of a 1 6  h photoperiod. 
Photoperiod was extended using 1 00 W incandescent lights ( 1 0  1-1mol m-2 s- 1 ) .  Plants 
held under ambient temperature conditions were housed in semi-transparent plastic 
tunnels that provided frost protection and unrestricted air-flow from both ends. 
Temperatures were monitored using Hobo shuttle loggers (model H8, Scott Technical 
Instruments, Hamilton, New Zealand) . Treatments were replicated using four blocks, 
each containing six individuals of 'Vibrance' and one of 'Scarlet Pimpernel' .  
Observations of plant development were made as for the controlled environment 
experiment. 
144 
Table 4. 1 Classification of stages of floral development in Metrosideros excelsa. 
Developmental stage Description Diagram 
1 :  Bract shedding ? Subtending bracts 
shed from cymule 
? Individual flower 
buds visible 
2: Tight bud ? Remaining 
bracteoles shed 
? Red petals visible 
between calyx lobes 
3 :  Flower opening ? Petals reflexing 
? Style and stamens 
expanding 
4 :  Anthesis ? Stamens and style 
fully elongated 
? Pollen shedding 
5 :  Stamen wilting ? Stamens starting to 
wilt 
6: Stamen abscission ? Stamen abscission 
occurnng 
7 :  Senescence 
? Petals abscising 
? All stamens and 
petals abscised 
? Style may have 
ab seised 
1 45 
4.2.4 Statistical Analyses 
Statistical analyses were conducted using SAS System statistical package (SAS 
Institute, Cary, N.C. ,  USA). For each experiment, the main and possible interaction 
effects of temperature and day length on the proportion of flowering plants and the 
number of inflorescences per plant were analysed using analyses of variances with the 
GENMOD procedure. The GENMOD procedure fits a generalised linear model to the 
data by maximum likelihood estimations, where the mean of a population is dependent 
on a linear predictor through a nonlinear link function. The response probability 
distribution was based on a Poisson distribution for count data, and a logistic regression 
for proportional data. An examination of differences in the rates of flower development 
was conducted using a repeated measures analysis of variance with treatment and time 
as the main factors . Differences between treatments were assessed using Tukey 
multiple comparison tests. Data were log transformed prior to analyses . 
4.3 Results 
4.3.1 Controlled environment experiment 
For the cultivar Vibrance, flowering after transfer to the forcing greenhouse occurred 
only in plants held for 1 5  weeks under cool (mean 1 5?C), short days ( 10  h) .  No plants 
of this cultivar flowered after being held under cool, long days ( 1 6  h), or under cold 
(mean 10?C) conditions with either long or short days (Figure 4. 1 A). Sixty-seven 
percent of 'Scarlet Pimpernel' plants flowered after being held for only 1 0  weeks under 
cool , short days before transfer to the forcing greenhouse. After 1 5  weeks, 33-67% of 
'Scarlet Pimpernel ' plants flowered in all treatment combinations (Figure 4 . 1 A). Both 
temperature and photoperiod main effects were significant (P<0.05), with cold and short 
days treatments having the higher proportion of flowering plants . Plants of either 
cultivar remaining in the forcing greenhouse or treated for only five weeks before 
transfer to the greenhouse did not flower. 
'Vibrance' plants in the cool, short day treatment bore 8-9 inflorescences per plant 
(Figure 4 . 1 B) .  In 'Scarlet Pimpernel ' ,  temperature and photoperiod significantly 
affected (P<0.05) the number of inflorescences per plant, with the cool, short day 
treatment giving rise to the greatest number of inflorescences (P<0.05) (Figure 4. 1 B) .  
146 
Figure 4. 1 Effects of temperature and photoperiod on flowering in plants of Metrosideros excelsa cvs. Scarlet Pimpernel (dark bars) and 
Vibrance (white bars). (A) Percentage of plants flowering, and (B) number of inflorescences per plant, in plants treated for 1 5  weeks under cool 
(mean 1 5?C) or cold (mean 10?C) temperatures in factorial combination with 10 h (SD) or 16 h (LD) photoperiods . (C) Percentage of plants 
flowering, and (D) mean number of inflorescences per plant, in plants grown continuously in greenhouses maintained at either ambient 
temperatures (Amb T) or under warm conditions (24! 1 7?C) in factorial combination with ambient daylengths (Amb DL) or a 1 6  h (LD) 
photoperiod. 
?UBJd;saouaosaJOJJUI 
(%) 6upaMOIJ SlU'eJd  
Q lU'eldfsaouaosaJOIJUI 
u (%) 6upaMOJJ S?UBJd 
__J 
0 
__J 
0 
147 
There was no interaction between the main effects for number of inflorescences per 
plant. 'Scarlet Pimpernel' inflorescences were significantly larger (P<O.O l )  in the cool, 
short day than in the cold, short day treatment (6.9 and 3 .4 cymules per inflorescence, 
respectively).  
Plants from the controlled environment experiment reached an thesis (Stage 4) in late 
spring (Julian day 290-340). Inflorescences of 'Scarlet Pimpernel' took 3-4 weeks to 
pass from Stage 2 (tight bud) to Stage 6 (stamen abscission), Stage 4 (anthesis) lasting 
5-6 days, regardless of treatment. However, there was a significant effect of treatment 
(P<0.05) on the time at which these stages of floral development took place (Figure 
4.2) .  Inflorescences of plants held for 1 0  weeks under cool, short days before transfer to 
the forcing greenhouse developed significantly earlier (by two weeks) than those of 
plants treated for 1 5  weeks under the same conditions. Plants treated for 1 5  weeks with 
cool, long days developed significantly later (by approximately five weeks) than those 
given the cool, short day treatment (Figure 4.2). However, there was no difference 
between the timing of floral development in plants treated with long or short days under 
cold conditions. The timing of floral development in 'Vibrance' plants treated with 
cool, short days was the same as that of 'Scarlet Pimpernel ' plants given the same 
treatment for 1 5  weeks. 
The diameter of 'Scarlet Pimpernel' buds that became floral after 1 5  weeks treatment 
with cool, short days was in the range 2 . 1 -4.0 mm (mean 2.6-3.0 mm size class) (Figure 
4 .3) .  By contrast, mean bud diameter for those that remained vegetative was centered 
on the 1 .6-2 .0 mm size class. The highest proportion of unbroken buds was in the 
smallest size class ( 1 . 1 - 1 .5 mm diameter) (Figure 4.3) .  Breaking buds of 'Scarlet 
Pimpernel' in the smaller size classes also tended to be vegetative rather than floral in 
the other controlled environment treatments, although the numbers of floral buds in the 
sample was low because the treatments were relatively ineffective for floral induction. 
4.3.2 Greenhouse experiment 
Temperature and daylength treatments significantly affected the proportion of plants 
that flowered (P<0.0 1 ) ,  and there was also a significant interaction between the main 
effects (P<0.05) in both cultivars. The highest proportion of plants that flowered 
occurred under conditions of ambient temperature and ambient day length ( 1 00% and 
148 
7 
6 -
.._. t: 
5 
1 
<J) E 
0-0 
<J) 
> 4 <J) u 
'-1-0 
<J) CO 3 -.:3 
U) 
2 
1 I 
240 
---?? -??--- ? ?----- ?-- ? - '""'???-?-?--? ???????-
- -G - Cool/SD 1 0  weeks 
? Cooi/SD 1 5  weeks 
--- Cool/LD 1 5  weeks 
----ia-Cold/SD 1 5  weeks f -? Cold!LD 1 5  weeks 
I 
I 
/ "Ill / 
? 
I 
? 
/ ;J---; 
-?  
eo-:-_ 
? 
._,.. 
T' 
260 280 300 
\Q-
-
Jul ian day 
7 
( 
I 
/ 
I T ? 
f 
?? ;-
/ 
/? 
!' 
320 340 360 
Figure 4.2 Mean (?SE) rate of floral development in p lants of Metrosideros 
excelsa ' Scarlet P impernel '  treated for 1 0  or 1 5  weeks i n  contro l led env ironments 
before being transferred to a forc ing greenhouse. P lants were grown under cool 
(mean 15?C) or cold (mean l 0?C) temperatures in factorial combination with 
10 h (SD) or 16 h (LD) photoperiods. 
1 49 
50 
? 40 
Bud s ize c lass (mm) 3.6-4.0 
Floral 
ot broken 
Figure 4.3 Percentage of Metrosideros excelsa 'Scarlet P impernel '  buds i n  each 
of seven bud s ize c lasses that were floral ,  remained vegetative, or did not b reak 
in p l ants transferred to a forcing greenhouse after 1 5  weeks treatment w ith cool 
( 1 5?C), short days (1 Oh) . 
1 50 
83% for 'Scarlet Pimpernel ' and 'Vibrance' , respectively) (Figure 4. 1 C). The 
proportion of 'Scarlet Pimpernel' plants that flowered was also high ( 1 00%) in the 
ambient temperature and long day treatment, although this was not the case for 
'Vibrance'  ( 1 2.5%). No flower buds developed under warm (24/ 17?C), long days for 
either cultivar. 
Temperature and day length significantly affected the number of inflorescences per plant 
(P<O.OO 1 ), with ambient temperature rather than warm conditions, and ambient 
daylength rather than long days, resulting in more inflorescences per plant in both 
cultivars. There were also significant interactions between the main effects (P<0.00 1 )  
in  both cultivars. Significantly more inflorescences per plant occurred in  the ambient 
temperature and ambient daylength treatment for both cultivars (Figure 4. 1 D). Number 
of cymules per inflorescence across all treatments was in the range 5 .8-7 .0. 
Both cultivars in the greenhouse experiment reached anthesis at approximately the same 
time in early summer (Julian day 355-365). In general , the time taken for inflorescences 
to pass through Stages 1 -7 was similar to that shown for inflorescence of plants from the 
controlled environment experiment (Figure 4.2). At no time did treatment have an 
effect on the rate of inflorescence development in 'Scarlet Pimpernel' (Figure 4.4 A). 
However, inflorescences of 'Vibrance' plants in the warm temperature/ambient 
daylength treatment were more advanced by - 10  days than those in the ambient 
temperature treatment with either ambient or short daylength (Figure 4.4B) .  These 
significant effects (P<0.05) were evident between 337 and 36 1  Julian days. 
As in the controlled environment experiment, 'Scarlet Pimpernel'  and 'Vibrance '  buds 
that initially had a mean diameter of 2 . 1 -2.5 mm were more likely to become floral than 
those of any other size class.  The proportion of buds that remained vegetative increased 
with decreasing bud size class (highest proportion below 1 .0 mm) .  The largest 
proportion of unbroken buds in both cultivars was usually below 1 .0 mm in diameter. 
1 5 1  
7 ---- ???-?- ?  - ---- -?  ----
6 
..... c: 
E 5 -0.. 
0 
<l) 
? 4 
-o 
4-0 
<l) 
? 3 
..... 
C/) 
2 
1 
7 
6 
..... 
? 5 E 0.. 
0 
<l) 
? 4 
u 
4-0 
<l) bJ) 
b5 3 -
2 
A 
-- --- ---- ------ -? ? ?  - -- -??? -- ? --- ? --?-?- ??-??- --- -- -
B 
? J ? 
I 
I 
/ 
I 
W I Amb DL 
......_ Amb T I Amb DL 
Amb T I LD 
1 ?,-----?-----,----------,---------------------? 
320 340 360 
Jul ian day 
1 5  35 
Figure 4.4 Mean (?SE) rate of  floral deve lopment in plants of  Metrosideros 
excelsa (A) 'Scarlet P impernel '  and (B) 'V ibrance' grown continuous ly  in 
greenhouses maintained at e ither ambient temperature (Amb T) w ith e i ther 
ambient day lengths (Amb DL) or a photoperiod of 16 h (LD) ,  and at a day I 
n ight temperature of 24/1 7?C (Warm) w ith ambient day length (Amb DL) .  
1 52 
4.4 Discussion 
Metrosideros excelsa 'Vibrance' was highly responsive to photoperiod. While a 
significant number of 'Scarlet Pimpernel ' plants flowered under both short days and 
long days, few or no 'Vibrance' plants flowered under long days. However, in both 
cultivars the intensity of flowering was significantly reduced by exposure to long days. 
Consequently, Metrosideros excelsa can be considered to be a facultative short-day 
plant. Temperature also had a significant effect on flowering, with the controlled 
environment cold treatment completely inhibiting flowering of 'Vibrance' and reducing 
the intensity of flowering of 'Scarlet Pimpernel ' .  A combination of lowered 
temperatures and short photoperiods are also important for floral induction in a number 
of temperate and tropical species within Myrtaceae. Shillo et al. ( 1985), for example, 
showed that Chamelaucium uncinatum required a four week treatment of short days and 
cool (20/14?C) temperatures for flower initiation and development to occur. Similarly, 
combinations of short days and lowered temperatures caused flowering in 
Leptospermum scoparium (Zieslin 1 985) and Hardenbergia (King 1998).  In contrast to 
M. excelsa, some species in Myrtaceae appear to be responsive to either photoperiod or 
temperature. Bolotin ( 1 975), for instance, reported that seedlings of Eucalyptus 
occidentalis (less than one year old) were capable of precocious flowering when grown 
under a long-day regime of 1 6  h or longer, whereas temporary exposure (4-6 weeks) to 
a relatively low temperature regime in E. lansdowneana was sufficient for promoting 
floral initiation, regardless of the daylength (Moncur 1 992). Moncur and Hasan ( 1 994) 
also suggested that low temperature rather than a change in photoperiod is a strong 
stimulus for flowering in E. n itens. 
The variability in flowering responses exhibited by the different cultivars of M. excelsa 
was not unexpected. Variation in responses may reflect the different parental origins of 
cultivars. Noack et al. ( 1 996), for example, investigated the influence of different low 
temperature treatments on the timing and degree of flowering in three New Zealand 
He be cultivars. For He be 'V ariegata' , the duration of the low temperature requirement 
for obtaining an equivalent number of flowers was longer than that in 'Inspiration' and 
'W aikiki ' ,  and was attributed to the long chilling period experienced in the natural 
environment by the parental stock of the former cultivar. King et al. ( 1 995) also 
reported on the flowering responses of different ecoytpes of Pimelea ferruginea grown 
under controlled conditions and found that selections from cooler southerly sites 
1 53 
flowered with a lower temperature optimum than those from northerly sites. Thus, 
between-cultivar differences in M. excelsa were more likely due to differences in 
cultivar responsiveness rather than to differences in the conditions under which the 
plants were grown before experimentation began. Plants of both cultivars were grown 
under identical conditions for several months during the previous year, including the 
time of the spring growth flush when potential floral meristems would have been laid 
down. 
The optimum conditions for floral initiation within the controlled environment 
experiment were Cool ( 1 7/ 14?C) temperatures and Short Daylength. In regards to the 
number of inflorescences per plant, 'Vibrance' induced under Cool/Short Day 
conditions showed a similar response to those grown in the ambient 
temperature/ambient daylength greenhouse environment, suggesting that a minimum 
requirement of 1 5  weeks was suitable for maintaining a comparable flower intensity. 
'Scarlet Pimpernel ' appeared to be more responsive to a range of treatments and 
durations ( 1 0  and 1 5  weeks) to induce flowering. However, in both cultivars the 
optimum number of inflorescences per plant was achieved with the 1 5  week Cool/Short 
Day length treatment. The effect of this treatment imposed during the controlled 
environment experiment may have been greater had the treatments been imposed for 
longer than 1 5  weeks, especially in the 'Vibrance' .  In addition, plants may have borne 
more flowers had the environment in the forcing house to which the plants were 
transferred been optirnised for floral development subsequent to floral initiation. 
However, because treatment effects were maintained during forcing, it can be 
considered that the forcing house environment was satisfactory for its purpose. For this 
reason, the lack of floral response in plants held continuously in the forcing house can 
be ascribed to the unsuitability of its warm, long days for floral induction, rather than 
for subsequent floral development. 
In field-grown plants of M. excelsa, floral initiation is observed over a 6- 1 2  week period 
during the time of most rapid decline in daylength in autumn (Sreekantan et al. 200 1 ) . 
It is therefore likely that floral initiation in our controlled environment experiment 
occurred before plants were transferred to the forcing house. The delay of floral 
development in 'Scarlet Pimpernel' plants in the cool, long day treatment relative to the 
cool, short day treatment may have arisen directly from a delay in floral initiation in the 
former treatment. 
1 54 
Exposure of plants to inductive conditions for up to 1 5  weeks in the controlled 
environment experiment advanced flower development by eight weeks compared to that 
of plants grown continuously under ambient conditions. Floral initiation was probably 
well advanced in both sets of plants before the plants from the controlled environment 
were transferred to the forcing greenhouse. Therefore, the difference in flowering time 
was due to a shorter floral developmental time between initiation and anthesis in plants 
transferred to the forcing house. It was interesting to note that 'Scarlet Pimpernel' 
plants transferred from the 10 and 15 week cool , short day treatment reached anthesis 
only two weeks apart. Similarly, Noack et al. ( 1 996) found that Hebe plants, subjected 
to longer low temperature treatments followed by a transfer to forcing conditions 
flowered more rapidly compared to those exposed to shorter low temperature 
treatments. 
In the greenhouse experiment, flowering intensity was greater in the ambient 
temperature and daylength environment, one in which vegetative growth was minimal 
during winter. The relationship between reduced vegetative growth and enhanced 
flowering is a common trend in woody species (Bernier et al. 198 1b) .  Developing floral 
meristems are intensive centres of growth, requiring assimilates and other metabolites. 
A reduction in shoot growth enables the plant to channel these nutrients to the growing 
bud (Halevy 1 987) .  
'Scarlet Pimpernel' plants from the Ambient Daylength and Ambient Temperature/ 
Long Day greenhouse treatments reached anthesis simultaneously despite their 
day length or temperature treatments. This lack of treatment response may be due to an 
overriding effect of irradiance, photosynthetic photon flux levels would have increased 
similarly across all greenhouse treatments. Although 'Vibrance' plants receiving the 
Ambient Temperature treatment reached anthesis at approximately the same time as 
those of 'Scarlet Pimpernel' ,  floral development in this cultivar was more accelerated 
by the Warm temperature treatment. For instance, 'Vibrance' inflorescences in the 
forcing Warm/ Ambient Day length environment reached an thesis and senesced 
significantly faster compared to the other environments, suggesting a greater cultivar?
specific sensitivity to temperature for flower opening than that of 'Scarlet Pimpernel' .  
King et al. ( 1 992) demonstrated a similar response in Pimeleaferruginea, whereby 
plants transferred to a warm environment (24/1 9?C) following seven weeks of cool 
155  
inductive conditions ( 1 5/ 10?C) accelerated floral development relative to those held 
continuously in the cool environment. Similarly, opening patterns in the Portulaca 
species from Japan are influenced strongly by temperature and, less so, by light 
(Ichimura and Suto 1 998). Rapid flower opening occurred when temperatures were 
increased to 20?C. This response was also intensified when light (as opposed to dark) 
conditions were applied. 
This study supported the hypothesis that the size of the bud during induction can 
significantly affect its competency to become an inflorescence. Interestingly, resting 
buds of an intermediate diameter at the time of their treatment were more responsive to 
inductive signals.  Such a positive correlation between bud size and probability of buds 
becoming floral has been observed for field-grown Metrosideros excelsa trees (Clemens 
et al. 1 998, Sreekantan et al. 200 1 ) . Sreekantan et al. (200 1 )  observed that larger buds 
were more likely to be floral and bear more cymules in each inflorescence than smaller 
buds. However, they also noted that some of the largest buds borne on vigorous shoots 
tended to be vegetative. They attributed this to the early development of leaves within 
the resting bud before the onset of inductive conditions .  A few studies on other species 
have shown a similar relationship between bud size and floral competency, such as in 
Iris (Doss and Christian 1979), Vaccinium macrocarpon (Patten and Wang 1 994) and 
Eryngium planum (Ohana and Weiss 1 998). Physiological processes underlying this 
phenomenon have not been reported, although in the case of Metrosideros this effect 
may be due to the timing of shoot developmental events during the previous vegetative 
growth flush (Clemens et al. 1 998,  Sreekantan et al. 200 1 ) .  Larger buds presumably 
contain more developmentally advanced meristems that are better able to respond to 
environmental signals than those in smaller buds. 
In conclusion, this study supported the hypothesis that flowering in Metrosideros 
excelsa can be manipulated by applying or withholding cool temperatures and short 
days, and by advancing flowering time using warm, long day conditions after inductive 
conditions have been applied. Cultural treatments that synchronise development of 
quiescent bud populations within plants could be used to increase the proportion of buds 
that become floral, provided inductive treatments are applied when buds are of an 
appropriate size. 
1 56  
Chapter 5 
Effect of irradiance on floral induction and development in 
Metrosideros excelsa 
5.1 Introduction 
The influence of temperature and photoperiod on flowering has been a recurrent theme 
in the literature over the past century. In many species, plant responses to changes in 
temperature and/or photoperiod can precede the onset of floral activity in meristematic 
regions. However, less emphasis has been placed on other environmental factors such 
as irradiance which can have a significant effect on flower number and quality in many 
species. In several myrtaceaous woody plants, the manipulation of temperature and/or 
photoperiod has been successful in promoting flowering in genera such as 
Leptospermum (Zieslin 1985),  Chamelaucium (Shillo et al. 1 985) and Eucalyptus 
(Moncur 1 992), whereas only in C. uncinatum (Dawson and King 1 993) has the effect 
of irradiance been examined. In this species, the level of irradiance applied during 
induction (up to 750 Jlmol m-2 s- 1 ) was linearly related to the number of flowers 
produced per plant. 
A common theme demonstrated in various woody plants has been this positive 
relationship between flower number and irradiance. Omamentals such as Begonia x 
hiemalis (Myster 1999), Hibiscus rosa-sinensis (Neumaier et al. 1987),  Kalanchoe 
blossfeldiana (Mortensen 1 994) and Rosa species (Anderson 1993) produce a greater 
number of flowers under high compared to low irradiance treatments. Associated with 
plants growing under these different irradiance treatments are changes in the status of 
leaf chlorophyll and carbohydrate concentrations, often being features associated with 
flowering ability and number (Jiao and Grodzinski 1 998, King and Ben-Tal 200 1 ) .  In 
some omamentals such as Leucospermum or Hamelia patens, exposure to low light can 
also affect flower morphogenesis or delay the time to anthesis, respectively (Jacobs and 
Minnaar 1 980, Armitage 1 995). For instance, flower stem length in various woody 
species such as Boronia heterophylla and R megastigma (Plummer et aL 1 998),  
1 57 
Begonia x hiemalis (Myster 1 999) and various cultivars of Rosa (Bredmose 1 993) were 
significantly increased when grown under low irradiance. 
The effects of varying levels of irradiance on flowering and vegetative growth has been 
documented in number of other woody perennials, such as Boronia megastigma 
(Roberts and Me nary 1989),  Garcinia mangostana (Wiebel et al. 1 994) and A verrhoa 
carambola (Marler et al. 1 994 ) .  The emphasis of these and other studies has been on 
manipulation of irradiance either outdoors or in greenhouse environments using 
artificial lighting or through the use of neutral density shade-cloth. However, the use of 
controlled and manipulated environments can provide the most satisfactory results for 
interpreting irradiance effects on flowering and vegetative performance by removing 
any confounding environmental factors, such as temperature. 
This study investigated the role of irradiance on floral and vegetative features under 
controlled inductive conditions using a range of artificially supplied photosynthetic 
photon fluxes (PPF) on adult pot plants of M. excelsa. The aim was to quantify the 
effects of irradiance on floral induction/initiation and subsequent development, and 
relate both flowering and vegetative growth patterns to physiological changes within the 
plant during induction. It is hypothesised that: 
Irradiance is positively related to the number of flowers produced in M. excelsa 
when applied in conjunction with treatments inductive for flowering. 
5.2 Materials and Methods 
5.2.1 Plant material 
Experiments were conducted using two commercially available cultivars ( 'Lighthouse' 
and 'Scarlet Pimpernel' )  of Metrosideros excelsa. Thirty-two mature 1 .5 m (in height) 
'Lighthouse' plants were obtained in August 1 998 from Lyndale Nurseries, Auckland, 
and were transplanted into 30 1 containers using a peat and pumice growing medium 
(80:20 v/v) supplemented with control-release fertil isers . Plants were maintained under 
ambient outside conditions and were irrigated daily. 
158  
fu addition, 32  micropropagated 'Scarlet Pimpernel' plants, obtained in August 1 997 
from Lyndale Nurseries Auckland Ltd . ,  grown in 50 mm square plastic pots were 
transplanted into 0.82-7 .2 1 containers containing a 90: 10 (v/v) growing medium 
composed of expanded clay granules ( 1 -4 mm diameter, Hydrotron, NZ Hydroponics 
futernational Ltd. ,  Tauranga) and vermiculite (4 mm grade, Revertex Industries Ltd).  
Each plant was supplied with 4.0 g controlled-release fertiliser. Plants were grown in a 
temperature controlled greenhouse (range 15 -23?C) for 9 months. Plants were trained 
up as a single-stem shoot by pinching all developing axillary buds, a technique shown to 
promote early vegetative phase change (Chapter 1 ) .  Upon reaching an adult vegetative 
state, as indicated by the attainment of leaf shape, size and texture characteristics 
indicative of adult plants, they were transplanted into 7 1 containers using a 80:20 (v/v) 
mixture of peat and pumice, and supplemented with Osmocote. The single-stem plants 
were transferred in June 1998 to an outdoor environment experiencing ambient late?
winter conditions where plants were allowed to form side branches. Both cultivars were 
maintained under these ambient conditions until being transferred into their 
experimental environments on 1 2  March 1999. 
5.2.2 Experimental environments 
Plants of both cultivars were transferred to controlled environment rooms to receive one 
of four irradiance treatments provided during inductive 8 h days, for 20 weeks. The 
four controlled environmental rooms were located at the New Zealand Controlled 
Environment Laboratory, The Horticulture and Food Research fustitute of New Zealand 
(HortResearch), Palmerston North. Plants assigned to each room were arranged in a 
completely randomised block design using four blocks each containing two replications 
of both cultivars. Plants were assigned to each block based on a stratified randomised 
sample of the total number of terminal axillary buds. 
Lighting in each of the rooms was provided by either two, four or six 1kW Metalarc 
lamps (8 h)  combined with either one, two or six l kW Tungsten halogen lamps (8 h) 
producing a PPF output of either 1 74, 567, 96 1 or 1 355 Jlmol m?2s- 1 per room (PPF 
increments of 394 units) .  fu each room, the temperature was maintained at (day/night) 
1 7/ 14?C with a vapour pressure deficit at 0 .4/0.3 kPa, a relative humidity at 79/8 1 %  and 
159 
a dew point at 1 3/ 1 1 cc. The concentration of C02 was maintained at 350 111 r 1 .  
Downward airflow through the plants was at a velocity of  0.3-0.5 m s? 1 ? Watering to 
container capacity was conducted for 10 min twice daily using a microtubule irrigation 
system. Plants were rotated every 7 d to ensure an even light exposure and to avoid any 
positional effects within the room. 
Concurrently, eight plants of each cultivar serving as the control treatment were placed 
outside under ambient conditions (early-autumn through to mid-winter) for 20 weeks at 
the Plant Growth Unit, Massey University, Palmerston North. The arrangement of 
plants in the ambient treatment followed the same criteria as for those in the controlled 
environments. Temperatures were monitored outside using a model H8 Hobo shuttle 
logger (Scott Instruments, Hamilton, NZ). The mean monthly day/night temperatures 
over the 20 week period for 12 March through to 30 July 1 999 were 2 1 .6/ 1 6?C, 
1 6.6/ 1 1 .9?C, 1 3 . 3/9 .3?C, 10/9.6?C and 10.2/8.4?C, respectively. 
Following 20 weeks exposure to the four irradiance treatments and the ambient 
treatment, all plants were transferred on 30 July 1 999 to a forcing greenhouse 
maintained under warm, LD conditions. The greenhouse was vented when temperatures 
exceeded 2 1  oc and heated below 1 5?C. Day length was extended in the greenhouse 
using 100 W incandescent lights ( 1 0  jlmol m-2 s- 1 ) .  Plants were maintained in this 
environment until plants had reached anthesis, during which time anatomical 
measurements were made on vegetative and floral shoots . 
5.2.3 Bud measurements 
At the start of the experiment, the diameter of all distal axillary buds was measured so 
as to determine the fate of those buds during (e.g. histological examination) and at the 
completion of the experiment (e.g. presence/absence of an inflorescence) .  Leaves 
subtending every pair of axillary buds from each plant were labelled. The initial 
diameter measurements were also used to classify the buds into one of three size classes 
from which later samples were taken for histological examination. The three size 
classes were defined as small (<1 .5  mm), medium ( 1 .6-2.0 mm) and large (>2.0 mm) 
buds. 
1 60 
5.2.4 Bud histology 
Bud samples from the three bud size classes were collected after 1 3 , 20 and 23 weeks 
from the start of the experiment, which occurred during and at the end of the application 
of the irradiance treatments, and after three weeks in the forcing greenhouse, 
respectively. During each of these sampling times, one distal axillary bud per size class 
was randomly selected and carefully removed from each 'Lighthouse' plant using a 
razor blade. Bud samples were individually transferred to 1 rnl Eppendorf tubes 
containing a FAA fixative solution and subsequently processed following the protocol 
described in Section 3 .2.4 for viewing under light microscope. 
5.2.5 Inflorescence measurements 
At 1 1  weeks after plants were transferred to the forcing greenhouse and most 
inflorescences (89%) were at Stage 1 (October 1 3- 1 5, 1 999), a number of records were 
obtained for each inflorescence. This included records of the number of inflorescences 
per plant from each cultivar. For each inflorescence, the bud size class (as determined 
at the start of the experiment) from which the inflorescence had been derived was 
determined. In addition, quantitative measurements were made along the axis of each 
inflorescence. This included count data for (a) the number of individual nodes bearing 
cymules (?5 mm in length), (b) the number of cymule pairs, and (c), if present, the 
number of fully expanded leaves along the actively growing vegetative axis originating 
from the terminal bud of the indeterminate inflorescence. 
Linear measurements using a hand-held digital caliper (Sylvac, Industrial Tooling Ltd, 
NZ) were made of the regions from which count measurements were conducted, 
included the length (a) from the base of the inflorescence peduncle to the point of 
attachment of the first cymule (referred to as the peduncle), (b) from the point of 
attachment of the first cymule through to that of most apical cymule (referred to as the 
inflorescence axis), and (c) of the actively growing vegetative axis originating from the 
terminal vegetative bud (referred to as the terminal vegetative axis). Qualitative data 
describing the level of activity of the terminal vegetative bud was used to classify the 
bud as either actively growing, aborted or not broken/broken (>4 bud scales shed) . 
, ,.. ,  1 0 1  
The stage of floral development was scored weekly starting from the time all cymules 
on inflorescences were macroscopically visible (Stage 1 )  through to anthesis (Stage 4) . 
Specific classification of stages of floral development were based on those in Table 4. 1 
of this thesis .  This included stages corresponding to the shedding of bracts sheathing 
the floral receptacles (Stage 1 ), expansion of floral receptacle girth (Stage 2), elongation 
of stamens/style (Stage 3), attainment of anthesis (Stage 4), onset of stamen wilting 
(Stage 5) ,  onset of stamen abscission (Stage 6) and abscission of all stamens (Stage 7) .  
5.2.6 Vegetative measurements 
On the eleventh week after plants had been transferred into the forcing greenhouse, the 
six vegetative shoots were selected randomly from each plant of both cultivars . The 
length and number of nodes were recorded on each shoot. 
5.2.7 Chlorophyll determination 
One leaf per plant was harvested between 0900- 1000 h from each 'Lighthouse' plant on 
week 20, immediately before plants were transferred to the forcing greenhouse. A 
representative leaf located on each plant' s  upper canopy was randomly selected on the 
second or third node proximal to an axillary branch. The mid-rib which lacks 
chlorophyll was discarded and a 0.2 g leaf blade sample was weighed and transferred 
into a glass cuvette containing 4 ml of dimethylformarnide (DMF) . Samples were 
chilled temporarily on wet ice before being transferred to a dark 4 oc refrigerator for 48 
h. Each sample was diluted with 1 1 10  DMF solution and the extinction coefficient was 
measured on a spectrophotometer at wavelengths 664 nm (A 664) and 647 nm (A 647) , 
corresponding to chlorophyll a and b, respectively (lnskeep and B loom, 1 985) .  
Chlorophyll content was measured in mg/g FW using the following formula: 
Chlorophyll a (Chl a) = 1 2.7*0.007+A 664 - 2.79* A 647 
Chlorophyll b = (Chi b) 20.7* A 647 - 4.62 * 0.007 + A  664 
Total Chlorophyll = Chi a + Chl b 
162 
5.2.8 Carbohydrate extraction and determination 
A second leaf sample was collected concurrently on 'Lighthouse' plants following the 
same selection and harvest criteria as described for the chlorophyll analyses.  Leaf blade 
tissue were snap frozen in liquid nitrogen and stored on wet ice in tin foil sachets. 
Subsequent processing, extraction and determination of soluble sugars (sucrose, 
fructose and glucose) and starch concentration follow the protocol described in Section 
3.2 . 1 0. The methodology was modified by Dr. Jocelyn Eason (The Crop and Food 
Research Institute) to accommodate the use of micro-amounts that could be read on a 
plate reader as opposed to a spectrophotometer. 
5.2.9 Statistical analyses 
All biometrical analyses were conducted using the SAS program (SAS Institute, Cary, 
N.C. ,  USA). One-way Analyses of Variances (ANOVA) were used to detennine the 
effect of irradiance treatment on the size of microscopic floral meristems and leaf 
concentrations of chlorophyll, starch and soluble sugars. Multivariate ANOV A were 
used to test the effect of irradiance and cultivar treatments on a number of floral and 
vegetative characteristics .  The main and possible interaction effects of irradiance and 
bud size on the proportion of flowering plants and the number of inflorescences per 
plant were analysed using ANOV As with the GENMOD procedure (procedure 
described further in Section 4.2.4). To test the effect irradiance and cultivar treatment 
on the rate of floral development, a repeated measure ANOV A was used using time as 
the repeated measure. Data was log transformed when necessary in order to normalise 
var1ances. 
5.3 Results 
5.3.1 Histological examination of buds 
Buds from 'Lighthouse' plants were harvested at 1 3 , 20 and 23 weeks from the start of 
the experiment in order to assess the effects of irradiance and bud size on the presence 
(or absence) of cyrnule primordia and, if present, on the level of floral meristem 
initiation and differentiation. At 1 3  weeks (seven weeks before removal from the 
1 63 
inductive irradiance treatments), a histological study of buds showed no evidence to 
suggest that floral initiation had occurred. There was a lack of cellular activity in the 
axils of bud scales (Plate 5 . 1  A). 
By week 20 (the end of the inductive irradiance treatments), there was a significant 
effect of irradiance environment on the proportion of buds with cyrnule primordia that 
were microscopically visible (P<0.05), although there was no effect of bud size 
(P>0.05) or interaction of bud size with irradiance environment (P>0.05) (Figure 5 . 1 A 
and Plate 5 . 1  B) .  Buds with cyrnule primordia were observed only in the 567, 96 1 and 
1 355 llrno1 rn?2 s? 1 controlled environments, although their proportions did not differ 
significantly between these three environments (P>0.05). 
At 23 weeks (third week in the forcing greenhouse), there was a significant interaction 
effect of bud size and irradiance environment (P<O.OS) on the proportion of harvested 
buds with cyrnule primordia, although no main effects of either treatment (P>O.OS) 
(Figure 5 . 1 B). Buds in the 567 llrnol rn?2 s? 1 PPF environment that were from the 
medium size class ( 1 .6-2.0 mm at the start of the experiment) had the highest proportion 
of buds with cyrnule primordia (50% of buds per plant), whereas low levels (6%) 
occurred in the 1 74 !lrnol rn?2 s? 1 PPF environment, and only from the large bud size 
class. No cyrnule primordia were observed at 23 weeks in the small and medium bud 
size classes from either the 1 74 llmol rn?2 s? 1 or ambient environment, or from the small 
bud size class from the 96 1 !lrnol rn-2 s? 1 environment. 
An examination of floral rneristem dimensions between the 20 and 23 week period 
showed that growth during this period was primarily directed towards elongation of the 
cyrnule primordia as shown by a significant increase in meristern length (P<O.OS) rather 
than width (P>O.OS) over that three week period. During this time span, the mean 
length and width dimensions of cyrnule primordia had increased from 1 30?5 1 11m to 
369?144 !-Lrn and from 1 63?45 !liD to 269?70 !liD, respectively. Meristern size (both 
length and width parameters) was significantly and positively correlated with the 
diameter of the bud at the time of harvest (both R2=0.487, P<O.OS) (Figure 5 .2  A-B) .  
164 
Plate 5 . 1  M icroscopic study of vegetat ive and flora l in i t iat ion and development i n  buds 
col lected at 20 and 23 weeks after the start of the experiment w ith  20 weeks treatment to 
d ifferent i nductive irrad iances. A :  Vegetative meristem showi ng a lack of f loral 
meristematic act iv ity in leaf ax i l s  at 23 weeks ( 1 74 f-tmOI m?2 s- 1 ) .  B: In it iat ion of floral 
primord ia  i n  scale ax i l s  at 20 weeks (567 f-tmOI m-2 s- 1 ) .  C :  Development of cymule  
primordia at 23 weeks (567 flmol m-2 s- 1 ) .  D :  Floral receptac l e  and subtending bract 
d ifferentiation ev idenl at 23 weeks (567 f-tmOI m?2 s- 1 ) .  Bar = 500 f-tm . AM = Apical 
meristem. B = Bract. CP = Cymule primordium. L = Leaf. R = F loral receptacle .  S = 
Scale. 
1 65 
A 
? ..... c ? 
0. 
I-Cl) 0. 
C/J E 
Cl) ..... C/J 
I-Cl) E 
? I-
0 
G: 
B 
50 
20 
1 0  
? ..... c ? 
0.. 
I-
Cl) 
0.. 
C/J E 
Cl) ..... C/J 
? ;:::: Cl) E 
? I-
0 
G: 
50 
30 
-?-?-- -?-? -
. l 
-----------------1 
,,.., .... ---
? 
- --?- ---- --? -- - -
Figure 5 . 1  Effect of  irradiance and ambient cond it ions (amb) on the cumu lat ive 
proportion of floral meristems per plant at  (A) 20 and (B) 23 weeks from the start of 
the experiment i n  three bud s ize c l asses of M. excelsa 'Lighthouse' . Week 20 
corresponded to the last week plants were held under inductive i rradiance cond it ions 
and week 23 to the th ird week plants were held u nder forc ing condit ions. 
1 66 
E' 
::l 
-
..J:: 
...... 
00 c: <!) 
......l 
1 200 
1 000 
800 
600 
400 
200 
0 
A Linear fit 
P<0.05 
r=0.698 
700 T - ----- ----" -"""" - - -1 B Linear fit 
600 l P<0.01 
500 l' r=0.698 
? 
? 
? 
? 
E' 
2 400 ? ? 
..c :?
300 l 
? 200 ? ? 
1 00 I 
0 ---------,---------.------------------?------?
4 5 6 7 8 9 
Bud s ize (mm) 
Figure 5.2 Corre lations between bud s ize at t ime of harvest w i th maximum 
(A) length and (B) w idth of floral meristems in Metrosideros excelsa 
Lighthouse' . Data pooled for buds col lected at 20 and 23 weeks from 
the start of the experiment. 
1 67 
At 23 weeks, on the terminal regions of the cymule primordia, three receptacles and 
opposing bracts on either side of the three receptacles were beginning to differentiate 
(Plate 5 . 1  C-D). Also by this time, the size of the floral meristems were significantly 
affected by irradiance, being larger, at least in terms of length, in the 567 J..lmol m?2 s? 1 
treatment thar1 in the other treatments (Figure 5 .3) .  
5.3.2 Proportion of plants flowering 
Flowering occurred in both cultivars and in all irradiance treatments (including control 
plants induced under ambient conditions) with the exception of 'Lighthouse' plants in 
the 1 74 J..lmol m-2 s- 1 PPF treatment. Both cultivars grown under the differing PPF 
treatments displayed a similar unimodal pattern of flowering, with the modal peak 
occurring at an irradiance treatment of 567 J..Lmol m-2 s- 1 (Figure 5 .4). The main effects 
of both irradiance and cultivar type were highly significant (P<0.00 1 )  although there 
was no significant interaction effect (P>0.05) .  Across all irradiance treatments (but not 
in the ambient controls), 'Scarlet Pimpernel ' consistently produced a greater number of 
inflorescences than 'Lighthouse' with both cultivars averaging the highest yields of 5 .6  
and 2 .5 inflorescences per plar1t, respectively, at the 567 jlmol m-2 s? 1 . There was 
sufficient data from the 'Scarlet Pimpernel' cultivar to test whether the size of buds at 
the start of the experiment had a significant effect on the proportion of inflorescences 
that developed within each environment. The main effects of irradiance (P<0.000 1 )  
and initial bud size (P<0.05) but not their interaction (P>0.05) significar?tly affected the 
proportion of buds that produced inflorescences (recorded at Stage 1 of floral 
development) (Figure 5 .5). The irradiance environment and bud size class with the 
highest proportion of inflorescences were in the 567 J..lmol m-2 s- 1 and large size class 
treatments,  respectively. Intermediate levels in the proportion of inflorescences from 
each size class occurred in the 96 1 jlmol m-2 s- 1 environment (<6%) and, to a lesser 
extent, in the 1 355 J..lmol m-2 s? 1 environment ( <4%) whereas, relatively few 
? inflorescences per bud size class ( <3%) occurred in the 1 7  4 jlmol m?2 s - 1  and ambient 
control environments. 
1 68 
1 20 --- - -? ?-- ??-- ?-- ---- - -? ? - - ----
1 00 
..-.. 
D Length 
D Width 
E 80 ..3-
v N 
Cl) 60 
40 
20 
1 74 567 96 1 1 355 amb 
Figure 5 .3  Effect of i rradiance treatments and ambient condit ions during 
floral i nduction on the mean (? SE) s i ze of  floral meristems at  23 weeks 
after the start of the experiment. P lants held under amb ient induct ive 
condit ions denoted "amb" .  
169 
--- ---- - - ---? 
SP 
1 74 567 96 1 1 355 amb Cul ti var 
Irrad iance (!-lmol m-2 s- 1 ) 
Figure 5.4 Effect of inductive i rradiance and ambient condit ions on the 
mean number of inflorescences per plant in Metrosideros excelsa 
'Lighthouse' (LH) and 'Scarlet P impernel ' (SP) . 
Cll <J) (.) t: <J) 
(.) Cll <J) .... 
0 
4-t: .--.. 
? - ? 4- -0 
>, 
(.) t: <J) ::l 
er <J) .... 
? 
4 
2 -
0 
1 74 
-?-- --
567 961 1 355 
' 
' cj1 -L;22 0 
amb 
<1 .6 Bud s i ze 
c lass (mm) 
Figure 5 .5 Effect of i rradiance and ambient conditions (amb) on the 
cumulat ive proportion of inflorescences per three bud s ize c lasses per 
p lant of Metrosideros excelsa ' Scarlet P impernel '  at 1 1  weeks after 
transference to the forcing greenhouse env i ronment. 
1 70 
5.3.3 Flowering time 
Both cultivars from all treatments reached anthesis at approximately 1 5  weeks after 
being transferred to the forcing greenhouse (rnid-November) . A repeated measures 
analysis of variance showed no significant effect of either inductive irradiance 
environment (P>0.05) or cultivar type (P>0.05) on the rates of floral development. 
Inflorescences took approximately seven weeks to progress from a stage where flower 
buds were visible and bracts had shed (Stage 1) through to anthesis (Stage 4). A further 
five weeks passed before all inflorescences had senesced (Stage 7) .  
5.3.4 Inflorescence morphology 
A comparative assessment of inflorescence morphological characteristics was 
conducted between cultivars and across irradiance treatments at 3 1  weeks ( 1 1  weeks 
after transference to the forcing greenhouse) when most (89%) of inflorescences from 
both cultivars had reached Stage 1 . With the exception to the length of the vegetative 
shoot arising from the terminal bud of the inflorescence, the inductive irradiance 
treatments had little to no effect on a number of inflorescence morphological 
characteristics and those that were significant were primarily cultivar specific (Table 
5 . 1 ) .  For example, measurement of several regions in inflorescences of 'Scarlet 
Pimpernel '  plants were significantly longer than those of 'Lighthouse' ,  including the 
mean length of the peduncle, inflorescence and overall terminating vegetative shoot 
regions of the inflorescence. Similarly, the mean number of individual cymule nodes 
(and cymule node pairs) per inflorescence and leaf pairs on the terminating vegetative 
shoot region were significantly greater in inflorescences of 'Scarlet Pimpernel' .  
The frequency of vegetative shoots elongating from the terminal bud of inflorescences 
within each environment varied between cultivars (Figure 5 .6 A-B) .  No fewer than two 
thirds of 'Scarlet Pimpernel' plants produced terminal vegetative shoots in all of the 
inductive environments. In contrast ,  100% of 'Lighthouse' plants produced terminal 
vegetative shoots only from the ambient inductive environment. The highest proportion 
of aborted terminal apices in inflorescences occurred in approximately 75- 1 00% of 
inflorescences in 'Lighthouse' and 6% in 'Scarlet Pimpernel from both the 567 and 96 1 
Jlmol m-2 s- 1 PPF environments. 
1 7 1  
Table 5 . 1 Effect of inductive environment and cultivar on inflorescence morphological 
characteristics .  Data are mean ? standard error of count and length measurements . 
Significance tests for cultivar (CV) and environment (ENV) at 5% level .  
Inflorescence Cultivar Si o-nificance 
characteristic 
Lighthouse Scarlet Pimpernel CV ENV 
Counts 
Cymule number 3 .6  ? 0.4 5 . 8  ? 0.5 <0.0 1 n.s .  
Cymule nodes 2.4 ? 0.2 3 . 6  ?0.2 <0.01  n.s .  
Terminal shoot nodes 1 .9 ? 0.5 4.2 ? 0.4 <0.00 1 n.s .  
Length (mm) 
Peduncle axis 9 .8  ? 0.9 1 2 .0 ? 0.5 n.s. n.s .  
Inflorescence axis 14 .8 ? 1 .6 2 1 .9 ? 1 .5 <0.0 1 n.s .  
Terminal shoot 7 .7 ? 2.2 1 5 .9 ? 1 .4 <0.01  <0.000 1 
5.3.5 Leaf chlorophyll concentrations 
Chlorophyll a and b, and total chlorophyll concentration were significantly affected by 
irradiance treatment (P<0.00 1 )  (Table 5 .2), showing a negative correlation with the 
level of PPF (r= -0.386, P<0.05) .  Chlorophyll concentration in plants induced under 
ambient conditions did not differ statistically from any of the irradiance treatments, 
except being significantly lower than plants grown at 1 74 jlmol m?2 s? 1 . 
5.3.6 Leaf carbohydrate concentrations 
Leaf total soluble sugars differed significantly in plants from the different irradiance 
environments, there being higher and lower concentrations, respectively, in the ambient 
control and the 1 74 jlmol m?2 s? 1 treatments than the other irradiance treatments (Table 
5 .2). This result was primarily attributed to differences in sucrose and, to a lesser 
extent, to fructose, concentrations between inductive treatments. The highest and 
lowest sucrose concentrations were in plants from the ambient control and 1 74 jlmol m?2 
s? 1 PPF environments, respectively, while intermediate concentrations were from the 
other treatments. Fructose concentration differed only between plants from the ambient 
control and the 1 74 jlmol m?2 s? 1 PPF environments, being significantly higher in the 
former. There were no differences in fructose levels among the other irradiance 
172 
? 
>-. 
u 
c 
Q) 
::l 
0" 
Q) 
..... (.I..; 
A Lighthouse 
-? - --- --- --?-??-. 
1 00 -
80 ?1 
B Scarlet P impernel 
------? 
1 00 -
80 
60 
amb 
amb 
I 
-? 
Growth 
Abort 
No Growth 
Figure 5 .6 Effect of i nductive env i ronment on the percentage of term inal  buds of 
i nflorescences that had broken and from which vegetative shoots were elongating 
(growth) ,  that had aborted (abort) , or that had not broken (No growth) in 
Metrosideros excelsa (A) 'Lighthouse' and (B) 'Scarl et Pimpernel assessed at 
Stage 1 of i nflorescence development. P lants held under ambient i nducti ve 
cond i tions denoted "amb" .  
1 73 
treatments. Of all sugars tested, only sucrose showed a relationship with chlorophyll 
concentration, being significantly and inversely correlated (R 2= -0.5 10, P<O.OO 1 ) .  
The concentrations of starch differed between environments (P<0.001 ) ,  being 
significantly higher in leaves collected from plants treated at and above a PPF of 567 
11-mol m?2 s? 1 in comparison to the other environmental treatments (Table 5 .2) .  Starch 
concentration in leaves of the control plants was significantly lower than those in all 
other treatments except in the lowest PPF ( 1 74 11-mol m?2 s? 1 ) environment. Overall ,  
starch concentration was positively correlated with PPF (R2=0.398, P<0.00 1 )  and 
negatively correlated with leaf total chlorophyll concentration (R2= -0.32 1 ,  P<O.O l ) . 
Table 5 .2 Effect of irradiance and ambient conditions during floral induction on leaf 
chlorophyll and carbohydrate concentrations in Metrosideros excelsa 'Lighthouse' after 
20 weeks in induction treatments. Data are mean concentrations for leaves harvested 
from each environment. Values followed by different letters (along rows) are 
significantly different according to a Tukey' s multiple range test at a 5% significance 
level .  
Ambient 
Variable control 
Chloroghyll (mg dm-2 ) 
Chlorophyll a 1 34 .6a 
Chlorophyll b 54.3a 
Total Chlorophyll 1 88 .9a 
Carbohydrates (mol g? 1 DW) 
Sucrose 0.028a 
Glucose 0.0 1 8a 
Fructose 0.056a 
Total Soluble 0. 1 03a 
Starch (mg g?1 DW) 1 2 .3a 
Inductive environment 
Irradiance (11-mol m?2 s? 1 ) 
1 74 
279.0b 
102. 1 b 
38 1 .2b 
0.0 1 2c 
0.0 1 2a 
0.036b 
0.07 1 c 
4 .3b 
567 
204. 1 ab 
73 .9ab 
278.0ab 
0.0 19b 
0.027a 
0.042ab 
0.097b 
26.7c 
96 1 
1 20 .3a 
46.9a 
1 67 . 1 a 
0.02 1b 
0.020a 
0.045ab 
0.088b 
42.4c 
Significance 
1 355 P-value 
102.9a <0.000 1 
42.4a <0.000 1 
145.3a <0.000 1 
0.022b <0.000 1 
0.0 14a n.s .  
0.045ab <0.01  
0.08 1 b <0.000 1 
33 .7c <0.000 1 
1 74 
E' E 
------
.c 
..._, 00 c 
a.> 
..._, 0 0 .c C/) 
-E 
? 
.c 
Oo 
c 
? 
0 
0 
.!: 
Cl) 
100 - -------? - - ---?-?? - ?  - -
80 
l 
60 
I}'"'-
? 
40 
20 
1 74 
0 Lighthouse 
D Scarlet Pimpernel 
567 961 1355 
Irrad i ance (f-!mol m?2 s-1 ) 
l 
amb 
Figure 5 . 7  Effect of i rrad iance and ambient cond i tions during floral 
i nduction on vegetative shoot length (mean ? SE) in two cult i vars 
of Metrosideros excelsa determined 1 1  weeks after transference to 
forc ing condit ions.  P lants held under ambient inducti ve  conditions 
denoted as " amb" . 
140 l 
1 20 j ?e (?mol m?' s ')  
100 I 7 
I 9
6 1 
8 1 355 80 I 
60 ? ? I 
---?? - ?-?--- -
? 
- - -
? 
? 
'9 
? 
-- -
? 
? 
-
40 1 Linear fit -20 
1 
P<O.OOO l 
r=0 .655 
0 
0 1 00 200 300 400 500 
Total ch lorophy l l  (mg dm-2) 
Figure 5 .8 Relationship between vegetat ive shoot length and leaf tota l 
chlorophy l l  concentration i n  Metrosideros excelsa 'L ighthouse' grown 
under d ifferent  i nduct ive irradiance treatments. 
1 75 
5.3. 7 Vegetative Growth 
There was a significant interaction effect of cultivar with inductive environment on the 
length of vegetative shoots assessed eleven weeks after transferring to the forcing 
greenhouse (P<O.OO I ) . In comparison to all other treatments, 'Lighthouse' plants 
within the lowest PPF treatment bore longer vegetative shoots (P<O.OO l ) , which was a 
result of both a greater number of nodes (P<O.OOI )  and an increased internode length 
(P<O.OO l )  (Figure 5 .7). Shoot growth in 'Scarlet Pimpernel' was less responsive to 
irradiance than in 'Lighthouse ' ,  and there were no significant differences (P<0.05) in 
the length, node number or internode length for this cultivar. In comparison with the 
plants in the ambient control environment, only 'Scarlet Pimpernel' plants within the 
lowest PPF treatment bore shoots that were significantly longer (P<O.OO I ) .  This was 
attributed to shoots having a greater number of nodes along the shoot axis (P<0.05), but 
not to the length of internodes (P>0.05) .  
Patterns in vegetative growth were associated with chlorophyll concentration as 
assessed from leaf tissue collected from 'Lighthouse' plants at 20 weeks. The mean 
length of vegetative shoots showed a strong positive correlation with total chlorophyll 
(R2=0.429, P<O.OO I )  (Figure 5 .8) .  
5.4 Discussion 
In the present study, the flowering response in two cultivars of Metrosideros excelsa 
was clearly not linearly related to irradiance. Both cultivars of M. excelsa responded 
similarly to the different PPF treatments, producing the highest proportion of 
inflorescences under an intermediate PPF level of 567 11mol m-2 s- 1 , whereas excessively 
high or low irradiance imposed during inductive treatments detrimentally affected 
subsequent floral development in M. excelsa. Thus, a unimodal PPF response in flower 
number suggests that inductive treatments with intermediate levels of PPF, in 
combination with short daylengths and cool temperatures, were essential for maximum 
flower yield in this species. Thus, the hypothesis that irradiance is positively related to 
the number of flowers produced was not upheld by this study. 
176 
A similar uni-modal flowering pattern in response to different irradiances has also been 
reported for Boronia heterophylla grown under ambient conditions. Plants of B. 
heterophylla maintained continuously outside (during cool SD conditions) under a 
shade cloth (75% sunlight or 1022 J.lmol m-2 s? 1 ) produced the highest yield of flowers 
(mean 347 flowers per stem), whereas no shading or shading treatments below 75% 
sunlight produced significantly fewer flowers (:=;; 2 1 8  flowers per stem) (Plummer et al. 
1 998).  This in contrast to a number other studies examining the effect of irradiance on 
flowering whereby a strong linear correlation between flower number and irradiance are 
typically reported (e.g. Bredmose 1993, Dawson and King 1993, Mortensen 1 994) .  The 
results are often attributed to a lack of a broad range of PPF levels, or to a failure to 
subject plants to supra-optimal levels of PPF, which might otherwise provide an 
accurate response curve for determination of conditions optimal for flowering. 
In general, flowering in many species, however, appears to be limited more under low 
than high irradiance (Dawson and King 1 993). The current study is in accordance with 
these observations to the extent that both cultivars produced a greater number of 
inflorescences in PPF treatments at and above 567 rather than at 1 74 J.lmol m-2 s- 1 . For 
example, the 'Lighthouse' produced no flowers that reached Stage 1 in floral 
development under the 174 J.lmol m-2 s? 1 treatment, whereas flowering was recorded in 
all other PPF treatments. In Chamelaucium uncinatum (also Myrtaceae), the number of 
flowers produced was linearly related to the PPF applied during 8 weeks of inductive 
conditions, yielding approximately 400 flowers per plant at -750 J.lmol m-2 s? 1 , whereas 
few to no flowers were observed below 1 60 J.lmol m-2 s- 1 (Dawson and King 1 993) .  
Low light levels have also been reported to decrease the flowering response in  other SD 
species, such as Chrysanthemum (De Jong 1 986), Pimelea ciliata (Slater et al .  1 994) 
and in various crop plants such as grape and kiwifruit (Morgan et al. 1 985) .  
In various species, floral induction and subsequent development at different irradiances 
have been strongly associated with carbohydrate status (Bernier et al. 1 98 1 a, Zieslin 
and Mor 1 990, King and Evans 1 99 1 ,  King and Ben-Tal 2001) .  In rose, for example, 
the lack of assimilates due to low light conditions has generally been assumed to cause 
flower abortions (Zieslin and Moe 1 985), and only a shift in assimilate partitioning 
within the plant (subtending leaves) without an increase in photosynthesis can promote 
177 
shoot or flower development (Maas and Bakx 1 997). The data for Metrosideros support 
the suggestion that there was a lack of an adequate supply of photoassimilates at 1 74 
11mol m?2 s- 1 for promotion of long-term development in floral meristems. This was 
evident by the sucrose and storage carbohydrate (starch) concentrations, which were 
significantly reduced in plants grown in that environment. The reduction in the number 
of inflorescences in plants exposed to high irradiance treatments did not appear to be 
directly related to differences in the concentration of carbohydrates, since similar values 
were reported in these treatments (PPF of 961 and 1 355 11mol m-2 s- 1 ) with that in which 
flower number was highest (PPF of 567 11mol m-2 s- 1 ) .  Therefore, this supports the 
hypothesis that while carbohydrates ,  such as sucrose, may have a 'florigenic' role 
during the flowering process (e.g. increased number of flowers under favourable 
irradiances), they are unlikely to be the only factor mediating this process (Bemier et al. 
1 993, King and Ben-Tal 200 1 ) . 
Evidence to suggest that floral buds in the other PPF environments may have undergone 
abortion comes from comparisons of the proportion of buds that were microscopically 
visible at an early stage of development (week three in the forcing greenhouse) with 
those at Stage 1 of development (week 1 1  in the forcing greenhouse). Within 
'Lighthouse' ,  approximately 6% of buds from the large size class in plants grown at 1 74 
11mol m-2 s- 1 contained floral meristems on the third week while in the forcing 
greenhouse. However, by the eleventh week, no inflorescences had reached Stage 1 of 
development. A between-cultivar comparison also suggests a similar fate . The 
proportion of buds that contained microscopically visible floral meristems (e .g. 
in 'Lighthouse')  with the proportion of buds that reached Stage 1 of floral development 
(e.g. in 'Scarlet Pimpernel' )  suggests that a considerable reduction occurred in the 
number of flowers that reached this later stage of development. For instance, this was 
clearly evident for medium-sized buds in the 567 11mol m-2 s- 1 environment for which 
approximately half of all buds in that size class contained floral meristems but only 
around 1 0% reached Stage 1 of floral development. If this response was not an artefact 
due to differences between cultivars, it would not be unreasonable to conclude that a 
large proportion of buds from different environments and bud size classes had 
undergone abortion while under the warmer forcing greenhouse conditions. It is  
possible that the considerably warmer conditions within the forcing greenhouse may 
1 78  
have inhibited floral development during a critical stage(s) of development. A similar 
suggestion has been made for developing floral primordia in Acacia pycnantha, 
whereby both low light and high temperatures can block different steps in the process of 
floral differentiation (Sedgley 1 985). Thus, transference of M. excelsa plants from the 
inductive conditions to a forcing greenhouse with lower mean temperatures may have 
increased the number of inflorescences that would have developed through to an thesis. 
Despite the possibility of floral abortions, irradiance clearly had a differing effect on 
buds of different size that eventuated into inflorescences. In ' Scarlet Pimpernel ' ,  the 
highest proportion of buds that produced inflorescences occurred in plants containing 
intermediate and large sized buds treated at 567 ).!mol m-2 s- 1 . Similarly, in field grown 
plants of M. excelsa, Sreekantan et al. (200 1 )  observed that flowering capacity was also 
dependent on buds being within a certain size class range during the inductive winter 
season, and that parts of the canopy containing large floral buds were those that were 
brightly lit. The effect of bud size in determining competency for flowering has also 
been acknowledged in various other species, including Iris (Doss and Christian 1 979), 
Vaccinium macrocarpon (Patten and Wang 1 994) and Eryngium planum (Ohana and 
Weiss 1 998) .  However, the physiological processes underlying this phenomenon have 
not been reported, let alone the influence of irradiance during induction. 
It was surprising that the proportion of flowers in plants treated to the ambient control 
environment was significantly reduced in comparison with those under optimal 
controlled irradiance conditions. The lack of a strong flowering response in this 
treatment was probably due to the comparatively lower temperatures experienced by 
these plants under ambient inductive cortditions (mean day/night of 10.2/8.4?C in July) 
in comparison with those in controlled inductive conditions (mean day/night of 
1 7/ 14?C). A reduction in temperature would, therefore, have reduced growth as evident 
in the decreased size and number of floral meristems at three weeks after transference 
into the common forcing greenhouse environment in comparison with those previously 
treated under controlled inductive conditions. Similar to the response in A. pycnantha, 
shifting the M. excelsa plants from the colder ambient conditions to the heated forcing 
greenhouse at a critical stage in floral primordia development may have led to the early 
abortion of flower buds (Sedgley 1 985) .  
179 
Irradiance can significantly affect the morphology of inflorescences, such as the effect 
of low-light in promoting flower stem length (Bredmose 1 993, Myster 1 999). In M. 
excelsa, the proportion of inflorescences with elongating tenninal vegetative shoots was 
considerably higher in 'Lighthouse' (but not 'Scarlet Pimpernel' )  plants grown under 
ambient inductive conditions ( 100%), in comparison with those from the other 
controlled inductive environments ( <20% ). However, this response is unlikely to be 
related to irradiance, otherwise, it probably would have been observed in plants grown 
in one or more of the controlled environments with different irradiances. It is likely that 
this response was due to the cooler temperature regime experienced by plants under 
ambient winter conditions, which may have provided chilling favourable for subsequent 
vegetative growth of the tenninal apex of the inflorescence. 
Vegetative shoot length was significantly affected by irradiance, at least in the 
'Lighthouse' .  By the eleventh week within the forcing conditions, 'Lighthouse' plants 
which had been grown under low light conditions ( 1 74 IJ.mol m?2 s? 1 ) had produced 
significantly longer shoots containing a higher number of nodes. The increased size of 
the shoots in 'Lighthouse'was probably due to the increased capacity of plants to 
produce photoassirnilates as a result of having obtained higher chlorophyll 
concentrations, and presumably a higher chloroplast density, while being grown in the 
low light inductive environment. Acclimation of plants to shady conditions has been 
reported widely in the literature, and include modifications to a number of leaf 
anatomical and morphological characteristics, such as the presence of larger leaves with 
reduced thickness, stomatal density and conductive tissue (Morgan et al. 1 985) .  
Moreover, physiological modifications can include increases in total chlorophyll 
content (Marler et al. 1 994, Marks and Simpson 1 999). These modifications allow for a 
maxirnisation of photosynthetically active area over the total plant mass, a strategy often 
employed in low light environments (Marler et al. 1 994 ) .  Therefore, these conditions 
probably provided an added advantage for vegetative growth once plants of 
'Lighthouse' were transferred to the relatively high light conditions in the forcing 
greenhouse. Moreover, the strong negative correlation between chlorophyll 
concentration and irradiance also suggests that plants transferred from the high light 
environments may have suffered from photoinhibition, a feature acknowledged to cause 
a lowering of chlorophyll concentrations and damage to the photosynthetic apparatus 
1 80 
(Krause 1 988 ,  Marks and Simpson 1999). The lack of a similar low-light response in 
'Scarlet Pimpernel' ,  in terms of an increase in vegetative shoot growth, may have been 
due the diversion of photoassimilates towards floral development, considering that some 
flowering was recorded in this cultivar in the low PPF treatment. This would be 
consistent with the diversion of assimilates towards stronger sinks such as those that 
favour floral growth, as postulated by several authors (Sachs and Hackett 1 983,  Kinet et 
al. 1985) .  
In conclusion, in inductive cool, short-day conditions, irradiance had a significant effect 
on not only the rate of floral primordium growth but also on the proportion of buds 
within a size class that produced floral meristems. Intermediate levels of irradiance 
( -567 11mol m-2 s- 1 ) clearly enhanced the number of inflorescences that were produced, 
while excessively low or high irradiance imposed during inductive treatments 
detrimentally affected subsequent floral development in M. excelsa. 
1 8 1  
Chapter 6 
The postharvest characteristics of Metrosideros excelsa as a cut flower 
6.1 Introduction 
Within the family Myrtaceae, the main postharvest and display problems are associated 
with ethylene-related abscission of whole flower and/or floral organs (Zieslin and 
Gottesman 1 983 ,  Joyce and Poole 1 993), and adverse water relations within cut flowers 
(J oyce and J ones 1 992, J ones et al. 1 993a, Burge et al. 1 996). For example, in cut 
flowers of Metrosideros collina stamen wilting and abscission readily occur in response 
to unfavourable holding solutions and to exogenous and/or endogenous ethylene (Sun et 
al. 2000). Generally, inclusion of sucrose to vase solutions at moderate to low 
concentrations (::;; 1 0%) can improve the flower quality and extend vase life (Jones et al. 
1 993a, Sun et al. 2000). However, the response of cut flowers to different sucrose 
solutions can be variable even in closely related species .  For instance, continuous 
application of 0.5-5% sucrose in a holding solution reduced vase life of Eucalyptus 
tetragona, whereas concentrations up to 2% had no effect in E. youngiana (Delaporte et 
al. 2000). In E. ficifolia, increasing concentrations up to 2% sucrose were associated 
with decreased stamen wilting, a factor limiting vase life (Sun et al. 200 1 ) .  
The addition of  biocides such as hydroxyquinoline (HQ) based compounds to holding 
solutions usually improves hydraulic conductivity within cut stems (Ichimura et al. 
1 999). Their suggested mode of action has been through the inhibition of vascular 
occlusions of microbial origin or deposition of organic substances (Put and Clerkx 
1 988). In M. collina, the vase solution that successfully reduced the level of stamen 
wilting was a 2% sucrose with 200 mg r1 HQ citrate (Sun et al. 2000). Whether similar 
combinations of additives and/or concentrations are effective in other Metrosideros 
species are not known. 
The sensitivity of cut flowers in Myrtaceae to ethylene can vary depending on ge!'lera 
and species. Cut flowers in M. collina, for example, were found to be highly sensitive 
to the effects of this hormone. Application of exogenously applied ethylene (0. 1 j..ll r 1 ) 
1 82 
resulted in 80% stamen abscission by day 5 after harvest, and higher concentrations 
caused whole flower abscission (Sun et al. 2000). In contrast, higher concentrations of 
ethylene (8 .6 111 r1 ) caused no floral abscission in Verticordia chrysantha, V. plumosa or 
V. densiflora but substantially induced flower, pedicel and leaf abscission in V. nitens 
(Joyce and Poole 1 993) .  Furthermore, environmental conditions such as high relative 
humidity (RH) can induce floral abscission possibly through the enhancement of 
endogenous ethylene effects (Zieslin and Gottesman 1 986). Thus far, there have been 
only a few studies examining the effect of RH on ethylene production and/or vase life 
characteristics .  
The use of chemical agents that inhibit ethylene action have successfully extended the 
vase life and preserved floral qualities in a wide variety of cut flowers. Application of 
silver thiosulfate (STS) has been used consistently and applied successfully to intact and 
excised flowers over recent decades (Cameron and Reid 1983,  Joyce and Beal 1 999). 
Relatively recent concerns of the toxic effects of silver and its release into the 
environment, however, have prompted a search for safe but equally effective 
alternatives (Halevy 1 994) . Thus far, 1 -methylcycloporopene ( 1 -MCP), a gaseous 
compound that also inhibits ethylene action appears to be a promising alternative to 
STS, and affords protection from ethylene in a number of flowering species (Serek et al. 
1 995b, Macnish et al. 2000) . In various genera within Mrytaceae, 1 -MCP prevented 
floral abscission and generally delayed ethylene-induced senescence, such as in 
Chamelaucium (Serek et al. 1995a) and Leptospermum (Macnish et al. 2000). In 
contrast, 1 -MCP applied to cut flowers of Metrosideros was relatively ineffective in 
preventing ethylene-induced abscission of flowers and stamens, and instead, stimulated 
endogenous ethylene production (Sun et al. 2000) . Whether this response was specific 
for M. collina or occurs within other congeneric species merits an investigation. 
The aim of the present study was to investigate the postharvest physiology and vase life 
characteristics of cut flowers of Metrosideros excelsa. In light of the current literature, 
it is hypothesised that: 
1 )  The vase life of M. excelsa is limited by adverse water relations and by the ethylene?
related abscission of whole flower and floral organs. Appropriate holding solutions 
1 83 
and inhibitors of ethylene action may, therefore, provide effective treatments for 
extending the vase life and minimising ethylene-related damage. 
6.2 Materials and Methods 
6.2.1 Plant material 
The postharvest experiments were conducted using inflorescences harvested from two 
cultivars of mature ( - 1 .5 m in height) Metrosideros excelsa plants, 'Lighthouse' and 
'Vibrance' .  Liner plants obtained from Lyndale Nurseries, Auckland, in August 1 998 
were potted up into 30 1 containers with an 80:20 v/v peat and pumice media, and 
supplemented with a slow-release (6 month) fertiliser. Plants were grown outside under 
ambient conditions at the Plant Growth Unit, Massey University, Palmerston North, 
New Zealand, and supplied daily with overhead sprinkler irrigation. Plants were 
arranged into four blocks containing approximately 1 0- 1 5  plants of each cultivar from 
which inflorescences were harvested over two consecutive summers, being late 
December 1 999/early January 2000 and December 2000 I January 200 1 .  
6.2.2 Harvest and experimental preparations 
Inflorescences containing three to four pairs of cymules were harvested between 0700 
and 0800 h from the experimental plants . Cut inflorescence stems were transported in 
buckets with distilled water to the laboratory in preparation for each experiment. 
Cymules selected included those containing a long pedicel (>5 mm) with three flower 
receptacles at late Stage 2 (tight bud with petals beginning to reflex) (See Table 4. 1 ) . 
Cymules were detached by hand from the floral stem. Each cymule was randomly 
selected and the base of the petiole cut under distilled water using a sterile razor blade, 
and assigned at random to a treatment. 
Each explant (individual cymule) was suspended individually in a 40 rn1 beaker 
supported by an aluminium foil cover so that the end of the pedicel was immersed in a 
holding solution. Standard holding solutions, unless otherwise stated, contained 2% 
sucrose and 200 mg r 1 of hydroxyquinoline citrate (HQC), as recommended by Sun et 
al. (2000). 
1 84 
6.2.3 Vase life room 
Throughout the experiments, cymules were maintained under controlled conditions in a 
vase-life evaluation room at the Plant Growth Unit, Massey University, following the 
conditions prescribed by Halevy ( 1976). The vase life evaluation room was maintained 
at a constant temperature of 20?C with a relative humidity (RH) at approximately 70%. 
Overhead lighting was provided by four cool white fluorescent tubes (36 W Philips 
TLD) ( 1 5  J.Lmol m-2 s- 1 ) on a 12 h photoperiod. 
6.2.4 Scoring flower condition 
Flower quality was assessed daily using several methods. The developmental stage of 
flowers on each cymule was scored based on a modified version of the classification 
scheme outlined in Chapter 4. This included stages that corresponded to the expansion 
of flower bud girth and petal reflexing (Stage 2), elongation of the stamens/style (Stage 
3), attainment of anthesis and stamens fully elongated (Stage 4), dehiscing of anthers 
and visibility of pollen (Stage 5), onset of stamen abscission (Stage 6), and abscission of 
all stamens and petals (Stage 7) .  Flower quality was assessed based on the level of 
wilting or abscission in stamens and petals (percentages) , and on the abscission on floral 
receptacles (numeric counts) . The vase life of cut cymules was defined as time taken 
for flowers to develop from the onset of Stage 2 through to the attainment of Stage 7.  
Flower mass, water uptake and the rate of transpiration of each cymule was measured 
daily. This was undertaken by recording the mass of the holding solution with and 
without the cymule, both before and after the addition of any supplementary solution, 
required as either part of the treatment or to replenish a depleted holding solution. 
Based on these data, cymule mass (fresh weight) was determined. Comparisons with 
data collected from the previous 24 h session enabled calculation of the difference in 
daily water uptake and transpiration. 
6.2.5 Experiments 1 A-C: Holding solutions :  effects on flower quality 
Cymules of 'Lighthouse' and 'Vibrance' were harvested at late Stage 2 and assigned to 
one of five treatments within five blocks based on a Randomised Block Design (RBD). 
1 85 
The holding solution treatments included ( 1 )  distilled water (control) ,  (2) 200 mg r1 
HQC solution, (3) 2% sucrose and 200 mg r1 HQC solution (referred to as 2% 
sucrose), (4) a 24 h pulse solution of 10% sucrose and 200 mg r1 HQC before 
transference to a 200 mg r 1 HQC solution (referred to as 10% sucrose pulse) , or (5) a 24 
h pulse solution of 10% sucrose and 200 mg r1 HQC before transference to a 2% 
sucrose and 200 mg r 1 HQC solution (referred to as 10% sucrose pulse, 2% holding). 
Cymules were held in the vase life evaluation room continuously and measurements on 
flower quality were conducted daily using the characteristics defined above. The effects 
of holding solution on the vase life of 'Lighthouse' and 'Vibrance' were examined in 
separate experiments due to the variation in flowering times between the cultivars . 
A third experiment was undertaken using cut cymules of 'Lighthouse' to examine 
whether the effects on flower quality were attributed to the HQC itself, or to the pH 
effect of HQC. The experiment was conducted using a completely randomised block 
design with two treatments and four blocks. The first series of treatments consisted of 
holding solutions of only distilled water (control) (pH 5 .0) and distilled water with HQC 
added ( 100, 200 and 300 mg r1) corresponding to pH levels of 3 .8 ,  3 .7  and 3 .6. The 
second series consisted of distilled water containing HQC ( 100, 200 and 300 mg r1 ) but 
with pH adjusted to 5 .0 using additions of sodium hydroxide (0. 1 M). Flower quality 
was scored based on the stages of floral development, level of wilting and stamen 
abscission over the duration of the experiment. 
6.2.6 Experiment 2: Effect of humidity on flower quality and endogenous ethylene 
production. 
The experiment was conducted using explants harvested from M. excelsa, 'Lighthouse' .  
The experiment was designed based a repeated measures randomised block design with 
two humidity environments and seven assessment (time) treatments within four blocks. 
The two environments consisted of either a bench top within the vase-life evaluation 
room (RH - 70%) or one of four glass aquarium enclosures (RH >90%). Each glass 
enclosure consisted of 50 L, semi-sealed aquarium tank outfitted with a narrow ( 1  x 10  
cm) opening on the top to provide unrestricted airflow and to prevent ethylene 
accumulation. Beakers containing two cymule explants were placed inside each 
enclosure along with 20 g soda lime (Carbasorb, BDH). An aquarium air pump was 
1 86 
used to discharge humidified air (>90%) into each enclosure. Hygrometers were used 
to ensure that the desired RH levels within each enclosure and in the vase-life room 
were maintained continuously. Air samples from both the environments were taken 
every 24 h to ensure that ethylene concentrations did not differ significantly from 
ambient. 
The experiment was conducted using eight assessment (time) treatments in order to 
quantify the level of exogenously produced ethylene at various stages of floral 
development. Daily, over the course of the eight days, beakers with two explants were 
transferred from either the bench or the aquarium enclosures to 0.5 1 jars . To each of 
the jars, 5 g of soda lime was added for absorption of respiratory C02, and lids were 
tightly sealed. Lids on each of the jars were outfitted with a gas-impermeable rubber 
plug (septum) that allowed the extraction of 1 rnl of internal air via a needle syringe at 
0, 24 and 48 h after being sealed. The concentration of ethylene in the air was measured 
using a gas chromatograph (Hewlett Packard 5890A, USA). Flower quality was 
assessed daily from cymules in both environments and within the glass jars . 
6.2.7 Experiment 3:  Effect of applied exogenous ethylene on flower quality and 
endogenous ethylene production. 
This experiment was arranged following a factorial design with four exogenous 
ethylene treatments and four subsequent incubation periods . Each explant of 
'Lighthouse' was held in a beaker placed in 1 l jar along with 10 g soda lime [and 1 0  g 
of Purafil (Ipsco Ltd. ,  Auckland) held in a paper sachet (control jars only)] . Jars were 
tightly sealed using lids outfitted with septa. Jars were treated with one of four 
concentrations of exogenous ethylene (0, 0. 1 ,  1 and 10  !J.l r 1 )  injected via the rubber 
septa and held in darkness for 24 hours. After ethylene treatment, beakers containing 
explants were transferred to a bench in the vase-life evaluation room. Explants were 
then transferred at time 0, 24, 48 and 96 h to clean 1 l jars containing only soda lime 
(5 g) for determination of ethylene production over 24 h. The vase-life characteristics 
of explants were assessed daily. 
1 87 
6.2.8 Experiment 4: Efficacy of 1-methylcyclopropene (1-MCP) and silver 
thiosulfate (STS) on ethylene-induced responses 
The effectiveness of STS and 1 -MCP was evaluated in explants of 'Lighthouse' 
subsequently treated with exogenous ethylene. The experimental layout followed a 
factorial design with two main factors, comprising four levels of an ethylene inhibitor 
treatment and two levels of an exogenous ethylene treatment. Replication consisted of 
two explants per treatment within each of the five blocks. 
6.2.8.1 Protection treatment 
Ex plants in holding beakers were sealed in  1 1 jars outfitted with septa through which 1 -
MCP (ElthylBloc, Yates, New Zealand Ltd.) was injected to provide one of three 
concentrations (0, 15 ,  and 1 50 nl r\ Additional explants were pre-treated with 2 mM 
STS sprayed to incipient run-off, and allowed to dry for 1 h before being placed in jars . 
Jars containing both 1 -MCP and STS treated explants were sealed and stored for 6 h. 
After 6 h, the j ars were opened and the gaseous 1 -MCP was dissipated for 1 h. 
6.2.8.2 Ethylene treatment 
The jars were re-sealed after adding 5 g soda lime and injected with ethylene to provide 
one of two concentrations (0 and 5 ?1 r1 ) and stored in darkness for 24 h. After 
incubation for 24 h, beakers with explants were removed from the jars and placed on a 
bench in the vase-life evaluation room, and vase life characteristics were monitored 
daily. In addition to the vase life characteristics described earlier, a visual measure of 
the level of petal in-rolling was recorded daily based on a 0- 100% scale, using 10% 
units. This range included petals that were fully- to semi-turgid (0-30% ) ,  partially 
turgid/desiccated ( 40-60% ), and completely desiccated (70- 1 00% ) .  
6.2.9 Statistical analyses 
Biometrical analyses were conducted using the SAS (SAS Institute, Cary, N.C. ,  USA) 
statistical program. A repeated measures analysis of variance was used to analyse daily 
1 88 
changes in stages of floral development, cymule mass, transpiration, water uptake and 
the proportion of organ abscissions. Percent data (stamen wilting level) was 
transformed using a log transformation. Treatments incorporating two or more main 
treatment set up in factorial combination were analysed using a factorial repeated 
measures. In all analyses,  a Least Square Means (LSM) test was used to compare 
differences between treatments at specific time points. 
6.3 Results 
6.3.1 Experiment 1 A :  Holding solutions: effects on flower quality in 'Lighthouse' .  
6.3.1.1 Floral development 
Holding solution had a small but significant effect on the stage of floral development 
over the 1 2  day experimental period (P<0.05) (Figure 6. 1 ) . All treatments reached 
Stage 4, 5 and 6 on Days 3, 5 and 7-9, respectively. Differences in the rate of flower 
development differed significantly only during Days 7 -9. On these days, cymules 
treated with both 10% sucrose pulse treatment showed a decrease in the rate of 
senescence compared with cymules held in a 2% sucrose solution. 
6.3.1.2 Water relations 
There was a main effect of holding solution treatment (P<0.05) and time (P<0.000 1 )  on 
water uptake but no significant interaction (P>0.05) (Table 6. 1 ). Generally, water 
uptake was lower in all treatments in comparison with controls, except in the 2%  
sucrose treatment. Data pooled for all treatments showed that water uptake declined 
steadily over the experimental period starting at 0.26 g day" 1 (Day 1 )  and decreased to 
0. 1 5  g day" 1 (Day 1 2) with a peak uptake on Day 4 of 0.39 g day" 1 compared with Day 1 
(P<0.05) (Figure 6.2 A).  
The main effects of holding solution treatment (P<0.01 )  and time (P<0.000 1 )  affected 
transpiration, although there was no significant interaction (P>0.05) (Table 6. 1 ) .  
Transpiration in the 2% sucrose treatment did not differ from that o f  control, and was 
lower in the other treatments. Generally, transpiration rates mirrored the response in 
1 89 
......, c <l) E 0.. 
0 
<l) 
> 
<l) 
"0 
ro ...... 0 
4-
4-
0 
<l) l:ll) 
ro ..... 
[/) 
7 -,-------- -? ?  . .  ---
6 
5 
4 
3 
2 
L 
;A 
J I - 
f 
0 1 2 3 
.'/? 
4 5 
..........._ Water 
-?-HQC alone 
---< 2% sucrose continuous 
--e-- 10% sucrose pulse 
- ? 1 0% sucrose pulse. 2% holding 
6 7 8 9 1 0  
Time after harvest (days) 
1 1  
Figure 6 . 1 Effect of ho ld ing solution on  the mean (?SE) stage of floral 
development in cut cymules of Metrosideros excelsa 'Lighthouse' . 
1 2  
1 90 
Table 6. 1 Effect of holding solution on mean water uptake and transpiration in cut 
cymules of M. excelsa 'Lighthouse' over the experimental period. Mean separation in 
columns by Least Square Means test, 5% level. 
Holding solution 
Water (control) 
HQC alone 
2% sucrose continuous 
10% sucrose pulse 
10% sucrose pulse, 2% 
holding 
Water uptake (g d.
1 ) 
0.3 1 a 
0. 1 9c 
0.26ab 
0.22bc 
0 .21  be 
Transpiration (g d-
1 ) 
0.33a 
0.20c 
0.28ab 
0.22c 
0.22bc 
water uptake (Figure 6.2 A), although the mean rate of transpiration was higher than 
that of water uptake from Day 6 onwards.  
Cymule mass was significantly affected by the different holding solutions over time 
(P<0.0001 ) .  All treatments displayed a uni-modal pattern with cymule mass peaking 
between Days 5-6 (Figure 6.2 B) .  Generally, differences in cymule mass between 
treatments occurred after Day 7, when the mass of HQC-treated cymules was lower in 
comparison with both pulse treatments (P<0.05), although these did not differ from the 
mass of control cymules. The start of the decline in cymule mass (for all treatments) 
occurred after Day 5, corresponding to the same day that the mean rate of transpiration 
exceeded water uptake (Figure 6.2 A) . 
6.3.1 .3 Stamen wilting 
Stamen wilting increased significantly between treatments over time (P<0.05) ,  with the 
onset of wilting occurring after Day 7 (Figure 6.3 A). Between Days 1 1 - 1 2, the level of 
stamen wilting in all treatments (except in 10% sucrose pulse) was inversely related to 
the proportion of stamen abscissions (Figure 6.3 B). Therefore, most treatments with 
low stamen abscission rates had the highest incidence of wilting (except the 10% 
19 1  
1 .  2 ?-?-----? ?- -?---- ?-.. ?-??--.. ??-? -?-"? .... -- ........ , .... _ ....... ,_, _____ ,, ____ ,.,_. _____ ? .. ? .......... ...... -.......... _ ............. -....... ........... ............ , .. ______ , .. , ......... ___ .. ,_, ............. ,. , .. , " 
A 
-+--Water uptake 
1 . 0 ---Transpi rat ion 
-- 0 . 8  
-o 
.!:9 
? 0 . 6 
0 . 2  
0 . 0  +---.---.--.,..--.,..---.----.----,----,----,--------,-----! 
1 . 6 - - -?-?-?------?---------- - ---???--?--- -- --?-- -- -
B 
1 . 4 
_._Water 
1 . 0 HQC alone 
2% sucrose continuous 
--+-- I 0% sucrose pulse 
-o-JO% sucrose pu lse. 2% holding 
0 . 8  +---.--,---,--,---,----r---,--,--,--.,------
0 2 3 4 5 6 7 8 9 1 0  1 1  1 2  
Time after harvest (days) 
Figure 6 . 2  Changes in (A) mean (?SE) water f lux (uptake and transp i ration) 
pooled for a l l  hold ing sol ut ion treatments and (B) mean (?SE) cymu le  mass 
for i ndiv idual treatments in cut cymules of Metrosideros excelsa 'L ighthouse'. 
192 
? 
c: 0 
(/') (/') 
u (/') .0 C<j 
c: (!) E C<j 
...... 
[/) 
1 0 0  
_ .... __ __ .... ...... -- - ----- ---.. ---- ----? ?----? ---- - - ---.. - - .. .  ___ ...... ........ ___ _ .. _ ........ . .. . .....  -... ............ _ . .. . ..... .. _ 
A 
8 0  ___._Water 
-?HQC alone 
2% sucrose continuous 
-+-- 10% sucrose pulse 
,- 1 0% sucrose pulse. 2% holding 
/ 
1 00 _ _ ___ _ ______  ...... ..... ...... .... ......... - ----- - ---- ---- -.. - - - - --... - .... ?--? ---------.. ? ? - ... . . ... ... , 
B 
8 0  
6 0  
4 0  
2 0  
0 2 3 
I 
4 5 6 7 8 
Time after harvest (days) 
T :r 
I -11 I 
9 1 0  1 1  1 2  
Figure 6.3 Effect of ho ld ing solution o n  mean (?SE) (A) stamen w i l ting 
and (B) abscission i n  cut cymu les of Metrosideros excelsa 'L ighthouse'. 
1 93 
sucrose pulse treatment). In particular, stamen wilting in the HQC solution was 
significantly higher in comparison with all other treatments (P<0.05) ,  attaining a 70% 
wilting level by Day 12 .  In comparison, only a 4% wilting level was recorded in the 
distilled water treatment (control). Generally, both the pulse ( 10% sucrose pulse and 
1 0% sucrose pulse, 2% holding) and the 2% sucrose treatments were higher than the 
control treatment (except Day 12), attaining a wilting level of 1 8-34% by Day 1 1 . 
The level of stamen wilting in all treatments pooled between Days 8- 1 2  was negatively 
correlated with cymule mass (P<0.00 1 ,  r = -0.537) . In particular, stamen wilting with 
solutions containing only HQC (HQC alone and 10% sucrose pulse) showed the 
strongest relationship with cymule mass compared to either the control (water) or the 
addition of sucrose (2% sucrose and the 1 0% sucrose pulse, 2% holding) (Table 6.2). 
However, the lack of a possible significant correlation in the control group was due, at 
least in part, to the lack of wilting data in that treatment. 
Table 6.2 Correlation of mean stamen wilting and cymule fresh weight between Days 8-
12 for each holding solution treatment for cut cymules of Metrosideros excelsa 
'Lighthouse' .  Symbol: n.s. not significant at 5% level. 
Holding solution 
Water (control) 
HQC alone 
2% sucrose continuous 
1 0% sucrose pulse 
1 0% sucrose pulse, 2% 
holding 
Total 
Pearson correlation F 
- value P - value 
-0.0 1 2  n.s. 
-0.9 1 5  <0.000 1 
-0.365 n.s. 
-0.627 0.005 
-0.332 n .s .  
-0.537 0.000 1 
194 
6.3.1.4 Floral abscission 
There was a significant interaction of the main effects of treatment and time on the 
proportion of stamen abscissions (P<O.OOO) that ensued from Day 7 onwards (Figure 6.3 
B) .  After Day 9, abscission rates from the HQC alone and 1 0% sucrose pulse 
treatments were lower than the 2% sucrose or the control water treatment (P<0.05) .  The 
1 0% sucrose pulse 2% holding treatment displayed an intermediate level of stamen 
abscission and did not differ from both of the HQC treatments (HQC alone and 1 0% 
sucrose pulse) or the 2% sucrose and control treatment (P>0.05). 
Petal abscission ensued in all treatments from Day 7 onwards, increasing up to a mean 
abscission rate of 44% of petals per cymule by Day 1 2. There was no significant 
interaction of the main effects of treatment and time on petal abscission (P>0.05) .  
6.3.2 Experiment 1 B :  Holding solutions: effects on flower quality i n  'Vibrance'. 
6.3.2. 1 Floral development 
There was no significant interactive effect of holding solution treatment and time on 
changes in floral development over the 1 6  day experimental period (P>0.05) .  Cymules 
from all treatments reached Stage 4 and 5 on Days 4-5 , and 6-8, respectively. Most 
cymules reached Stage 6 (onset of stamen abscission) by Day 14, with the exception of 
those in the HQC solution, since stamen abscission was absent in this treatment 
throughout the experimental period. 
6.3.2.2 Water relations 
Mean water uptake did not differ significantly between holding solutions over the 1 5  
day experimental period (P>0.05) .  Over this time period, water uptake per cymule 
averaged for all treatments peaked on Day 5 (0.50 g dai1 ) before decreasing steadily to 
0.08 g dai1 by Day 1 5  (Figure 6.4 A). 
195 
-o 
0 .  7 1--?-- -
1 A 
0 .6 1 
0 .5  ? 
I 
?0.4 -i 
X 
::s ....... I ? 0 .3  _j 
ro 
:3 
0 .2  
I 
I 
0 . 1 l 
0 
-+- Water uptake 
-Transpiration 
0.9 ---- ???--- ...... ________ __ ____ __ - - -- 1 
B 
0.8 
-- 0.7  
? 
Cl) Cl) ro E 
Cl) 
::s E >, 
u 0.5 
.............. Water 
HQC alone 
f T -
0.4 - 2% sucrose continuous 
............... 1 0% sucrose pulse 
1 0% sucrose pulse. 2% holding 
0 1 2 3 4 5 6 7 8 9 10 1 1  1 2  13 14 15 16  
T ime after harvest (days) 
Figure 6.4 Changes in (A) mean (?SE) water f lux (uptake and transpiration) 
pooled for a l l  hold ing solut ion treatments and (B) mean (?SE) cymule mass for 
ind i v idual treatments in cut cyumles of Metrosideros excelsa 'Vibrance'. 
1 96 
There was no significant interaction effect of treatment and time on the rate of 
transpiration (P>0.05) .  Mean transpiration per cymule peaked on Days 5-6 at a rate of 
approximately 0.5 1 g day" 1 and subsequently declined and remained relatively constant 
at ea. 0.24 g day" 1 for the duration of the experiment (Figure 6.4 A) . Transpiration rates 
exceeded water uptake after Day 5 .  
Cymule mass differed significantly between holding solutions over time (P<0.0001 ) .  
All treatments showed a uni-modal pattern in  weight change with a mean peak of 0.72 g 
per cymule (pooled data) on Day 5 before declining to 0.52 g day" 1 by Day 1 5  (Figure 
6.4 B) .  From Day 7 onwards, cymule mass in the HQC holding solution declined 
significantly (loss of 0.05 g day" 1 ) in comparison with all other treatments (loss of 0.03 
g day" 1 ) (P<0 .05) .  After Day 10, cymule mass from the 10% sucrose pulse treatment 
also decreased significantly in comparison to the control, 2% sucrose and the 1 0% 
sucrose 2% sucrose pulse treatments (P<0.05) .  
6.3.2.3 Stamen wilting 
Treatments associated with lowest incidence of stamen abscission produced the highest 
levels of wilting over time (P<0.000 1 )  (Figure 6.5 A). This included higher levels of 
wilting in both the pulse and constant HQC solutions throughout most of the 
experimental period. Initially, the highest rates of stamen wilting were recorded in the 
HQC solution, which reached a maximum wilting level of 76% by Day 10 .  Up until 
Day 1 3 ,  both the HQC and 10% sucrose pulse solutions were significantly higher (mean 
88% wilting level) compared to control and sucrose treatments (mean 3 1 %  wilting 
level) (P<0.05). From Day 1 3  onwards, the level of wilting in both sucrose treatments 
(2% sucrose and the 10% sucrose pulse 2% holding) did not differ from the HQC 
treatment (P>0.05). 
Stamen wilting for all treatments pooled was strongly correlated with cymule mass 
during the days that wilting occurred (Day 4- 1 6) (r = -0.78 1 ,  P<0.000 1 ) . For individual 
treatments, the level of wilting showed a strong negative correlation with cymule mass 
(Table 6.3) .  In particular, both the HQC and 1 0% sucrose pulse treatments showed the 
strongest linear fit (r <-0.900). This corresponded to the lower mean levels of cymule 
1 97 
? 
..__., 
bJ) t: 
.._. 
? 
t: (J) 
E 
ro 
.._. 
r:/) 
1 00 - -?
A 
80 
60 
40 -
20 
--+--Water 
--y-- HQC alone 
2% sucrose continuous 
---+-- 1 0% sucrose pulse 
?--->-- 1 0% sucrose pulse. 2% holding 
I 
-I 
/ 0 +-----,,-- ,- '!r--,---, -?-F-- .. -
1 00 -.-- -------
80 
.Q 60 
(/) (/) 
u (/) ..Cl ro 
? 40 
E 
ro ..., 
r:/) 
20 
B 
T 
I 
0 1 2 3 4 5 6 7 8 9 1 0  1 1  1 2  1 3  14  1 5  1 6  
Time after harvest (days) 
Figure 6.5 Effect of holding solution on mean (?SE) (A) stamen w i l t ing 
and (B)  absciss ion in cut  cymules of Metrosideros excelsa 'Vibrance' . 
198 
mass per day recorded for HQC and 1 0% sucrose pulse after Day 5 and 10, respectively 
(Figure 6.4 A). 
Table 6.3 Correlation of mean stamen wilting and cymule fresh weight between Days 4-
16 for each holding solution treatment for cut cymules of Metrosideros excelsa 
'Vibrance' .  
Holding solution Pearson correlation F - P - value 
value 
Water (Control) -0.724 <0.000 1 
HQC alone -0.929 <0.000 1 
2% sucrose continuous -0.676 <0.000 1 
10% sucrose pulse -0.906 <0.000 1 
10% sucrose pulse, 2% -0.699 <0.0001 
holding 
Total -0.78 1 <0.000 1 
6.3.2.4 Floral Abscission 
The onset of stamen abscission occurred from Day 7 onwards in all treatments, except 
within the HQC holding solution where stamen abscission remained absent ( < 1 %) 
during the entire 1 6  day experimental period (Figure 6.5 B) .  There was a significant 
interaction between the main effects of treatment and time on the proportion of stamen 
abscissions (P<0.00 1 ) .  By the end of the experimental period (Days 1 3- 1 6), the highest 
levels of stamen abscission occurred in both the pulse and constant 2% sucrose 
solutions in comparison with both HQC solutions where stamen abscission was 
significantly lower (P<0.001 ) . Stamen abscission in the water control treatment did not 
differ from both HQC treatments. 
Petal abscission did not differ significantly between treatments over time (P>0.05) .  The 
level of abscission was low for all treatments by Day 1 6, averaging a loss of 4% of 
petals per treatment. 
199 
6.3.3 Experiment 1 C:  Holding solutions: effect of HQC and pH on flower quality 
in ' Lighthouse'. 
The highest levels of stamen abscission occurred after Day 6 in the control solution 
containing only water (pH 5.0) in comparison with all solutions containing HQC 
(P<0.0 1 )  (Figure 6.6 A). By Day 10 onwards, approximately 50% of all stamens in the 
control treatment had abscised in comparison to HQC-treated cymules (mean ::; 5%),  
regardless of pH adjustment. There was also a strong significant interaction effect of 
holding solution treatment and time on stamen wilting (P<0.0001 ) .  Holding solutions 
that showed the least amount of stamen abscission had the highest level of wilting and 
vice-versa (Figure 6.6 B). 
6.3.4 Experiment 2 :  Effect of humidity on flower quality and endogenous ethylene 
production. 
6.3.4.1 Floral development and abscissions 
Flower quality was assessed in explants maintained continuously in environments with 
either high (>90%) or intermediate ( -70%) levels of relative humidity (RH) over the 8 
day experimental period. Environment had a significant effect on the stage of floral 
development over time (P<O.O 1 )  (Figure 6. 7) .  A higher humidity level (>90%) slowed 
flower bud opening by 2 days in comparison to cymules at 70% RH. For instance, 
cymules maintained at RH 70% reached Stage 4 (anthesis) on Day 4, whereas those at 
90% RH only reached this stage on Day 6. 
Flower (as opposed to stamen) abscission occurred only in the high RH environment. 
The onset of flower abscission in this 90% RH environment occurred on Day 5 (mean 
63% abscission) . By Day 7, 100% of flowers in this environment had abscised. No 
flower abscission was recorded for cymules maintained in the 70% RH environment 
over the 8 day experimental period. 
200 
? 
c: 0 
Cll Cll 
(.) Cll .rJ ro 
c: Q) 
E 
ro 
_, if) 
? 
tl.() 
c: 
_, 
? 
c: 
Q) 
E 
;:l if) 
1 00 ---- --??---------- - --- ------------?----- - - ' 
A 
80 
HQC 
concentration (ul 1 ' 1 )  - pH 
60 - ---o - 5. 1 
40 
20 
0 
1 00 
80 
60 
40 
20 
0 
? 1 00 - 5. 1 
-+-lOO - 3.8 
--'l 200 - 5. 1 
-,;:- 200 - 3.7 
300 - 5. 1 
? 300 - 3.6 
B 
0 1 2 3 
-- ----?-?--?-?--
? 
T 
2 ? ,._ ' ....-? 
4 5 6 
-? ----- -i---"'::; i;/ lx? . I ; ;;t 
w l :t' 
;f --; r 
7 8 9 1 0  
T ime after harvest (days) 
Figure 6 .6  Effect of HQC and pH ho ld ing solutions on mean (?SE) 
(A) s tamen absc ission and (B) wi l ti ng in cut cymu les of lv!etrosideros 
excelsa 'Lighthouse'. 
1 1  
201 
7 ----------- -- ?- -?------- - --- ---?--- - - ? - -
.... 6 c: Q) E 0. 
0 
g: 5 Q) u 
0 
Env i ronment 
---+- RH > 90% 
---- RH = 70% 
1 2 3 4 5 6 7 
Time after harvest (days) 
8 
Figure 6 .7 Effect of re lati ve humidity (env i ronment) on mean (?SE) stag. 
development in cut cymules of Metrosideros excelsa 'Lighthouse' . 
202 
Stamen abscission occurred in cymules maintained only in the 70% RH environment 
starting from Day 5 and proceeding through to the end of the experiment (Table 6.4). 
Similarly, petal abscission was also recorded for cymules only in the 70% RH 
environment (Table 6.4). 
Table 6.4 Effect on humidity on percentage of stamen and petal abscission in cut 
cymules of Metrosideros excelsa 'Lighthouse' . Days in which all flowers in a treatment 
had abscised are indicated by a ' - ' .  
Abscission Environment: Days 
type Relative 
humidity (%) 4 5 6 7 8 
Stamen 70 0 0.25 7.4 26 33 . 8  
> 90 0 0 0 
Petal 70 0 0 1 3 .3  13 .3  1 3 . 3  
> 90 0 0 0 
6.3.4.2 Water relations 
Relative humidity significantly affected the rate of water flux (water uptake and 
transpiration) in cymules, as indicated by a strong interaction effect of environment and 
time (P<O.OOO l )  (Figure 6.8 A). Comparatively, water flux was lower for cymules 
maintained in the high humidity environment. 
The extent of water flux within each of the environments affected cymule mass.  A high 
(>90%) RH environment significantly reduced the water flux in the cut cymules 
(P<O.O l )  (Figure 6.8A). Cymule mass was also affected and was significantly lower in 
the high RH environment (P<O.Ol )  (Figure 6.8 B). Within the 70% RH environment, 
mean transpiration exceeded mean water uptake after Day 5, which coincided with a 
decline in mean cymule mass. In contrast, transpiration in the >90% RH environment 
remained lower than water uptake and cymule mass subsequently increased with time. 
203 
1 .4 
1 . 2 
1 
-o 
?0.8 
>< 
:::1 
0.2 
------------ -------? 
A Water uptake 
-e- RH > 90% 
? RH = 70% 
Transpiration 
- - o - - RH > 90% 
- - o - - RH = 70% 
:e . -:1 - - - - - - -i- - - - - - -? 
., 
I 
1 
I I 
:. - - - - - - :e- - - - - - - - - - - - -
0 +-----?-------,------.------.------?------------?
1 .5 
1 .4 
1 .3 
:? 
C/J 1 .2 C/J ? 
E 
Cl) 
E 1 . 1  
>-, 
u 
1 .0 
0.9 
0 .8 
0 
B 
1 2 3 4 
Env i ronment 
?RH > 90% 
_._RH = 70% 
5 6 
T ime after harvest (days) 
F igure 6.8 Effect of rel ative humidi ty on mean (?SE) (A) water f lux 
(water uptake and transp i rat ion) and (B)  cymule mass in  cut  cymules 
7 
of Metrosideros excelsa 'L ighthouse' . For clarity purposes, standard 
errors bars were omitted for transpiration data al though they were s im i l ar 
to those for water uptake. 
204 
6.3.4.3 Endogenous ethylene production 
Measurements of the amount of ethylene emanated in jar headspaces did not differ for 
cymules that had been held in the different RH environments (P>0.05) .  Based on data 
pooled for both environments, the amount of ethylene produced differed across 24 h 
incubation periods (P<O.OOO l ) , with the highest levels produced on Days 5-6 (Figure 
6.9). Similarly, 48 h incubation periods also differed between each consecutive 
enclosure days (P<O.O 1 )  although the highest levels were recorded on Day 4.  
Comparisons of the amount of ethylene produced between 24 and 48 h per enclosure 
date was also statistically significant (P<0.0 1 ) .  On enclosure days 4-5, ethylene levels 
approximately doubled between the 24 and 48 h incubation periods (Figure 6.9). 
Generally, the rise in ethylene on enclosure day 4 coincided with the shift of cymules 
from Stage 4 (anthesis) at 24 h to Stage 5 (dehiscence of anthers) by 48 h. On enclosure 
day 5, cymules at 24 h were at Stage 5 and shifted to Stage 6 (onset of stamen 
abscission) by 48 h. Therefore, the highest ethylene emanation period occurred either 
just before or during the time when anthers dehisced. 
6.3.5 Experiment 3: Effect of exogenous ethylene on flower quality and endogenous 
ethylene production. 
6.3.5.1 Floral development 
Changes in floral characteristics were analysed for cymules pooled for all times of 
observation after ethylene treatment (0, 24, 48 and 96 h), since responses between these 
treatments were similar. There was no significant effect of exogenous ethylene 
treatment (0- 1 0  J.Ll r 1 ) on the stage of floral development over the 9 day experimental 
period (P>0.05). In all ethylene treatments, cymules reached Stage 3 and 4 on Days 2-3 
and Day 4, respectively. Cymules exposed to exogenous ethylene treatments 0 and 0. 1 
J.Ll r1 reached Stage 5 on Day 9, whereas those treated with 1 .0 and 10  J.Ll r1 of ethylene 
failed to reach this stage, and underwent flower abscission instead. 
205 
0.5 
2-oA 
<l) 
::l 
E >, u 
..... 
g_o.3 
c 0 
?.;::; ? c ? 
? 0.2 
<l) c <l) 
>, ..;:: 
? 0. 1 
Incubation period 
D 24 hours 
048 hours 
0 1 2 3 4 5 6 
Enclosure date (consecutive days) 
Figure 6.9 Mean (?SE) headspace ethylene concentration above cut 
cymules of Metrosideros excelsa 'L ighthouse' at 24 and 48 h after 
incubation per enclosure day .  
206 
6.3.5.2 Abscission 
The amount of flower abscission was associated with the concentration of exogenous 
ethylene (Table 6.5) .  However, in the 0 (control) or 0. 1 !J.l r 1 of ethylene treatments, 
flower abscission was relatively low during the 9 day experimental period. Thus, 
stamen abscission occurred only in the 0 and 0. 1 j..tl r 1 treatments, although the level of 
abscission from both treatments was relatively low (mean 3 . 1 %  per cymule on Day 9) ? 
and did not differ significantly (P<0.05). 
There was a significant main effect of ethylene concentration on petal abscission 
(P>O.OOO l ) . By Day 6, cymules treated with 10 !J.l r 1 of ethylene incurred the lowest 
levels of petal abscission (mean 2.5% petals per cymule) compared with 0 (control), and 
0. 1 and l j..tl r 1 treatments (pooled mean 1 6.7% petals per cymule). The proportion of 
abscised petals in the 10 j..tl r1 treatment was low due to the high abscission rate of 
whole flowers in that treatment. 
Table 6 .5 Effect of exogenously applied ethylene on percentage of abscised flowers . 
Mean separation in columns by Least Square Means test, 5% level. 
Exogenous Days 
ethylene (J-Ll r 1 )  3 4 5 6 7 8 9 
0 (control) oa oa oa oa 6.3a 1 9a 1 9a 
0. 1 oa oa oa oa oa 25a 3 1  a 
1 .0 Oa 56b 75b 75b 75b 8 1  b 8 1  b 
10.0 56b 88c 94c 94c 100c 100c 1 00c 
6.3.5.3 Wilting 
There was a significant interaction of ethylene treatment and time on the level of stamen 
wilting (P<O.OO 1 )  (Table 6.6). The wilting response in the 0 and 0. 1 j..tl r 1 exogenous 
ethylene treatments were similar, reaching an approximate 5 1 %  wilting level by Day 9.  
207 
The low levels of wilting in the 1 .0 and 10 ?I r 1 ethylene treatments were due to the 
high level of flower abscission (?75%) from Days 5-9 in these treatments . 
Table 6 .6 Effect of exogenously applied ethylene on percentage of mean stamen wilting. 
Mean separation in columns by Least Square means test, 5 %  level. 
Exogenous Days 
ethylene (gl r I )  5 6 7 8 9 
0 (control) 0.6a 6.3a 1 9 .4a 35.0a 50.0a 
0. 1 oa 8 . 1  a 23. 1 a 40.6a 52.8a 
1 .0 oa oa 1 .9b 1 .9b 1 2 .5b 
10.0 oa 1 .3a 5 .6ab 
6.3.5.4 Endogenous production of ethylene after exposure to exogenous ethylene 
Exposure of cyrnules to exogenous ethylene had no effect on subsequent ethylene 
production. There was no significant interaction of sampling time (0, 24, 48 or 96 h) 
and ethylene level (0. 1 ,  1 .0 or 10  111 r 1 ) on ethylene production (P>0.05) .  Ethylene 
concentration per 24 h of incubation averaged 0.23?0. 1 ?1 r1 per day. 
6.3.6 Experiment 4: Effect of 1-MCP and STS on ethylene-induced responses 
6.3.6.1 Water relations 
There was significant three-way interaction of time, ethylene, and preventative 
treatment (P<0.05) on both water uptake and transpiration. In the absence of ethylene 
(0 ?1 r1 ethylene treatment), water flux (water uptake and transpiration) was higher in 
the STS treated buds in comparison with all other treatments (P<0.0 1 )  (Figure 6. 1 0  A).  
However, in the presence of ethylene (5 ?1 r 1 ) ,  water flux was higher in STS treated 
cymules on all days except Day 4 after treatment (Figure 6. 10  B) .  On this day, water 
flux did not differ significantly from either the 0 or 1 50 nl r1 1 -MCP treatments 
(P>0.05) ,  but was higher than the 1 5  nl r1 1 -MCP treatment (P<0.05) .  
208 
2.0 
0.5 
......--------- -------------- , _ _  --- --?- ------ ?-? -
A 0 fll 1 " 1  ethylene 
- - - -- - -- _- -
1 -MCP (n l  1 " 1 ) 
_._ 0 
-__,-- 1 5 
1 50 
_._2mM STS 
0.0 +-----,-----,-----.-----,---------! 
2.0 
1 .5 -
0.5 
1 
B 
-- -?---?-?- ------ ----- - ----?---- --- ----------------?-- - -
2 
5 fll 1 " 1  ethylene 
?-----"---;-? r-- ----, 
3 4 
Time after treatment (days) 
5 
- 'P I 
6 
Figure 6. 1 0  Effect of preventat ive treatment on mean (?SE) water f lux  
(water uptake and transpi ration) for cut cymules treated w ith either (A)  0 
or (B) 5 fll l - 1 exogenous ethy lene. For c larity purposes, standard error 
bars are omitted for transpirat ion data a l though they were s im i l ar to those 
for water uptake. Water uptake = sol id  l ines, transp i ration = dashed l i nes. 
209 
Treatments with either 0 or 5 nl r 1 ethylene did not differ in their effects on the fresh 
weight of cymules over time (P>0.05). The use of a preventative treatment did, 
however, influence the weight of cymule over time (P<0.000 1 )  (Figure 6. 1 1 ) .  
Generally, cymules that were not treated with a preventative treatment (control) 
maintained a lower cymule mass over time than treated cymules. Application of STS 
resulted in a higher cymule mass per day in comparison with cymules treated with 
either 0 or 15 nl r 1 1 -MCP (P<0.0 1 ) . The weight of STS treated cymules was also 
higher than those treated with 1 50 nl r 1 1 -MCP, but only on Days 5, 7-8 (P<0.0 1 ) .  
6.3.6.2 Stamen wilting 
There was a significant interaction of the main effects of preventative treatment and 
time on the level of stamen wilting (P<0.000 1 ), although there was no significant 
interaction of ethylene treatment and time (P>0.05) (Figure 6. 1 2). Generally, wilting 
was inversely correlated with cymule mass and, therefore, treatments in which the mean 
cymule mass was comparatively higher over time showed the least amount of wilting 
and the weakest linear fit (Table 6.7). In this case, 2 mM STS and 1 50 nl r 1 1 -MCP 
treated flowers showed the lowest levels of wilting between days 5 - 8 , followed by 1 5  
nl r 1  1 -MCP treated flowers, and the highest levels occurred i n  the 0 nl r 1 1 -MCP 
(control) treatment (P<0.05). 
Table 6.7 Correlation of stamen wilting and cymule fresh weight of cut cymules of M. 
excelsa pre-treated with preventative treatments of either 1 -MCP or STS before 
application of exogenous ethylene (data pooled for 0 and 5 1-11 r1 ethylene) . 
Preventative Pearson correlation P - value 
treatment F - value 
0 nl r 1 1 -MCP -0.927 <0.000 1 
1 5  nl r1 1 -MCP -0.792 <0.000 1 
1 50 nl r1 1 -MCP -0.669 <0.000 1 
2 mM STS -0.7 10  <0.000 1 
Total -0.777 <0.0001 
210  
1 .2 -1 ?------- ---?-?- .. --- ?- ?--? ----.. ??---?--? -?? -- ........... -?- .. . 
1 . 1  . 
1 .0 
? --- 0.9 (/) (/) C<:l 
E 
<l) 0.8 -
:::l 
t 0.7 -
u 
0.6 
0.5 
1 -MCP (n l  1 " 1 ) 
---+- 0 
-->-- 1 5 
___.___ 1 50 
---+- 2 mM STS 
0.4 +-------?-------,------?------?--------?----? 
0 1 2 3 4 5 
Time after treatment (days) 
Figure 6 . 1 1  Change in mean (?SE) cymule mass fol low ing exposure to a 
preventat ive treatment and exogenous ethy lene (0 and 5 !J.l 1 " 1 ethy lene data 
pooled) . 
6 
1 00 ---- ---? - ?--- --- --- ---- ----
80 
? 
00 60 c 
? 
c 
<ll 40 E 
.::l ifl 
20 
1 -MCP (n l  1 " 1 ) 
---+- 0 (control) 
-+-- 1 5  
-+-- 1 50 
---+- 2 mM STS 
-/ 
I 
o +.----??????==???----?--?--?--? 
0 1 2 3 4 5 6 
Time after treatment (days) 
7 8 9 
Figure 6. 1 2  Effect of preventative  treatment on mean (?SE) stamen w i l ting .  
Data pooled for  cut  cymules treated w ith 0 and 5 !J.l 1 " 1 ethy lene data pooled. 
2 1 1 
6.3.6.3 Stamen abscission 
,There was a significant interaction of ethylene treatment, preventative treatment and 
time on the level of stamen abscission (P<0.0001 ) .  In the absence of an exogenous 
ethylene application (0 Jll r ' ethylene) , cymules treated with a preventative treatment of 
STS showed a high level of stamen abscission (mean 29% per cymule) on days 7-9 
(Figure 6 . 1 3  A). This was followed by intermediate levels in the 0 and 1 5  nl r ' 1 -
MCP treatments (mean 10% per cymule) (P<0.05). The lowest levels of stamen 
abscission occurred in the 1 50 nl r' 1 -MCP treatments (< 1 %  stamen abscissions per 
cymule) (P<0.05). 
Preventative treatments of 1 5, 1 50 nl r ' 1 -MCP and 2 mM of STS afforded significant 
protection against stamen abscission following an application of 5 Jll r' ethylene in 
comparison with 0 nl r ' 1 -MCP (P<0.05) (Figure 6. 1 3  B) .  There was, however, no 
difference between the 2mM STS, 1 5  and 1 50 nl r' 1 -MCP protective treatments 
(P>0.05) .  By day 9, only 7% of stamens per cymule had abscissed in protected flowers, 
whereas 32% of stamens per cymule had abscissed in the controls (0 nl r' 1 -MCP) . 
6.3.6.4 Flower and petal abscission 
An exogenous ethylene application of 5 Jll r' of ethylene was sufficient to cause flower 
abscission, although only in some treatments (Figure 6. 14) .  By the end of the 
experimental period, the highest incidence of flower abscission was in the control 
flowers (mean 88% of cymules in the 0 nl r' 1 -MCP treatment), followed by 25% and 
1 3% in the 1 5  and 1 50 nl r ' 1 -MCP treatments, respectively. No flower abscissions 
were recorded in the 2mM STS treatment throughout the experimental period. 
Petal quality was assessed on Day 9 based on a visual score of the severity of petal in?
rolling (0- 1 00%) in intact (not abscised) flowers that had been exposed previously to 5 
Jll r ' ethylene (Figure 6. 1 5) .  The highest severity of petal in-rolling was recorded in 
control plants (mean 95% per cymule) followed by both 1 -MCP treatments (mean 50% 
per cymule). Petals in STS plants remained turgid and showed no indication of in?
rolling (mean < 5% per cymule) by Day 9 .  
2 1 2  
50 -,- - -- ?- -- --?-?--1 A 
I 
40 _, 
? I - ! 
.2 30 j r/J r/J 
u r/J .0 ? 
1 0  -
_._ 0 (control )  
1 5  
-e- 1 50 
2 mM STS 
50 -.-----
40 
? 
c 
. 2  30 r/J Cl) 
u Cl) .0 C<:l 
? 20 -E 
C<:l 
-' 
(/) 
1 0  
B 
0 J-l.l l - 1  ethy lene 
T T 
---- -
--=-
5 J-l.l 1 " 1  ethy lene 
? 
?I?i----i 
0 +,----?-----? --- : -===;?-----,---------?,----------? 
0 1 2 3 4 5 6 
Time after treatment (days) 
7 8 9 
Figure 6 . 1 3  Change i n  mean (?SE) stamen abscission i n  cut cymules after a 
preventative pre-treatment of 1 -MCP and STS before appl ication of either 
(A) 0 and (B) 5 J-l.l 1 " 1  of exogenous ethy lene. 
2 1 3  
c 
0 
VJ VJ 
(.) VJ 
..0 
Cl:) 
.... 
Q) 
;::: 
0 
ii: 
? 
bJJ c 
0 .... ' c 
? Q) 
0... 
1 00 
80 
60 
40 
20 
0 
-?- ---
Time after treatment (days) 
--?------ ?-
5 
Figure 6 . 1 4  Change i n  mean flower absciss ion w ith time fol lowing exposure 
to 5 fJ.l r 1  exogenous ethy lene in pre-treated cymules of Metrosideros 
excelsa 'Lighthouse' w i th 1 -MCP and STS. 
-.. ??-
0 
--- ---
I 
- - - -?-- -? 
1 5  
-
' 
.. - -- --- -- ?-????----...... -
D 
D 
T 
1 
1 50 
Preventat ive treatment  (concentration) 
---- ??--? 
1 -MCP (nl 
STS (mM) 
..... 
2 
? r t )  
l 
? 
l I ? 
Figure 6 . 1 5  Effect on mean (?SE) petal in-rol l i ng on Day 9 fol lowing exposur 
to 5 fJ.l r
1 
exogenous ethylene in pre-treated cymules of Metrosideros 
excelsa 'L ighthouse' w i th 1 -MCP and STS. 
2 14  
6.4 Discussion 
In this study, treatments that induced relatively low levels of stamen abscission in both 
cultivars of Metrosideros excelsa incurred relatively high levels of wilting, and vice 
versa. This study supported the hypothesis that the vase life of M. excelsa was limited 
by both adverse water relations and by ethylene-related abscission of whole flowers and 
flower organs. 
Floral senescence in the family Myrtaceae is strongly related, at least in part, to the 
water status within cut-flowers (Joyce and Jones 1992, Burge et al. 1 996, Sun et al. 
2000) .  In the closely related Metrosideros collina, Sun et al. (2000) demonstrated that 
when stamen abscission did not occur in response to endogenous and/or exogenous 
ethylene, wilting of stamens ensued. In cut cymules of M. collina, the general effect of 
supplementation of sucrose on stamen wilting was concentration specific, a response 
that was also inversely and negatively correlated with cymule mass. In the current 
study, the use of relatively low and/or short-term doses of sucrose (2% supplied 
continuously or the 10% sucrose pulse 2% holding) did not appreciably increase the 
level of wilting above that of control water. However, this response may have occurred 
had higher sucrose concentrations been used, as observed in M. collina where 
concentrations above 10% caused a higher rate of wilting, presumably through a 
reduction in solution osmotic potential (Sun et al. 2000) . 
Inclusion of HQC in holding solutions can effectively inhibit or depress microbial 
growth and subsequently improve water relations within the cut flowers (Put and Clerkx 
1 98 8) .  The findings with regards to the addition of HQC in vase solutions of M. excelsa 
were not consistent with those reported for the closely related genera Chamelaucium 
(Joyce 1 988), Leptospermum (Burge et al. 1996) or for M. collina (Sun et al. 2000). 
HQC added to holding solutions of M. excelsa significantly increased flower wilting, 
especially when applied continuously and, to a lesser degree, after a 1 0% sucrose pulse. 
Specifically, by the end of the experimental period, wilting was highest in 'Lighthouse' 
following treatment with HQC alone (mean ?70%) and for 'Vibrance' with both HQC 
applied continuously and after a 1 0% sucrose pulse (?76%). This effect on water 
relations was most apparent in 'Lighthouse' , whereby water uptake and transpiration of 
cut-cymules treated with HQC was considerably lower than in control water. 
2 1 5  
Moreover, the rapid decline in fresh weight was strongly and negatively correlated with 
the degree of wilting, showing the tightest linear fit with HQC solutions. This suggests 
that the water status of HQC treated cut-cymules in 'Lighthouse' directly affect stamen 
wilting. 
HQC is generally considered non-toxic to most cut flowers even at concentrations 
effective for controlling microbial growth (Van Doom 1997). However, cut cymules of 
M. excelsa appear to be very sensitive to this compound when applied alone 
(irrespective of pH), but less so in the presence of sucrose. Sucrose dissolved in vase 
water (without a biocide) is conducive to bacterial growth and is subsequently 
associated with the formation of stem occlusions (Put and Clerkx 1 988). Furthermore, 
solutions with dissolved sugars (as opposed to water) are less easily transported through 
stems due to a lower osmotic potential of the solution (V an Doom 1997). Considering 
that water uptake in the 2% sucrose holding solution (with 200 mg r 1 HQC added) was 
significantly higher than in the solution containing HQC alone, it might be inferred that 
some level of protection from microbial occlusions resulted from the addition of HQC. 
Therefore, the wilting response by HQC may have occurred irrespective of its actions in 
preventing microbial growth in cut cymules of M. excelsa. In the woody cut flower 
Banksia coccinea, addition of HQS to a vase solution was considered detrimental, 
causing a reduction in vase life and acceleration in the opening of florets (Delaporte et 
al. 1 997). Similarly, HQ based compounds have also been reported to cause leaf 
damage or stem browning in Chrysanthemum (Gladon and Staby 1 976) and Gypsophila 
(Katchansky 1979) and petal discolouration in marguerite daisy (Byme et al. 1 979). 
High levels of relative humidity (RH) can affect various aspects of flower development 
(Zieslin and Gottesman 1 983 ,  Jones et al. 1993a, Burge et al. 1 998). In M. excelsa 
flowers, high RH not only slowed the rate of floral development ( eg. stamen unfolding 
and expansion) but ultimately induced a flower abscission response at an early stage in 
development, occurring before the onset of anthesis (Stage 4). It was observed that 
flowers under the high RH environment remained turgid,  even after abscission. Zieslin 
and Gottesman ( 1983) also reported a similar response in excised flowering shoots of 
Leptospermum scoparium under comparable conditions .  Conditions with high RH 
( 100%) induced whole flower abscission (approximately 25% of fresh weight of stems), 
although flowers remained turgid. In contrast, at low RH (50-70%) no floral abscission 
2 1 6  
in L. scoparium occurred and flowers eventually shrivelled while petals underwent in?
rolling (Zieslin and Gottesman 1 983) .  
Changes in  RH may also affect the water balance within cut flowers and subsequent 
flower qualities (Burge et al. 1 998). In harvested inflorescences of hybrid Limonium 
'Chorus Magenta' , water uptake was significantly depressed under high RH conditions 
(98- 100%) averaging a 10 ml difference in water uptake per any given day. Fresh 
weight of inflorescences, however, showed the opposite response being relatively 
higher at high RH, although no explanation was given for the inverted response (Burge 
et al. 1 998). In M. excelsa, cut cymules held under high RH also showed lower rates of 
water flux, but lower cymule fresh weights in comparison with those treated to low RH. 
This response is likely attributed, at least in part, to the differences in atmospheric and 
holding solution water potential, whereby the comparatively higher water potential of 
the atmosphere would depress rates of transpiration and subsequently reduce water 
uptake (V an Doom 1 997). Consequently, an overall reduction in water uptake and 
transpiration would account for a low fresh weight as observed in the high RH 
environment. Furthermore, even though fresh weight was comparatively lower under 
the high RH conditions, it continued to increase with time and maintained a net positive 
water balance (transpiration did not exceed water uptake) which would explain the 
turgidity of stamens at the time of abscission. 
It was surprising that there was no significant difference in the amount of endogenous 
ethylene produced by cymules held in the different RH environments. This is in light of 
the fact that 1 00% of cymules had abscised under conditions of high RH, whereas no 
abscission occurred when RH was low. The lack of maintenance of high humid 
conditions within the sampling jars held for 48 h may partly account for these 
observations. Alternatively, high humidity conditions may increase flower 
susceptibility to ethylene action. Zieslin and Gottesman ( 1983) suggested that 
endogenous ethylene was involved in flower abscission in the closely related L. 
scoparium, with higher moisture levels required for the enzymatic activity of the 
abscission process. 
Endogenous ethylene production and/or sensitivity of the plant tissue to ethylene 
usually increase with tissue age (H?yer 1 996b, Woolf et al. 1 995, 1 999). For example, 
217  
floral bud abscission in intact pot plants of Camellia becomes increasingly sensitive to 
exogenous ethylene with time, an effect that was also mediated by temperature (Woolf 
et al. 1 995,  1 999). In cut cymules of M. excelsa, endogenous levels of ethylene 
increased with flower age, showing a rise in production on Day 4 of incubation (0.4 )..ll r 
1 per 48  h) followed by a subsequent decline. Similar peaks in ethylene production have 
been reported in various woody flowering genera used in the cut-flower industry, such 
as in Leptospermum (Zieslin and Gottesman 1983), Telopea (Faragher 1 986) and 
Boronia (Macnish et al. 1 999). In contrast, flower sprigs of Chamelaucium uncinatum 
showed no endogenous ethylene peak during vase life. Research into the respiratory 
behaviour of flowers during senescence may elucidate as to whether Metrosideros is 
climacteric .  It was interesting to note, however, that the highest levels of ethylene 
production occurred during the time of anther dehiscence (Stage 5), which precedes the 
stage during which stamens abscise (Stage 6). This contrasts to the response of non?
climacteric species in which senescence is not likely to be regulated by ethylene (Olley 
et al. 1 996) . 
Exogenous ethylene appears to be involved in flower abscission of various genera 
within the family Myrtaceae, including Leptospermum (Burge et al. 1 996), 
Chamelaucium (Joyce 1 989), and Metrosideros (Sun et al. 2000), in addition to a 
number of Australian woody ornamentals from other families (Macnish et al. 2000). 
Cut cymules of M. excelsa were extremely sensitive to exogenous sources of ethylene. 
Ethylene concentrations greater than 0. 1 )..ll r ' applied for 24 h were sufficient to induce 
whole flower abscission from Day 3 onwards. A similar response was also effective in 
hastening floral abscission in cut cymules of M. collina (Sun et al. 2000) . Exposure of 
cut cymules to exogenous ethylene up to 10  )..ll r ' (for 24 h) did not, however, stimulate 
the production of endogenous ethylene, at least not for cymules pulsed with ethylene at 
Stage 2 of development (receptadc ex.pa.1sicn a.'1d sta__rt of petal reflexing). 
1 -Methylcyclopropene ( 1 -MCP) and silver thiosulfate (STS) afforded significant 
protection from the deleterious effects of ethylene, reducing the level of whole flower 
and stamen abscission in cut cymules of M. excelsa. This is consistent with the efficacy 
of STS and/or 1 -MCP in protecting flowers from ethylene-induced abscission in the 
closely related genera Chamelaucium (Serek et al. 1 995b), Verticordia (Joyce and Poole 
? 1 8  
1 993) and Leptospennum (Macnish et al. 2000). In this study, the level of protection 
conferred by 1 -MCP was slightly less than that of STS (when treated with exogenous 
ethylene), but effective up to a maximum dosage of 1 50 nl r1 of 1 -MCP. 
Concentrations of around 1 50 nl r1 of 1 -MCP in Metrosideros are, however, considered 
to be close to saturation (Sun et al. 2000). The effectiveness of STS in reducing the 
level of stamen abscission was similar for cut cymules of M. collina (Sun et al. 2000). 
However, application of 1 50 nl r1 1 -MCP in the latter species conferred only short-term 
protection and subsequently enhanced endogenous ethylene production (Sun et al. 
2000) . Adverse ethylene emanation responses have been reported with the use of STS 
in L. scoparium, although protection against ethylene-induced abscission was still 
conferred (Zieslin and Gottesman 1 986). It was surprising that in the present study 
STS-treated cymules produced a higher level of stamen abscission in the absence rather 
than presence of an exogenous ethylene treatment, and may reflect an unexplained 
adverse response to STS in the absence of exogenous ethylene. 
Within the scientific literature, the efficacy of STS has primarily been attributed to its 
role as a potent inhibitor of ethylene action (Serek et al. 1 995b, Sisler and Serek 1 997). 
However, evidence from this and various other studies suggest that this compound may 
also play a role in improving water relations within cut-flowers, subsequently 
improving flower quality and delaying flower senescence (Joyce 1 989, Burge et al. 
1 996) .  In the case of M. excelsa, 2mM of STS applied as a spray significantly doubled 
the level of water flux compared to unprotected cymules. This result was irrespective of 
whether 0 or 5 Jll r1 of exogenous ethylene was applied, although the effect was slightly 
more pronounced in the former case. Nevertheless, the fresh weight of cymules treated 
with 2mM STS was higher in comparison with controls (either 0 or 5 Jll r1 of ethylene), 
whilst cymule fresh weight in all treatments was inversely related to the level of wilting. 
Joyce ( 1989) also noted a :similar improvement in fresh weight of STS-treated cut 
flowers of C. uncinatum in the presence (9.4 Jll r1) or absence of ethylene. Thus, the 
relative turgidity of petals and low stamen wilting in STS-treated M. excelsa cymules 
can not be accounted for solely by the ethylene-inhibiting action afforded by STS, since 
the response appears to be coupled by the attainment of a more favourable water 
balance. A decrease in petal wilting (senescence) in cut flowers of L. scoparium was 
2 19  
also attributed to an STS-related improvement in water relations rather than to an anti?
ethylene response (Burge et al. 1 996). 
The water relations and flower quality (petal and stamen wilting) of M. excelsa flowers 
appeared to be less affected by 1 -MCP than STS treatments. In comparison with 
unprotected cymules (for ethylene treatments of 0 or 5 J.tl r1 ), 1 -MCP treatments 
showed a marked reduction in the rate of decline of cymule fresh weight and stamen 
wilting, although water conductivity did not differ from control treatments. In excised 
miniature potted roses, 1 -MCP also improved flower fresh weight despite not conferring 
protection from ABA-induced senescence (petal in-rolling) (Muller et al. 1 999). 
Whether 1 -MCP plays a role in water relations in other cut flowers merits further 
attention. 
In conclusion, the hypothesis that the vase life of cut flowers of Metrosideros excelsa is  
limited by the changes in water relations within the cymule, and by the presence of 
endogenous and exogenously ethylene which are primarily associated with abscission of 
floral organs, is supported by this study. These findings are consistent with those for 
closely related genera. A recommendation of the most favourable vase solution varied 
depending on the flower attribute in question (e.g. wilting or abscission). By  the end of 
the experimental periods, distilled water or a solution with 2%  sucrose (with or without 
a pulse) consistently produced the lowest levels of wilting for both cultivars, 
approximately less than 50%. The 10%-HQC alone was effective in reducing wilting 
but only in 'Lighthouse' .  Furthermore, applications of 2 mM STS and 1 50 nl r1 of 1 -
MCP conferred the greatest protection against exogenous and/or endogenous ethylene. 
Considering the different effects of 1 -MCP on the floral vase life of M. excelsa and M. 
collina, future studies may wish to investigate its effects in other congeneric species. 
220 
Chapter 7 
General discussion 
Establishment of a floricultural crop within the ornamental industry requires an 
extensive scientific knowledge of the factors underlying the promotion and maintenance 
of flowering. Several diverse and novel approaches were undertaken in this study that 
addressed three factors that potentially limit the successful expansion of Metrosideros 
excelsa as an ornamental container and cut flower crop. The three primary concerns 
addressed in this study were the lack of scientific information on (a) promoting 
vegetative phase change, (b) controlling floral induction and subsequent development, 
and (c) limits to the postharvest potential of M. excelsa as a cut flower. 
The study of phase change, despite many decades of research in this field, continues to 
pose a formidable and significant challenge for many biologists studying this 
phenomenon. One of the primary limitations in this field has been the paucity of 
information regarding the accurate characterisation of the morphological and 
physiological processes occurring during phase change, as shown for Arabidopsis 
thaliana (Telfer et al. 1997, Orkwiszewski and Poethig 2000) . Work on woody plants 
also poses a challenge given the lengthy period of each phase and the inability to obtain 
flowering during the juvenile phase in almost all woody plants. This study sought to 
focus efforts on M. excelsa by addressing the fundamental problem in this species, 
which is the reversion to a juvenile form following micropropagation. The hypothesis 
that limiting growth to a single stem successfully promotes and accelerates maturation 
was supported by this study. The use of a novel technique for capturing the optical and 
dimensional properties of leaves was successfully employed in this study, and provided 
a tool fer accurately characterising the gradual transition in leaves from a juvenile 
through to an adult form. However, as acknowledged in this and other studies, the 
attainment of adult leaf characteristics does not imply reproductive competency. 
The spatial and temporal changes in individual morphological and anatomical features 
occurring in leaves of plants of M. excelsa undergoing phase change is in accordance 
with the model proposed by Hackett and Murray ( 1 997), e.g. that phase change is a 
phenomenon controlled by a multiple, possibly overlapping, set of switches rather than 
221  
by a single master-switch. This was evident by differences in the timing of expression 
of leaf traits, such as size and shape characteristics and the accumulation of a tomentum 
on the abaxial leaf surface. Support for the multiple-switch regulatory pathway is also 
in accordance with similar observations in other woody species, such as Picea sitchensis 
(Steele et al. 1 989) and Eucalyptus globulus (James and Be11 200 1 ) . Such a conclusion 
was also reached for the model species, maize (Bongard-Pierce et al. 1 996, 
Orkwiszewski and Poethig 2000) and Arabidopsis (Telfer and Poethig 1 998). Thus, the 
search for reliable markers of phase change has been ongoing, and is possibly 
complicated by the fact that changes in individual phenotypical traits may be 
independently regulated. The notion that carbon isotope discrimination may be used as 
a reliable marker of phase change was not borne out of this study. Rather, this attribute 
appeared to reflect predominantly the physiological status of the plant following a shoot 
restriction treatment. 
Although vegetative phase change in rejuvenated, single-stemmed plants of M. excelsa 
was promoted by limiting shoots to a single stem, this was also accompanied by a 
number of physiological changes. However, these physiological changes such as the 
down-regulation of various photosynthetic parameters and the accumulation of starch in 
source leaves appeared to be symptomatic of a disruption in source and sink relations . 
These phenomena were also apparent in branched plants, which may have also resulted 
from changes in source-sink relations. Thus, the confounding of these physiological 
and phase change effects along the shoot axis of single-stemmed plants complicates the 
interpretation of processes that relate to either or both phenomena. There have been 
relatively few studies examining phase change and its relationship to physiological or 
biochemical factors (e.g. carbohydrate concentrations) during this process. One study, 
using tobacco mutants with low expression of Rubisco, suggest that attainment of a 
critical source strength is importa.Tlt in preventing a delayed vegetative phase change 
response (Tsai et al. 1997). However, Tsai et al. ( 1997) note that other factors, other 
than carbohydrates, may also be important. A further understanding of the role of 
genetic factors and their relation to various processes associated with phase change 
would contribute greatly to our understanding in this field. 
The roles played by environmental factors such as temperature, daylength and 
irradiance are critical for the commencement of the flowering process in Metrosideros. 
222 
Most studies examining flowering suggest that changes in either temperature and/or 
photoperiod are critical for flowering in myrtaceous species (Shillo et al. 1 985,  Zieslin 
and Gottesman 1 986, Day et al. 1 994a), although little attention has been given to the 
role of irradiance during this process. Clearly, the hypothesis suggesting that 
modification in temperature and daylength during induction were necessary for the 
commencement of flowering was upheld by this study. Moreover, irradiance also 
played a significant role in affecting floral development, although prior to this study its 
effects in a myrtaceous species had not been fully described over a range of irradiance 
levels.  These results are, therefore, not only scientifically interesting but also important 
from a commercial perspective, given the necessity for the accurate timing, quantity and 
quality of flowers as demanded by the consumer. In this study, the optimal conditions 
recommended for increased flowering yields can be obtained by growing plants under 
inductive cool (mean 1 5?C), short-days ( 1 0  h) for a minimum of 10-20 weeks using an 
irradiance of 567 J.lmol m-2 s-1 . The timing of anthesis can also be accurately predicted, 
occurring at approximately 1 1  weeks after transference to a warm (mean 1 8?C) 
greenhouse following a 20 week floral induction period. 
Interestingly, the size of the bud at the time of induction appeared to be a crucial factor 
in determining its competency for undergoing floral initiation and subsequent 
development. This relationship has been investigated in a few other species (Patten and 
Wang 1 994, Ohana and Weiss 1 998). As also demonstrated in field-grown plants of M. 
excelsa, larger buds were more likely to become floral, and may have had more 
developmentally advanced meristems that were better able to respond to environmental 
signals than those of smaller buds (Sreekantan et al. 200 1 ) . Thus, the size of the buds 
may serve as a reliable marker of their competency to respond to florally inductive 
conditions. Future research in this area would profit from a further understanding of the 
physiological tactors associated with bud size and flowering. 
The hypothesis that the vase life of cut flowers of M. excelsa is limited by adverse water 
relations and by the ethylene-related abscission of whole flower and floral organs was 
supported by this study. It was not surprising that holding solutions that contained 
sucrose were effective in extending vase life in M. excelsa, considering that similar 
favourable responses have been described in cut flowers of M. collina and various other 
223 
species within Myrtaceae (Burge et al. 1 996, Sun et al. 2000, 2001 ) . However, the 
addition of HQC when applied alone or after a 10% sucrose pulse detrimentally affected 
flower quality, causing premature wilting and subsequently shortening of vase life. 
Trials into alternative biocides would be strongly recommended given that holding 
solutions containing sucrose can provide a favourable medium for bacterial growth 
(Van Doom 1997). There has been some suggestion, however, that STS may serve as 
an alternative but effective biocide (Ohk:awa et al. 1 999). Clearly, STS enhanced water 
uptake in cut flowers of M. excelsa, which may have been through the inhibition of 
intra-vascular microbial activity. 
The potency of different ethylene concentrations on flower quality was clearly 
demonstrated in this study. Exposure to exogenous ethylene levels above > 1 .0 I-ll r '  
detrimentally reduced the vase life in  cut cymules of  M. excelsa. Based on  an ethylene?
response classification scheme proposed by W oltering ( 1 987) for intact flowers from 
various ornamental species, the results from this study suggests that cut flowers of M. 
excelsa could be classified as being relatively ' sensitive' to the effects of exogenous 
ethylene. Inhibition of ethylene-mediated damage, however, was possible provided the 
cut flowers were treated with appropriate inhibitors of ethylene action. In particular, 
STS clearly afforded significant protection against ethylene-related abscission of floral 
organs. However, given the recent concerns of the environmental effects of silver, 
treatment with 1 -MCP may provide an effective and safer alternative to STS, despite a 
slight reduction in flower petal quality. 
In conclusion, the objectives as initially stated in this thesis and specified in the sub?
contract with Massey University by the Crop & Food Research Institute were addressed 
and successfully achieved. The current thesis has, therefore, advanced our biological 
understanding of vegetative phase change, the environmental control of flowering and 
the postharvest characteristics of cut flowers in M. excelsa. These findings are critical 
for the establishment of M. excelsa within the floricultural industry as an ornamental 
container plant and cut flower. 
224 
References 
Abeles, F.B . ,  Morgan, P.W., Saltveit, M.E. ( 1992) Ethylene in Plant Biology. Second 
Edition. San Diego, Academic Press. 
Amasino, R.M. ( 1 996) Control of flowering time in plants. Current Opinion in Genetics 
and Development 6 :  480-487. 
Amthor, J .S . ,  Koch, G.W., Bloom, A.J. ( 199 1 )  C02 inhibits respiration in leaves of 
Rumex crispus L. Plant Physiology 98: 757-760. 
Anderson, N.E. ( 1 993) Effects of mobile shade on growth and development of pot plants. 
Acta Agriculturae Scandinavica 43 : 1 25- 1 28.  
Anon. (2000) Horticultural Monitoring Report: Floriculture. Ministry of Agriculture and 
Forestry, New Zealand. Retrieved September 0 1 ,  200 1 ,  http://www.maf.govt.nz/ 
MAFnet/publications/farm-monitoring/fmhortOO/httoc.htm. 
Anon. (200 1a) Module 1 - An overview of the world horticultural industry. School of 
Agriculture and Horticulture, University of Queensland, Australia. Retrieved 
September 0 1 ,  200 1 ,  http://www.library.uq.edu.au/gattonleres//modl .html 
Anon. (200 1b) Horticulture: industry overview. Retrieved September 0 1 , 200 1 ,  
http://www .careers.co.nz/industry/i 1_agr/i 1 d_hor.htm. 
Armitage, A. M. ( 1 995) Photoperiod, irradiance, and temperature influence flowering of 
Hamelia patens (Texas Firebush). HortScience 30: 255-256. 
Aspinall, D. ( 1980) Role of abscisic acid and other hormones in adaptation to water 
stress. In Adaptation of plants to water and high temperature stress .  Turner, N.C., 
Kramer, P.J. (eds) . pp. 155- 1 72.  Wiley, New York. 
Atherton, J.G. ( 1987) Manipulation of flowering. Butterworths, London. 
Bailey, D.G. ( 1 995) Pixel calibration techniques. Proceedings of the New Zealand Image 
and Vision Computing '95 Workshop, Lincoln: 37-42. 
Bailey, D.G., Hodgson, R.M. ( 1988) VIPS - digital image processing algorithm 
development environment. Image and Vision Computing 6: 1 76- 1 84. 
Bauer, H. ,  Bauer, U. ( 1 980) Photosynthesis in leaves of the juvenile and adult ivy (Hedera 
helix). Physiologia Plantarum 49: 366-372. 
Bemier, G.  ( 1988) The control of floral evocation and morphogenesis. Annual Review of 
Plant Physiology and Molecular Biology 39:  1 75-219 .  
225 
Bernier, G., Kinet, J.M., Sachs, R.M. ( 1 98 1 a) The physiology of flowering. Vol . I. The 
initiation of flowers. CRC Press, Boca Raton, FL. 
Bernier, G., Kinet, J .M., Sachs, R.M. ( 1 98 1b) .  The physiology of flowering. Vol . II. 
Transition to Reproductive Growth. CRC Press, Boca Raton, FL. 
Bernier, G., Havelange, A., Houssa, C., Petitjean, A., Lejeune, P. ( 1993) Physiological 
signals that induce flowering. The Plant Cell 5: 1 147- 1 1 55 .  
Berthon, J.Y., Boyer, N . ,  Gasper, T .  ( 1987) Sequential rooting media and rooting 
capacity of Sequoiadendron giganteum in vitro: peroxidase activity as a marker. 
Plance Cell Report 6: 341 -344. 
Beyer, E.M. ( 1 976) A potent inhibitor of ethylene action in plants . Plant Physiology 58:  
268-27 1 .  
Bieleski, R.L., Reid, M.S. ( 1992) Physiological changes accompanying senescence in 
the ephemeral daylily flower. Plant Physiology 98: 1 042- 1049. 
Blazquez, M.A., Green, R., Nilsson, 0., Sussman, M .R., Weigel, D. ( 1998) 
Gibberellins promote flowering of Arabidopsis by activating the LEAFY promoter. 
Plant Cell lO:  79 1 -800. 
Bleecker, AB., Kende, H. (2000) Ethylene: a gaseous signal molecule in plants. 
Annual Review of Cell Developmental Biology 16: 1 - 1 8 . 
Boardman, N.K. ( 1 977) Comparative photosynthesis of sun and shade plants. Annual 
Review of Plant Physiology 28: 1 55- 1 77. 
Bodson, M.  ( 1983) Effect of photoperiod and irradiance on floral development of young 
plants of a semi-early and a late cultivar of azalea (Rhododendron sp. ) .  Journal of 
American Society of Horticultural Science 1 08:  382-386. 
Bolotin, M. ( 1975) Photoperiodic induction of precocious flowering in a woody species, 
Eucalyptus occidentalis Endl. Botanical Gazette 1 36: 358-365 .  
Bonal, D. ,  Sabatier, D., Montpied, P . ,  Tremeaux, D. ,  Guehl, J.M. (2000) Interspecific 
variability of 81 3C among trees in rainforests of French Guiana: functional groups 
and canopy integration. Oecologia 1 24: 454-468. 
Bond, B .J .  (2000) Age-related changes in photosynthesis of woody plants . Trends in Plant 
Science 5 :  349-353. 
Bongard-Pierce, D .K., Evans, M.M.S. ,  Poethig, R.S. ( 1 996) Heteroblastic features of leaf 
natomy in maize and their genetic regulation. International Journal of Plant Sciences 
1 57: 33 1 -340. 
226 
Bonnet-Masimbert, M. ( 1 987) Floral induction in conifers: a review of available 
techniques. Forest Ecology & Management 1 9 :  1 35- 146. 
Borochov, A. ,  Woodson, R. ( 1989) Physiology and biochemistry of flower petal 
senescence. Horticultural Reviews 1 1 :  1 5-43 .  
Borochov, A . ,  Spiegelstein, H . ,  Philosoph-Hadas, S .  ( 1997) Ethylene and flower petal 
senescence: Interrelationship with membrane lip catabolism. Physiologia 
Planarum 100:  606-612. 
Bowman, J.L., Alvarez, J . ,  Weigel, D., Meyerowitz. E.M., Smyth, D.R. ( 1 993) Control 
of flower development in Arabidopsis thaliana by APETALAJ and interacting genes. 
Development 1 1 9 :  72 1 -743 . 
Brand, M.H., Lineberger, R.D. ( 1992) In vitro rejuvenation of Betula (Betulaceae): 
morphological evaluation. American Journal of Botany 79: 6 1 8-625 . 
Bredmose, N. ( 1 993) Effects of year-round supplementary lighting on shoot development, 
flowering and quality of two glasshouse rose cultivars. Scientia Horticulturae 54: 69-
85. 
Brink, R.A. ( 1 962) Phase change in higher plants, and somatic cell heredity. Quarterly 
Review Biology 37: 1 -22. 
Brink, R.A., Styles, D.E., Axtell, J .D. ( 1968) Paramutation: directed genetic change. 
Science 1 59 :  1 6 1 - 1 70. 
Brown, K.M. ( 1 997) Ethylene and abscission. Physiologica Plantarum 100: 567-576. 
Brown, J .H. ,  Legge, R.L., Sisler, E.C., Baker, J.E.,  Thompson, J .E. ( 1986) Ethylene 
binding to senescing carnation petals .  Journal of Experimental Botany 37: 526-534. 
Brugnoli, E., Hubick, K.T., von Caernmerer, S . ,  Wong, S.C., Farquhar, G.D. ( 1 988) 
Correlation between the carbon isotope discrimination in leaf starch and sugars of 
C3 plants and the ratio of intercellular and atmospheric partial pressures of carbon 
dioxide. Plant Physiology 88: 141 8 - 1424. 
Burge, G.K., Bicknell, R.A., Dobson, B .G. ( 1 996) Postharvest treatments to increase 
water uptake and the vase life of Leptospermum scoparium Forst. New Zealand 
Journal of Crop and Horticultural Science 24: 371 -378. 
Burge, G.K., Morgan, E.R., Konczak, I., Seelye, J .F. ( 1998) Postharvest characteristics of 
Limonium 'Chorus Magenta' inflorescences. New Zealand Journal of Crop and 
Horticultural Science 26: 1 35- 142. 
227 
Bum, J.E., Bagnall, D.J., Metzger, J.D., Dennis, E.S. ,  Peacock, W.J. ( 1 993) DNA 
methylation, vemalization, and the initiation of flowering. Proceedings of the 
National Academy of Sciences, USA 90: 287-29 1 .  
Byme, T.G., Famham, D.S. ,  Pyett, L.S. ( 1979) Postharvest studies with marguerite daisies. 
Florist Nursery Report 1 -4. 
Cameron, A.C., Reid, M.S. ( 1983) Use of silver thiosulfate to prevent flower 
abscission from potted plants. Scientia Horticulturae 1 9: 373-378. 
Camprubi, P . ,  Nichols, R. ( 1 978) Effects of ethylene on carnation flowers (Dianthus 
caryophyllus) cut at different stages of development. Journal of Horticultural Science 
53 :  1 7-22. 
Carswell, F.E., Day, J.S. ,  Gould, K.S ., Jameson, P.E. ( 1 996) Cytokinins and the 
regulation of plant form in three species of Sophora. New Zealand Journal of Botany 
34: 1 23- 1 30. 
Chalmers, P.M. ( 1 992) The adaptive significance of juvenile versus adult leaves in 
Eucalyptus globulus ssp. globulus. Unpublished Honours Thesis. University of 
Tasmania, Hobart. 
Chattaway, M.M. ( 1 948) The development of tyloses and secretion of gum in heartwood 
formation. Australian Journal Science Research Biology 108:  227-240. 
Choinski, J .S . ,  Wise, R.R. ( 1 999) Leaf growth and development in relation to gas 
exchange in Quercus marilandica Muenchh. Journal of Plant Physiology 1 54: 302-
309. 
Clearwater, M.J., Gould, K.S. ( 1 993) Comparative leaf development of juvenile and 
adult Pseudopanax crassifolius. Canadian Journal of Botany 72: 658-670. 
Clemens, J., Jameson, P.E., Bannister, P., Pharis, R.P. ( 1 995) Gibberellins and bud 
break, vegetative shoot growth and flowering in Metrosideros collina cv. Tahiti. 
Plant Growth Regulation 16:  1 6 1 - 1 7 1 .  
Clemens, J . ,  Jameson, P.E., Henriod, R.E. ( 1 998) Hastening and controlling flowering in 
Metrosideros. Combined Proceedings International Plant Propagators ' Society 48, 
80-83. 
Cockayne, L. ( 19 1 1 )  Observations concerning evolution derived from ecological studies 
in New Zealand. Transactions and proceedings of the New Zealand Institute 44: 
1 -50. 
Cockayne, L. ( 1 928) The vegetation of New Zealand. Die Vegetation der Erde Vol. 1 4. 
Second edition. Leipzig, Wilhelm Engelmann. 
228 
Coen, E.S. ,  Meyerowitz, E.M. ( 1 99 1 )  The war of the whorls: Genetic interactions 
controlling flower development. Nature 353: 3 1 -37. 
Cordell, S . ,  Goldstein, G., Mueller-Dombois, D. ,  Webb, D. ,  Vitousek, P.M. ( 1 998) 
Physiological and morphological variation in Metrosideros polymorpha, a dominant 
Hawaiian tree species, along an altitudinal gradient: the role of phenotypic plasticity. 
Oecologia 1 1 3 :  1 88-196.  
Criley, R.A. ( 1998) Leucospermum: botany and horticulture. Horticultural Reviews 22:  
27-90. 
Damesin, C. ,  Rambal, S . ,  Joffre, R. ( 1998) Seasonal and annual changes in leaf 81 3C in 
two eo-occurring meiterranean oaks: relations to leaf growth and drought 
progression. Functional Ecology 1 2, 778 - 785. 
Davis, L.J. ( 1 99 1 )  Early induction of flowering in kiwifruit seedlings. Acta Horticulturae 
297: 23 1 -235. 
Dawson, J .W. ( 1 968) The vegetative buds of the New Zealand species of Metrosideros. 
New Zealand Journal of Botany 6: 240 - 242. 
Dawson, LA., King, R.W. ( 1 993) Effects of environment and applied chemicals on the 
flowering and form of Geraldton Wax (Chamelaucium uncinatum Schauer). 
Scientia Horticulturae 54: 222-246. 
Day, J .S. ,  Gould, K.S. ( 1997) Vegetative architecture of Elaeocarpus hookerianus. 
Periodic growth patterns in divaricating juveniles. Annals of Botany 79: 407-41 6. 
Day, J .S . ,  Loveys, B .R., Aspinall, D. ( 1994a) Environmental control of flowering of 
Boronia megastigma (Rutaceae) and Hypocalymma angustifolium (Myrtaceae). 
Australian Journal of Botany 42: 2 1 9-229. 
Day, J.S. ,  Loveys, B .R., Aspinall, D. ( 1994b) Manipulation of flowering and 
vegetative growth of brown boronia (Boronia megastigma Nees.)  and white 
myrtle (Hypocalymma angustifolium Endl .) using plant growth regulators. Scientia 
Horticulturae 56: 309-320 
Day, J.S. ,  Loveys,  B .R., Aspinall, D. ( 1995) Cytokinin and carbohydrate changes during 
flowering of Boronia megastigma. Australian Journal of Plant Physiology 22: 
57-65. 
Day, J .S . ,  Gould, K.S. ,  Jameson, P.E. ( 1997) Vegetative architecture of Elaeocarpus 
hookerianus. Transition from juvenile to adult. Annals of Botany 79 : 6 1 7-624. 
Day, J .S . ,  Gould, K.S. ,  Jameson, P.E. ( 1998) Adventitious root initiation, plasticity, 
and response to plant growth regulator treatments of seeding, juvenile, and adult 
Elaeocarpus hookerianus plants. New Zealand Journal of Botany 36: 477-484. 
229 
De Jong, J. ( 1986) Adaptation to Chrysanthemum morifolium to low light levels. Scientia 
Horticulturae 28: 263-270. 
Delaporte, K.L., Klieber, A., Sedgely, M. ( 1997) To pulse or not to pulse? Improving the 
vase life of Banksia coccinea by postharvest treatments. Australian Horticulture 95: 
79-84. 
Delaporte, K.L., Klieber, A., Sedgley, M. (2000) Postharvest vase life of two flowering 
Eucalyptus species. Postharvest Biology and Technology 19:  1 8 1 - 1 86. 
Diekman, J .  ( 1 997) Elucidating cytokinin response mechanisms using mutants. Plant 
Growth Regulation 23: 33-40. 
Diekman, J . ,  Ulrich, M. ( 1 985) A novel cytokinin-resistant mutant of Arabidopsis with 
abbreviated shoot development. Planta 195 :  440-449. 
Donovan, L.A., Ehleringer, J .R. ( 199 1 )  Ecophysiological differences among juvenile 
and reproductive plants of several woody species. Oecologia 86: 594-597. 
Donovan, L.A., Ehleringer, J .R. ( 1992) Contrasting water-use patterns among size and 
life-history classes of a semi-arid shrub. Functional Ecology 6: 482-488.  
Donovan, L.A. ,  Ehleringer, J .R.  ( 1994) Carbon isotope discrimination, water-use 
efficiency, growth, and mortality in a natural shrub population. Oecologia 1 00: 347-
354. 
Doss, R.P., Christian, J.K. ( 1979) Relationships between bulb size, apex size and flowering 
bulbous Iris cv. Ideal . Physiologia Plantarum 45 : 2 1 5-2 18 .  
Durkin, DJ .  ( 1 979) Effect of millipore filtration, citric acid, and sucrose on peduncle 
water potential of cut rose flower. Journal of American Society of Horticultural 
Science 1 04: 860-863. 
Dutkowski , G.W., Potts, B .M. ( 1 999) Geographical patterns of genetic variation in 
Eucalyptus globulus ssp. globulus and a revised racial classification. Australian 
Journal of Botany 47: 237-263. 
Eason, J.R., DeVre, L. ( 1995) Ethylene-insensitive floral senescence in Sandersonia 
aurantiaca (Hook.). New Zealand Journal of Crop and Horticultural Science 23:  447-
454. 
Ehleringer, J.R., Hall, A.E., Farquhar, G.D. ( 1 993) Stable isotopes and plant carbon/water 
relations. Academic Press, San Diego, USA. 
Einset, J .W. ( 1986) Citrus stylar abscission in tissue cultures. Acta Horticulturae 1 79:  621 -
630. 
230 
Endress, P.K. ( 1994) Diversity and evolutionary biology of tropical flowers. Cambridge 
University Press. 
Eris, A., Barut, E. ( 1 993) Decreasing severity of alternations using girdling and some 
plant regulators in olive. Acta Horticulturae 329: 1 3 1 - 1 33.  
Evans, L.T. ( 1 969) The induction of flowering: some case histories . Cornell University 
Press, Ithaca, New York. 
Evensen, K.B . ( 1 99 1 )  Ethylene responsiveness changes in Pelargonium x domesticum 
florest. Physiologia Plantarum 82: 409-41 2. 
Evensen, K.B . ,  Page, A.M., Stead, A.D. ( 1 993) Anatomy of ethylene-induced petal 
abscission in Pelargonium x hortorum. Annals of Botany 7 1 :  559-566. 
Faragher, J.D. ( 1986) Post-harvest physiology of waratah inflorescences (Telopea 
speciosissima, Proteaceae ). Scientia Horticulturae 28: 27 1 -280. 
Faragher, J.D. ( 1989) A review of research on postharvest physiology and horticulture of 
Australian native flowers. Acta Horticulturae 261 :  249-256. 
Farquhar, G.D., Richards, R.A. ( 1984) Isotopic composition of plant carbon correlates 
with water-use efficiency of wheat genotypes. Australian Journal of Plant 
Physiology 1 1 : 539-552. 
Farquhar, G.D., O'Leary, M.H., Berry, J.A. ( 1 982) On the relationship between carbon 
isotope discrimination and intercellular carbon dioxide concentration in leaves. 
Australian Journal of Plant Physiology 9: 1 2 1 - 1 37 .  
Farquhar, G.D., Ehleringer, J .R., Hubick, KT. ( 1 989) Carbon isotope discrimination 
and photosynthesis. Annual Review of Plant Physiology and Molecular Biology 40: 
503-537. 
Felix, G., Grosskopf, D.G., Regenass, M., Basse, C.W.,  Boiler, T. ( 199 1 )  Elicitor-
induced ethylene biosynthesis in tomato cells - characterisation and use as a bioassay 
for elicitor action. Plant Physiology 97: 1 9-25. 
Fluhr, R. ,  Mattoo, K.A. ( 1996) Ethylene - Biosynthesis and perception. Critical 
Reviews in Plant Science 15 :  479-523. 
Foley, J.G., Van Dam, A. ( 1982) Fundamentals of interactive computer graphics. 
Addison Wesley. 
Fortanier, E.J. ,  Jonkers, H. ( 1 976) Juvenility and maturity of plants as influenced by 
their ontogenetic and physiological aging. Acta Horticulturae 56: 37-44. 
23 1 
Frydman, V.M.,  Wareing, P.F. ( 1973a) Phase change in Hedera helix L. I. Gibberellin-
like substances in the two growth phases. Journal of Experimental Botany 24: 1 1 3 1 -
1 1 38 .  
Frydman, V.M., Wareing, P.F. ( 1973b). Phase change in Hedera helix L. II.  The 
possible role of roots as a source of shoot gibberellin-like substances. Journal 
Experimental Botany 24: 1 1 39- 1 148.  
Frydman, V.M., Wareing, P.F. ( 1974) Phase change in Hedera helix L. ID. The effects 
of gibberellins, abscisic acid and growth retardants on juvenile and adult ivy. 
Journal of Experimental Botany 25 : 420-429. 
Fulton, T. ( 1998) National Climate Laboratory Report (Experiment No. 580). The 
Horticulture & Food Research Institute of New Zealand, Palmerston North. 
Furutani, S .C. ,  Johnston, E., Nagao, M. ( 1 989) Anthesis and abscission of blue jade 
vine flowers treated with ethephon and AOA. HortScience 24: 1042. 
Garner W.W., Allard A.H. ( 1 920) Effect of the relative length of day and night and 
other factors of the environment on growth and reproduction in plants . Journal of 
Agricultural Research 1 8 :  553-606. 
Gerrish, G. ( 1989) Comparing crown growth and phenology of juvenile, early mature, and 
late mature Metrosideros polymorpha trees . Pacific Science 43 : 2 1 1 -222. 
Gezelius, K., Ericsson, A., Hallgren, J .E., Brunes, L. ( 198 1 )  Effects of bud removal on 
nutrient contents and carboxylase activity in Scots pine (Pinus silvestris) seedlings. 
Physiologia Plantarum 5 1 :  1 8 1 - 1 88 .  
Gladon, R.J., Staby, G.L. ( 1976) Opening of  immature chrysanthemums with sucrose and 
8-hydroxyquinoline citrate. Hortscience 1 1 : 206-208 . 
Godley, E.J. ( 1985) Paths to maturity. New Zealand Journal of Botany 23 :  687-706. 
Goebel, K. ( 1 900) Organography of plants. Part I. General organography. English 
translation by Balfour, I.B. New York: Hafner. 
Goszczynska, D., Zieslin, N .  ( 1993) Abscission of flower peduncles in rose (Rosa x 
hybrida) plants and evolution of ethylene. Journal of Plant Physiology 1 42:  
2 14-2 1 7. 
Greenwood, M.S .  ( 1 995) Juvenility and maturation in conifers: current concepts. Tree 
Physiology 1 5 :  433-438 .  
Greenwood, M.S . ,  Hutchinson, K.W. ( 1 993). Maturation as developmental process .  In 
Clonal Forestry: Genetics, Biotechnology and Applications .  Ahuja, M.R., Libby, W.J. 
(eds) .  pp 14-33 . Springer-Verlag, New York. 
232 
Greenwood, M.S ., Hopper, C.A.,  Hutchison, K.W. ( 1 989) Maturation in larch, IT. 
Effects of age on photosynthesis and gene expression in developing foliage. Plant 
Physiology 94: 1 308- 1 3 1 5 . 
Greer, D.H., Halligan, E.A. (2001 )  Photosynthetic and fluorescence light responses for 
kiwifruit (Actinidia deliciosa) leaves at different stages of development on vines 
grown at two different photon flux densities. Australian Journal of Plant Physiology 
28: 373-382. 
Griffin, A.R., Whiteman, P., Rudge, T., Burgess, LP., Moncur, M. ( 1993) Effect of 
paclobutrazol on flower-bud production and vegetative growth in two species of 
Eucalyptus. Canadian Journal of Forest Research 23: 640-647. 
Grulke, N.E., Miller, P.R. ( 1 994) Changes in gas exchange characteristics during the 
life span of giant seqoia: implications for response to current and future 
concentrations of atmospheric ozone. Tree Physiology 14:  659-668. 
Hackett, W.P. ( 1976) Control of phase change in woody plants. Acta Horticulturae 56: 
143- 154. 
Hackett, W.P. ( 1985) Juvenility, maturation, and rejuvenation of woody plants. 
Horticultural Reviews 7: 109- 1 55. 
Hackett, W.P.,  Sachs, R.M. ( 1 966) Flowering in Bougainvillea 'San Diego' .  Journal of 
American Society of Horticultural Science 88 :  606. 
Hackett, W.P., Murray, J.R. ( 1992) Maturation and rejuvenation in woody plants. Acta 
Horticulturae 3 14:  1 95-203. 
Hackett, W.P., Murray, J. R ( 1 997) Approaches to understanding maturation or phase 
change. Biotechnology of ornamental plants. Cab International, Wallingford, UK. pp. 
73-86. 
Hackett, W.P., Cordero, R.E., Srinivasan, C. ( 1987) Apical meristem characteristics 
and activity in relation to juvenility in Hedera. In Manipulation of Flowering. 
J.G. Atherton, J.G. (ed). Butterworths, London. 
Halevy, A.H. ( 1 976) Treatments to improve water balance of cut flowers. Acta 
Horticulturae 64: 223-230. 
Halevy, A.H. ( 1985) CRC Handbook of flowering, Vol. ll. CRC Press, Boca Raton, FL. 
Halevy, A.H. ( 1986) CRC Handbook of Flowering, Vol V. CRC Press, Boca Raton, FL. 
Halevy, A.H. ( 1 987) Assimilate allocation and flower development. In Manipulation of 
Flowering. J.G. Atherton,  J.G. (ed). pp 363-378.  Butterworths, London.  
233 
Halevy, A.H. ( 1989) CRC Handbook of Flowering, Vol VI. CRC Press, Boca Raton, FL. 
Halevy, A.H. ( 1 994) The use of plant bioregulators in ornamental plants. Acta 
Horticulturae 394: 37-43 . 
Halevy, A.H., Mayak, S. ( 1 98 1 )  Senescence and postharvest physiology of cut flowers 
- Part 2. Horticultural Reviews 3 :  59- 143 .  
Halevy, A.H., Whitehead, C.S. ( 1989) Pollination-induced corolla abscission and 
senescence and the role of short-chain fatty acids in the process. In Cell Separation in 
Plants . Osbome, D.J. ,  Jackson, M.B. (eds). pp. 35 :  22 1 -23 1 .  NATO ASI Series, 
Springer Verlag, Berlin. 
Hall, A.E.,  Findell, J.L., Schaller, G.E., Sisler, E.C., Bleecker, A.B . (2000) Ethylene 
perception by the ERSI protein in Arabidopsis.  Plant Physiology 1 23 :  1449- 1457. 
Hanba, Y .T. ,  Miyazawa, S.l. , Terashima, I. ( 1999) The influence of leaf thickness on 
the C02 transfer conductance and leaf stable carbon isotope ratio for some 
evergreen tree species in Japanese warm-temperate forests . Functional Ecology 1 3 :  
632-639. 
Hand, P. ,  Besford, R.T., Richardson, C.M., Peppitt, S.D. ( 1 996) Antibodies to phase 
related proteins in juvenile and mature Prunus avium. Plant Growth Regulation 20: 
25-29. 
Hansen, D.H. ( 1 996) Establishment and persistence characteristics in juvenile leaves and 
phyllodes of Acacia koa (Legurninosae) in Hawaii. International Journal of Plant 
Sciences 1 57: 1 23- 1 28 .  
Harlevy, P.C. ,  Sharkey, T.D. ( 199 1 )  An improved model of C3 photosynthesis at high C02: 
Reversed 02 sensitivity explained by lack of glycerate reentry into the chloroplast. 
Photosynthesis Research 27: 1 69- 1 78.  
Haslemore, R.M.,  Roughan, P.G.  ( 1 976) Rapid chemical analysis of some plant 
constituents. Journal of Food, Science and Agriculture 27: 1 1 7 1 - 1 1 78 .  
Heimburger, C .C., Fowler, D.P. ( 1969) Precocious flowering in some pines of the 
lariciones group. Silvae Genetica 1 8 : 146- 1 50. 
Helier, A. ,  Borochov, A., Halevy, A.H. ( 1994) Factors affecting the growth and flowering 
of Coleonema aspalathoides. Scientia Horticulturae 58:  325-334. 
Hewett, E.W. ( 1993) New horticultural crops in New Zealand. In New Crops. J .  Janick, J. 
Simon, J.E. (eds). pp. 57-64. Wiley, New York. 
234 
Hilioti, Z. , Richards, C. ,  Brown, K.M. (2000) Regulation of pollination-induced 
ethylene and its role in petal abscission of Pelargonium x hortorum. Physiologia 
Plantarum 1 09:  322-332. 
Hirayama, T., Alonso, J.M. (2000) Ethylene captures a metal ! Metal ions are involved 
in ethylene perception and signal transduction. Plant and Cell Physiology 4 1 :  
548-555. 
Horrell, B .A., Jameson, P.E., Bannister, P. ( 1 989a) Propagation and regulation of 
phase change in some New Zealand heteroblastic species. Combined Proceedings 
International Plant Propagators ' Society 39: 268-274. 
Horrell, B .A., Jameson, P.E., Bannister, P. ( 1989b) Growth promotion of ivy (Hedera helix 
L.) by paclobutrazol. Plant Growth Regulation 8: 309-3 14. 
Horrell, B .A., Jameson, P.E., Bannister, P. ( 1 990) Responses of ivy (Hedera helix L.) 
to combinations of gibberellic acid, paclobutrazol and abscisic acid. Plant 
Growth Regulation 9: 107- 1 17 .  
Hll)yer, L. ( 1996a) Critical ethylene exposure for Capsicum annuum 'Janne' is  dependent 
on an interaction between concentration, duration and developmental stage. 
Journal of Horticultural Science 7 1 :  62 1 -628. 
Hll)yer, L. ( 1996b ). Critical ethylene exposure for Hibiscus rosa-sinensis is dependent on 
an interaction between ethylene concentration and duration. Postharvest Biology 
and Technology 9 :  87-95. 
Hrib, J . ,  Chaturvedi, H. C., Dobry, J. ( 1 980) Development of shoot tips of Sambucus 
sp. grown in vitro. Phytomorphology 30: 266-270. 
lchimura, K.,  Suto, K. ( 1 998) Environmental factors controlling flowering opening and 
closing in a Portulaca hybrid. Annals of Botany 82: 67-70. 
Ichimura, K., Kojima, K., Go to, R. ( 1 999) Effects of temperature, 8-hydroxyquinoline 
sulphate and sucrose on the vase life of cut rose flowers. Postharvest Biology and 
Technology 1 5 :  33-40. 
Imbault, N. ,  Tardieu, I., Joseph, C., Zaerr, J.B.,  Bonnet-Masimbert, M. ( 1988) Possible 
role of isopentenyladenine and isopentenyladenosine in flowering of Pseudotsuga?
menziesii, endogenous variations and exogenous applications. Plant Physiology & 
Biochemistry 26: 289-296. 
Inskeep W.P., Bloom, R.P. ( 1 985) Extinction coefficients of chlorophyll a and b in N,N?
dimethylformamide and 80% acetone. Plant physiology 77: 483-485. 
Ito, S . ,  Suzaki, T. ( 1 990) Morphology and water relations of leaves of Eucalyptus 
globulus sprouts. Bulletin of the Kuyshu University Forests 63:  37-53 .  
235 
Jacobs, G. ( 1 985) Leucospermum. In CRC Handbook of Flowering, Vol. Ill. Halevy, 
A. H., (ed). pp. 283-286. CRC Press, Boca Raton, Florida. 
Jacobs, G., Minnaar, H.R. ( 1 980) Light intensity and flower development of 
Leucospermum cordifolium. Hortscience 1 5 :  644-645. 
Jackson, D .I. , Sweet, G.B . ( 1 972) Flower initiation in temperate woody plants. 
Horticultural Abstracts 42: 9-24. 
James, A.C., Mantell, S .H. ( 1 994) Characterization of developmental phases of the 
woody perennial shrub, Solanum aviculare Forst. New Phytologist 1 27 :  59 1 -600. 
James, A.C., Mantell, S .H. ( 1 997) Characterisation of developmental phases of the 
woody perennial shrub, Solanum aviculare Forst. New Phytologist 1 27 :  591 -600. 
James, S .A., Bell, D.T. ( 1 996) Leaf orientation in juvenile Eucalyptus camaldulensis. 
Australian Journal of Botany 44: 1 39-156. 
James, S .A.,  Bell, D.T. (2000a) Influence of light availability on leaf structure and 
growth of two Eucalyptus globulus ssp. globulus provenances. Tree Physiology 
20: 1007- 10 1 8 .  
James, S .A., Bell, D .T. (2000b) Leaf orientation, light interception and stomatal 
conductance of Eucalyptus globulus ssp. globulus leaves. Tree Physiology 20: 
8 1 5-823. 
James, S .A.,  Bell, D.T. (200 1 )  Leaf morphological and anatomical characteristics of 
heteroblastic Eucalyptus globulus ssp. globulus (Myrtaceae). Australian Journal 
of Botany 49 : 259-269. 
James, S .A., Smith, W.K., Vogelmann, T.C. ( 1 999) Ontogenetic differences in  
mesophyll structure and chlorophyll distribution in Eucalyptus globulus ssp. 
globulus. American Journal of Botany 86: 1 98-207. 
Jameson, P.E., Horrell, B .A. ( 1988) Studies on plant growth regulation and phase 
change in Hedera helix and New Zealand heteroblastic species. In Proceedings of 
the Plant Growth Regulator Society of America. Cook, A.R. (ed). pp 205. Ithaca, 
New York. 
Jensen, T.E., Valdovinos, J.G. ( 1968) Fine structure of abscission zones. Ill. 
Cytoplasmic changes in abscising pedicels of tobacco and tomato flowers. Planta 
83 :  303-3 1 3. 
Jiao, J. ,  Grodzinski, B .  ( 1 998) Environmental influences on photosynthesis and carbon 
export in greenhouse roses during development of the flowering shoot. Journal of 
American Society of Horticultural Science 1 23:  1 08 1 - 1088 .  
236 
Jiao, J., Gilmour, M., Tsuj ita, M.J., Grodzinski, B .  ( 1 989) Photosynthesis and carbon 
partitioning in Samatha roses. Canadian Journal of Plant Science 69: 577-584. 
Johansen, D.A. ( 1 940) Plant Microtechnique. McGraw-Hill Book Company, 
NY, USA. 
Johnson, K.A., Ronowicz, K. (2000) Morphological and genetic variation in 
Christmas bush (Ceratopetalum gummiferum Sm.) a new Australian cut flower. 
Acta Horticulturae 522: 87-96. 
Jones, C.S . ( 1 999) An essay on juvenility, phase change, and heteroblasty in seed plants. 
International Journal of Plant Sciences 160: S 1 05-S 1 1 1  
Jones, M., Sedgley, M. ( 1993) Leaf waxes and postharvest quality of Eucalyptus foliage. 
Journal of Horticultural Science 68: 939-946. 
Jones, R.B .,  Faragher, J .D., Van Doom, W.G. ( 1993a) Water relations of cut flowering 
branches of Thryptomene calycina (Lindl . )  Stapf. (Myrtaceae). Postharvest 
Biology and Technology 3: 57-67. 
Jones, R.B., Truett, J.K., Hill, M. ( 1993b) Postharvest handling of cut immature 
Eucalyptus foliage. Australian Journal of Experimental Agriculture 33 :  663-667. 
Jonkers, H. ( 197 1 )  An international experiment on juvenility in apple. Euphytica 20: 57-
59. 
Jordan, G.J. ,  Potts, B .M., Wiltshire, R.J.E. ( 1 999) Strong, independent, quantitative 
genetic control of the timing of vegetative phase change and first flowering in 
Eucalyptus globulus ssp. globulus (Tasmanian Blue Gum). Heredity 83: 1 79- 1 87.  
Joyce, D.C. ( 1 988) Postharvest characteristics of Geraldton Wax flowers. Journal of 
American Society of Horticultural Science 1 3 : 738-742. 
Joyce, D.C. ( 1989) Treatments to prevent flower abscission in Geraldton Wax . 
HortScience 24: 391 .  
Joyce, D.,  Haynes, Y. ( 1989) Postharvest treatment and shipping of Western 
Australian flora. In 'WA Flora Summit, Proceedings' . pp. 43-65. Geraldton Mid?
West Development Authority: Geraldton, WA. 
Joyce, D.C., Jones, P.N. ( 1992) Water balance of the foliage of cut Geraldton 
waxflowers. Postharvest Biology and Technology 2: 3 1 -39. 
Joyce, D.C.,  Poole, M.C. ( 1993) Effects of ethylene and dehydration on cut flowering 
stems of Verticordia spp. Australian Journal of Experimental Agriculture 33 :  489-
493. 
237 
Joyce, D.C., Beal, P.R. ( 1 999) Cutflower characteristics of terminal flowering tropical 
Grevillea: a brief review. Australian Journal of Experimental Agriculture 39: 78 1 -
794. 
Joyce, D.C., Jones, R., Faragher, J. ( 1 993) Postharvest characteristics of native 
Australian flowers. Postharvest News and Information 4: 6 1N-67N. 
Joyce, D.C., Shorter, A.J . ,  Joyce, P.A., Beal, P.R. ( 1995) Respiration and ethylene 
production by harvested Grevillea 'Sylvia' flowers and inflorescences. Acta 
Horticulturae 405 : 224-23 1 .  
Ka-Ipo, R.,  Sakai , W.S.,  Furutani, S .C., Collins, M. ( 1989) Effect of postharvest 
treatment with antitransparants on the shelf life of H. psittacorum cv. Parakeet cut 
flowers. Bulletin of the Heliconia Society International 4:  1 3- 14. 
Katchansky, M. ( 1979) Bud opening and pulsing of Gypsophila flowers with TOG II. 
Hassadeh 59: 23 1 2-23 14. 
Kerstetter, R.A., Poethig, R.S. ( 1 998) The specification of leaf identity during shoot 
development. Annual Review of Cell Developmental Biology 14: 373-398. 
Kester, D.E. ( 1976) The relationship of juvenility to plant propagation. Combined 
Proceedings International Plant Propagators ' Society 26: 7 1 -84. 
Kinet, J.M. ( 1993) Environmental, chemical, and genetic control of flowering. 
Horticultural Reviews 1 5 :  279-334. 
Kinet, J.M., Sachs, R.M., Bernier, G. ( 1985) The physiology of flowering. Vol . Ill. 
The development of flowers. CRC Press, Boca Raton, Florida. 
King, R.W. ( 1998) Dual control of flower initiation and development by temperature and 
photoperiod in Hardenbergia violacea. Australian Journal of Botany 46: 65-74. 
King, R.W., Evans, L.T. ( 199 1 )  Shoot apex sugars in relation to long-day induction of 
flowering in Lolium temulentum L. Australian Journal of Plant Physiology 1 8 :  1 2 1 -
1 35 .  
King, R.W., Ben-Tal, Y. (2001 )  A florigenic effect of  sucrose in  Fuchsia hybrida i s  
blocked by gibberellin-induced assimilate competition. Plant Physiology 1 25 :  
488-496. 
King, R.W., Dawson, LA., Speer, S .S .  ( 1992) Control of growth and flowering in two 
Western Australian species of Pimelea. Australian Journal of Botany 40: 377-388 .  
238 
King, R.W., Pate, J.S.,  Johnston, J. ( 1 996) Ecotypic differences in the flowering of 
Pimeleaferruginea (Thymelaeaceae) in response to cool temperatures .  Australian 
Journal of Botany 44: 47-55. 
Koevenig, J .  L., Sillix, D. ( 1973) Movement of !AA in sections from spider flower 
( Cloeme hassleriana) stamen filaments. American Journal of Botany 60: 23 1 -235. 
Kolb, T.E.,  Stone, J.E. (2000) Differences in leaf gas exchange and water relations 
among species and tree sizes in an Arizona pine-oak forest. Tree Physiology 20: 
1 - 1 2. 
Kotoda, N. ,  Wada, M., Komori, S . ,  Kidou, S .I., Abe, K., Masuda, T., Soejima, J. (2000) 
Expression pattern of homologues of floral meristem identity genes LFY and AP 1 
during flower development in apple. Journal of the American Society for 
Horticultural Science 125 :  398-403. 
Krapp, A., Stitt, M. ( 1995) An evaluation of direct and indirect mechanisms for the 
"sink-regulation" of photosynthesis in spinach: Changes in gas exchange, 
carbohydrates, metabolites, enzyme activities and steady-state transcript levels 
after cold-girdling source leaves. Planta 195:  3 1 3-323. 
Krause, G.H. ( 1988) Photoinhibition of photosynthesis. An evaluation of damaging and 
protective mechanims. Physiologia Plantarum 74: 566-574. 
Krishnamoorthy, H.N., Nanda, K.K. ( 1 968) Flower bud reversion in Impatiens 
balsamina under non-inductive photoperiods. Planta 80: 43-5 1 .  
Kull, 0., Koppel, A. ( 1987) Net photosynthetic response to light intensity of shoots 
from different crown positions and age in Picea abies (L.) Karst. Scandinavian 
Journal of Forest Research 23 : 450-456. 
Langton, F.A. ( 1977) The responses of early-flowering chrysanthemums to daylength. 
Scientia Horticulturae 7: 277-289. 
Lauteri, M., Brugnoli,  E., Spaccino, L. ( 1 993) Carbon isotope discrimination in leaf soluble 
sugars and in whole-plant dry matter in Helianthus annuus L. grown under different 
water conditions. In Stable Isotopes and Plant Carbon-Water Relations .  Ehleringer, 
J.R.,  Hall, A.E., Farquhar, G.D. (eds). pp. 93- 108.  Academic Press, New York. 
Lawson, J.R., Poethig, R.S . ( 1 995) Shoot development in plants: time for a change. 
Trends in Genetics 1 1 : 263-268. 
Levy, Y.Y., Dean, C. ( 1998) The transition to flowering. The Plant Cell 1 0: 1 973-
1 989. 
Lyrene, P .M. ( 198 1 )  Juvenility and production of fast -rooting cuttings from blueberry 
shoot cultures. Journal of American Society for Horticultural Science 1 06: 396-398.  
239 
Maas, F.M., Bakx, E.J. (1997) Growth and flower development in roses as affected by 
light. Acta Horticulturae 4 1 8 :  1 27- 1 34. 
Macnish, A.J . ,  Joyce, D.C., Hofman, P.J . ,  Simons, D.H. ( 1999) Involvement of ethylene in 
postharvest senescence of Boronia heterophylla flowers. Australian Journal of 
Experimental Agriculture 39:  9 1 1 -9 1 3. 
Macnish, A.J., Joyce, D.C., Hofman, P.J., Simons, D.H., Reid, M.S. (2000) 1 -
Methylcyclopropene treatment efficacy in preventing ethylene perception in banana 
fruit and grevillea and waxflower flowers. Australian Journal of Experimental 
Agriculture 40: 4 7 1 -48 1 .  
Marks, T.R., Simpson, S.E. ( 1999) Effect o f  irradiance on shoot development in vitro. 
Plant Growth Regulation 28 : 1 33- 142. 
Marler, T.E., Schaffer, B., Crane, J.H. ( 1 994) Developmental light level affects 
growth, morphology, and leaf physiology of young Carambola trees.  Journal of 
American Society of Horticultural Science. 1 1 9: 7 1 1 -7 1 8 . 
Marousky, F.J. ( 1 972) Water relations, effects of floral preservatives on bud opening, and 
keeping quality of cut flowers. HortScience 7: 1 14- 1 16. 
Mathooko, F.M., Inaba, A., Nakamura, R. ( 1998) Characterization of carbon dioxide 
stress-induced ethylene biosynthesis in cucumber (Cucumis sativus L.) fruit. Plant 
Cell Physiology 39: 285-293. 
Mayak, S . ,  Halevy, A.H., Katz, M. ( 1972) Correlative changes in phytohormones in 
relation to senescence processes in  rose petals. Physiologia Plantarum 27: 1 -4. 
Mayak, S., Halevy, A.H., Sagie, S., Bar-Yosef, A., Bravdo, R. ( 1 974) The water 
balance of cut rose flowers. Physiologica Plantarum 32: 1 5-22. 
McKenzie, R.J., Lovell, P.H. ( 1992) Flower senescence in monocotyledons: a taxonomic 
survey. New Zealand Journal of Crop and Horticultural Science 20: 67-7 1 .  
McKenzie, M.J., Veit, B. ,  Walton, E. ,  Jameson, P.E., Clemens, J. ( 1 997) MEL: A LEAFY?
like DNA sequence isolated from Metrosideros excelsa. Genebank submission. 
Accession No. AF007869. 
Meilan, R. ( 1 997) Floral induction in woody angiosperms. New Forests 14 :  1 79-202. 
Mercier, H.,  Endres, L. ( 1999) Alteration of hormonal levels in a rootless epiphytic 
bromeliad in different phenological phases. Journal of Plant Growth Regulation 
1 8 : 1 2 1 - 125. 
Michaels, S .D.,  Amasino, R.M. (2000) Memories of winter: vemalization and the 
competence to flower. Plant, Cell and Environment 23: 1 145- 1 1 53.  
240 
Moe, R., Smith-Eriksen, A. ( 1986) The effect of ethephon and STS treatment on 
flower malformation and flower bud abscission in Begonia x cheimantha Everett. 
Acta Horticulturae 1 8 1 :  1 55-160. 
Moe, R. ( 1990) Effect of day and night temperature alternations and of plant growth 
regulators on stem elongation and flowering of the long-day plant Campanula 
isophylla Moretti . Scientia Horticulturae 43 : 29 1 -305 . 
Moncur, M.W. ( 1 992) Effect of low temperature on floral induction of Eucalyptus 
lansdowneana F. Muell. & J. Brown subsp. lansdowneana. Australian Journal of 
Botany 40: 1 57- 167. 
Moncur, M.W.,  Boland, D.J. ( 1989) Floral morphology of Eucalyptus melliodora 
A.Cunn. ex Shau. and comparisons with other Eucalyptus species . Australian Journal 
of Botany 37: 125-35.  
Moncur, M.W.,  Hasan, 0. (1 994) Floral induction in Eucalyptus nitens. Tree 
Physiology 14: 1303- 1 3 1 2. 
Mondal, M.H.,  Williams, A.B., Brenner, M.L. ( 1978) Effects of sink removal on 
photosynthesis and senescence in leaves of soybean (Glycine max L.) plants. Plant 
Physiology 6 1 :  394-397. 
Monteuuis, 0. ( 199 1 )  Rejuvenation of a 1 00-year-old Sequoiadendron giganteum through 
in vitro meristem culture. I. Organogenic and morphological arguments. Physiologia 
Plantarum 8 1 :  1 1 1 - 1 15 .  
Mor, Y .  ( 1 987) Plant growth regulators in rose petals. Horticultural Reviews 9: 53-67. 
Mor, Y., Halevy, A.H. (1979) Translocation of 14C-assimilates in roses. 1 .  The effect 
of the age of the shoot and the location of the source leaf. Physiologia Plantarum 
45 : 1 77-1 82. 
M or, Y . , Halevy, A.H. ( 1984) Dual effect of light on flowering and sprouting of rose 
shoots. Physiologia Plantarum 6 1 :  1 19- 124. 
Morgan, D.C., Rook, D.A., Warrington, I.J . ,  Tumbull, H.L. ( 1 983) Growth and 
development of Pinus radiata D. Don: the effect of light quality. Plant, Cell and 
Environment 6: 691 -70 1 .  
Morgan, D .C. ,  S tanley, C.J. ,  Warrington, I.J. ( 1 985) The effects of simulated daylength 
and shade-light on vegetative and reproductive growth in kiwifruit and grapevine. 
Journal of Horticultural Science 60: 473-484. 
241 
Mortensen, L.M. ( 1994) Effects of day/night temperature variations on growth, 
morphogenesis and flowering of Kalanchoe blossfeldiana v. Poelln. at different C02 
concentrations, daylengths and photon flux densities. Scientia Horticulturae 59: 233-
241 .  
Muller, R., Stummann, B .M., Andersen, A.S . ,  Serek, M .  ( 1999) Involvement of ABA 
in postharvest life of miniature potted roses. Plant Growth Regulation 29: 143-
1 50. 
Munoz, C.E., Davies, F.S., Sherman, W. B. ( 1 982) Hydraulic conductivity and 
ethylene production in detached flowering peach shoots. Hortscience 17 :  226-228. 
Murray, J.R., Hackett, W.P. ( 199 1 )  Dihydroflavonol reductase activity in relation to 
differential anthocyanin accumulation in juvenile and mature phase Hedera helix 
L. Plant Physiology 97: 343-35 1 .  
Myers, D.A., Thomas, R.B ., DeLucia, E.H. ( 1 999) Photosynthetic responses of 
loblolly pine (Pinus taeda) needles to experimental reduction in sink demand. 
Tree Physiology 19 :  235-242. 
Myster, J. ( 1 999) The effects of temperature alterations, irradiance level, photoperiod, and 
day extension light quality on morphogenesis, growth, and flowering of Begonia x 
hiemalis Fotsch. Gartenbauwissenschaft 64: 206-2 1 3 . 
Myster, J. ,  Moe, R. ( 1 995) Effect of diurnal temperature alternations on plant morphology 
in some greenhouse crops a mini-review. Scientia Horticulturae 62: 205-2 1 5 .  
Nakano, H.,  Muramatsu, S . ,  Makino, A., Mae, T .  (2000) Relationship between the 
suppression of photosynthesis and starch accumulation in the pod-removed bean. 
Australian Journal of Plant Physiology 27 : 167-173 .  
Nautiyal, P.C., Ravindra, V. ,  Joshi, Y.C. ( 1 998) Net photosynthetic rate in peanut (Arachis 
hypogaea L.): Influence of leaf position, time of day, and reproductive-sink. 
Photosynthetica 36: 1 29- 1 38.  
Navarro, L. ,  Roistacher, C.N. ,  Murashige, T. ( 1 975) Improvement of shoot tip grafting 
in vitro for virus-free citrus. Journal of the American Society of Horticultural Science 
100:  471 -479. 
Nell, T.A. ( 1 994) Controlling ethylene extends flower life. FloraCulture International 
December: 40-42. 
Neumaier, E.E., B lessington, T.M., Price, J .A. ( 1987) Effect of light and fertilizer rate 
and source on flowering growth and light quality of Hibiscus. HortScience 22: 902-
904. 
242 
Newman, J .P., Dodge, L.L., Reid, M.S. ( 1 998) Evaluation of ethylene inhibitors for 
postharvest treatment of Gypsophila paniculata L. HortTechnology 8 :  58-63 . 
Niinemets, U., Kull, 0., Tenhunen, J.D. ( 1 999) Variability in leaf morphology and 
chemical composition as a function of canopy light environment in coexisting 
deciduous trees.  International Journal of Plant Science 1 60: 837-848.  
Njoku, E. ( 1958)  Effects of gibberellic acid on leaf form. Nature 182: 1 097- 1 098.  
Noack, L.E., Warrington, I.J., Plummer, J .A. ,  Andersen, A.S. ( 1 996) Effect of low?
temperature treatments on flowering in three cultivars of Hebe Comm. ex Juss. 
Scientia Horticulturae 66: 103- 1 1 5 .  
Nunez-Elisea, R. ,  Davenport, T.L. ( 1995) Effect of leaf age, duration of cool 
temperature treatment, and photoperiod on bud dormancy release and floral 
initiation in mango. Scientia Horticulturae 62: 63-73 . 
Nunez-Elisea, R., Davenport, T.L. , Caldeira, M.L. ( 1993) Bud initiation and 
morphogenesis in 'Tommy Atkins' mango as affected by temperature and triazole 
growth retardants. Acta Horticulturae 34 1 :  192- 1 98. 
Oberholster, S .D. ,  Peterson, C.M., Dute, R.R. ( 1 99 1 )  Pedicel abscission of soybean: 
cytological and ultrastructural changes induced by auxin and ethephon. Canadian 
Journal of Botany 69 : 2 1 77-21 86. 
Ohana, 0., Weiss, D. ( 1998) Environmental and physiological factors regulate 
Eryngium planum flowering. Israel Journal of Plant Sciences 46: 47-5 1 .  
Ohkawa, K., Kasahara, Y., JungNam, S .  ( 1 999) Mobility and effects on vase life of silver?
containing compounds in cut rose flowers . Hortscience 34: 1 1 2- 1 1 3 .  
Oleson, P .0. ( 1 978) On cyclophysis and topophysis. Silvae Genetica 27 : 1 73- 178 .  
Oliphant, J.L. ( 1 988) Consideration of juvenility and maturity in the micropropagation of 
the Metrosideros species. Acta Horticulturae 227: 482-484. 
Oliphant, J.L. ( 1 990) Juvenility and maturity of woody species in New Zealand. In Plant 
Aging: Basic and Applied Approaches. Rodriguez, R. (ed). pp. 37-42. Plenum Press, 
New York. 
Oliphant, J.L. , C lemens, J. ,  Baird, H.E. ( 1 992) Accelerated maturation after 
micropropagation in Metrosideros cultivars by plant growth regulators and other 
cultural practices. Acta Horticulturae 3 14: 22 1 -225. 
243 
Olley, C.M., Joyce, D.C., Irving, D.E. ( 1 996) Changes in sugar, protein, respiration, 
and ethylene in developing and harvested Geraldton waxflower (Chamelaucium 
uncinatum) flowers. New Zealand Journal of Crop and Horticultural Science 24: 
1 43- 150. 
Oren-Shamir, M., Nissim-Levi, A. ( 1999) Temperature and gibberellin effects on 
growth and anthocyanin pigmentation in Photinia leaves. Journal of Horticultural 
Science and Biotechnology 74: 355-360. 
Orkwiszewski, J.A.J., Poethig, R.S . (2000) Phase identity of the maize leaf is determined 
after initiation. Proceedings of the National Academy of Sciences of the United States 
of America 97 : 1063 1 - 1 0636. 
Orlovich, D.A., Drinnan, AN., Ladiges, P.Y. ( 1 996) Floral development in the 
Metrosideros group (Myrtaceae) with special emphasis on the androecium. Telopea 6: 
689-7 19. 
Osorio, J. ,  Pereira, J.S. (1 994) Genotypic differences in water use efficiency and ()1 3C 
discrimination in Eucalyptus globulus. Tree Physiology 14: 1 547- 1 556. 
Park, B .H., Pearson, S .  (2000) Environmental regulation of flowering time in heliotrope 
(Heliotropium arborescens L. cv. Marine) . Scientia Horticulturae 85 :  23 1 -241 .  
Parletta, M.A., Sedgley, M .  ( 1 996) Chemical and environmental manipulation of 
ornamental Acacia Mill. Species for pot plant production. Scientia Horticulturae 67 : 
235-246. 
Patten, K.D., Wang, J. ( 1994) Leaf removal and terminal bud size affect fruiting of 
cranberry. HortScience 29: 997-998. 
Patterson, M.T., Rundel, P.W. ( 1993) Carbon isotope discrimination and gas exchange in 
ozone-sensitive and -resistant populations of Jeffrey pine. In Stable Isotopes and 
Plant Carbon-Water Relations. Ehleringer, J.R., Hall, A.E., Farquhar, G.D. (eds). pp. 
2 1 3-242. Academic Press, New York. 
Payne, W.H. ( 1997) New cultivars now released for horticulture. Australian Plants 19 :  
20 1 -202. 
Pelaz, S . ,  Ditta, G.S., Baumann, E., Wisman E.,  Yanofsky, M.F. (2000) B and C floral 
organ identity functions require SEPALLATA MADS-box genes. Nature 405: 
200-203. 
Perrin,Y., Doumas, P. ,  Lardet, L., Carron, M.P. ( 1 997) Endogenous cytokinins as 
biochemical markers of rubber-tree (Hevea brasiliensis) clone rejuvenation. Plant 
Cell Tissue & Organ Culture 47 : 239-245.  
244 
Pharis, R.P., Tomchuk, D., BeaU, F.D., Rauter, R.M., Kiss, G. ( 1 986) Promotion of 
flowering in white spruce (Picea glauca) by gibberellin A, auxin (naphthaleneacetic 
acid), and the adjunct cultural treatments of girdling and Ca(NO) fertilization. 
Canadian Journal of Forest Research 16: 340-345. 
Plummer, J.A., Wann, J.M., Orifici , R., Spadek, Z.E. ( 1 998) Effects of photon flux 
density on photosynthesis, growth, flowering, and oil content in Boronia. 
Australian Journal of Agricultural Research 49: 79 1 -797 . 
Poethig, R.S. ( 1 990) Phase change and the regulation of shoot morphogenesis in plants. 
Science 250: 923-930. 
Poethig, R.S. ( 1 997) Leaf morphogenesis in flowering plants. The Plant Cell 9 :  1 077- 1087. 
Porat, R., Borochov, A., Halevy, A.H., O'Neill, S .D. ( 1 994) Pollination-induced 
senescence of Phalaenopsis petals .  Plant Growth Regulation 15 :  129- 1 36 .  
Poss, J .A.,  Grattan, S .R., Suarez, D.L., Grieve, C.M. (2000) Stable carbon isotope 
discrimination : an indicator of cumulative salinity and boron stress in Eucalyptus 
camaldulensis. Tree Physiology 20: 1 1 2 1 - 1 1 27.  
Potts, B.M.,  Wiltshire, R.J.E. ( 1 997) Eucalypt genetics and genecology. In Eucalypt 
Ecology: Individuals to Ecosystems. Williams, J.E., Woinarski , J.C.Z. (eds). pp. 56-
9 1 .  Cambridge University Press, Cambridge. 
Pritchard, S .G. ,  Peterson, C.M., Prior, S.A., Rogers, H.H. ( 1997) Elevated carbon dioxide 
differentially affects needle ultrastructure and phloem anatomy in Pinus palustris: 
interactions with soil resource availability. Plant, Cell and Environment 20: 461 -47 1 .  
Put, H.M.C., Clerkx, A.C.M. ( 1988) The infiltration ability of microoganisms Bacillus, 
Fusarium, Kluyveromyces and Pseudomonas spp. into xylem vessels of Gerbera 
cv. 'Fleur' and Rosa cv. 'Sonia' cut flowers: a scanning electron microscope 
study. Journal of Applied Bacteriology 64: 5 1 5-530. 
Racchi, M .L., Chiusi, A.P. , Giannini, R. ( 1996) Catalase isozymes as biochemical markers 
of different developmental stages in cypress (Cupressus sempervirens L). Canadian 
Journal of Forestry Research 26: 1 629- 1 633. 
Reboucas, A., Jose, S .  ( 1 997) Effect of girdling treatments on flowering and production of 
mango trees cv. Tommy Atkins. Acta Horticulturae 455: 1 32- 1 34. 
Reeves, P.H.,  Coupland, G. (2000) Response of plant development to environment: 
control of flowering by daylength and temperature. Current Opinion in Plant Biology 
3 :  37-42. 
245 
Reid, M.S. ,  Wu, M. ( 1992) Ethylene and flower senescence. Plant Growth Regulation 
1 1 : 37-43 . 
Reuter, D.J. ,  Robinson, J.B. ( 1 986) Plant Analysis. An Interpretation Manual. Inkata Press, 
Melbourne. 
Richards, D. ( 1 985) Effects of cytokinin application and environment on growth and 
flowering in Boronia heterophylla F. Muell. Scientia Horticulturae 27: 325-334. 
Robbins, W.J. ( 1957) Gibberellic acid and the reversal of adult Hedera to a juvenile state. 
American Journal of Botany 44: 743-746. 
Roberts, N.J., Menary, R.C. ( 1 989) Environmental interaction between daylength, 
night temperature, and photon flux density on growth and flowering in Boronia 
megastigma Nees. Journal of Horticultural Science 64: 597-604. 
Roberts, N.J. ,  Luckman, G.A., Menary, R.C. ( 1993) In vitro flowering of Boronia 
megastigma Nees. and the effect of 6-benzylaminopurine. Journal of Plant 
Growth Regulation 12: 1 17- 1 22 .  
Robinson, L.W., Wareing, P.F. ( 1969) Experiments on the juvenile-adult phase change 
in some woody species. New Phytologist 68: 67-78.  
Robinson, S.A.,  Osmond, C.B. ,  Giles, L. ( 1 993) Interpretations of gradients in delta 
1 3C value in thick photosynthetic tissues of plants with crassulacean acid 
metabolism. Planta 190: 27 1 -276. 
Rogler, C.E.,  Hackett, W.P. ( 1975) Phase chage in Hedera helix: Stabilization of the 
mature form with abscisic acid and growth retardants. Physiologia Plantarum 34: 
1 4 1 - 1 52.  
Ross, S.D., Pharis, R.P. ( 1976) Promotion of flowering in the Pinaceae by 
gibberellins. I. Sexually mature, non-flowering grafts of Douglas-Fir. 
Physiological Plantarum 36: 1 82-1 86. 
Rottmann, W.H., Meilan, R., Sheppard, L.A., Brunner, A.M., Skinner, J .S . ,  Ma CaiPing, 
Cheng ShuPing, Jouanin, L., Pilate, G., Strauss, S.H. (2000) Diverse effects of 
overexpression of LEAFY and PTLF, a poplar (Populus) homolog of 
LEAFY/FLORICAULA, in transgenic poplar and Arabidopsis. Plant Journal 22: 235-
245. 
Rundel, P.W., Esler, K.J., Cowling, R.M. ( 1999) Ecological and phylogenetic patterns 
of carbon isotope discrimination in the winter-rainfall flora of the Richtersveld, 
South Africa. Plant Ecology 1 42:  1 33- 148.  
Sacalis, J .N. ( 1 993) Cut flowers: postproduction care and handling. Second Edition. Bael, 
Batavia, IL. USA. 
246 
Sachs, R.M. ( 1 977) Nutrient diversion: A hypothesis to explain the chemical control of 
flowering. HortScience 12 : 220-222. 
Sachs, R.M., Hackett, W.P. ( 1 983) Source-sink relationships in flowering. In Strategies of 
Plant Reproduction. pp 263-272. Meudt, W.J. (ed). Allanheld, Osmun, Granada. 
Salisbury, F.B. ,  Ross, C.W. ( 1 992) Plant Physiology. Fourth Edition.Wadsworth 
Publishing Co. Belmont, California. 
Salmon, J.T. ( 1 980) The native trees of New Zealand. A.H & A.W. Reed Ltd. ,  Wellington. 
Sandquist, D.R., Schuster, W.S .F., Donovan, L.A., Phillips, S .L., Ehleringer, J.R. ( 1 993) 
Differences in carbon isotope discrimination between seedlings and adults 
of Southwestern Desert perennial plants. Southwestern Naturalist 38:  2 1 2-2 17 .  
Sawada, S . ,  Arakawa, 0. ,  Muraki, 1., Echigo, H., Miyashita, M., Iwafune, M., Kasai, M. ,  
U suda, H. ( 1 999) Photosynthesis with single-rooted Amaranthus leaves. I .  
Changes in the activities of RuBP- 1 ,5-bisphosphate carboxylase and 
phosphoenolpyruvate carboxylase and the amounts of intermediates in photosynthetic 
metabolism in response to changes in the source-sink balance. 
Plant and Cell Physiology 40: 1 143- 1 1 5 1 .  
Schwabe, W.W. ( 1 976) Applied aspects of juvenility and some theoretical considerations. 
Acta Horticulturae 56: 45-56. 
Sedgley, M. ( 1985) Some effects of temperature and light on floral initiation and 
development in Acacia pycnantha. Australian Journal or Plant Physiology 1 2: 
1 09- 1 1 8. 
Sedgley, M. ( 1998) Banksia: new Proteaceous cut flower crop. Horticultural Reviews 22: 
1 -25. 
Serek, M. ( 1 993) Ethephon and silver thiosulfate affect postharvest characteristics of Rosa 
hybrida 'Victory Parade' .  HortScience 28:  199-200. 
Serek, M. ,  Sisler, E.C.,  Reid, M.S. (1 994) Novel gaseous ethylene binding prevents 
ethylene effects in potted flowering plants. Journal of the American Society for 
Horticultural Science 1 1 9: 1 230- 1 233 .  
Serek, M.,  Sisler, E.C. ,  Tirosh, T. ,  Mayak, S.  ( 1995a) 1 -methylcyclopropene prevents 
bud, flower, and leaf abscission of Geraldton Waxflower. HortScience 30: 1 3 1 0. 
Serek, M.,  Sisler, E.C.,  Reid, M.S. ( 1995b) Effects of 1 -MCP on the vase life and 
ethylene response of cut flowers. Plant Growth Regulation 1 6: 93-97. 
Serres, R., McCown, B .  ( 1 993) Rapid flowering ofmicrocultured cranberry plants. 
HortScience 29: 1 59- 1 6 1 .  
247 
Sexton, R., Roberts, J.A. ( 1 982) Cell biology of abscission. Annual Reviews of Plant 
Physiology 33:  1 33- 1 62. 
Shillo, R., Weiner, A., Eini, S. ( 1 98 1 )  Wax Flower as a pot-plant. Hassadeh 6 1 :  1 334-
1336, 1 5 1 8-1 522. 
Shillo, R., Weiner, A., Halevy, A.H. ( 1984) Inhibition imposed by developing flowers on 
further flower bud initiation in Chamelaucium uncinatum Schauer. Planta 1 60 :  
508-5 1 3. 
Shillo, R., Weiner, A., Halevy, A. ( 1985) Environmental and chemical control of 
growth and flowering of Chamelaucium uncinatum Schauer. Scientia 
Horticulturae 25: 287-297. 
Shillo, R., Ronen, A., Mathan, E., Moldavshy, R. ( 1 997) Advancing and spreading out 
flowering of out-door cultivation in the Geraldton Wax Flower by short day 
treatment. Acta Horticulturae 453: 61 -69. 
Shu, Z.H. ,  Sheen, T.F. ( 1987) Floral induction in axillary buds of mango (Mangfera 
indica L.) as affected by temperature. Scientia Horticulturae 3 1 :  8 1 -87. 
Singh, K., Moore, K.G. ( 1 994) Sites of ethylene production in flowers of sweet pea 
(Lathyrus odoratus L.). Scientia Horticulturae 58 :  35 1 -355. 
Sisler, E. C., Serek, M. ( 1 997) Inhibitors of ethylene responses in plants at the receptor 
level: Recent developments. Physiologia Plantarum 1 00: 577-582. 
Sisler, E.C., Serek, M. ( 1999) Compounds controlling the ethylene receptor. Botanical 
Bulletin of Academia Sinica 40: 1 -7 .  
Slater, A.T., Franz, P.R., Thompson, W.K. (1 994) The effects of photoperiod, 
temperature and photon flux density on flowering in Pimelea ciliata. Australian 
Journal of Botany 42: 575-585. 
Smith, R., Greenwood, M.S.  ( 1 995) Effects of gibberellin A417, root pruning and 
cytokinins on seed and pollen cone production in black spruce (Picea mariana). 
Tree Physiology 1 5 :  457-465. 
Snowball, A.M. ( 1989) Morphological and immunological studies on the juvenile-mature 
phase change in citrus. Ph.D. thesis. Faculty of Agriculture, University of 
Sydney, NSW, Australia. 
Snowball, A.M., Halligan, E.A., Mullins, M.G. ( 1 988) Studies on juvenility in citrus.  
In Proceedings of the Sixth International Citrus Congress. March 6- 1 1 ,  1 988 .  
Goren, R . ,  Mendel, K .  (eds). Tel Aviv, Israel, Balaban Publishers. 
248 
Snowball, A.M., Zeman, A.M., Tchan, Y.T., Mullins, M.G., Goodwin, P.B.  ( 1 99 1 )  
Phase change in citrus : immunologically detectable differences between juvenile 
and mature plants. Australian Journal of Plant Physiology 1 8 : 385-396. 
Snowball, A.M., Warrington, I.J., Halligan, E.A., Mullins, M.G. ( 1994) Phase change 
in citrus: The effects of main stem node number, branch habit and paclobutrazol 
applications on flowering in citrus seedlings. Journal of Horticultural Science 69 : 
1 49- 1 60. 
Southerton, S .G., Strauss, S .H., Olive, M.R., Harcourt, R.L., Decroocq, V.  Zhu, X., 
Llewellyn, D.J., Peacock, W.J., Dennis, E.S. ( 1 998) Eucalyptus has a functional 
equivalent of the Arabidopsis floral meristem identity gene LEAFY. Plant Molecular 
Biology 37: 897-9 10. 
Sparks, J .P., Ehleringer, J.R. ( 1997) Leaf carbon isotope discrimination and nitrogen 
content for riparian trees along elevational transects. Oecologia 1 09: 362-367 . 
Sreekantan, L. ,  Mckenzie, M.J., Jameson, P.E., Clemens, J. (2001 )  Cycles of floral and 
vegetative development in Metrosideros excelsa (Myrtaceae). International 
Journal of Plant Science 1 62: 7 1 9-727 . 
Srivastava, P.S. ,  Iqbal, M. ( 1994) Correlative mechanisms and controls of flower 
development. In Growth patterns in vascular plants. Iqbal, M. (ed). pp. 3 1 3-343 . 
Dioscorides Press, Oregon, USA. 
Staby, G.L., Basel, R.M., Reid, M.S. ,  Dodge, L.L. ( 1 993) Efficacies of commercial 
anti-ethylene products for fresh cut flowers. Horttechnology 3 :  199-202. 
Stead, A.D., Moore, K.G. ( 1983) Studies on flower longevity in Digitalis. The role of 
ethylene in corolla abscission. Planta 157:  1 5-21 . 
Stead, A.D., Van Doom, W.G. ( 1 994) Strategies of flower senescence: A review. In 
Molecular and Cellular Aspects of Plant Reproduction. Scott, R.J., Stead, A.D. (eds). 
pp. 2 1 5-237. Cambridge University Press, Cambridge. 
Steele, J.J. ,  Coutts, M.P., Yeoman, M.M. ( 1989) Developmental changes in Sitka 
spruce as indices of physiological age. I. Changes in needle morphology. New 
Phytologist 1 1 3 :  367-375. 
Steeves, T.A., Sussex, I.M. ( 1989) Patterns in plant development. Second edition. 
Cambridge University Press. 
Stein, O.L. ,  Foskett, E.B . ( 1 969) Comparative developmental anatomy of shoots of 
juvenile and adult Hedera helix. American Journal of Botany 56: 546-55 1 .  
249 
Steward, G.R., Tumbull, M.H., Schmidt, S. ,  Erskine, P.D. ( 1995) 1 3C natural abundance in 
plant communities along a rainfall gradient: a biological integrator of water 
availability. Australian Journal of Plant Physiology 22: 5 1 -55. 
Stitt, M., Schulze, E.D. ( 1 994) Plant growth, storage, and resource allocation: from flux 
control in a metabolic chain to the whole-plant level. In Flux control in biological 
systems. Schulze, E.D. (ed). pp. 57-1 1 8 . Academic Press, New York. 
Sun, J., Jameson, P.E., Clemens, J. (2000) Stamen abscission and water balance in 
Metrosideros flowers. Physiologia Plantarum 1 1 0: 27 1 -278. 
Sun, J., Jameson, P.E., Clemens, J. (200 1 )  Water relations and stamen abscission in cut 
flowers of selected Myrtaceae. Acta Horticulturae 543 : 1 85-1 89. 
Sussex, I. ( 1 976) Phase change: physiology and genetic aspects . Acta Horticulturae 56: 
275-280. 
Taiz, L., Zeiger, E. ( 1 998) Plant Physiology. Second edition. Sinauer Associates, 
MA, USA. 
Telfer, A., Poethig, R.S.  ( 1998) Hasty: a gene that regulates the timing of shoot maturation in 
Arabidopsis thaliana. Development 125 :  1 889- 1 898.  
Telfer, A., Bollman, K.M.,  Poethig, R.S.  ( 1997) Phase change and the regulation of trichome 
distribution in Arabidopsis thaliana. Development 124:  645-654. 
Tsai, C.,  Miller, A., Spalding, M., Rodermel, S .  ( 1 997) Source strength regulates an 
early phase transition of tobacco shoot morphogenesis. Plant Physiology 1 1 5 :  
907-9 14. 
Thomas, B . ,  Vince-Prue, D. ( 1 984) Juvenility, photoperiodism and vemalization. In 
Advanced Plant Physiology. Wilkins, M.B . (ed). pp. 408-439. Longman Scientific, 
Essex. 
Thomas, B ., Vince-Prue, D. ( 1 997) Photoperiodism in plants. Second edition. Academic 
Press, San Diego, CA. 
Van Altvorst, A.C., Bovy, A.G. ( 1 995) The role of ethylene in the senescence of 
carnation flowers, a review. Plant Growth Regulation 16: 43-53. 
Van Doom, W.G. ( 1997) Water relations of cut flowers. Horticultural Reviews 1 8 : 1 -85.  
Van Doom, W.G., Woltering, E.J. ( 1 99 1 )  Developments in the use of growth 
regulators for maintenance of post-harvest quality in cut flowers and potted 
plants. Acta Horticulturae 298 :  1 95-209. 
250 
Van Doom, W.G., Stead, A.D. ( 1997) Abscission of flowers and floral parts. Journal 
of Experimental Botany 48 :  821 -837. 
Van Meeteren, U., De Proft, M. ( 1 982) Inhibition of flower bud abscission and ethylene 
evolution by light and silver thiosulphate in Lilium. Physiologia Plantarum 56: 
236-240. 
Veinbrants, N. ( 1 972) Effects of succinic acid-2,2-dimethyl hydrazide (Alar) on scoring 
on growth and flower initiation of young apple trees. Australian Journal of 
Experimental Agronomy and Animal Husbandry 1 2: 89-95. 
Venglat, S .P., Sawhney, V.K. ( 1 996) Benzylaminopurine induces phenocopies of 
floral meristem and organ identity mutants in wild-type Arabidopsis plants. Planta 
198: 480-487. 
Vince-Prue, D. ( 1 985) Photoperiod and hormones. In Encylopedia of Plant Physiology. 
Pharis, R.P., Reid, D.S. (eds). pp. 1 1 :  308-364. Springer, Berlin. 
Visser, T. (1 964) Juvenile phase and growth of apple and pear seedlings. Euphytica 1 3 : 
1 1 9- 129. 
Visser, T. ( 1973) The effect of rootstocks on growth and flowering of apple seedlings. 
Journal of American Society of Horticultural Science 98:  26-28. 
Voesenek, L.A.C.J., B lom, C.W.P.M. ( 1996) Plants and hormones: an ecophysiological 
view on timing and plasticity. Journal of Ecology 84: 1 1 1 - 1 19 .  
Vogel, J.P., Schuerman, P. ,  Woeste, K. ,  Brandstatter, I . ,  Kieber, J.J. ( 1998) Isolation and 
characterization of Arabidopsis mutants defective in the induction of ethylene 
biosynthesis by cytokinin. Genetics 149: 4 17-427. 
V on Stern, K. ( 1 96 1  ). Uber den Erfolg einer uber drei Generationen gefuhrten Auslese 
auf fruhes Bluhen bei Betula verrucosa. Silvae Genetica 10: 48-5 1 .  
Walcroft, A.S ., Whitehead, D. ,  Silvester, W.B., Kelliher, F.M. ( 1997) The response of 
photosynthetic model parameters to temperature and nitrogen concentration in Pinus 
radiata D. Don. Plant, Cell & Environment 20: 1 338 - 1348. 
Wareing, P.F. ( 1 959) Problems of juvenility and flowering in trees. Journal of the Linnean 
Society of London Botany 56: 282-289. 
Wareing, P.F., Frydman, V.M. ( 1976) General aspects of phase change with special 
reference to Hedera helix L. Acta Horticulturae 56:  57-68. 
Way, R.D. ( 197 1 )  Hastening the fruiting of apple seedlings. Journal of American Society 
of Horticultural Science 96: 384-389. 
25 1 
Webster, B.D., Chiu, H.W. ( 1975) Ultrastructural studies of abscission in Phaseolus: 
characteristics of the floral abscission zone. Journal of the Amercian Society for 
Horticultural Science 1 00: 6 13-6 1 8. 
Webster, A.D., Atkinson, C.J., Vaughan, S .J . ,  Lucas, A.S. ( 1 997) Controlling the shoot 
growth and cropping of sweet cherry trees using root pruning or root restriction 
techniques. Acta H orticulturae 45 1 :  643-65 1 .  
Weigel, D. ( 1998) From floral induction to floral shape. Current Opinion in Plant Biology 
1 :  55-59. 
Weigel, D., Meyerowitz, E.M. 1994. The ABCs of floral homeotic genes. Cell 
78: 203-209. 
Weigel, D., Nilsson, 0. ( 1995) A developmental switch sufficient for flower initiation 
in diverse plants. Nature 377: 495-500. 
Weigel, D., Alvarez, J., Smyth, D.R., Yanofshy, M.F., Meyerowitz, E.M. ( 1992) 
LEAFY controls floral meristem identity in Arabidopsis. Cell 69: 843-859. 
Weiss, D. ,  Ohana, 0. ( 1996) Flowering control of Ixodia achillaeoides. Scientia 
Horticulturae 65: 59-64. 
Whitehead, C.S. ,  Vasiljevic, D. ( 1993) Role of short-chain saturated fatty acids in the 
control of ethylene sensitivity in senescing carnation flowers. Physiologia 
Plantarum 88:  243-250. 
Wiebel, J., Chacko, E.K., Downton, W.J .S . ,  Ludders, P. ( 1994) Influence of irradiance 
on photosynthesis, morphology and growth of mangosteen (Garcinia mangostana 
L.) seedlings. Tree Physiology 1 4: 263-274. 
Williamson, V.G., Milbum, J.A. ( 1995) Cavitation events in cut stems kept in water: 
implications for cut flower senescence. Scientia Horticulturae 64: 2 19-232. 
Williams, J.E., Brooker, M.I.H. ( 1997) Eucalypts, an introduction. In Eucalypt Ecology: 
Individuals to Ecosystems. Williams, J.E., Woinarski, J.C.Z. (eds) .  pp. 1 - 1 6. 
Cambridge University Press, Cambridge. 
Wiltshire, R.J.E., Potts, B .M., Reid, J .B.  ( 1 99 1 )  Apaedomorphocline in Eucalyptus. IT. 
Variation in seedling morphology in the E. risdonii/E. tenuiramis complex. 
Australian Journal of Botany 39: 545-566. 
Wiltshire, R.J.E., Potts, B .M., Reid, J .B .  ( 1 998) The genetic control of reproductive 
and vegetative phase change in the Eucalytpus risdonii I E. tenuiramis complex. 
Australian Journal of Botany 46: 45-63. 
252 
Wirthensohn, M.G., Sedgley, M., Ehmer, R. ( 1996) Production and postharvest treatment 
of cut stems of Eucalyptus L. Her. foliage. HortScience 3 1 :  1 007- 1 009. 
Woltering, E.J . ( 1987) Effects of ethylene on ornamental pot plants: a classification. 
Scientia Horticulturae 3 1 :  283-94. 
Woltering, E.J., Van Doom, W.G. ( 1 988)  Role of ethylene in senescence of petals: 
Morphological and taxonomical relationships. Journal of Experimental Botany 
208: 1 605- 16 16. 
Woodson, W.R. ( 199 1 )  Gene expression and flower senescene. In Genetics and Breeding 
in Ornamental Species. Harding, J. ,  Sing. F., Mol, J .N.M. (eds). pp. 3 1 7-33 1 .  Kluwer 
Academic Publishers, Dordrecht, Netherlands. 
Woodson, W.R., Hanchey, S .H.,  Chrisholm, D.N. ( 1985) Role of ethylene in the 
senescence of isolated Hibiscus petals. Plant Physiology 79 : 679-683. 
Woolf, A.B ., Clemens, J. ,  Plummer, J .A. ( 1992) Selective removal of floral buds from 
Camellia with ethephon. HortScience 27: 32-34. 
Woolf, A.B .,  Clemens,  J., Plummer, J .A. ( 1995) Leaf maturity and temperature affect 
the selective removal of floral buds from Camellia with ethephon. Journal of 
American Society of Horticultural Science 1 20: 6 14-621 .  
Woolf, A.B . ,  Plummer, J.A., Clemens, J .  ( 1 999) Maturity and temperature influence 
ethylene-promoted organ abscission in Camellia. New Zealand Journal of Crop 
and Horticultural Science 27: 33-4 1 .  
Wu, M.J., Van Doom, W.G., Reid, M.S. 199 1 .  Variation in the senescence of carnation 
(Dianthus caryophyllus L.) cultivars. I. Comparison of flower life, respiration and 
ethylene biosynthesis. Scientia Horticulturae 48: 99- 107 .  
Wiinsche, J.N., Palmer, J.W., Greer, D.H. (2000) Effects of  crop load on fruiting and 
gas exchange characteristics of 'Braebum' /M.26 apple trees at full canopy. 
Journal of American Society of Horticultural Science 1 25 : 93-99. 
Yang, S .F., Dong, J .G. ( 1 993) Recent progress in research of ethylene biosynthesis. 
Botanical Bulletin of Academia Sinica 34: 89- 1 0 1 .  
Young, T.W. ( 1 977) Effect of branch girdling on yield of severely pruned 'Brewster' 
lychee trees .  Proceedings of the Florida State Horticultural Society 90: 25 1 -253.  
Zaerr, J .B ., Bonnet-Masimbert, M.  ( 1987) Cytokinin level and flowering in Douglas-fir. 
Forest Ecology & Management. 1 9 : 1 15- 1 20. 
253 
Zer, H. ,  Fahn, A. ( 1992) Floral nectaries of Rosmarinus officinalis L. Structure, 
ultrastructure and nectar secretion. Annals of Botany 70: 39 1 -397. 
Zhang, H. ( 1998) Cytokinins and phase change in Pin us radiata: Morphological, 
physiological and molecular studies. Ph.D. thesis. Plant Biology, Massey 
University. New Zealand. 
Zhang, H., Horgan, K.J., Reynolds, P.H.S ., Norris, G.E., Jameson, P.E. (200 1 )  Novel, 
cytokinins: The predominant forms in mature buds of Pinus radiata. Physiologia 
Plantarum 1 1 2:  127- 1 34.  
Zieslin, N. ( 1 985) Leptospermum scoparium. In CRC Handbook of Flowering 
Vol. ill. Halevy, A.H. (ed). pp. 280-282 .  CRC Press, Boca Raton, Florida. 
Zieslin, N. ,  Gottesman, V. ( 1983) Involvement of ethylene in the abscission of flowers 
and petals of Leptospermum scoparium. Physiologia Plantarum 58 :  1 14-1 1 8 .  
Zieslin, N. ,  Moe R. ( 1 985) Rosa. In CRC Handbook of Flowering Vol. IV.  Halevy, A.H. 
(ed). pp. 214-225. CRC Press, Boca Raton, Florida. 
Zieslin, N. ,  Gottesman, V. ( 1 986) Environmental factors involved in growth, flowering and 
post-harvest behaviour of flowers of Leptospermum scoparium. Israel Journal of 
Botany 35: 1 0 1 - 108.  
Zieslin, N. ,  Mor, Y.  ( 1990) Light on roses. A review. Scientia Horticulturae 43 : 1 - 14. 
Zimmerman, R.H. ( 1 972) Juvenility and flowering in woody plants: A review. 
HortScience 7: 447-455 .  
Zimmerman, R.H. ( 1 977) Relation of pear seedling size to length of the juvenile period. 
Journal of American Society of Horticultural Science 102: 443-447. 
Zimmerman, M.H. ( 1983) Xylem Structure and the Ascent of Sap. Springer-Verlag, Berlin. 
Zimmerman, R.H., Hackett, W.P., Pharis, R.P. ( 1985) Hormonal aspects of phase change in 
precocious flowering. In Encyclopedia of Plant Physiology. Pharis ,  R.P., Reid, 
D.M. (eds). pp. 1 1 : 79- 1 1 5 . Springer-Verlag, Berlin .  
254 
Appendix 1 :  Wax Infiltration Procedure 
Day 1 
Wash out fixative with 70% ethanol .  
Day 2 
70% OH TBA (for 2 h) - 300 mls distil led water, 500 mls ethanol (95%) , 200 mls  TBA 
transfer, 
85% OH TBA (for 2 h) - 1 50 mls distil led water, 500 mls ethanol (95%) , 350 mls TBA 
transfer, 
95% OH TBA (for 2 h) - 450 mls ethanol (95%) , 550 mls  TBA 
transfer, 
1 00% OH TBA + Erythron .  Dye (for 2 h) - 750 mls  TBA, 250 mls absol .  ethanol, + pinch 
of Erythrosine dye. 
transfer, 
1 00% TBA overnight. 
Day 3  
At 9am, renew the 1 00% TBA solution, again at 1 and 4 pm and leave overnight. 
Day4  
9-10am TBA/Paraffinol (50:50) 
10 am - midday - 1/z fi l l  new containers with wax chips.  Pour paraffin mix in with 
samples. Place in oven at 60?C. 
12  pm - replace with pure melted wax 
4 pm - replace with pure melted wax 
At least 4 changes of pure wax. Leave over two days. Store in  oven (60?C) .  
Day 7  
Embed samples & section the next day. 
Abbreviations: OH = absolute ethanol ,  TBA = 2-Methylpropan-2-ol tert-Butanol . 
255 
1-:l 
? 
Appendix 11 :  Staining Schedule : Safranin/Fast Green 
Histoclear 
0 
1 0  mins. 
1--
Histoclear Absolute 
Abs.-OH Ethanol 
0 
1--
0 
5 mins. 5 mins. 
Water 
Wash off 
Excess Stain 
0 
Clove Oil l I Clove Oil 
OH:His. 
L::e I n L::ec?. 
-
Absolute 
Ethanol 
0 
5 mins. 
70% 
Ethanol 
G 
5 mins. 
Histoclear 
+ -OH 
0 
5 mins. 
..._ 
95% 
Ethanol 
0 
5 mins. 
0.5% 
Pie. Acid 
0 
10  secs. 
Histoclear 
? 
5 mins. 
-
85% 
Ethanol 
0 
5 mins. ....._____ 
95% 
Ethanol 
0 
Histoclear 
?? 
5 mins. 
1--
' 70% 
Ethanol 
[!] 
5 mins . 
95% 
Ethanol 
liU 
10  secs. 
Mount coverslips 
with DPX onto 
slides (-
20 
?,.,: