NITROGEN RESPONSIVENESS IN +RLBY

by Ghodratollah Fathi (M.Sc. Agronomy) (University of Tarbiat Moddares Iran)

Thesis submitted for the degree of Doctor of Philosophy

Department of Plant Science Waite Agricultural Research Institute The University of Adelaide

July 1994 I

TABLE OF CONTENTS Page

Abstract v Declaration x Acknowledgments xi List of Figures xü List of Tables xiv List of Plates xviü List of Appendices xix

CHAPTER. 1. GENERAL INTRODUCTION L-4

CHAPTER. 2. LITERATURE REVIEV/ 5-43 2.I Introduction 5

2.2 Growth of winter cereals in Meditenanean environments 7 2.2.I Major features of the Mediterranean environment 7 2.2.2 Likely occuffence of water stress in Mediterranean environment 9 2.2.3 Effects of climatic factors on crop growth and yield 1 0 2.2.4 Seasonal availability of N in the Meditenanean-type environments t2 2.2. 5 Interaction between water and N 13 2.2.6 Conclusion I6

2.3 Nitrogen 17 2.3. I Nitrogen in the soil t7 2.3.2 Nitrogen uptake and utilisation by the crop l8 2.3.3 Crop responses to N uptake 20 2.3.3 Response curves 20 2.3.3.2 Effects of N on growth, grain yield and yield components 22 2.3.3.3 Effects of N on grain protein concentration 23 2.3.4 Grain quality and N 24 2.3. 4. 1 Grain protein concentration and malting quality 24 2.3. 4.2 Relationship between grain yield and grain protein concentration 25 2.3. 5 Factors affecting crop responses to N 27 2.3.5.1 Form and timing 28 2.3.5.2 Moisture 30 2. 3. 5.3 Genotype 30 ll

2.3.6 Nitrogen use efficiency indices 31 2. 3.7 Conclusion 33

2.4 Genetic variation in responsiveness to N 34 2.4. L Nitrogen uptake 35 2. 4.2 Nitrogen remobilisation 36 2.5 Conclusion 4I

CHAPTER. 3. NITROGEN RESPONSTVENESS OF BARLEY CULTTVARS 44-83

3.1 Introduction 44 3.2 Materials and Methods 45 3.2. I Sites 45 3.2.2 Treatments and experitnental design 45 3.2.3 Establishment of the experiment 45 3.2.4 Measurements 46 3.2.4.1 Dry matter measurements 46 3. 4. 2. 2 Nitrogen analysis 48 3. 4.2.3 Statistical analysis 48 3.2. 5 Response index (RI) 49

3.3 Results 50 3.3.I Weather 50 3.3.2 Level of production in the different experiments 50 3.3.3 Average responses to N fertiliser at each site 50 3.3. 4 Varietal responses to N 53 3.3.4. 1 Dry matter at 10 weeks (DMro) 53 3.3. 4.2 Dry matter at anthesis 53 3. 3. 4. 3 Dry matter at maturity 53 3.3.4.4 Grain yield 59 3.3. 4.5 Yield components 63 3.3. 4.6 Grain nitrogen concentration 70 3.3.4.7 Response index 74 3. 3. 4.8 Nitrogen harvest index 74 3.3. 4.9 Grain N concentration and total shoot N at anthesis 74

3.4 Discussion 76

CHAPTER. 4. DIFFERENCES IN NruRATE UPTAKE AND ASSIMILATION AMONG BARLEY CULTTVARS 84-TL7 4. I Introduction 84

4.2 Preliminary investigations on nitrate uptake 84 üi'

4.2. I Materials and Methods 84 4.2.1.1 Experiment 1: Short term uptake of nitrate 84 4.2. I.2 Experiment 2: Response to different concentrations of nitrate 87

4.3 Results 89 4.3. I Experiment 1 89 4.3.2 Experiment 2 93 4.4 Discussion 94 4. 5 Effect of genotypes on nitrate uptake and assimilation in barley cultivars 96 4.5.1 Materials and Methods 96 4.5.1.1 Plant material 97 4.5. L.2 Growth conditions 97 4.5. I.3 Measurements 99 4.5. I.4 Data analysis 99 4. 5.2 Result r01 4. 5.2. I Experiment 3 101 4.5.3.2 Experiment 4 106 4. 5.3.3 Experiment 5 109 4.6 Discussion 113 4.7 General conclusions tt6

CHAPTER. 5. REMOBILISATION OF DRY MATTER AND N DURING GRAIN FILLING IN BARLEY

CULTTVARS 1 18-148

5. 1 Introduction 118 5.2 Materials and Methods 119 5.2.I Plant culture 119 5.2.2 Treatments il9 5.2.2. 1 Nitrogen 119 5.2.2.2'Water 120 5. 2.2.3 Plant harvest t20 5.3 Results r20 5. 3. 1 Growth at 10 days after anthesis (Hr) r20 5. 3. 1. 1 Dry matter production 120 5.3. 1.2 Grain dry weight t2l 5. 3. 1. 3 Tiller number t2r 5.3. 1.4 Plant N t2l 5.3.2 Growth at maturity (H2) r23 5.3.2. I Dry matter production r23 lv

5.3.2.2 Grain yield 126 5. 3. 2. 3 Yield components r27 5.3.2. 4 Plant N at maturity (H2) r29 5.3.3 Dry matter and N remobilisation r33 5.3.3. 1 Root dry matter and N remobilisation 133 5.3.3.2 Shoot dry matter and N remobilisation 136 5.3. 4 Contribution of shoot dry matter remobilisation to grain yield and N remobilisation to grain N yield 136 5.4 Discussion r42

CHAPTER. 6. GENETIC VARIATION IN THE RESPONSE OF BARLEY CULTWARS TO NITROGEN r49-r82 6.I Introduction 149 6.2 Materials and Methods 149 6.2.I Experiment 1: Northfield t49 6.2. I. 1 Measurements t49 6.2.2 Experiment 2: Charlick 151 6.2.2. I Measurements 151 6. 2. 2. 2 Measures of responsiveness t52 6. 2. 2. 3 Statistical analysis t52 6.3 Results r53 6.3.1 Weather 153 6.3.2 Level of production r54 6.3.3 General responses to N r54 6.3. 4 Relationship between grain yield and GPC r54 6.3. 5 Correlation between grain yield and other factors 156 6. 3. 3 Principal component analysis r62 6.3.3. 1 Northfield 1991 r62 6.3.3.2 Charlick 1993 168 6. 3. 7 Responsive indices t73 6. 3. 7.1 Correlation between principal components and responses indices 173 6.4 Discussion r79

CHAPTER. 7. GENERAL DISCUSSION 183-190't c|. APPENDICES rsl'-20Å'.? .,'Ì REFERENCES 209-226 ,,f;ril'- ¡ - v ABSTRACT

South Australia is the major barley producer in Australia but only 3O-4OVo is of malting quality. Farmers are being encouraged to apply nitrogen (N) fertiliser to barley to improve yields. However, the South Australian environment is characterised by the frequent occurrence of postanthesis water deficit which can increase grain nitrogen concentration

(GNC), especially when high rates of N are used. High GNC is not desirable for malting quality. It is important therefore, to develop varieties which have a high grain yield response but a low GNC response to N fertiliser to improve the chance of predicting malting quality when N fertiliser is applied. A number of studies were conducted to examine responses to N in the field, genetic variation in nitrate uptake and post-anthesis mobilisation of dry matter and N.

Initial studies of the responses of malting barley cultivars to N application were conducted at

4 sites over 2 years. Six cultivars (Clipper, Stirling, Weeah, Schooner, Chebec, SkifÐ

were grown at eight rates of N (0-105 kgN/ha, applied in increments of 15 kgN/ha).

Responses to N varied between sites, however consistent genetic differences between barley

cultivars were found. The semidwarf cultivar Skiff was the most responsive cultivar, while

the tall cultivar Weeah tended to be the least responsive cultivar; the other cultivars were

intermediate in their response. In general, the yield responses to N were correlated

positively with dry matter at anthesis, ears/m2 and kernels/m2, although this relationship

varied between cultivars. Addition of N fertiliser often resulted in smaller grain size,

although there were differences between cultivars. The kernel weights of Skiff and Stirling

were lower and more variable those that of the other cultivars. The experiment indicated that

although a semidwarf cultivar like Skiff may show a greater yield response to N, the

consistency in malting quality may be lower because of its more variable kernel weight.

Yield and GNC responses to N at these 4 sites varied considerably between yeafs, with the

greatest yield response (and the lowest GNC response) occurring in the year with the wettest

spring. GNC was increased at the highest rates of N, however the experiments showed that vl it is feasible to add moderate rates of N (<45 kgN/ha), to improve yields without greatly increasing GNC.

Nitrate concentrations in the soil at the start of the growing season are high because of mineralisation of organic N during autumn and the addition of N fertiliser. It may be useful to exploit this N as much as possible. Therefore, nitrate uptake, assimilation and dry matter production among a range of cultivars were examined in a series of experiments utilising a hydroponic system. Seedlings were gro,wn at rates of nitrate which ranged from 0.25mM to

1.OmM and the experiments lasted for between 20d and 26d. Significant genetic differences in growth and nitrate uptake were identified. The cultivars Skiff and Franklin consistently produced large seedlings which took up large quantities of nitrate from solution whereas

Stirling, Schooner and Triumph produced small seedlings and took up small amounts of nitrate. However, apart from differences based on seedling vigour, there was some evidence that a group of cultivars which had the Victorian cultiva¡ Research as a common parent was more efficient physiologically in assimilating nitrate. For comparable amounts of nitrate taken up from solution, total dry matter production in this group of cultivars was consistently greater than the other cultivars examined. Results from this work established that genetic differences in nitrate uptake exist between cultivars which in most cases were related to the size of the plant, especially the root system. The field studies also showed same differences in early growth between cultivars which were consistent with the differences in nitrate uptake in the hydroponic study. However, the importance of greater nitrate uptake by the seedling (as demonstrated by glasshouse hydroponic experiment) and consequently of early growth, to grain yield was not clearly established because early vigour was not always beneficial to yield.

Most of the N in the grain at maturity is derived from N mobilised from green tissue during grain filling. Studies on postanthesis N remobilisation were conducted in a glasshouse experiment with 2 rates of N (equivalent to 50 and 100 kgN/ha) with and without postanthesis water stress. Stress induced by applying the half of quantity of water supplied to the well watered pots at each watering and harvests were made 10d after anthesis and at maturity. Results from this experiment were generally similar to the field: Skiff had the vll highest response to N at both levels of water stress, V/eeah showed no response to N and

Clipper and Chebec were intermediate. Skiff had a high tiller and ear production while in

Weeah tiller and ear number were low. Kernel weight in Skiff, Chebec and Stirling were reduced by low moisture and N while kernel weight in V/eeah was not. Weeah had low shoot dry matter remobilisation during grain filling, while shoot dry matter remobilisation in

Stirling was high compared with the other cultivars. Nitrogen remobilisation also differed significantly between cultivars, with Skiff showing a high N remobilisation and Chebec a low N remobilisation. Under well watered conditions, the differences in yield responsiveness to N between cultivars was positively correlated with differences in shoot dry matter remobilisation. However, there was no correlation under postanthesis stress. N remobilisation increased with the addition of fertiliser, but this increase was not associated with an increase in the grain N yield. Despite the significant differences between cultivars in remobilisation of dry matter and N in this study, these were not strongly associated with the response in grain yield or grain N yield. Although this study showed there is variation in kernel weight and kernel weight stability between cultivars, this was not related to the ability to remobilise dry matter from the vegetative plant parts: Skiff, which had a variable kernel weight also had a large response in the amount of dry matter remobilised while Weeah remobilised little dry matter and had a less variable kernel weight. Similarly, postanthesis

stress strongly affected the responses of GNC to N and caused an increase in GNC, but differences in remobilisation of N were not associated with GNC responsiveness of the

different cultivars.

This experiment also showed that when the supply of N was low, postanthesis water stress

reduced yields less than at the higher N rate, and also it did not affect kernel weight or GNC

as much. The result may help explain the conservative use of N fertiliser by farmers:

although when postanthesis water stress was low there were large yield responses to N and

GNC was low, when grown under stressful conditions the potential losses in malting

quality from reduced grain size and increased GNC were much lower.

Further experiments to examine a wider range of cultivars were conducted at 2 sites

(Northfield and Charlick) in 2 years. Seventy eight cultivars were grown at 0 and 50 vlll kgN/ha (Northfreld) or 0 and 45 kgN/ha (Charlick). Measurements of agronomically useful characteristics were made during year and the data analysed using principal component analysis (PCA). There was a large response in grain yield to N at Charlick but little response at Northfield. The yields responses of the cultivars at the 2 sites were not correlated, which reflected different environmental limitations to growth and yielcl. For example, boron toxicity was evident at Northfield but not at Cha¡lick. Grain yield responses and GNC responses to N fertiliser rwere independent at both sites, indicating that it should be possible to select for high yield and low GNC responses (as well as high yield and high

GNC response). PCA showed that date of flowering and height were the 2 most important components relating to grain yield. At both sites, the grain yield response to applied N was correlated negatively with the amount of growth and height of the cultivar when no N was applied; that is, tall cultivars which produced more dry matter at 0 kgN/ha tended to be less responsive. At both sites GNC responses were not significantly correlated with any of the principal components. The results of this field experiment showed a strong environmental effect on responsiveness, suggesting that screening should be done in the target areas in order to identify and develop genotypes that are highly and consistently responsive to N, particularly in environments where available moisture affect postanthesis is variable. The results of the Northfield experiment also indicated that cultivars which showed greater symptoms of boron toxicity when no N was applied tended to be more responsive to N.

Boron uptake is affected by N, so adding N at Northfield may therefore have been associated with alleviating boron toxicity especially in cultivars with greater sensitivity to high soil boron. This result, which showed the possible importance of boron toxicity in influencing response to N, highlights that further investigations of the interactions between

N and other nutrients is needed.

The experiments conducted in this thesis have established that genetic differences exist in the response of barley cultivars to the application of N fertiliser, although the differences

between cultivars in grain yield and GNC responses at different experimental sites

demonstrated the large influence of seasonal variability on responses to N; this affects

assessment of genetic variability in f,reld experiments. The importance of postanthesis stress

to GNC and yield suggest that other management practices, such as sowing time and rate of rx sowing, which influence the level of postanthesis stress, are necessary to help discriminate between cultivars. The results of the initial field experiments showed that reliable genetic differences occurred between the semidwarf cultivar Skiff and the tall cultivar Weeah. The importance of height to responsiveness was confirmed when a larger number of cultivars was examined at 2 sites. However, the results with Skiff also suggested that a reduction in height may also adversely affect quality, although more extensive experimentation is required to confirm this. Therefore the importance of height to yield responsiveness to N

and to quality of malting barley needs to be further investigated. However, it is probable that a semidwarf cultivar should be able to be selected with larger and more stable grain size.

Growing a large number of lines or cultivars at 2 N levels in the field appears to be a feasible method of screening for responses of yield and GNC to N. However, because of the large

G x E effect, it is necessary to conduct such work in the target environment. The use of

hydroponic studies in addition to the f,reld evaluation may also provide useful information on

the responsiveness of cultivars to N. x

DECLARATION

I hereby declare that the thesis presented here contains no work which has been accepted for the award of any other degree or diploma in any university or other tertiary institution and, to the best of my knowledge and belief, contains no material previously published or written by another person, except where due reference has been made in the text.

I give consent to this copy of my thesis, when deposited in the University Library, being available for loan and photocopying.

Fathi XI

ACKNOWLEDGMENTS

I wish to express my sincere thanks and deep sense of gratitude to my supervisors Dr G K

McDonald (Main supervisor) and Dr R Lance (Co-supervisor) for their support, constant encouragement, excellent advice, guidance, patience and constructive criticism over the period of experimental work and the preparation of this thesis. It was a great opportunity for me to work with and learn much from them.

I would like to thank Dr Cha¡les Oti-Boateng for his help during experiments and writing of the thesis and Ms Lynne Giles and Mr Emiel Storken with assistance on computer software and statistical advice.

I wish to gratefully acknowledge Mr Anthony Smith, Mr Geoff Dean and Peter Cassidy for their help, assistance and friendship,

I would like to express my appreciation to Mr Barry Felberg and Ms Vada Habner for help in providing research materials and facilities, Mrs Ruth Ellickson and Mrs Helen Taylor for their suggestions during my preparation of this thesis and the photographer Ms Jennie

Groom for her help in taking and processing colour photographs and slides.

My thanks are due to the Head, students and staff in the Departments of Plant Science,

Waite Campus for their encouragement and companionship.

I wish to gratefully acknowledge the generosity of Ministry of Culture and Higher

Education of the Islamic Republic of kan for awarding a scholarship to allow me to pursue my studies at the University of Adelaide.

Finally, I wish to express my deepest gratitude to my wife and my sons, Mehdi and

Mohsen, for their faith and patience during our stay in Adelaide. xll LIST OF FIGURES

Figure 2.1. Relationship between malt extract, diastase activity and grain protein concentration 6

Figure 2.2 Stages in the assimilation of nitrate by plants 19

Figure 2.3. Examples of responsc curvcs of plants to applied N fertiliser 2l

Figure 2.4. General relationship between grain yield, grain protein concentration and N application for rainfed wheat 26

Figure 3.1. Average response to N fertiliser in (a) dry matter at anthesis, (b) dry matter at maturity, (c) grain yield and (d) grain N concentration at 4 sites (Northfield 1990, Nuriootpa 1990, Northfield 1991 and Charlick 1991) 52

Figure 3.2. The effect of N fertiliser on the dry matter production after ten weeks in six cultivars of barley at Northfield in 1991 and Charlick in 1991 55

Figure 3.3. The effect of N rate on the dry matter production at anthesis of six cultivars of barley at Northfield in 1990, Northfield in 1991 and Charlick in 1991 57

Figure 3.4. The grain yield responses of six cultivars of barley to N fertiliser at Northfield, 1990, Northheld, 199I, and Charlick, 1991 60

Figure 3.5. The relationship between grain yield and dry matter production at anthesis for six cultivars of barley at three sites in 1990 and 199 1 . The sites are Northfield in 1990, Northfield in 1991 and Charlick in 1991 62

Figure 3.6. The relationship between grain yield and ea¡ number for six cultivars of barley at three sites in 1990 and 1991. The sites are Northfield in 1990, Northfield in 1991 and Charlick in 1991 66

Figure 3.7. The relationship between dry matter production at anthesis and the number of kernels/m2 for six cultivars of barley at Northfield in 1990, Northfield in 1991 and Charlick in 1991 67

Figure 3.8 The relationship between the number of kernels per m2 and grain yield for six cultivars of barley at Northfield in 1990, Northfield in 1991 and Charlick in 1991 68

Figure 3.9. The effect of N rate on the kernel weight of six cultivars of barley grown at Northfield in 1990, Northfield in 1991 and Charlick in 1991 69

Figure 3.10. The effect of N fertiliser rate on the grain N concentration in srx cultivars of barley at Northfield in 1990, Northfield in 1991 and Cha¡lick in 1991 72

Figure 3.1 1. The relationship between grain N concentration and grain yield for six cultivari of barley at Northfield in 1990, Northfield in 1991 and Cha¡lick in 1991 73

Figure 3.12. The relationship between the total amount of N in the shoot xlll

at anthesis and the grain N concentration for six cultiva¡s of barley at Northfield in 1990, Northfield in 1991 and Charlick in 1991 77

Figure 4.1 Relationship between total plant dry weight and (a) nitrate uptake and (b) nitrate accumulation for 6 barley cultivars 92

Figure 4.2. Interaction between nitrate levels and cultivar in (a) Plant dry weight and (b) nitrate accumulation and (c) relationship between nitrate accumulation and total plant dry weight for 4 barley cultivars 95

Figure 4.3. Experiment 3: Relationship between increase in dry weight and nitrate uptake over 6 days (a), and (b) total dry weight and total nitrate uptake over 26 days for l0 barley cultivars ro4

Figure 4.4. Relationship between nitrate uptake and nitrate assimilation efficiency in Experiment 3 (a), 4 (b), 5 (c) for barley cultivars 105

Figure 4.5. Experiment 4: Relationship between increase in dry weight and nitrate uptake over 6 days (a), and (b) total dry weight and total nitrate uptake over26 days for 9 barley cultivars 108

Figure 4.6. Experiment 5: Relationship between increase in dry weight and nitrate over 6 days (a), and (b) total dry weight and total nitrate uptake over 26 days for 8 barley cultivars T12

Figure 5.1. Relationship between shoot dry matter remobilisation and grain yield, N remobilisation and grain N yield in well watered and water stressed conditions for 6 barley cultivars 140

Figure 5.2. Relationship between increase in grain yield and shoot dry matter remobilisation, increase N remobilisation and grain N yield in well watered and water stressed conditions of six barley cultivars t4t

Figure 6.1 The correlations between Charlick and Northfield for (a) absolute yield response and (b) relative yield response 159

Figure 6.2. The relationship between grain yield and grain protein concentration in mean (a), absolute (b) and relative response (c) at Northfield 1991 and Charlick 1993 (d, e, Ð for 78 barley cultivars 160

Figure 6.3. The correlation between PC1,PC2 and grain yield increase at Northfield and Charlick r67

Figure 6.4. The average values of RI2 and NEI for 78 barley cultivars at 2 sites t74

Figure 6.5 The correlation between PC1 and PC2 and R12, NEI at Northheld t77

Figure 6.6. The correlation between PC1, PC2 and R12 and NEI at Cha¡lick 178 xIV LIST OF TABLES

Table 1.1 Grain protein concentrations of standard malt samples of the Australian Barley Board 1

Table 3.1. Characteristics of the sites used for experiments 46

Table 3.2. Meteorological data at four sites 1990 and 1991 47

Table 3.3. Average values of production of some factors at the four sites 51

Table 3.4. Summaries of analysis of variance for responses to N in six cultivars of barley 51

Table 3.5. Summaries of analysis of variance of dry matter and tillers/m2 at 10 weeks (DM1g, TNlg), at anthesis (DMa, TNa) and dry matter at maturity (DMnt 54

Table 3.6 Dry matter response at anthesis to N of six barley cultivars in 1990 and 1991 as estimated by quadratic regression 56

Table 3.7. The main effect of Cultivar on dry matter at maturity, earslmz, kernels/m2, kernel weight, grain N concentration and grain yield at four sites 58

Table 3.8. Summary of analysis of variance for grain yield and yield componentat 3 sites 6I

Table 3.9. Grain yield response to N fertiliser rate of barley cultivars in 1990 and 1991 as estimated by quadratic regression 6I

Table 3.10. The simple linear correlation coeff,rcients between grain yield, yield components and grain N concentration 65

Table 3.1 1. The intercept and slope of the regression between kernel weight and N rate at three sites 70

Table 3.12. Summary of analysis of variance for grain N concentration, grain Nyield and N harvest index of six barley cultivars at 3 sites 7I

Table 3.13. Relationship between grain N concentration and N rate for six barley cultivars at 3 sites 7l

Table 3.14. Mean responsiveness index at 3 sites 74

Table 3.15. The main effects of Cultivar and N on NHI at different sites 75

Table 3.16. The intercept and slope of the correlation between grain N concentration and total shoot N at anthesis for Northfield 1990 and Northfield 1991 75

Table 4.1 Compositions of solutions used for hydroponic studies of nitrate uptake and assimilation 86

Table 4.2. Root, shoot dry and total plant dry weight and relative growth rate of eight cultivars of barley grown in the presence of a lmM nitrate solution. Plants were 20 days old and harvest two (Hz) occurred 34h. after harvest one (Ht) 89 XV

Table 4.3. Total nitrate uptake, accumulation and assimilation of barley cultivars 9l

Table 4.4. Growth responses of four barley cultivars to different levels of nitrate 93

Table 4.5. The barley cultivars used in the studies of nitrate uptake 98

Table 4.6. Experiment 3: Effect of lmM nitrate supplied hydroponically on root, shoot and total dry matter production and growth rate of ten cultivars of barley r02

Table 4.7. Experiment 3: Total nitrate content, increase in nitrate uptake nitrate uptake per root dry weight, and indices of N use efficiency of 10 cultivars of barley grown with lmM nitrate 103

Table 4.8. Experiment 4: Effect of lmM nitrate supplied hydroponically on root, shoot and total dry matter production and growth rate of nine cultivars of barley 106

Table 4.9. Experiment 4: Total nitrate content, increase nitrate uptake and nitrate uptake per root dry weight 109

Table 4.10. Experiment 5: Effect of lmM nitrate supplied hydroponically on root, shoot and total dry matter production and growth rate of eight cultivars of barley 110

Table 4.11. Experiment 5: Total nitrate content, increase in nitrate uptake and nitrate uptake per root dry weight IL3

Table 4.12. Interaction between Cultiva¡ and N in plant dry weight at 10 weeks after sowing at 0 and 45 kgN/ha in f,reld experiment (Chapter 3) 116

Table 5.1. The influence of two levels of N on root, shoot dry weight, grain dry weigh¡ and tiller number of six barley cultiva¡s 10 days after anthesis 122

Table 5.2. The influence of two levels of N on root N concentration, root N content,shoot N concentration, shoot N content and GNC of six barley cultivars 10 days after anthesis 124

Table 5.3. The influence of two levels of N on root and shoot dry weight of six barley cultivars at maturity r23

Table 5.4. The effect of N and water stress on root and shoot dry weight of six barley cultiva¡s at maturity 125

Table 5.5. The summary of analyses of variance of grain yield and yield components t26

Table 5.6. The influence of N and water on grain yield of six barley cultivars at maturity t26

Table 5.7 The influence of two levels of N on ear number per pot of six barley cultivars at maturity r27

Table 5.8. The effect of N and water stress on kernel number per pot of six barley cultivars at maturity t28

Table 5.9 The effect of N and water stress on kernel weight xvl

of six barley cultivars at maturity t28

Table 5.10. The influence of two levels of N on root N concentration, root N content, shoot N concentration and shoot N content of six barley cultivars at maturity 130

Table 5.11 The effect of N and water stress on root N content of six barley cultivars at maturity 131

Table 5.12. The effect of N and water stress on shoot N concentration of six barley cultivars at maturity 131

Table 5.13 The effect of N and water stress on shoot N content (grain + straw) of six barley cultivars at maturity r32

Table 5.14. The effect of two levels of N and water stress on GNC of six cultivars of barley at maturity 134

Table 5.15. The effect of two levels of N and water stress on grain N yield of six cultivars of barley at maturity 134

Table 5.16. The effect of two levels of N and water stress on dry matter remobilisation from the root of six cultivars of barley 135

Table 5.17. The effect of two levels of N and water stress on root N remobilisation of six cultivars of barley 135

Table 5.18. The effect of two levels of N and water stress on shoot dry matter remobilisation of six cultiva¡s of barley 137

Table 5.19. The effect of two levels of N and water stress on remobilisation of N in six cultivars of barley at maturity 138

Table 5.20. The effect of cultivar on dry matter and N remobilisation of six barley cultiva¡. Negative value is mean amount of dry mater remobilised and positive sign is mean no remobilisation of dry matter 138

Table 5.21. The effect of two levels of N and water stress on the contribution of shoot dry matter remobilisation to grain yield of six cultivars of barley at maturity t39

TabIe 5.22. The effect of two levels of N and water stress on the contribution of N remobilisation to grain N yield of six cultivars of barley at maturity t39

Table 5.23. The relative contribution of shoot dry matter remobilisation to grain yield and N remobilisation to grain N yield during grain f,rlling in six barley cultivars 142

Table 6.1. The monthly distribution of rainfall at Northfield in 1991 and Charlick in 1993 157

Table 6.2. Average values of production and measurements of some factors at two sites r57

Table 6.3. Summary of analyses of variance of data at Northfield in 1991 158

Table 6.4. Summary of analyses of variance of data at Cha¡lick in 1993 158 xvll

Table 6.5 Simple linear correlations between measured attributes and grain yield at Northfield 1991 161

Table 6.6 Simple linear correlations between measured attributes and grain yield at Cha¡lick in 1993 161

Table 6.7 Eigenvectors of eight principal components from Principal Component Analysis of some attributes of barley cultivars grown with 0 kgN/ha at Northfield 1991 t63

Table 6.8 Eigenvectors of eight principal components from Principal Component Analysis of some attributes of barley cultivars grown with 50 kgN/ha at Northfield 1991 r64

Table 6.9 Eigenvectors of eight principal components from Principal Component Analysis of the increase in the values of some attributes of barley cultivars when 50 kgN/ha was applied at Northfield 1991 r66

Table 6,10. The intercept and co9{lcçry9f regressio-n_analysis of responses in grain yield, GPC and grain N yield against PC at Northfield in 1991 r68

Table 6.11 Eigenvectors of seven principal components from Principal Component Analysis of some attributes of barley cultivars grown with 0 kgN/ha at Charlick 1993 r69

Table 6.12. Eigenvectors of seven principal components from Principal component analysis of some attributes of barley cultivars grown with 45 kgN/ha at Cha¡lick 1993 r70

Table 6.13. Eigenvectors of seven principal components from Principal Component Analysis of the increase in the value of some attributes of barley cultivars when 45 kgN/ha was applied at Cha¡lick 1993 t72

ang regres sio_n_ analy-si s Table 6.14. The interc ept coefficigtt _of of responses in grain yield, GPC and grain N yield against PC at Charlick 1993 t7l

Table 6.15. Summary of analysis of va¡iance and averages of three responsive indices at two sites r73

Table 6.16. The intercept and regression coefficient of regression analysis of Principal Component Analysis with response indices at 2 sites 176 xvur

LIST OF PLATES Plate 3.1 Effect of N rate on the Plant height of Skiff, Stirling, V/eeah and Schooner at Northfield in 1991. Yields of Skiff and Stirling were responsive to N while Weeah and Schooner were not responsive 79

Plate 4.1 The hydroponic system which was used in the glasshouse to identify differences in nitrate uptake between barley cultiva¡s for Experiment I and 2 88

Plate 4.2 Method used to examine differences between genotypes in nitrate uptake and assimilation in barley cultivars in Experiments 3,4 and 5 100

Plate 6.1 A general view of the field experiment at Charlick in 1991 showing responses to N 155 xlx

LIST OF APPENDICES

Appendix 3.1 Measurement of total N concentration using the Kjeldahl method 191

Appendix Table 3.1. Origin and cha¡acteristics of 6 barley cultivars used in field experiments 191

Appendix Table3.2. The residual mean squares of models chosen to fit dry matter at anthesis against N 192

Appendix Table 3.3. The residual mean squares of models chosen to frt grain yield against N 192

Appendix Table 3.4 The residual mean squares (x10-3¡ of models chosen for fitted grain N concentration against N r93

Appendix Table 3.5. Coefficient of the fitted equations for the relationships between kernel number and dry matter at anthesis, and between kernel weight and kernel number at Northfield 1990, l99l,Charlick l99l t93

Appendix Table 3.6. Coefficient of fitted equations for the relationships between yield components at Northfield 1990, 1991 194

Appendix Table 3.7. Coefficient of the fitted equations for kernel weight and kernel number against N at Northfield 1990, 1991, Cha¡lick 1991 195

Appendix Table 3.8. Coefficient of fitted equations for the relationships between yield component against N at Charlick 1991 196

Appendix Table 3.9 Mean of parameters measurement at Northfield in 1990 r97

Appendix Table 3.10. Mean of parameters measurement at Northfield in 1991 198

Appendix Table 3.11. Mean of parameters measurement at Charlick in 1991 199

Appendix 4.1. Nitrate determination in plant tissue 201

Appendix Table 5.1. Balance between root and shoot total N of 6 barley cultivars 202

Appendix 6.1. Waite Accessions and pedigrees of lines used in the N response trials at Northf,ield in 1991 and at Charlick in 1993 203

Appendix Table 6.2a. Simple linear correlations between some parameters of barley grown at 0 kgN/ha atNorthfield 1991 204

Appendix Table 6.2b Simple linear correlations between some parameters of barley grown at 50 kgN/ha + xx

at Northfield 1991 205

Appendix Table 6.3a. Simple linear correlations between some of barley grown at 0 kgN/ha parameters at Charlick 1993 206

Appendix Table 6.3b. Simple linear correlations between some of barley grown at 45 kgN/ha parameters at Charlick 1993 207

Appendix Figure 3.1. The interaction between cultiva¡s and N for N harvest index of 6 barley cultivars at Northfield 1991 200 I

T CHAPTER f.; l|_.s'' I T'll l':tJi.,','.Ìi rr : I :]: i! : . :;t¡- .:...;1;. .i ¡ I j,".¡ GENERAL INTRODUCTI ":.ìt. i::r:t r"-r. A? :.,j.;,.... ,..l !li;;i.1:.:'i;-;,',..ii l L'i 1

Barley has been a major crop in southern Australia, and South Australia in particular, since the early days of settlement. South Australia is the major producer of barley in Australia, and the during the 5 year period, 1985-1990, the average area sown to barley for grain in

South Ausralia was 954,000 ha. The average production was 1,471,000t and the average grain yield was 1.54 t/ha (ABARE 1990). Barley has two major end uses- malting and feed.

Malting barley attracts significant premiums on both the local and export markets however, only about 30-407o of the barley produced in South Australia is of malting quality although approximately half of the atea sown to barley is sown to malting varieties. One of the imporønt quality parameters of malting barley is a low petcentage of protein in the grain, but in recent years there has been an upward trend in the grain protein percentage in the

Austalian Barley Board's standard samples (Table 1.1).

Table 1.1. Grain protein concentrations of standard malt samples of the Australian Barley Board (Data courtesy of the Australian Barley Board)

Grade Season Protein (7o)

Clipper 1983184 LO.2 malt 1984/85 t0.7

Schooner 1985/86 10.3 malt 1986187 10.6 1987/88 10.4 r988/89 11.0

The cereal belt of South Australia has a Mediteranean environment which presents special problems in breeding for malting quality. Studies have been done th¡oughout the world in

many different environments on the effects of agronomic factors on the malting quality of

barley, but in South Australia few such studies have been carried out. Barley is gtown as a

dryland crop in South Australia, but yields are often low in comparison with wheat. An

important reason for this is the place of barley in the rotation and the attendant problems of

poor N nutrition. In Australia, barley is generally included in two types of rotations; either 2 as a stubble-sown, second, or occasionally third, crop in a long totation or as a first crop after a legume pasture ley in a short rotation. Fallow often precedes the first crop in the long rotation which is generally wheat, but barley is only rarely sown on fallow land. In these rotations yield of barley can be limited by N deficiency.

Nitrogen deficiency is becoming a important factor limiting both grain yield and grain protein in cereals gfown in South Australia as the intensity of cropping increases and soil fertility, in general, declines (McDonald 1989, Reuter 1989, Xu et al.; L99L, L992). Larger amounts of

N a¡e being applied to barley as large increases in yield have been shown (Xa et al. l99l).

In wheat the responses to applications of N fertiliser however, valies with environmental conditions and freld management practices (Russell 1968a, 1968b) and similar variability can be expected with barley. The climate (especially rainfall) and soil type are the major environmental variables affecting the responses to N fertiliser under dryland farming systems in South Australia. If seasonal conditions are unfavourable for large yield responses, grain protein will increase, often to levels unacceptable for malting quality. As well, kernel weight and grain size may be reduced if the crops mature under high levels of moisture stress, a situation that may occur if the rate of N is too high.

High protein in malting barley has been associated with several major processing problems during malting and brewing. High-protein barley requires longer steeping times, and its erratic germination is difficult to control (Burger and LaBerge 1985). The total N content of the grain is also of some importance; this is determined by the environment in which the crop is grown. The malting premium has emphasised the need for an improved understanding of the management options that can increase the probability of farmers achieving malting grade for their grain. Apart from the effects of seasonal weather on grain N, agronomic practices will have a major role in determining whether a crop of barley achieves malting quality (Smith 1990). The applicarion of N fertiliser (both the rate and timing) is one such agronomic practice.

Sometimes farmers in South Ausralia who wish to grow malting badey do not apply high rates of nitrogenous fertiliser because of the risk of increased grain protein levels. These 3 farmers may suffer a yield penalty that in many cases is not compensated by the premium paid for the malting grade. Results from N response trials conducted in barley at different sites in South Australia by the Department of Agriculture in 1989 and 1990 showed that grain protein usually increased with added N but that in very deficient situations, when large yield responses were achieved, added N could result in a decrease or no change in grain protein (Jefferies 1990). Applying N fertilisers to barley crops is essential to maintain the yield of the crop in the low nutrient Íueas of South Australia. However, because high rates of N can increase grain N concentration, the amount of N applied should be matched to the

barley variety being used, and to climatic and soil conditions.

In Australia, a considerable amount of work has been conducted on the effects of rotation,

site and season on the response to N fertiliser, particularly in wheat. The yield response to

applied N of many of our varieties and advanced lines of malting barley is not known.

There appears to be little doubt that crop genotypes can differ in various aspects of uptake

and metabolism of N (eg Anderson 1985). In winter cereals, varietal differences in grain

yield response to applied N have been demonstrated in the field under a wide range of

conditions (Gardener and Rathjen L975; Fischer and Wall 1976; Power and Alessi 1978;

Krentos and Orphanos 1979). Genotypic differences in uptake rates of nitrate have been

reported for maize (Cacco et al.1983;Pan et at.1987),barley (Perby and Jensen 1983) and

wheat (Woodend et al. 1986). Interactions betrreen N rate and variety have been reported in

a number of studies with cereals (Hibberd et al. 1978: Stanford et al. L973) but apart from

work of Birch and Long (1990) and Bi¡ch et al. (L993) in Queensland and Doyle and

Kingson (lggz)in northern NSW there is little recent information about the responsiveness

of Australian barley cultivars to applied N. A differential cultivar response to N fertiliser for

grain yield and grain protein in barley suggests a non-predictable response when N

trearments are applied (McGuire et al. 1979). Therefore, newly developed varieties may

achieve mar

commonly adopted by farmers. In order to define the maximum benef,rts from adopting a

new variety, the most appropriate management techniques or options for each variety need to

be identified. It would also be useful to identify varieties which use applied N more

efficiently. 4

Therefore, to improve the chance of producing high yields of malting quality barley when N is applied, it is important to identify varieties which have a high grain yield response but low grain N concentration response to N fertiliser. The response in kernel weight will also be important because grain plumpness is another important characteristic of malting quality.

Variation in responses of barley cultiva¡s to applied N was examined in a preliminary experiment at Arthurton on the Yorke Peninsula in South Australia in 1985 (Y,lheelet et al.

1987). Results of this study also indicated that variability between cultivars in grain yield and protein responses to N fertiliser exists.

The overall N response in barley is more complex than for many of the other elements, but generally it can be described in three terms:

(i) Uptake of N (mainly nitrate) from the soil

(ü) Vegeøtive growth related to grain yield

(iii) Remobilisation of N and partitioning of dry matter and N to ttre grain

The work reported in this thesis was therefore undertaken to investigate the response of a range of barley cultivars to different levels of N. Firstly, the pattern of responses of 6 barley cultiva¡s to 8 rates of N fertiliser was examined in a series of the field experiments in order to facilitate further work. The influence of N fertiliser on vegetative and generative growth during the growing season was measured in the first experiment. Secondly, the differences in nirate uptake and assimilation among different barley cultiva¡ groups at the seedling stage was investigated using a hydroponic system. Thirdly, the interaction between N and water stress on yield and grain protein was studied in the glasshouse. Nitrogen and dry matter remobilisation in a number of barley cultivars under well-watered and \tater-stressed conditions was examined in a pot experiment. Lastly, the genetic variability among 78 barley cultivars in terms of grain yield and grain protein concentration responses to N was investigated in two field studies at different sites and years. The overall objective of the work was to conduct a series of preliminary experiments that would provide information on

the variation in N responsiveness in yield and protein among barley genotypes which could

be used to determine the direction of future experimental work. 5 CHAPTER 2 LITERATURE REVIEW

2.1. Introduction

This review examines the environmental influences on the growth and yield of barley, the seasonal availability of nitrogen (N) and its importance to crop production in South

Australian agriculture. The environmental and genetic factors that influence responses to N fertiliser are discussed also. However, information about the responsiveness of malting barley to N fertiliser is limited, so this review will discuss cereals more generally, with particular reference to wheat and the extent of genetic diversity among winter cereals for response to N.

In South Australia the area where malting quality barley can be grown successfully year after year is relatively small. The soil and climate are important factors determining the character of malting barley (Spanow 1972). Production of malting barley is generally confined to the cool, moist parts of the cereal belt where water and heat stress a¡e not severe.

Malting quality is a complex factor, dependent on the end result of the action and interaction of a hormone-enzyme-substrate complex. Barley has to have specific characteristics for it to be of malting quality. Maltsters require plump grain which is relatively low in protein (up to approximately 1L.8 Vo), high in starch and which has a high even rate of germination. A high protein concentration in the barley decreases beer quality because of the effects of the protein on the head and lacing of the beer. The level of N in the grain is therefore an important factor determining quality and is widely used as the major consideration in the commercial assessment of grain when awarding premiums for malting barley. A high grain

N level means a lower carbohydrate content in the grain which in tum means that there will be less malt extract.

The variety of barley has a major role in determining grain quality and therefore the malting grade of barley (Sparrow L972;Peterson and Foster 1973;Burger and LaBerge 1985; Smith

1990). Rapid modification is desirable and has been found to be determined largely bV Þ- glucanase levels (Henry 1989). A high rate of p-glucanase production may be more 6

important than a low B-glucan content. There is no correlation between barley p-glucan and

malt p-glucan (Mact eod et al. 1993). Another important facror is diastatic power (Dp). Dp

has been found to fluctuate *r[|ä19ç¡r¿jlJ!;.fü.,rÄrtrmental conditions under which the

barley is grown (Arends et at. 19931 pttC)tas a strong positive correlation with Dp

(Arends et aL. 1993; Nedel et aL. 1993); unfortunately, a negative correlation between Dp and malt extract precludes the use of high GNC varieties to increase DP (Fig 2.1). Low

GNC also causes low soluble protein in the wort which can affect the growth and

metabolism of yeast. When GNC is high the soluble protein in wort becomes high and causes cloud protein in beer which reduced the storage life-beer.AA Bendelow ( 1964) has also "/ found the proportion of B-amylase activity to be a useful cha¡acter for screening hybrid lines

for potential malting quality. The differences in enzyme concentration in different varieties

and the influences of these enzymes on malting quality emphasise the importance of variety

in determining malting quality. However, the yield and protein response to applied N of

many of south Australia va¡ieties and some advanced lines is not known.

Malting Quality

()¿+ Relationships and Criteria 300

öó 0') çAL oô I 200 :¿ O B1 co 80 o) 1' U'' (g o79 q') Cg 100 t- .g x78 o LU "/" Exlract o\\o 77 Diaslase (WK)

76 0 't 7 B I 0 11 12 13 14 Grotí*, "/" Protein ^

Figure 2. l. Relationship between malt extract, diastase activity and grain protetn concentration (Redrawn from Lance and Macleod 1993, Schooner cultivar). 7 In southern Australia barley is grown throughout the cereal districts in rotation with wheat and legumes. It is frequently grown after wheat with little additional N or following a period of legume-dominant pasture. Under field conditions barley often suffers from a period of N defîciency especially when grown as a second cereal crop, and the chance of a significant N response is high. Despite these observations, grain yield and protein responses to increasing levels of N in malting barley cultivars has received relatively little attention until recently. Moisture deficits are usually *iäi;|Jlr*" tn the growing season in southern Australia and this may interact strongly with Nnto determine grain yield and grain protein concentration. Both yield increases and decreases due to N application are known, depending on water supply. For example Barley and Naidiu (1964) showed that applications of N to wheat increased the rate of soil moisture depletion and consequently increased water stress during the grain filling stage. An understanding of the N cycle and the interaction between water and fertiliser N are important in efficiently managing N.

2.2. Growth of winter cereals in Mediterranean environments

Barley in southern Australia is grown as a rainfed crop. There are many studies that show that responses to N depend on the availability of moisture during the growing season and the incidence of environmental stress (Barley and Naidiu L964; Blum and Dnuel 1990; Van

Oosterom et al. 1993). Therefore it is important to understand the limitations of the climate when examining N responsiveness.

2.2.L. Major features of the Mediterranean environment

Less than 2Vo of the earth's land surface has a Mediterranean climate and these are in five,

widely-separated regions between 30o and 40o in both the southern and northern hemisphere

(Leeuwrik 1974). Mediterranean environments are found in-the Mediter-ranean basin, central o0 SowlL AÊxca.- and southern California, central Chile, south-west Cape Province, south west of Western

Australia and in the southern part of South Australia (Gibbon 1981). The major features of

this climate are mild, wet winters and hot, dry summers. Annual rainfall varies considerably

and ranges between 275 and 900 mm. More than 65Vo of the annual rain falls in winter

(Smith and Harris 1981; Hamblin et al. 1987; and Buddenhagen 1990). Dry areas

frequently receive their annual precipitation in sporadic but heavy falls with consequent rapid 8 runoff and erosion. The length of the growing season is defined by the availability of

moisture for plant growth. It depends on the locality and season, but ranges from five to

seven months (Prescott and Thomas 1949). Soils of the Mediterranean areas are variable

but are generally calcareous and alkaline, although in coastal areas some soils have no

calcium carbonate in the profile and are almost neutral in pH (Leeuwrik 1974). The

principal cereal crops of this region are wheat and barley. The variation in rainfall tends to

put severe restrictions on the stability of agricultural production (Blum and Dnuel 1990; Van

Oosterom et aI .1993).

General descriptions of the climate of southern Australia are given by Leeper (1970),

Gentilli (1971) and Nix (1975). Total annual rainfall in the cereal belt of southem Australia

varies from 300 to 600 mm. Most rain falls during the winter, and the April-October rainfall

comprising about 80Vo of the annual total. The growing season is determined by the

beginning of effective rains, and its length varies from 4 to 6 months. Droughts occur

frequently in the drier regions. In the cereal districts growth during winter is sometimes

restricted by low temperatures. High temperature and strong winds during the grain f,rlling

period in spring sometimes reduce cereal yields (French and Schultz 1984).

The general pattern of temperature and rainfall in the Mediterranean environment is very

much influenced by topography and proximity to the sea. Average monthly temperatures

during winter are between 4'C and 10"C with frost-free periods varying from 2 to 12 months

(Aschmann 1973). In summer, average monthly temperatures frequently rise above 27"C

and maximum daily temperatures in excess of 38oC are not uncommon.

Sunshine is uniformly high, above 3000 hours per year, with little year to year variation.

This creates high potential evapotranspirtion which greatly exceeds rainfall during most of YÀrw delay in its development tmfio increased

productivity (Arnon L979; Y anOosterom et al. 1993). 9 2.2.2. Likely occurrence of water stress in Mediterranean environment

The major features of the environment in the Meditenanean region were described in section

2.2.I. In this environment lack of adequate rainfall and also distribution of rainfall are important constraints to winter cereal production (Ceccarelli et al. I99I). Patterns of rainfall in the winter cereal growing regions in Australia range from the classical Mediterranean-type distribution, where rain is received only during the winter months, through to areas where rainfall is uniformly distributed during the year and where cereals are grown during the winter using stored soil moisture (Kassam 1981). The va¡iability in seasonal rainfall largely influences the levels of productivity of cereal crops. For example, Cornish (1950) found that 70-80Vo of the variance in yield of wheat in South Australia over 296locations was accounted for by variation in seasonal rainfall.

The arrival of the opening rains across southern Australia is unpredictable. They may fall any time between April to June (Nix 1975) and when they occur they are sufficient for substantial germination, emergence, establishment and growth of seedlings. Rainfall in the winter months (June-August) is usually adequate for plant growth but moisture stress of unpredictable severity, duration and timing can occur any time after mid-September (French and Schultz 1984), during stem elongation, flowering and grain filling phases (Hamblin er qI. l98l). The uncertainly of the start and end of the growing season are the dominant factors in crop production in dryland areas of southern Australia.

The supply of water is also related to the soil water reserves available to plants (Ritchie

1981), since only a small amount of water can be stored in crop plants relative to the rate of transpiration. Soil moisture at sowing plus precipitation during the growing season sets an environmental limit over which the plant has no control. Under Mediterranean conditions, the total water content of the soil profile increases during the winter period, reaching a maximum in late winter-ea¡ly spring. During spring, when rainfall declines and evaporative demand increases the soil profile dries, through crop extraction of water or direct evaporation from the soil surface or both (Cooper 1983; Gregory et aI. 1984). The depth of wetting during winter varies from site to site and season to season, depending on the total 10 rainfall and its distribution wi[þin the. growth season as well es lype of soil. Adaptive sncL æe v/uplì¡ry, >Ï vøríLã, ÞÞw3 livne- a+L &-l};ü;v.r nule,, cropping strategies. in turn are influenced by the probable timing of stress periods. A

The yield potential of cereals in Mediterranean areas is greatly determined by the incidence of winter rainfall (Van Oosterom et al. 1993), and is modified by solar radiation and temperature. Wheat and barley crops in Mediterranean environments are not generally subjected to severe water stress during the early period of growth because during the winter months moisture supply from rainfall is high and evapotranspiration rates are low (section

2.2.I). In spring, evapotranspiration increases as the vapour pressure deficit of the air increases and the crop become more dependant on soil moisture reserves. Crop water stress increases concurrently. The rate of development of stress depends on the amount and pattern of rainfall during the season and on soil type, but it is the development of terminal water stress which greatly influences grain yield and the responsiveness of crops to N.

2.2.3. Effects of climatic factors on crop growth and yield

The physiology of yield in cereals has been the subject of many research papers (eg, Evans

I975; Austin and Jones 1975; Evans and Wardlaw I976). The effects of climate on the growth and yield of crops depends on the likely time of stress in relation to the stage of development of the crop. In cereals, there is considerable variation for time of sowing both between and within seasons. Winter cereals grown in a¡eas with Meditelranean climate are sown in autumn with the opening rains. At sowing time the amount of moisture and N in the soil is enough for seedling establishment. Generally the growth of cereals is divided into pre-and post-anthesis phases. Pre-anthesis growth refers to the phases of seedling establishment, tillering and stem elongation up to flowering, while post-anthesis growth refers to the period from flowering to maturity which includes grain development and growth.

The rate and duration of tiller emergence is affected by temperature, photoperiod and

nutritional status, partcularly N. There is also much genetic variation in tillering habit (Austin 1939). In wheat and barley ear initiation occurs during the tillering phase.

Development of the floral organs occurs during the period of rapid vegetative growth, while 11 development of spikelets and florets occurs during stem elongation and when death of tillers is occurring. Grain growth occurs when vegetative growth is essentially complete and occurs under conditions of increasing water and heat stress.

Over much of the cereal belt of southern Australia productivity of winter cereals is limited not only by low or uncertain rainfall, but also by high temperatures in spring and summer and sometimes by low temperatures in winter. There is a strong interaction between temperature and moisture stress in the crop growth environment. There is generally a rapid rise in temperature in the spring to levels where temperatures during the day are above the optimum for the growth of temperate grasses. There are three main consequences of this rapid increase in temperature: firstly, respiration rates are markedly increased, leading to lower net assimilation and dry matter accumulation rates; secondly, the rate of physiological development is increased and the process of senescence occurs more rapidly; thirdly, and perhaps most importantly, saturation vapour pressure deficit of the atmosphere increases,

leading to a reduction in the efficiency of water use due to the inverse relationship between

carbon fixation and vapour pressure deficit (Tanner and Sinclair 1983).

Under a Mediterranean-type climate, the post-anthesis environment is usually hot and dry

and photosynthesis is therefore limited. Consequently, yield depends to a large extent on

the translocation of pre-anthesis assimilates to the grain, although the degree of this

dependence varies with season and site (Papakosta and Gagianas 1991). It has been shown

that during very severe post-anthesis drought, grain yield of wheat is related to biomass

growth in the late vegetative stage of growth (Fischer 1981).

(< rs-lo'c) The effects of low temperature on the productivity of winter-grown cereals vary with

geographical area. Low temperatures in the Meditenanean region may restrict crop growth

rates and the resultant poor ground cover increases evaporative loss from bare soil surfaces,

which reduces the water available for crop growth and reduces water use efficiency (Cooper

1983). However low temperatures in the post-anthesis stage are an advantage to grain

filling and low grain protein concentration. Respiration and transpiration rates ate reduced, t2 water use ef|lciency may increase and the potential for assimilate partitioning into grain can be increased.

Freezing damage ," d:äto1ä:ü *U yield has been observed at several stages of "."e^ growth.Frostdamag'ffiisarecognisedhazardofmanyofthesemi-a¡id environments. When frosts occur at about -3oC in late stem elongation, booting, or heading stages, severe yield reductions may result (Single I975). At these stages of growth there is little opportunity of recovery from frost damage as crops enter periods of high temperature and water deficit.

2,2.4. Seasonal availability of N in the Mediterranean-type environment

The cycling of N in plants during growth and development and the efficient recovery of N in the soil-plant system has been a topic of considerable interest over the years (Harper et al.

1937). The dynamics of N supply. are particularly important in the rainfed agricultural systems of the Mediterranean-type climates. Many of the soils are deficient in nutrients, particularly N (Jackson 1977; Sillanpaa 1982; Brown et al. 1987) and the normal soil fertility needs to be supplemented with N from fertilisers or from N fixed by legumes. The crops requirement for N in the Mediterranean-type climate varies with location, cropping system and year-to-year weather conditions. Soil factors along with climate and cultural factors determine the capacity of the soil to provide available nitrogen to plant (Russell

1968a).

The amount of mineral N in the soil available for uptake by plants shows a characteristic seasonal pattern which affects the N nutrition of crops. At sowing in autumn or early in the winter, the amount of N in the soil is often large. The amount depends on previous crop history which affects the accumulation of organic N. Cultivation, together with the

relatively warm temperatures and moist soil in autumn cause an accumulation of mineral N.

As rainfall increases during winter, losses of mineral N by leaching can be large (Bell 1979;

Ellington and Reeves,1990). Pa¡t of the N is also taken up by the plants. This reduction in

the concentration of available soil N continues until grain filling and then slowly increases

because the increasing summer temperatures result in high mineralisation rates and because 13 of lower uptake by plants. The main factors influencing the mineralisation of N are the availability of soil moisture, amount of organic N and the length of the period of the growing season.

Losses of mineral N from the root zone depend on the amount of mineral nitrogen in the profile during the latter part of the rainfall season. Rainfall of high intensity during this period, if capillary conductivity is large, may cause leaching of mineral nitrogen from the root zone, particularly if the soil is shallow. In southern Australia, subsoil accumulation of nitrate is common, but it usually remains within the root zone of crops (Greenland 197l) because rainfall is insufficient to cause leaching beyond the roots.

Researchers have found that not all of the soil and applied fertiliser N can be accounted for at the end of a growing season, and have suggested that some N is lost from soils in the

volatile form as ammonia (NH3), dinitrogen (NZ), nitric oxide (NO), nitrogen dioxide

(NOZ), nitrous oxide (NZO), and various amines (Harper et al.1983; Dabney and Bouldin

1985). Losses may also occur from the plant system through the release of gaseous NH3

(Wetselaar and Farquhar 1980; Parton et aI. 1988). The use of N fertilisation to overcome

soil N deficits is still relatively uncommon in many dryland areas because of the economic

risk associated with variable responses (Vlek et al. l98l). Nitrogen may also become

unavailable to the plant due to immobilisation, particularly in management systems when

there is high organic residue at the soil surface, such as in conservation tillage or forage

system (Doran 1980; Jansson and Persson 1982).

The extent to which these various processes take place in a particular soil, depends on the

agro-climatic conditions under which the soil is found and it is important to time fertiliser N

applications to coincide with periods of greatest need by the crop, to ensure acceptable grain

yields and to minimise fertiliser use.

so;l 2.2.5. Interaction between water andnN so;t In the environment of southern Australia, water undnN are frequently the two most critical

factors in limiting crop production. A strong interaction between the two is known to exist t4 (Young et al.1967; Singh and Prihar 1978; Eck 1988; Engel 1991) and one may improve the use efficiency of the other to a certain extent (Singh et aI. 1975; Kanemasu et al. 1983;

Benbi 1990). An optimum combination of water and N is therefore essential for enhancing their water use efficiency and maximising crop yields. Van Keulen (1977) for example, found with upland rice that the efficiency of N utilisation, in terms of grain yield per unit of

N absorbed, was about one-third of what was already found under irrigated conditions because moisture shortage at the end of the growing period accelerated senescence of the vegetative plant parts. Photosynthesis therefore declined rapidly and the grains could not be filled to the same degree as under inigated conditions. This resulted in grains with high protein concentration and low weights. This interaction is greatly modified by soil type, root distribution pattern and other factors that influence moisture availability and use, eg. root diseases.

There are many ways by which water use efficiency can influence N efficiency in cereals.

Firstly, added water may increase root growth and thus increase absorption of N; secondly soil moisture affects mineralisation of N from the soil organic matter; thirdly, N is absorbed by plants largely through mass flow which requires water; fourthly water movement into the soil is required to move N fertiliser into the root zone to make it available but excess water also removes soil N by denitrification and leaching (Sander et aI. 1987). As water use efficiency increases, the response to N fertilisers will be improved and will be reflected in

higher crop yields. This response to N differs between varieties and season (Sander et al.

1987).

Nitrogen is a mobile element in the soil and moves from colloids to roots mostly by mass

flow (Russell 1980). Where moisture in the soil is low, the response to nitrogen fertiliser

may be reduced with reduced flow of N to roots. The concentration of N in the plant tissues

then drops. Low moisture levels also reduce growth and total dry matter yield, thereby

decreasing the demand for N by the crop. Considerable quantities of N fertiliser accumulate

as nitrate in the soil in dryland regions, particularly when high rates of N are applied (Olson

et aL 1916). Under dryland conditions a high proportion of the fertiliser remains in the soil

and will probably be available for subsequent crops. Craswell and Strong (1976), working 15 on vertisols in Queensland showed that if crop growth was limited by dry conditions, the fertiliser N not absorbed remained in the soil at harvest as nitrate, although the amount will vary. However, ro-e 15N balance experiments in dry regions (Myers and Paul I97I:

Olson 1930) have shown that I6-36Vo of applied 15N may be lost and as little as 5Vo may remain as nitrate-N in the soil at harvest.

Under dry conditions N placed at 45 cm depth was less susceptible to loss and more available to the wheat crop than N at 15 cm depth (Craswell and Strong 1976). Work done in other climatic regions also shows that fertiliser N in the subsoil is more available than surface applications if the surface soil dried up during crop growth (Garwood and Williams

196l). Therefore, although the utilization of N during the season in which it is applied may be low, there may be some residual effects in the following seasons.

In areas of limited precipitation, a balanced nutrient supply enables the crop to make more efficient use of the available moisture (Arnon 1975) although the total water use may not be greatly altered. In Mediterranean environments crop yields are assumed to be limited frequently by available moisture. Crop yields decline with decreases in annual precipitation, and the high variability in yields between seasons are related to the amount and distribution of rainfall (Harmsen et aI. 1983). Aspinall et aI. (1964) demonstrated that a nutrient- deficient plant growing poorly used water at the same rate as a nutritionally balanced plant,

although it produced considerably lower yields and hence its water use eff,rciency was lower.

Work by Aktan (1976, cited by Bolton 1981) confirms these findings, and indicates that at

different levels of soil moisture at seeding, applied N must be varied in order to balance the

moisture and N supply for maximum water use efficiency. He showed with barley that

water use efficiency can be decreased markedly by nitrogen def,rciency'

The highly variable moisture supply, typical of dryland regions, requires that fertiliser

requirements be tailored to the season if efficient use of water is to be achieved. Brown

( 1971) reported that the addition of N to crops of dryland wheat in Syria increased yields,

water use efficiency and also increased total water use by increasing the depth of water

extraction from the soil. Brown et al. (1987) previously showed with barley that application t6 of N fertiliser increased soil water extraction rate during early spring and total water use by about 10-20 mm. Efficient use of water by crops given fertiliser earlier in growth was associated with longer root systems, particularly below the surface 15 cm (Brown et aI.

1987).

Efficient use of N and water are therefore strongly related. Growth under N deficient conditions implies a slower rate of accumulation of dry matter and leaf area. Under such conditions, direct soil evaporation is greater than under non deficient conditions where a closed canopy is reached earlier. The amount of moisture available for transpiration is therefore smaller under N deficient conditions (Van Keulen 1981).

High levels of applied N fertiliser may however decrease yields due to lodging and haying off. Haying off is attributed to high rates of available soil N stimulating early vegetative growth and water use, thereby depleting soil moisture reserves and inducing water stress in the crop. This decreases harvest indices because of water stress during the latter part of the growing season (Stonier I975). In semi-arid regions, variability in precipitation and the associated variability in the degree of moisture stress may account for up to 85 per cent of the variability in the yield of wheat (French and Schultz 1984). In order to explain the variable effects of N on the yield of dryland barley, information on the magnitude of moisture stress and its effect on plant development is needed.

2.2.6. Conclusion

Analysis of climatic information in Mediterranean environment suggests that the major

limitations to yield are soil moisture, the length of the growing season and N availability

during the growing season. However, N deficiencies have to be considered within the

context of water deficiency. The use of N fertiliser by farmers in dryland areas is influenced

by the small responses under low rainfall conditions, the high year to year variability in

crop yield and response, and the consequent economic risk. When the use of N fertiliser is

inhibited by low rainfall, and climatic variability, more attention needs to be given to

alternative strategies. One way is to develop adapted varieties which will be responsive to N

fertiliser even in stressful environments. In the past, most of the work in Australia has I7 concentrated on agronomic aspects of N fertiliser rather than an examination and exploitation of likely genetic variability.

2.3. Nitrogen 2.3.1. Nitrogen in the soil

Nitrogen is frequently a major limitation for both food production and protein content. It is an essential nutrient in plant metabolism and constitutes I.5-5Vo of plant dry matter (Haynes

1936). It is a constituent of all proteins, of many metabolic intermediates involved in synthesis and energy transfer and of nucleic acids. The metabolism of N is well documented in the reviews of McKeel (1962) and Bray (1983).

Nitrate and ammonium are the most important forms of N utilized by cereals although urea is also supplied to plants as fertiliser. Nitrate however is the most common form of inorganic

N in the soil because under most soil conditions ammonium fertiliser is rapidly nitriired to nitrate by soil organisms (Haynes and Goh 1978; Hageman 1979).

When supplies of soil water are adequate, N is most commonly the key limiting factor for crop production. On the average, about 99Vo of. soil N is organically bound (Rosswall

1976) and accumulation of soil N closely follows that of soil organic matter. Detailed reviews of organic N compounds in soil have been provided by Kowalenko (1978) and

Parsons and Tinsley (1975). The organic matter in the surface layers of soils usually has a

C/1.{ ratio of 10 to 12 andthe ratio decreases with depth (Oades 1989). However, since the soil N system is dynamic, any change to the environment (e.g. a change of climate) may lead to a new equilibrium level for soil N. Thus, the N content of soils is very diverse, ranging from <0.17o in desert soils to over 27o in highly organic soils (Haynes 1986). The total N content of Australian soils varies widely, ranging from 0.01 to O.\Vo in the surface horizon of normal mineral soils and up to 2Vo in swamp and fen soils (V/illiams and

Raupach 1983). The N in organic matter becomes available to plants very slowly, with only

2-3Vo of it converted to available forms in a year. The conversion into forms available for plant uptake is controlled by the activity of soil microbes. In many instances this natural 18 supply is inadequate for crop growth and N fertiliser must be supplied to non-legume crops to achieve optimum yields.

2.3.2. Nitrogen uptake and utilisation by the crop

Nitrogen uptake from the soil by plants depends on both the demand by the plant and the availability of N in the soil. Total N uptake by a crop involves a period of slow accumulation in the early part of the growing season, when tissue N concentration is high, followed by a rapid linear rate of accumulation that coincides with rapid plant growth

(Tinker 1978: Pearson and Muirhead 1984). In this period of rapid growth the N concentration of tissue decreases. For field crops, the rate of uptake during the rapid growth phase can be extremely high (3-5 kg N ha-l day-l;Tinker 1978; Remy and Viaux 1982;

Olson and Kurtz l9S2). The rate of N accumulation decreases late in the season and continues at a decreased rate until maturity. Net N uptake when N fertiliser is applied early in the season is generally completed by anthesis (Stonier 1962; Smith et al'1989b).

Current views on the fate of nitrate in plants are summarised in Fig.2.2. Nitrate uptake occurs as a result of active influx and concurrent efflux back to the external medium. Since attraction between nitrate and soil colloids is negligible, nitrate is mobile and is readily carried to plant roots by mass flow (Russell 1980; Stevenson 1982). When potential uptake exceeds the supply from mass flow, the concentration of N at the root surface is lowered and the process of diffusion begins.

Following uptake by roots, a series of independent transformations occur. Nitrate may be

stored in the roots, reduced and synthesised into amino acids by root tissues, or transported

across root cells and deposited in the xylem for movement into the shoots. Amino acids

formed there can move back to roots or to grain via the phloem (Pate I97l:'Pate et aI.1979).

Once the reproductive phase begins in cereals mobilisation of products from vegetative to the

reproductive tissue also begin (Pate and Layzell1981). MacKown and van Sanford (1986)

for example, showed with wheat in USA that leaves contributed 4OVo, glumes 22Vo, culm

22Vo androots I6Vo of remobilised N to the grain. Simpson et al. (1983) also showed with

wheat that N in the grain could be derived entirely from the redistribution of N from L9 vegetative organs, although the conditions under which this was done were not stated.

Environmental conditions would be expected to influence the relative contribution of the

different plant parts to the grain.

The mobility of N within the plant is an important characteristic of N metabolism. The

response of many grain crops to applied N depends as much on N redistribution within the

plant as on uptake during the rapid phase of growth. Translocation of N during grain

development affects grain yield and grain protein concentration (Pearson and Murihead

1 984). For instance, by anthesis wheat can contain about SOVo of the N that is present in

above-ground parts at maturity (Dalling et al. I976; Austin et al.1977).

storage NO; 4l IV NO; ----+ NO; ---+ NHi ----+ Amino Acids

SHOOT

NOt No; ----+ No; --+ runi ---+ Amino Ac¡ds 1l NO; E XTERNAL storage ROOT MEOIUM

Figure 2.2. Stages in the assimilation of nitrate by plants (Pilbeam and Kirkby 1990)

The total N uptake in plants increases prior to heading and flowering, but substantial losses

can occur after this (Storrier 1962; Farquhar et al. 1980; Parton et al. 1988). Volatilisation

of N from the plant is the most important mechanism for loss of N around flowering (Parton

et al. 1988). Errors involved in sampling through the loss of dead leaf and stem material

may be another cause of the reported reductions in N at maturity. 20 2.3.3. Crop response to N uptake

The response of a plant to N is an integration of many complex mechanisms, many of which are not measured or understood in most field experiments. Yield responses of plants to N fertiliser application may occur as an increase in dry matter yield, protein yield, quality improvement, or some other feature. The simple response of plants to applied N, when N is the major factor limiting growth is increased dry matter yield with increasing rates of N up to a maximum, after which yield stays constant or declines with further application N. Plant growth is however, affected by many environmental factors and these can greatly modiff the responses to applied N. Environmental factors also influence the availability and losses of

N from the soil. Therefore the response to N fertiliser, even when N is limiting, can be highly variable, depending on the other environmental factors.

2.3.3.1. Response curves

Where there are sufficient rates of N, the fitting of the data in a simple mathematical expression may be justified (Wood 1980). Describing yield responses mathematically allows data to be summarised into a series of fitted curves, thereby facilitating comparisons between sites and other treatments.

No single response model is likely to fit all the experimental results (Sparrow 1979; Wood

1980). Some yield responses to N curves are illustrated in Fig2.3. The modelling of yield

responses to N can become quite complex; different cultivars of one crop grown in the same

soil can have different response patterns. When N fertiliser experiments are carried out over

a wide range of locations and conditions no single response model is likely to fit all the

results. For example, Sparrow (1979) found that of eight models tested against N response

of spring barley in 83 trials, the inverse quadratic, linear over linear, and two intersecting

straight lines represented the yield-fertiliser relationship well although no one model fitted

best at every site. However, the commonly used model for cereals are quadratic and

modified Mitscherlich. 2I

POLYNOMIALS v straight line v quadrat ic

Y=a+bx y=a+bx+cx2

X X v v square root c ubic

I y=a+bx2+cx y=a+bx+cx2+dx3

X X

EXPONENTIALS v v M it scherlich modified Mitscherlich

Y=a-bcx y = a - bcx - d x

X X

Figure 2.3. Examples of response curves of plants to applied N fertiliser (Redrawn from Wood (1980). 22

In Australian experiments reported in the literature most researchers have used the quadratic model: Mason and Rowland (1990), Anderson et aI. (199I) in Western Australia; Doyle and

Shapland (1991) with wheat in NSW; Birch and Long (1990) in Queensland with barley.

Quadratic and Mitscherlich models have also been used for barley in New South Wales

(Dolye and Kingston 1992).

2,3.3.2. Effects of N on growth, grain yield and yield components

The overall yield response to applied N has been outlined. Nitrogen fertilisation often increases barley grain yield through its effect on various yield components (Littler et al.

1969; Nuttall 1973; Calder and Macleod 1974; Penny and Freeman 1974;Macleod et aI. t9t5).

Applications of N fertiliser to winter cereals increases tiller production and tiller survival so

that more tillers are available to form ears (Pushman and Bingham 1976; Doughetty et aI.

1979; Camberato and Bock 1990b). Leaf area usually increases with N applications mainly

through an increase in number of tillers, number of leaves, and also to increased leaf size

(Langer and Liew 1973; Spiertz and Ellen 1978; Pearman et aI. 1977). Leaf area duration or

longevity of leaves is also extended, increasing the potential for higher rates of crop

photosynthesis (Pearman et aL 1977; Thomas et al.1978; Spiertz and van der Haar 1978).

However, increasing respiration rates have also been reported with increasing N application

( Pearman et aI. 1977; Spiertz and Van der Haar 1978). The overall effect of N application

is therefore an increase in the source capacþ of the plant.

ofler anfuEis Sink capacitynis determined by the number and size of grains and their rates of growth.

Nitrogen generally increase the rate of floret development, the number of fertile florets

(Langer and Hanif l9l3; Thomas et aI. 1978) and the number of grains set (Langer and

Liew 1973; Camberato and Bock 1990a). N-induced yield increases are often the result of

increases in ear density (Remy and Viaux 1982).

Where ear number is little affected by N, an increase in ear weight can be induced by N

application (Spratt and Gasser 1970; Gasser and Iordanou 1967). Such an increase can be 23 due to an increase in the number of spikelets per ear and more grains per ear (Langer and

Liew 1973;Pearman et aI.1977).

Nitrogen application can also decrease, have no effect (Pushman and Bingham 1976;

Camberato and Bock 1990a) or increase kernel weight (Atkins et aI. 1955; Gately l97I;

Spiertz and Ellen 1978) depending on competition among individual grains for assimilates and N (Langer 1979). The effect of N on yield also depends on the growth stage at which

N is applied. Addition of N at the beginning of tillering enhances tiller formation and spikelet initiation (Darwinkle 1983). When applied at the onset of stem elongation, ear number is increased. The number of fertile spikelets, grains per spikelet and single grain weight are also increased when N is applied at ear emergence (Darwinkle 1983). Indeed barley has only one grain per spikelet so the number of fertile spikelets and grain weight are even more susceptible to variation. Application of N is not always beneficial to yield in

small grains. Additions of N tend to increase the height of a crop (McNeal'e/ aI. I97I), and the elongation of basal stem internode enhances the danger of lodging with consequent

decreased yield.

2.3.3.3. Effects of N on grain protein concentration

The rate and duration of both'starch and protein deposition are essentially independent

events, controlled and influenced by different factors. The energy costs for protein

deposition are high, at least twice the energy required for deposition of equivalent amounts

of starçh (Penning de Vries et al. 1974). The rate of grain protein concentration is

considerably influenced by the environment, and N nutrition and its balance with available

water is particularly important. Water stress decreases the amounts of sucrose in grain oC) which is convertt/r to starch. High post-anthesis temperature (>30 may cause premature "r* cessation of starch deposition in the endosperm, even in well watered crops (Bhullar and

Jenner 1986). N accumulation is less temperature sensitive than starch accumulation

(Bhullar and Jenner 1985) and so moderately high temperatures often result in an increased

grain protein concentration (Jenner et al.l99l). 24 A major effect of fertiliser N on crop quality is the increase in grain protein concentration

(Benzian and Lane 1979,1981; Olson and Kurtz 1982). High grain protein concentrations are undesirable in malting barley so the yield-protein relationship is very important. Grain protein concentration usually increases linearly with N rate up to a maximum irrespective of optimum nitrogen rate for yields (Benzian et al. 1983) so that under conditions where N can depress crop growth and yield, there can be a negative relationship between grain yield and grain protein concentration (Evans and V/ardlaw 1976; Novoa and Loomis 1981).

The duration and amount of protein deposition are determined by the balance between source and sink events during grain filling (Jenner et aI. l99l). A large proportion of stored N in the leaves is deposited in the chloroplast as Ribulose Bisphoshate Carboxylase Oxygenase.

Grower management strategies for malting barley attempt to maximise grain yield and kernel plumpness and keep grain protein low. Often, management strategies which maximise grain yield will not optimise grain protein and malting quality (Lauer and Partridge 1990).

Insufficient N can reduce grain yield, grain protein and quality below acceptable levels, while excessive N usually produces undesirable high protein levels and small grain

(Baldrige et al.1985).

2.3.4. Grain quality and N 2.3.4.1. Grain protein concentration and malting quality

Malting quality is a complex character and there are no universally accepted criteria of quality. Many factors are used to evaluate malting quality, but grain protein percentage is a very important parameter. The maximum acceptable protein percentage of midwestern six- rowed malting barley in the USA is l3.5%o on a dry-weight basis, but a lower percentage is preferred (Peterson and Foster 1973). In Australia the desirable concentration of protein is

about 10.5 to LI.SVo, but will vary for individual users (Henry 1990). The current upper

limit for malting grade 2 of the Australian Barley Boa¡d is 1 1.8 7o (source ABB 1993).

The nitrogenous constituents of barley grain influence the yield of the extract and the

proportion of malt and wort (Sparrow 1972). The extract is one of the first factors a brewer

considers in assessing a malt. It is a measure of soluble materials after mashing the malt and 25 the value of the extract generally ranges from77 to SlVo of the soluble material @urger and

LaBarge 1985). There is a close inverse relation between grain protein concentration and malt extract (Bishop 1930a; Sparrow I972;McGtire et al. 1979; Burger and LaBarge 1985;

Smith 1990; Fig. 2.1). It is assumed that any reduction of extract by protein is simply due to protein replacing carbohydrate in the grain. A high positive correlation has been found between grain protein and malt diastatic activity (Lance and Macleod 1993) (Fig 2.1).

High protein concentration in malting barley has undesirable effects on the malting and brewing processes. High protein barley requires longer steeping times, and its erratic germination is difficult to control. Malting losses are increased due to erratic rootlet and acrospire growth and also increased soluble protein in the brewing product (Deckard et al.

1986). Wort and beer colour increase due to brewing reactions that occur during kilning and wort boiling. Also high protein in malt contributes to increased haze formation in beer

(Peterson and Foster 1973).

2.3.4.2. Relationship between grain yield and grain protein concentration Fertiliser N has been, and still is, used to supplement soil N to increase yield.

Unfortunately, the use of supplemental N is not always associated with increased quality

(Deckard et aL 1986). Many studies have however shown negative correlations between

grain protein concentration and overall yield (Evans and Wardlaw 1976; Sinclair and de Wit

1975; Halloran 1981; Loffler et al. 1982; Cox et aI .1985; Stoddard and Marshall 1990;

Jenner et al. I99l).

The response of yield and protein to N supply is strongly influenced by the environmental

conditions, especially the quantity and timing of water available to the crop and the amount

of N in the soil. Responses in grain yield and grain N concentration to inçreased N supply

are shown schematically in Fig. 2.4. The relationship betwee¡ yield and proteiq percentage

changes according to soil N leve-l and can vary from negatiye to positive. When conditions

favour a large response to N (eg low soil N levels, or favourable moisture conditions), grain

N concentration may fall or remain unchanged (part A, Fig.2.Ð. In this part of the general

response there may be a negative relationship between yield and grain N concentration, or 26 grain N concentration may differ little over a range of yields. Eventually, grain N concentration will begin to increase along with grain yield and there will be a positive correlation between yield and grain N concentration (part B, Fig. 2.4). At high rates of N, or when grain yield is limited by factors other than N (eg water), the yield response is low or sometimes negative while grain N concentration continues to increase (part C,Fig.2.4).

In this case of the response there may be a negative correlation between grain yield and grain

N concentration.

6 5

Yield 1Z 4

"< - _c J: I as8 I .ç Prolein G ^c¿6 ò (9 4

0 0 A ö C lncreasing nitrogen applicatron

Figure 2.4. General relationship between grain yield, grain protein concentration and N application for rainfed wheat (from Perry and Hillman 1991).

The relationship between -grain yield and grain N concentration among cultivars grown under the same conditions is frequently negative (part A and C, Fig.2.4). A number of

attempts have been made to explain this apparent negative relationship between grain yield

and grain protein. Dilution and energy constraint effects are most commonly given as

causes for the negative relationship between yield and protein (Stoddard and Marshall

1990). Protein synthesis requires approximately twice the energy for starch synthesis

(Penning de Vries et al. 1914). Under conditions o{1lrruting energy therefore, proteins may

be produced at the cost of yield. Thus, given equal amounts of photosynthate, high protein-

low carbohydrate cultivars were predicted to produce less grain yield than low protein-high

carbohydrate cultivars (Bhatia and Robson 1976). The negative correlation between grain

protein concentration and yield can therefore be attributed to the higher energy costs of

protein synthesis (Jenner et al. l99l). Dilution is a more consistent and more powerful d¡rr¡'\q anir- Q\\7.41 cause for the negative relationship of protein to yield. Under good growln$ conditionf protein and starch deposition proceed simultaneously and grain protein concentration varies within a fairly narrow range of +2Vo during development (Bechtel et al. 1982). Under poor conditions yield is depressed and grain protein concentration elevated through drought and high temperature effects. Drought reduces the availability of sucrose which a grain may convert to starch (Brooks et aI. 1982). High temperature also suppresses the conversion of sucrose to starch (Bhullar and Jenner 1986). Therefore water deficit and high temperatures during grain filling, relatively common occurrences in the Australian wheat-belt, prevent the dilution of protein by starch.

obiecù'Y¿5 Another possible explanation for the negative relationship is the breedingmethsds used for improving crop cultivars. These methods have resulted in the inadvertent improvement of harvest index without the improvement of biomass (Kramer 1978). According to Kramer

(1978) more than two-thirds of the N used to synthesise grain protein will be in the plant at anthesis. Altering the plant environmentally or genetically to increase or decrease harvest index may affect the remobilisation of this N and subsequently grain protein concentration.

A high protein variety of wheat was found to have a lower kernel weight than a low protein

variety by Donovan et aI. (1977). The difference in weight and protein cohcentration were

due to different levels of carbohydrates, other than starch in the low protein variety. The

relationship between kernel size and grain protein concentration however is variable. Not

only can N effect kernel weight and grain protein concentration directly, but N can also

influence kernel number which tends to confound the correlation between grain weight and

protein concentration.

2.3.5. Factors affecting crop responses to N

Nitrogen fertiliser is occasionally applied to malting barley in southern Australia but the

responses are often variable. Factors affecting responses of plants to applied N have been

reviewed extensively elsewhere (Cooke 1982: Keeny I982a; Olson and Kutrz 1982), so

only the most important factors are discussed. 28 2.3.5.1. Form and timing

Many physiological studies utilising liquid culture methods have shown that species differ widely in response to nitrate and ammonium forms of N (Mason 1968). Nonetheless, most agronomic resea¡ch has shown that the two forms of fertiliser N a¡e virnrally interchangeable

(Tinker 1978) because in most well-drained soils ammonium is transformed to nitrate within a period of days. The choice of the form of N is therefore normally an economic rather than an agronomic decision.

To maximise the use of fertiliser N by crops, N should ideally be applied when it is required by the crop. For cereal crops, the most effective time for N application (tillering to stem elongation) is often related to the latest time compatible with the period of rapid N uptake by the plant (Miller et aI. 1975; Dougherty et al. 1979). Such an application time reduces opportunities for N losses by leaching, NH3 volatilisation or denitrification. Furtherrnore, applied N becomes available throughout the period of grain formation without being used earlier for production of unnecessary vegetative growth (Olson and Kurtz 1982); shorter plants with higher grain:straw ratios are therefore produced. Work by Olson and Kurtz

(1982) also showed that grain protein concentration increur{*n.n N is side-dressed rather ^ than applied at planting. However, the effects of timing of fertiliser application on crop yields are often small compared to those for rate of application (Needham 1982).

Research with barley (Widdowson et aI. 196l: Bowerman and Harris 1974; Easson 1984), wheat (Ellen and Spiertz 1975,1980), and rice (De Datta et al. 1972) has generally confirmed that applications of fertiliser N early in the season enhances total N uptake and utilisation for grain and protein yields provided there is sufficient rainfall to carry it to the root system. Apptication of fertiliser N when delayed, can result in lower yields and fertiliser N recovery by crops (Jung et aI. 1972; Easson 1984) even though some vegetative response may be obtained (Smith et al. 1989a).

Australian farmers often enquire late in winter as to whether it is suitable to apply fertiliser N

to crops sown in June which look yellow. Such enquires show the need to examine the

effect of a late season N application on wheat production. Variable responses to delayed 29 application of N have been reported (Littler 1963; Elliot et al. 1985; Randall et al. l99O).

The later the time of application, the greater the risk that the added N will not increase grain yield, even though some vegetative response may be obtained (Smith et al.I989a). Much of the variability in yield response with post-sowing applications of N is related to variability in rainfall.

In South Australia, Elliot et al. (1985) found that split applications of N were desirable where heavy winter rainfall caused temporary N deficiency during tillering. Generally, most

N fertiliser is applied with the seed at sowing in the higher rainfall areas but there may be an additional topdressing to deficient crops. When barley is grown after wheat, crop yields are low because yield can be limited by N deficiency. Although applications of large amounts of N at sowing is unwise when the crop has limited capacity to use it and N losses can be high particularly in high rainfall areas, split application at planting and as top dressing may be necessary. However, this late N application may increase grain protein concentration which is not desirable for malting barley. The current recommendation for malting barley in

South Australia is to apply N at sowing or at early tillering. In combination with early sowing this has the greatest chance of increasing grain yield without greatly increasing grain protein concentration (Holden and Jefferies 1988).

Applications of N after flowering can contribute to increasing the protein content of wheat (Smith etal. I989a; Randall etaLL990)andRandaIIetaI. (1990),regarditasauseful

option for improving the marketability of the wheat grain. Nitrogen fertilisation of malting

barley must however be approached with care; sufficient N to maximise yield may increase undes¡raHe. grain protein concentration to aneþjeetiend level. Therefore N application at the presowing

or seedling stage may be desirable to obtain low grain protein concentration as well as

increase yield. Variation of the time of N application between sowing and different stages of

crop development offers the option of adjusting application rates to match soil and seasonal

conditions and crop demand for N. As N must be moved to the root zone for effective

utilisation by the plant, the strategy must depend on imminent rainfall. 30 2.3.5.2. Moisture

Generally, there is interaction between applied fertiliser N and moisture for yield (eg Spratt and Gasser 1970; Pushman and Bingham t976; Singh et a\.1979). The effects of moisture on the N response of crop has been discussed in section 2.2.2. In Australia, the variability in the response of wheat to N is mainly due to the unreliability of seasonal rainfall (Russell

1968a; Taylor et al. L978). During winter, rainfall and/or stored water in the soil are high while atmospheric vapour pressure deficits are low. Crops are therefore not subjected to water deficit in early growth. The later stages of growth however, are characterised by increasing water deficits due to reductions in rainfall as well as an increased evaporative demand. Increasing vegetative g¡owth due to high N supply may however deplete soil water quickly and severely restrict grain yield. Grain protein concentration may increase under such conditions. Also higher rainfall areas in general have higher yield potentials which require more N and greater N rates.

2.3.5.3. Genotype

Different cultivars respond differently to N. Studies with wheat in South Australia found that grain yield in cultivars showed a similarresponse to N in dry matter except one cv.Gabo which showed no response (Alston and Lungley 1975). Va¡ietal differences in the recovery and uptake of N was also found by Woodurff (L972). He concluded that both factors were related to the maintenance of the leaf area after anthesis. Gardener (1972) also reported differences in the response of different barley cultivars to N application. He found the variation in varietal yields with N reflected variation in ear numbers.

Several factors are responsible for genotypic differences in responses (Clark 1983; Mengel

1983) and they can be divided into two major groups:

(i) Plant growth rate

The rate of plant grorilth is an obvious factor affecting N requirement. High-yielding

cultivars are characterised by rapid growth rates resulting in high production of plant material

per unit of land area or time. Such production can only be attained when matched by an

adequare supply of N. High production of leaves and stalks is usually associated with high 31 grain yield potential for high-yielding corn lines. These lines therefore require high rates of fertiliser N to obtain maximum yields (Agboola 1972; Mengel 1983).

(üi) Patterns of development

In reality, N often increases harvestable yield through is effect on various yield components, most of which are effected by the phenological development of the crop. In small grain crops leaf area is usually increased by N applications (Langer and Liew 1973; Spiertz and

Ellen 1978) due to an increased number of tillers, and therefore leaves, and also to increased leaf size. Thus, the potential photosythetic capacity of the crop is raised and the rate of photosynthesis can be increased (Spiertz and van der Haa¡ 1978). Birch and Long (1990) found with barley that total tiller production increased and tiller survival decreased with increased rates of N fertiliser. This results is consistent with other reports (e.g. Barley and

Naidu ß6/). The large early response to applied N was partly due to an increase in tiller numbers, many of which senesced as stress increased later in the season; however, maintaining adequate N nutrition later in the season can reduce tiller senescence.'Whether a more controlled pattern of tillering could result in a greater grain yietd response to N should be investigated further.

N application early in crop deveþment promotes tillering, leaf growth, the number of fertile

florets per spike and the subequent grain set. However, the extent of the effects of N will be

determined in part by the developmental pattern of the cultivar. For example, cultiva¡s with

a long pre-iniation period may produce a large number of tillers. In such cultivars, the

response to N may be a large increase in the number of tillers and subsequently in ears. The

dry matter and potential yield response to N may be large, but in order to achieve this

potential it may be necessary to maintain a high N supply throughout the pre-anthesis period

to minimise tiller senescence, and to reduce the possibility of water stress near flowering,

particulary if the later time of ea¡ iniation is also associated with delayed flowering.

After the boot stage applied N has little, if any, effect on grain number, although it may

increase total dry matter of the crop and, in some circumstances, gtain weight (Spiertz and

van de Haar 1978). Grain protein concentration is increased with N applications at this stage 32 (Smith et al. 1989a). However, if an N application at this stage results in an increase in kernel weight, then grain protein concentration could conceivably fall. Nedel et aI (1993) with barley found that N fertilisation had a positive effect on the spike number, and a tendency for a negative effect on kernel weight. These results are supported by those found in wheat by Bruckner and Morey (1988). They also found that the number of kernels per spike was not affected by N in 1987, but was affected in 1989; however, responses were nor different for semidwarf and. standard height (STD) genotypes. The number of grains per head and individual grain weight increasing with high N rate (150-200 kgNlha) with different for 3 barley cultivars, despite reduction in grain size (Birch and l-ong 1990).

(üi) Capacity for upøke and metabolism

Genotypes of various crop plants differ greatly in their capacity to produce dry matter at a given level of N supply and in the amount of dry matter produced per unit of N absorbed

(Chavalier and Schrader L977; Reed and Hagenen 1980). The differences in genotyPes are not only due to their different capacities to absorb N but also in the translocation and partition of N within the plant (Clark 1983). For some genotypes of wheat (Halloran and

Lee 1979; Fjetl er at. 1984) and corn (Rodriquez L977) there is no close relationship

between the proportioning of total plant N in grain at harvest and the total amount of N

accumulated in plants. However, high-protein genotypes of wheat generally translocate N

to grain from vegetative plant parts more readily and completely than do normal cultivars

(LaI et al. L978; Fjell er at. t984). Some high-protein genotypes of wheat also require

continued assimilation of N by leaves during grain development (Mikesell and Paulsen

1971; McNeal et al.l97l).

2.3.6. Nitrogen use efficiency indices range To assess of the efficiency of N fertiliser application and utilisation by crops over a of The environments and cropping practices it is necessary to have some indices of efficiency. (Craswell efficiency of N in a cereal crop can be defined in terms of 3 efficiency parameters

and Godwin 1984; Harmsen 1984): (i) Agronomic efficiency (kg/kg)=(YF-YO)/F

(ä) Apparent recovery (Vo)=1gg* (NF-N))/F 33

(üi) Physiological eff,rciency (kg,/kg)=(Yp-YO)/Nf-NO) where Y¡ and Yg are the grain yields of the fertilized and unfertilised crops respectively,

N¡' and Ng are the N contained in the grain and straw (kg/ha) of the fertilised and unfertilised crops respectively, and F is the amount of fertiliser N applied (kglha). These pa¡ameters are normally based on measurements of N uptake in the above-gtound plant parts and the calculation of apparent recovery depends on the assumption that both the fertilised and unfertilised crops absorb the same amount of N from the soil.

Both agronomic and physiological efficiencies are based on grain yield. The apparent recovery reflects the efficiency of the crop in obtaining fertiliser nitrogen from the soil, while the physiological efficiency is a measure of the ability of crops to utilise N in the plant for the synthesis of grain yield. Agronomic efficiency is the product of the apparent recovery and the physiological efficiency and thus estimates the overall efficiency of the system.

Increases in either or both the apparent recovery and physiological efficiency will increase the agronomic efficiency (Novoa and Loomis 1981; Craswell and Godwin 1984).

Agronomic efficiency generally decreases with increasing rates of fertiliser applied

(Harmsen et al.1983).

Another index for measuring the efficiency of N utilisation is the index 'physiological efficiency of uptake N', which is defined as the ratio of grain produced to the total N absorbed by the above-ground plant parts ($ain plus straw) at maturity (Novoa and Loomis

1981). The 'physiological efficiency of uptake N' of Novoa and Loomis (1981) differs

from the physiological efficiency (see üi above).

physiological efficiency (see (iii) above) depends on crop variety and environmental

conditions during the growing season. There are signifîcant differences between cereal

genotypes in their efficiency to convert N into grain yield. The efficiency of fertiliser

application can be assessed by the apparent recovery which is a more suitable indicator than

the agronomic efficiency, which also includes a crop factor (physiological efficiency). 34

Apparent recovery

Since crop growth rate depends on N availability, apparent recovery can be seen to be a very dynamic component of agronomic eff,rciency. Effrciency of recovery is therefore influenced by climate through effects on losses, the availability of N and through effects on crop gïowrh (Craswell and Godwin 1984). The redistribution of N from roots to shoots will also influence the apparent nirogen recovery since only N uptake in the aerial portion of the crop is normally considered in the calculation of recovery (Campbell et al . 1977). Various climatic factors affect the amount of volatile N loss from plant tops (Wetselaar and Farquhar

1980) and the amount of nitrogen lost via root exudation (Russell1977). These losses, together with losses through environmentally influenced senescence, will affect both recovery efficiency and physiological effrciency.

Physiological efficiency

Any factor that affects plant growth will influence the physiological efficiency parameter since plant growth and nitrogen uptake are closely related. Environmental stresses are geatly influence the partitioning of crop dry matter between süaw, roots and grain and thus affect physiological efficiency. The physiological efficiency also varies between plant genotype, as demonstrated by a comparison of tall and short wheat varieties (Alessi ef ø1. tg7g). Tinker and Widdowson (1982) suggested that improved varieties have played an important role in not only increasing crop yields but also improving N fertiliser efficiency.

2.3.7. Conclusion

In Mediterranean a¡eas N is often the major factor interacting with available water resources which limis production. However, by using artificial fertilisers or legumes, a shortage of N can be overcome more easily than a shortage of \ilater. The predominant positive impact of fertiliser N on crop quanrity and quality is in its increase in the grain yield and protein content of grain. At the crop level, the method and timing of the supplies of \ilater and N have a strong impact on N use efficiency. The N economy of crop production involves considerable complexity and one must take a systems view when considering improvements in management practices. Much information has accumulated on the physiological and 35 level of knowledge currently biochemical aspects of N utilisation in cultivated crops but the in response to N' available is frequently inadequate to explain genotypic variation

2.4. Genetic variation in responsiveness to N for N use efficiency in There is considerable evidence in support of genetic variability lg82' Paccaud et al' 1985; different crops (Epstein 1972; Clark and Brown 1980; Clark from a wide range of characters perby and Jensen 1986; Ctark 1990). This variability arises differences in root growth have which may be under genetic control. Genetically-controlled been attributed to differences in the been reported (Hackett 1g6g; Epstein rg72) and have relative grolilth rate (Chapin and demand for mineral elements caused by differences in Glass and Perley Bieleski lgLz),in uptake and transport (Doddema and Telkamp 1979; 1976; Whiteaker et al' L976)' 1980; Glass et ¡¿¡. lgSl), and in use-efficiency (Gerloff efficiency of above-ground dry Bruetsch and Estes (1g76) studied the variation in N use efficiency of nitrogen use ranged mattef production in 12 corn genotypes and found that the the importance of the genetic from 59 to 82Vo of kg biomass/kg N absorbed, showing between populations within component. strongly heritable va¡iation between species and added N has also been found by genotypes in the response of leaf a¡ea and shoot weight to (1983) with barley and by De Datta Goodman (1977) wlthlotiwnsp., by Perby and Jensen

and Broadbent (1990) with rice.

for improved nitrogen use effrciency Studies have shown that there is the possibility to select et 1984)' in rice (Broadbent er al' traits in maize (Muruli and Paulsen 1981; Anderson al' and in oats (Isfan 1993)' Wide 1987) in barley (Isfan 1990), in triticale (Isfan et al' l99l) but the genetic inheritance of these genotypic difference among cereals for N effrciency exist

plant responses is not well understood'

appears to exist among many Therefore considerable genetic variation in N use efficiency physiological factors' In terms of the final important cereals and this is due to a range of be comprised of 2 components - total N amount of N in the grain, this can be considered,to the grain' taken up and the extent of is remobilisation to 36 2.4.1. Nitrogen uptake

Several workers have reported differences among genotypes in the uptake of nitrate in different crops: barley (Perby and Jensen 1983; Perby and Jensen 1984; Isfan 1990), maize

(Cacco et at. 1983; Pan et al. 1987) and wheat (Woodend et al. 1986). The differences have been reported to be under genetic control (Epstein 1972; Clark 1983). V/ork by Perby and

Jensen (1983) with 17 barley cultivars showed that va¡iation irt root size (diameter and length) and morphology could account for differences in the uptake of nitrate by different barley cultivars. High nitrate uptake has also been associated with a more extensive root system in maize (Reed and Hageman 1980) and in barley (Hackett 1968). Nitrate accumulation was higher in later-maturing than in earlier-maturing barley genotypes (Perby and Jensen 1984).

Genotypic differences in dry matter yields have also found to influence nitrate uptake rates in sorghum. Genotypes differed in dry matter yields, N content and dry matter yields per unit

N (Franca 1981, cited by Cla¡k 1990). Genotypes with higher dry matter yields and higher

N contents have been found to have lower nitrate uptake rates per unit root weight compared to genotypes with lower dry matter yields and lower N contents. Uptake rates of N decreased after 34days of age, and total N absorbed per plant increased as plants aged with a maximum at about 44 days of age (Franca 1981).

Hoener and Deturk (1938) found differences in N uptake and assimilation in high and low protein lines of corn. They however, found signifrcant differences for N translocation from leaves to stalks ¿rmong corn genotypes. With an increase of N to 200 p.p.m in solution culture, the low protein corn gained in N content, but the high protein lines failed to make further appreciable gains. Howevêr there was a superiority of the high protein line in assimilation as well as absorption at 100 p.p.m N. Thus it may be possible to select lines with more efficient nitrate assimilation.

Cla¡k (1990) with 4 wheat cultivars (2 bread and2 durum) found postanthesis uptake of N

was greater under moist than under dry environments. There were cultivar differences in

uptake of N, and any differences observed were related to variations in plant dry matter. 37

Nitrogen uptake from soft dough to hanrest was greater for bread wheats than for durums.

Average plant N concentrations of the bread wheats were greÍìter than those of the durums at all growth stages except harvest. This did not cause differences in pre-anthesis uptake because dry matter accumulation was greater in the durums than in the common wheats

(Cutforth et at.1988). Therefore the differences in plant N concentrations were largely due to dilution effects.

Bloom and Chapin (1981) grew barley cultivars under identical conditions. A barley cultiva¡ selected on cold soils had higher ammonium and nitrate influx rates at low N concentration and at loril temperatures than cultiva¡s from warmer environments. Therefore, the genetic differences between these barley cultivars is related to their selection background-

The experiments reviewed indicate that there are genotypic differences in nitrate upøke benveen cereals crops and cultivars. Most of the genetic variation in nitrate uptake in barley,

wheat and sorghum is related to root size, morphology of root and dry matter yield. In

addition it has been found that nitrate uptake is also related to differences between low and

high protein lines of corn and wheat. The extent of cultivar differences in Australian malting

barleys however, has not been described adequately.

2.4.2. Nitrogen remobilisation

Most of the N present in the grain at maturity comes from N mobilised from the leaves,

stems and roots. Differences in the contribution of N mobilised to grain from tissue have

been reported (Dalling et al. 1976. Simpson et al. 1983; Spiertz and Ellen L978). Under

conditions when the post-anthesis soil N and water supply is high, at least 50Vo of the gfain

N yield will still be derived from the mobilisation andredistribution of N from the vegetative

organs (Spiertz and Ellen 1973). However, in those environments where the post-anthesis

supply of soil N is low, which includes most of the rainfed crops of Australia, N

redistributed from vegetative organs can contribute to more than 807o of grain N yield

(Dalling 1985). The leaves and stem are the most important reserves of N. Each may

contribute 30Vo of the protein deposited in the grain. Contribution from the roots is small

(lTTo,Dalling et at. 1976). The glumes of wheat may contribute tíVo which is important 38 because they are a site of transfer of N from xylem to phloem but this source may be insignificant in barley because the glumes in barley are smaller than wheat.

Translocation and N metabolism during grain development determines yield and grain protein concentration because nitrate uptake declines after anthesis and therefore in many dryland environments post-anthesis N uptake is negligible. The grain N therefore depends largely on mobilisation of N from vegetative parts. Halloran and Lee (1979) defined translocation efficiency for N as the ratio of total grain N to total head N and found that it ranged from 75 to 94Vo among wheat cultivars. N translocation from shoot to grain was about 67Vo under irrigation andT5Vo under dryland conditions (McNeal et aI. L968).

Remobilisation of N is a major determinant of the overall N use efficiency of the whole crop

(Huffaker and Rains 1978; Novoa and Loomis 1981). It represents an adaptation for high efficiency in the use of scarce supplies of N. It may therefore be possible to select crop species that are highly efficient in remobilisation of N and thus utilise mineral N from the soil more efficiently. Genotypic differences in N remobilisation efficiency have been noted in wheat (McNeal et at. L968;Dubois and Fossati 1981; Loffler and Busch L982;Paccaud et al . 1985), triticale (LaLetal.L978), maize (Hay et al. L953; Beauchamp et al' L976,

Chevalier and Schrader 1977) and barley (Perby and Jensen 1987).

McNeal et al. (1968) suggests varietal differences in grain N concentration in wheat could not þg attributed to difference in the percentage of N that was translocated from top $owth to the grain but there is a very close relation between grain N content and Otal N amount of the N translocation ro grain. Y/ork by Paccaud et al. (1935) with ten wheat cultivars

suggests that the observed genetic va¡iation for grain protein concentration is related to N

harvesr index (l.tHI). Cultivars with the highest N partitioning efficiency showed the

highest yields and the lowest grain protein concentration. NHI was different btween wheat

varieties and it has been suggested as a useful selection criterion to improve grain yield while

maintaining grain protein concentration (Loffler and Busch 1982). However LoffTer et al'

(1985) later suggested that further research was required to determine if early generation grain selection for N utilisation components would be worthwhile. Selection for high 39 protein concentration and grain protein yield without reducing yield of parents on the basis of factors such as NHI may be useful (Desai and Bhatia 1976; Loffler et al. L985).

Genetic differences in N remobilisation in 4 inbred corn was reported by Beauchamp et al.

(L976). They showed that with 4 inbred corn lines there are differences in percent N in leaves, stalks and ear at 4,24 and 28 days after silking, but there was also much unexpected variability amongst the inbreds from one year to the nexq which indicates that the proPortion of N ranslocated from different plant parts to the ear depends to a considerable extent on envfuonmental conditions. Also Chevalier and Schrader (1977) found that the concentration of remobilised N in the stems and leaf sheaths to be higher in the inbreds than in the hybnids derived from them. The difference in N partitioning was reflected in the percentage of total

N found in the ear. Among hybrids lines, total N was highest in roots, stem and leaf sheath of A632 x Oh43, but total ea¡ N was not appreciably higher in this genotype. In contrast, low levels of total N were found in all plant parts of W182E x A632 and W182E x OM3, but percentages of total plant N found in ea¡ fractions were higher in these genotlTes than in other hybrids. There was no relation in reduced N concentration between palents and Fl progenies indicating that N partitioning is not a simply inherited trait. Lal et al. (1978) found in their study with triticale and wheat that the translocation of N from the total top growth ranged from 49.9 ro 58.4Vo in triticale varieties and from 53.5 to 61.07o in wheat.

McNeal et at. (1966) considered a variety which translocatedTOTo or more plant N to grain as generally quite efficient under irrigated conditions. Both triticale and wheat cultivars differed greater in their capacity to accumulate and redistribute N. The loss owing to translocation from different plant parts to kernels va¡ied depending upon the variety. The percent 6anslocation was maximum from lower leaves and minimum from spike chaff (see

Lal et ¿t. Table 3). In barley, Perby and Jensen (1937) suggested that N partitioning to tillers was also important to dry weight production, and high external N supplies normally increase the number of tillers, although genotypes differed in this aspect.

In contrast, Clark (1990) suggested translocation did not vary among bread and durum wheat cultiva¡s. Absence of significant variation in uptake and translocation of N and nutrient utilisation efficiencies was süongly related to plant and grain dry matter amounts 40 matter was and partitioning. The amount of N translocated per unit of dry accumulated lowest in irrigated inversely related to water availability. It was greatest in dryland, and than in conditions. Similarly, retranslocation of carbohydrates tends to be greater in dry (1972) eight wheat moist environments (Biding er et al. Lg77). Also McNeal et at. with for high lines, found genotypic differences in N translocation, but there was no differences that grain protein compared to low grain protein concentration gloups. He suggested into grain which concengation was dependent on the amount of carbohyd¡ate translocation and the ability of the was influenced by the number of kemels present as carbohydrate sinks, kernel weight are plant to translocate carbohydrates to kernels. Both kernel number and affected by environmental conditions, but a¡e also under genetic control'

use (Gasser and The translocation of reduced N will influence the efficiency of nitrogen to 784o in wheat Iordanou 1967). Translocation efficiency was found to fange from 227o translocation efficiency varieties by Dalling et at. (1g75) and Hagem î et at. (1g76). A high N use efficiency (Huffaker with a large capacity for uptake and assimilation, should improve

and Rains 1978).

efficiency that Fischer and Wall (1976) pointed out that improvements to agronomic va¡ieties were related to occurred through the introduction of higher-yielding wheat This was achieved improved physiological efficiency rather than increase in N uptake' less dry matter is through the introduction of semi-dwarf varieties where proportionally The higher physiological apportioned to straw than in ull varieties (ie higher harvest index)' et aI' (1977) and in efficiency of short varieties was also demonstrated in wheat by Pearman

rice by Sanchez et al. (L973a).

appfoaching 507o with However, Fischer (1981) has argued that, since ha¡vest index was physiological efficiency via new semi-dwarf wheat varieties, further improvements in (1980) also found that wheat shows a increased ha¡vest index is limited. Ellen and Spiertz basal N at sowing' For malting higher physiological efficiency with late applied N than with stage or tillering should be barley, an application of N at moderate fates at the seedling but for feed barleys' where a adequate for acceptable level of grain protein concentration, 4l high protein concentration is desirable, a second dose of N fertiliser may be added after ear emergence to increase grain protein concentration (Johnson ¿f al.1976). Thus, protein can be managed by adjusting the timing of N application to ensure maximum physiological effrciency.

At the whole plant level, N efficiency is intimately related to carbon metabolism and partitioning. Leaf area and tillering responses to N are important since they result in increased source and sink capacities. Remobilisation of protein N also seems to be of great importance, and McNealet al. (1966) suggested that movementof.TÙVo of the N from leaves and stems to grain would represent good effrciency.

At the crop level the method and timing of the supplies of \üater and nitrogen have a strong impact on N use efficiency. Genotypic differences in remobilisation may be influenced by the environment. The level of mobilisation increases with stress and so in stressful envi¡onments the importance of mobilisation tends to be great. In rainfed crops the amounts of dry matter and N remobilised during the grain frlling perid can be relatively large and can increase with greater level of post-anthesis stress @idinger et al. L977; Fischer 1979; Clark

1990; Pheloung and Siddique 1991). Work by Gonzalez Ponce et al. (1993) with winter barley under different climatic conditions showed that the N translocation efficiency was high in the L986fl season, a year with suffrcient rainfall in the pre-anthesis period, and very little rain in the post-anthesis period and high temperatures occurring at the final grain filling stage. Ea¡lier, Gallagher et at. (1975) achieved high N translocation from the vegetative pafrs of the plant to barley grain during the post-anthesis period under conditions of severe drought. Therefore, sress will affect the remobilisation of dry matter and N and their allocation to the grain. Genotypic differences in translocation efficiency which have been reported may therefore be related to the sensitivity of the cultivars to stress.

Suggested differences between cultivars in remobilisation of N with different flowering time

(eg Beauch amp et at. 1976) may give a false indication of the level of genetic variability in remobilisation. Differences in maturity will often result in differences in the level of post-

anthesis s6ess, which may affect the amount of remobilisation (see above). Differences in 42 N remobilisation is under genetic control, but environmental factors also can affect mobilisation of N. The mechanisms by which some plant genotypes are: (i) able to grow and produce yield under conditions of N stress and (ii) also to mobilise higher amounts of N under similar conditions of N supply need further investigations.

2.4. Conclusions This review has shown that yietds of crops grown under rainfed conditions in the

Mediterranean environment are often limited by available \vater. Much of the variability in yields between seasons appears to be explained by differences in the amount and distribution of rainfall between seasons, although actual yields of crops in farmers fields are generally much lower than potential yields, and are determined by the length of the growing season and the genetic potential of the crop. Yields of unfertilised wheat crops in FAO experiments conducted in the Meditenanean region were in the range of 0.9 to 1.5 t/ha, whereas yields of fertilised crops were in the range of 1.5 to 2.3 tlha (FAO 1981). These results show that nutrient deficiencies are widespread in the Mediterranean countries and are a yield-limiting factor in many instances. In most cases fertiliser application increased the water use efficiency of the crop. To optimise the use of water and N under conditions of limited moisture supply, primary attention must be given to the efhcient use of \tater. This may be accomplished in two ways:

(i) storing maximum amounts of moisture in the soil profile by improved conservation

practices; and

(ii) by making maximum use of stored water and rainfall through improved crop

management pracuces

The response to N fertilisers will improve with increasing water use efficiency. Fertiliser

applications in semi-a¡id, rainfed regions where precipitation is seasonal and variable

should be based on the crurent moisture status of the soil and curent moisture supply rather

than long-tetm averages.

The overall yield response to applied N is mediated by the effects of N on various yield

components. However, application of N may lead to an increase, decrease or no effect on 43 yield, depending on time of N application and on grain protein concentration. It is important therefore to consider the objective for which N is being applied; for example, in malting barley one should apply N pre-sowing or at the seedling stage whereas for feed or food barley a second application of N fertiliser during stem elangation or anthesis, in addition to the early application may be required to enhance grain protein concentration.

In addition to the time of application of N, many other factors interact to influence the response of a crop to applied N. Among the most important are the form, and the method of application, "N-supplying power" of the soil, availability of soil water, influence of plant genotype, and the presence or absence disease and pests.

Different cultivars also respond differently to N and the selection of cultiva¡s in N fertiliser experiments can affect the more general interpretation of the results. Nutritional requirements also vary with species and cultivars. These differences may be under genetic control and may be at the level of uptake, Eansport and assimilation phases. Most studies have shown that the differences among cereals in N use efficiency relate to either uptake, translocation and assimilation of N particularly nitrate or some contribution of all of those.

Despite considerable work on fertiliser responses in Australia there has been relatively few experiments which have specificatly examined varietal differences in responsiveness in barley. Work from other crops and other environments shows that genetic differences can be substantial and so it is important to examine this aspect with malting barley. In view of ttris, the present study was planned with the following aims:

(i) to examine the responses of barley cultivars to N fertiliser in a series of freld

experiments and quantify the genetic variability between these some of barley cultivars

(ii) to examine N and dry matter remobilisation during grain filling among barley

genotypes

(iii) to study generic differences in nitrate uptake and assimilation between barley

cultivars. 43ï

ADDENDUM TO LITERATURE REVIEW of the examiners of this thesis' This addition to the Literature Review was fequested by one

2.2.6 DeveloPment of cereals role in the determination of yield' The pattern of development of a cereal crop plays a critical and number of the yield It has a direct effect on yield by its influence on the size yield through its components. As well, the pattern of development indirectly affects the timing of key stages of influence on the pattern of dry matter production and on the temperatufe and moisture development in relation to the seasonal changes in genotype to N may be influenced by environment. Therefore, the responsiveness of a barley

the the pattern of development in barley'

2.2.6.L The pattern of development of barley size or mass' In contrast' Growth is a quantitative process measured as an increase in In general' phenology (the development is a programmed qualitative change in form' growth) can be considered as the development though discrete and well defined stages of to maturity' It is greatly modified by developmental timetable of the plant from germination and abiotic stfess and their environmental factors such as photoperiod, temperature

interactions with a cultivar's sensitivity to ttrese factors.

plants' They produce a terminal The winter cereals, wheat and barley, are determinate leaf primordia are formed. The spikes of inflorescence after the initiation of which no more nodes bearing spikelets arranged wheat and barley consist of a number of reproductive Barley has a determinate' annual alternately on two sides of an elongated axis, the rachis' apical meristem in most shoots g¡owth habit cha¡acteristed by the synchronous switch of the spikelet primordia' The from the production of leaf primordia to the initiation of key factor in its adaptabitity to a range of phenological characteristics of a barley variety is a of dry matter throughout the life environments. It also has a gteat effect on the partitioning shoot a¡e initiated at the shoot apex and of the plant and on grain yield. All the organs of the determine the final number of grains and the rate and duration of the formative pfocesses

hence the yield potential of a plant (Kirby 1985)' 43ä

2.2.6.2 Developmental stages for barley

Development of barley can be described by a number of ca¡dinal points: germination (and emergence), ear initiation, anthesis and physiological maturity. These points provide a developmental framework which influences the partitioning of dry matter and the development of yield potential in the crop. The life cycle of the barley plant can also be partitioned into a series of phases by reference to the development of the shoot apex and the spike.

I-eaf and tiller production

The shoot apex present in the embryo contains a number of leaf primordia. At germination growth is resumed and initiation of new leaf primodia recommences. From 7 to 15 leaf primordia are initiated, depending on genotype and the effects of photoperiod and vernalisation on the time of floral iniation. Cultiva¡s which have a short pre-initiation perid

initiate fewer leaves on the main stem.

Tillering in winter cereals occurs in a systematic manner. Primary tillers arise directly from

leaf axils on the main shoot, usually from the a,rils of the coleoptile and the fust three leaves;

secondary tillers arise from the axils of leaves of the primary tillers. Frequency of tiller

formation in the axil of the coleoptile depends on genotype, temperature, photoperiod and

light intensity (Cannell 1969b), but the development of tiller buds appears to be regulated

internally by the plant hormone balance (Johnston and Jeffcoat 1977; Sharif and Dale 1980;

Koranteng and Matthews 1982). The number of tillers initiated is related to the time of ea¡

iniation and is related to the number of leaves because they arise from leaf axils. Conditions

which reduce the time to floral iniation can reduce tiller production. Long daylengths for

example, which can promote ear initiation, suppress tiller production (Kirby and Appleyard fertile l9S0). Quick maturing cultivars have a lower capacity to produce tillers and therefore

ears, which may limit their capacity to respond to N.

The number of tillers produced increases rapidly soon after crop emergence, reaches a

maximum, and subsequently declines. Shoot number usually stabilises by the time of spike ßin emergence. The number of tillers formed depends on the genotype, the level of fertility and the the incidence and severity of abiotic stresses (Cannell L969a: Simmons et al. L982).

Nitrogen nutriúon plays an inportant role in the development of tillers after tiller initiation and on their survival to form fertile ears (see 2.3.3.2).

Development of tillers is compressed in time compa¡ed to development of the main shoot.

Tillers usually produce fewer leaves than the main shoot possibly because their period of leaf initiation is shorter. Another consequence of the compressed growth period of tillers is that

initiation and flowering of spikes of tillers becomes relatively synchronous with the main

shoot (Fletcher and Dale L977,Hay and Kirby 1991) although there is considerable variation

in the synchrony of tiller production (Hay and Kirby 1991). The importance of the relative

rates of tiller development to yield potential and yield stability in barley has not been studied

intensively but it may be an important characteristic in environments that are inherently

variable and in which inra-plant competition for photoassimilate is intense.

The importance of tillering to grain yield is not fully resolved. Positive associations between

grain yield and spike number (Rasmusson and Cannell 1970) and kernels per unit area

(Dyson lg77) have been interpreted as favoring higher tiller production. Other researchers

have maintained that low tiller production @onald 1968) or low tiller mortality (Jones and

Kirby 1977; Kirby and and Jones 1977) is desirable for high grain yield. Most researchers

however, agree that complete elimination of tillering would reduce the crop's ability to

compensate for poor field stands and variable weather conditions.

Floral iniation and spike growth

The inflorescence of barley is a spike. When the transition to floral initiaúon occtlls, the rate

of primodium appearance increases about three fold. Double ridges can be detected when

one-third to one-half of the spikelet primodia have been initiated. During the period of

spikelet initiation rhe more distal primordia have the highest growth rates (Kirby 1977)- \\e

number of rachis nodes at which spikelet primordia are initiated ranges from l0 to 45,

depending on genotype and environmental conditions (Kirby 1977;Kltchen and Rasmusson

1983). Generally the rate of leaf initiation is about two and a half times greater than that of 43 iv the leaf emergence rate, therefore initiation of spikelet primordia begins when fewer than one-half of the leaves on a shoot have emerged.

Spikelet initiation ceases when awn primordia are formed on the most advanced spikelets.

After the maximum number of spikelet primodia is initiated, the differences in growth rates among spikelets disappear and all grow at the same rate (Scott et al.1983). Thus the relative sizes estabtished during the period of primordium initiation ate maintained. When the maximum number of spikelet primodia is attained, the embryonic spike of a two-ro\ted barley weights about 0.2 mg and is about 20 mm long but at anthesis the spike weights abour 200 mg and measures about 80 mm in length (Kirby L977). Therefore the period between maximum number of primordia and anthesis is a time of very rapid growth and it becomes an increasingly important sink for photosynthate within the growing plant.

Restriction of growth caused by environmental stress will therefore reta¡d the development

of the spikelet and reduce the yield potential.

Anthesis

Anthesis, or flowering, is the shedding of pollen following the dehiscence of the anthers. In

barley (as well as wheat and oats) pollen is shed inside the floret before the anthers a¡e

extruded and so the crop is almost entirely self-pollinated. Anthesis sta¡ts in the central

spikelets of the ear and the tip and base of the ear flower later. Tiller ea¡s flower a little laær

than those on the main stem and flowering in the whole crop may last for a number of days.

Adverse conditions such as low or high temperature stress or water stress around flowering

can affect pollen production and viabitity and fertilisation and so reduce seed sel In dryland

agriculture, management practices such as time of sowing, choice of cultiva¡ and fallowing

are often aimed at reducing the levels of stress prior to and during flowering. Time of

anthesis is determined by the genetic and environmental controls on floral initiation and

subsequent reproductive growth, and the general rate of gowth as affected by temperature

and water deficit. 43v Kernel growth

Dry maner accumulation by the developing kernel is initially slow, then increases to a nearly

constant rate when kernels attain a mass of approximately 4 mE (Gallagher et al. 1975).

Maximum kernel growth rates range from 0.9 to 2.2 mg kernel per day (Gallagher et ø1.

1975 Scott ¿r al. 1983), with kernels positioned centrally in the spike having the highest

growth rares (Walpole and Morgan l97l). Although kernel weights for given cultivars

varies between sites and years, compared with kernel number, it is a relatively stable

component of yield in barley (Gallagher et al. L975; Austin et al. L98O). This is important in

malting faney because of the requirement for plump grain for malting grade barley.

Delayed sowing, and hence delayed anthesis will reduce the length of the kernel filling

period and reduce kemel size. However, there is a degree of compensation between kernel

grotwth rate and the length of the grain filling period because the two are often inversely

related (Gallagher et at.1975) which minimises the changes in kernel weight. As well, the

stability in kernel weight is partly attributed to the ability of the plant to mobilise assimilate

stored in the stem and other vegetative tissue and transport it to the developing grain. Such

mobilisation can compensate for a shortage of current photoassimilate. However, in wheat

there is considerable variation in the ability to remobilise stem reserves to the grain and this

variation may be related to plant height (Blum 1983, Anderson and Smith 1990, Hossein er

at. 1990): there is little information relating to genotypic variation in barley.

2.2.6.3 Environmental effects on development

Changes in sowing date, or sowing location (Bauer et al. 1988) can dramatically alter the

duration of the different development phases, and to differing degrees depending on the cultivar used. Fischer (1984) considers that the major components of the environment

which affect development in such cases are temperature (vernalising and non-vernalising

temperatures), photoperiod and water stress. However, there are indications that other

factors such as level of nutrition, plant density, radiation and CØ can modify responses

(Rawson Lggz),but these effects a¡e small. Genotypes differ in their sensitivity to the major

factors and also appear to differ in thei¡ inrinsic rates of development (Flood and Halloran

1984). These differences can be exploited to develop cultiva¡s suited to envi¡onments with 43 vi different lengths to the growing season. The ability to alter the pattern of development to maximise r€sponses to N may also be feasible but has received relatively little attention.

Control of flowering time

Time of flowering is a critical stage of development in dryland cereal crops. The rate of development of a crop is influenced by the time it is sown and agtonomic practices such as time of sowing and the choice of cultivar are often determined largely by the most appropriate time for crops to flower. The onset of flowering in most genotypes of the world's annual crops is influenced by a plant's response to vernalisation, temperature and photoperiod; the relative importance of each of these varies throughout the life of the plant

(Ellis er al.1992).

Vernalisation response is useful in controlling development where the winters are reliably cool to cold, but the response will vary from season to season and between sites because of the annual differences in temperature. The annual cycle of daylength is a more reliable environmental signal and carefully tuned photoperiodic responses offer a precise way to time flower development. That is not easy, given the enormous range of possible planting dates at different latitudes. However, control of plant development by photopoeriod responsiveness widens the geographic range of individual cultivars and may also be a valuable adaptation to variably hot and dry environments since photoperiodic requirement could restrain more rapid development in the warmer conditions of drier ye¿¡rs. From an agronomic viewpoint time of flowering is probably the most critical stage of the crop plans development, but time of flowering is the end result of the rate of development of the apex from germination to spike emergence.

Factors affecting shoot apex development

Barley is a quantitative long day species with daylength sensitivity differing between cultivars (Tew and Rasmusson 1978; Ba¡ham and Rasmusson 1981). Little detailed work has been conducted in Australia on barley development. In surveys of wheat cultiva¡s used in Ausralia it has been found that, in general, cultivars adapted to southern Ausralia have little vernalisation response and rely primarily on photoperiod and temperature reponses to 43 vtl control their patterns of development (Davidson 199????). Water stress during the spike initiation phase can also affect the rate of primodium initation (Nicholls and May 1963).

Photoperiod. In Australia, barley is generally sown in late autumn or winter, when average daylength is near its minimum value, and it develops in an environment of lengthening days.

In most species, there is a pre-inductive phase beginning from sowing during which plant development is insensitive to photoperiod, followed by an inductive phase during which developmental rates a¡e influenced by photoperiod, which probably ends at flower initiation but before the first flower appearance (Roberts and Summerfield 1987). It is well

êstablished that shortening the photoperiod within the inductive range increases the total number of both leaf and spikelet primordia initiated (Rahman and Wilson 1977b). Short photoperiodic regimes delay the onset of spikelet initiation and prolongs the duration of

spikelet primordium initiation, more than compensating for the reduction in the rate of primordium iniation, which in most cases results in more spikelet primordia (Pinthus and

Nerson 1984; Craufurd and Cartwright 1989). Rate of spikelet initiation increases in long

days (Kirby 1969; Fairey et al.l97s),while duration of spikelet initiation is reduced by long

days (Kirby and Appleard 1980). The opposing effects of daylength on rate and dr¡ration do

not completely compensate for one another, and fewer spikelets are initiated in long days.

In addition to affecting the rate and duration of both leaf and spikelet primordia, photoperiod

also has a distinct effect on spike morphogensis. In a longer photoperiod the time taken

from sowing to flowering is shorter compared to a short photoperiod, due to a faster rate of

development (Holmes 1973). There a¡e relatively few studies about the effect of

photoperiod on the floret primodium initiation phase. It is reported by Langer and Hanif

(Lg73) that shorter days (12-15 hours compared with 24 hours) increased the duration of

floret primordium initiation in wheat while reducing the rate of floret primordium initiation

and floret number. Kirby and Appleyard (1980) with barley found that the duration of ear

initiation was decreased by long days for barley but the duration of the ear growth phase was

unaffected. They also reported genotypes responded differently to photoperiod for number

of leaves per plant, grains per ear and weight per grain. 43 vüi

Vernalization. In some species, germination of seed, initiation of flowering, or bud break require or are hastened by prolonged exposure to temperatures below to 4oC. The process involved in these varied responses is called vernalization. It is generally accepted that exposure to low vernalising temperatures after seed imbibition can affect the rate of development of cereals during the vegetative stages (i.e. until floral initiation), but it is not coÍtmon to expect that the time from initiation to anthesis will be affected by this factor

(Halse and Weir 1970). However, it has been postulated that vernalisation during the ea¡lier

stages of crop $o\tth can influence development byond the vegetative phase, at least in strongly responsive genotypes (Halloran and Pennell 1982). Rahman's (1980) data showed

that some genotypes of wheat had a quantitive response to vernalisation during the vegetative

phase (fust double ridge) and a qualitative response thereafter (before producing a terminal spikelet). Thus there is considerable variation among genotypes in the response to

vernalization that can show itself differently in the duration of both pre-and post-double

ridge phase and that, while the effecs are commonly greater on the earlier phase, the relative

effecs on the later phase can be smaller or greater.

Water stress and apical developmenf. Water stress is a pervasive feature of Mediterranean

environment (see 2.2.2). Although the probability of stress is greatest in spring and

surilner, periods of water stress can occur at any time during the growing season. There a¡e

three important stages to be considered when the general effects of water stress on barley

growth and development are discussed in relation to the specific problem of grain formation

and crop yield. The fi¡st phase is when the potential grain number is determined during

spike initiation and inflorescence development. The second phase is that of anthesis and

fertilisation, which determines the proportion of the potential grain number which are set'

and the third is the period of grain frlling.

Despite ttre importance of apical development during the vegetative phase (leaf primordium

initiation period), prior to floral initiation in cereals, the effect of water stress on this phase

and on the transition from the vegetative to the reproductive phase have received

considerably less attention. There is no clear evidence of any pronounced effects of water

srress on the final number of vegetative primordia. Nicholls and May (1963) working with 43 ix barley, found no effect on the number of vegetative primordia by water stress imposed ten days before spikelet initiation, whereas, it has been shown with sorghum that a moderate stress before spikelet initiation could decrease leaf number by up to three leaves (rù/hiteman and Wilson 1965). Oosterhuis and Carn¡¡right (1983) reported that there \ryas no significant effect of water stress on final leaf number of wheat compared to well-watered plants when the water stress was imposed at late vegetative phase just seven days before onset of spikelet

initiation phase. The same results \ilere reported by Frank et al. (L987) for spring wheat.

Previous research related spikelet number and grain yield to plant growth during the apex

vegetative phase. These researchers suggested that factors that stimulate gfowth such as

cooler temperature, increased N fertiliser and availability of water prolong the time required

to reach double-ridge stage and increased yietd potential (Rahman et al. 1977c; Frank and

Bauer 1982).

Temperature and apícal development. Temperature has been considered the main

environmental factor which determines the rate of crop development possibly because all

plants and processes of development Íue sensitive to it (Frank et al. 1978: Morrison ¿r ¿1.

1989). It is widely recognised that development accelerates as temperature increases, and

linea¡ relationships between the rate of development and mean temperature have frequently

been reported (Morrison et al. 1989). Genotyes vary signific,antly in their degree of

sensitivity to temperature (Halse and \üeir 1970). It has been found that high temperatures

shorten the leaf and spikelet initiation phases. High temperatures also accelerate the rate of

ear development resulting in a reduction in the f,rnal number of spikelets and hence the

number of grains (Warringtonet al. 1977; Frank et a|.1987). The increase in rate of

spikelet initiation is most rapid between2lo and 30o C. At higher temperatures the number

of spikelets initiated is reduced. Berween the appearance of double ridges on the wheat shoot

apex and emergence of the flag leaf (i.e. during floral development when the number of

florets is determined) no period of growth appears to be significantly more sensitive to high greater is temperture than any other but the longer the plants stay at a high temperaturc the

the reduction in grain number (Rawson and Bagga 1979). There is good evidence for a

reduction in grain number per ear associated with high temperature during the stage of rü/ardlaw booting (Saini and Aspinal lg82). Dawson and (1989) from their experiments 43x with 28 cultivars of wheat reported that exposure to high temperature during anthesis generally had a small effect on grain set. Some cultivars showed an increase in grain set, whereas, others showed a decrease. They concluded that part of this va¡iation may be related to the finding that under high temperatures, high humidity could reduce grain set. However, few attempts have been made to compare the effects of temperature during different stages of barley development from sowing to ear emergence, and its subsequent effect on final yield.

Nevertheless, temperature can have a significant effect on the size and number of the yield components which will affect a crops ability to respond to improved N fertility.

2.2.6.4 Effects of development on the pattern of dry matter accumulation

The development of a barley plant affects the pattem of accumulation and partitioning of dry matter. While the apex is in the vegetative stage, gowth is associated with the production of leaves and tillers, and dry matter production and its rate of increase is low. During this stage of the growrh cycle, the rate of dry matter production is affected by the amount of light intercepted, which in turn depends on the number and size of the leaves (Gallagher and

Biscoe 1978, Hay and Walker 19S9). Therefore the rate of leaf and tiller appearance g¡eatly affects the early vigour of the crop.

The time when dry matter production increases is partly affected by the time when the leaf canopy is sufficiently large to intercept most of the available light (French and Schultz

1984), but it is also related to the stage of development of the crop. The beginning of stem elongation in wheat generally coincides with the formation of ttre terminal spikelet (Hay and

Kirby 1991) and so the timing of this will also influence the pattern of dry matter accumulation. As well, the developing ear becomes an increasingly large sink for photosynthate (Fischer 1979) and so its development influences the partitioning of dry matter in the growing plant. Competition between the main stem and ears during this time, particularly if the plant is growing under water or nutrient stress, can become intense with a large number of the tillers senescing (Darwinket 1983, Hay and Kirby 1991). Therefore, variation in early vigour and in subsequent growth between genotypes, which may influence 43 xi

differences in the rates of development responses to N and water use effieicency' may reflect as well as inherent differences in photosynthetic production'

yield potential 2.2.6.5 The effects of the pattern of development on number of yield components: the eaf Grain yield in barley can be analyzedprimarily into a grains/eat' and the mean grain population density (number of ears/unit area), number of independently from the others' weight; each of which can, to a certain extent' vary between the yield components However, there is also a high degree of compensation

because of their sequential development'

of crop development' Ear The yield components afe determined at different stages but conditions around the time population density depends on the number of tillers formed' winter wheat, especially N supply) of terminal spikelet formation (in winter cereals such as to carry ears (George and Skinner determines the proportion of these tillers which survive cereals there would appear to be 1gg4; Kirby and Appleyard 1984). However in spring by the timing of N fertilisation (Hay much less scope for increasing ear population density

and Walker 1989).

per ear) is determined primarily during the The second component (i.e. number. of grains plant develops thfough an orderþ series period of spikelet initiation. The spike of the cereal genetic and environmental factors (Hay of morphogenic events, whose timing depends on the components of the spike's and Kirby 1991). These events determine, in sequence, number of kernels per spikelet' In potential yield capacity: the number of spikelets and the the yield potential per ear is therefore barley, which has only a single floret per spikelet, the proportion of these that a¡e fertile' The affected by the number of spikelets produced and period prior to differentiation of the potential spikelet number is determined during the period between spikelet initiation and the end terminal spikelet, and kernel number during the photoperiod' water stress and of anthesis. Environmental conditions such as temperature' may affect the expression of the yield nutrient stress, prevailing during a particular phase (Grieve et ø1. lggz). Although yield is directly component thar is dominant ar that time thus' is largely dependent on ea¡lier events and' determined after anthesis, the yield potential 43 xii storage capacity of the spike may respond to environmental conditions almost to maturity

(Evans et al.1975).

Finally, mean grain weight is determined primarily by the quantity of assimilate available for transport to the ear between anthesis and maturity. This, in turn, depends upon leaf area duration after anthesis, the photosynthetic activity of the ear, as well as source/sink relationships (Hay and Walker 1989). However, in seasons when leaf area duration is low

because of drought, for example, substantial amounts of assimilate stored in the stem and

leaf sheaths before anthesis can become available for grain filling (Lawlor et al. L98l).

There is an association between the development of the apex and the formation of the

components of yield. Hudson (1934) in his sea¡ch for stages which are most important in

building yietd potenrial in wheat found that the perid benveen terminal spikelet initiation and

anthesis was of paramount importance. Others have confirmed this association

experimentally (Fischer 1985; Siddique et aI.1989), and noted that this is because there is a

better relationship of grain yield to grain number per r# than to individual grain weight. Il

as suggested in many papers, there is association between particular developmental stages

and the yield components (Rawson and Bagga 1979) and if there a¡e associations between

the durations of phases and absolute yield (Craufurd and Ca¡twright 1989), then it becomes

important to be able to manipulate the duration of these phases to customise cultivars for

specific environments. Indeed, plant breeders have been doing this for centuries. 44 CHAPTER 3 RESPONSIVENESS OF BARLEY CULTIVARS TO NITROGEN

3.1. Introduction

There has been an increase in the use of N fertiliser in the production of malting barley across southern Australia as farmers try to increase their profitability by improving crop yields. In malting barley, the level of N in the grain is an important determining factor of quality (see section 2.3.3.2), and grain N concentration (GNC) is sensitive to high rates of

N fertiliser. A high grain N level is undesirable because there will be a lower carbohydrate content in the grain, which in turn means there will be less malt extract. Lockhard (1989) estimates that for each l%o increase in grain N there is as much as a 27o reduction in the amount of extract. A preferred level of N in the grain is considered to be no greater than

1.60-l.65Vo which relates to protein levels of 10.0-10.5Vo (Atherton 1989).

It is possible to increase grain yield and still maintain GNC within the desirable range, but the extensive affay of field experiments conducted with wheat and barley across southern

Australia indicates that responses in grain yield and protein are variable. Much of the variability is due to environmental effects, but little attention has been paid to genotypic variation. In malting barley grown in farming systems with improved N fertility, it is desirable to have cultivars that have a high grain yield response to N and a low response in

GNC. It is also desirable to have varieties that maintain kernel weight because grain size

(plumpness) is important in malting quality. Genetic variability in responses to N does exist: for example, Birch and Long (1990) found that grain protein concentration increased with added N (0-200 kgN/ha) and that there were different responses between cultivars.

Doyle and Kingston (1992) and Doyle et aI. (1993) also studied the effects of sowing rate and sowing date on yield, kernel weight and grain protein concentration in barley cultivars and found different responses between cultivars. Apart from this work, there is relatively little published information available on the range responsiveness among cultivars of malting barley in Australia, especially in southern Australia. Therefore, the experiments described in this Chapter were conducted to: 45 (i) characterise and quantify the grain yield and GNC response in a number of

different barley cultivars to various rates of N; and,

(ii) identiff possible causes of these differing responses.

Data on vegetative and reproductive responses were collected to quantify the response of the plants to N fertiliser and to examine the interactions between cultivars and N rate on plant growth, grain yield and GNC.

3.2. Materials and Methods 3.2.1. Sites

The experiments were conducted in South Australia at Northfield and Nuriootpa in 1990 and at Northfield and Charlick Experiment Station, Strathalbyn in 1991. The characteristics of the 4 sites used are given in Tables 3.1and3.2.

3.2.2. Treatments and experimental design

The experiments were conducted in the field using a split plot design with four replicates.

The treatments were:

(i) Six, 2-row malting barley cultivars: Clipper, Stirling, Weeah, Schooner, Chebec,

and Skiff.

(ii) Eight N levels: 0, 15, 30,45,60,'75,90, 105 kgN/ha.

The origin and characteristics of the barley cultivars are shown in Appendix 3.1. The six

cultivars are representative of malting barleys that have been grown throughout southern

Australia. The main plots were barley cultivars while levels of N were subplots. Nitrogen

was broadcast by machine as urea (467o N) 6 weeks after sowing. A basal dressing of

superphosphate (20 kgP/ha) was drilled with the seed at sowing.

3.2.3. Establishment of the experiment

Each cultivar was sown in rows 15 cm apart. In 1990 plot sizes were 8 rows wide by 12 m

long while in 1991 they were 10 rows by l0 m at Northfield and 10 rows by 5 m at

Charlick. The experiments were sown on 19 June (Northfield, 1990), 20 July (Nuriootpa, 46 1990), 17 June (Northfield, 1991) and 16 July (Charlick, 1991). In both years, prior to sowing, germination tests were made on 59 of seed and sowing rates were adjusted to give

145 seeds/m2. Seeds were sown at a depth of 25-35 mm and within 10 days 90Vo of. the seedlings had emerged.

Table 3. 1 . Characteristics of the sites used t'or experiments

Northfield Nuriootpa Northfield Charlick 1990 1990 1991 1991

Rainfall (mm) Annual 434 503 498 480 April-October 347 391 426 409

'Wheat Previous crops Lupin Wheat Barley Wheat Wheat Wheat Volunteer pasture

Soil type Silty Sandy Silty Red-brown black earth red-brown black earth earth earth

pH(CaClz) 0-20 cm 7.4 5.5 7.6 7.0 20-4O cm 7.6 6.1 7.9 7.6 Nitrate-N at sowing (ttglg) 0-20 cm 15.1 16.8 t9.9 18.7 2O-4O cm t7.2 19.9 22.6 23.6

3,2.4. Measurements 3.2.4.1. Dry matter measurements

Early growth (10 weeks after sowing)

At 2 sites, Northfield and Charlick in 1991, plots were sampled at approximately 10 weeks

after sowing to assess early crop growth. Measurements were based on a quadrat size of.2

rows x 0.5 m at 2 locations per plot (total sample area 0.30 m2). Samples were taken at

Northfield in 1991 at Zadoks growth stage 21 (beginning of tillering) and at Charlick in

l99I at growth stage 30 (Zadok et aI. L974)). Tiller numbers were counted and plants were

dried for 24h at 80"C and dry weight measured. Table 3.2. Meteorological data atfour sites 1990 and 1991

Rainfall (mm/month) Mean temperature ('C) Month 1990 1991 Ave.a 1990 Ave.b lggl Ave.c 1990 l99l Ave. 1990 Ave.e 1991 Ave.e

Jan. 17.3 15.4 22 18.0 l8 30.6 20 24.3 23.7 23 21.4 2T 2t.l 2t Feb. 10.6 0.0 22 20.o 20 0.0 2t 22.7 23.4 23 21.3 2T 19.9 20 Mar. 0.8 9.7 29 24.0 24 9.4 25 22.6 20.2 2l 18.7 t9 17.7 t9 Apr. 17.2 32.2 50 43.0 43 27.4 39 18.0 17.7 l8 15.2 15 16.4 16 May. 13.2 lt.2 63 58.0 58 18.6 55 15.7 14.4 l5 16.7 12 12.6 13 Jun. 69.8 103.5 52 51.0 51 99.3 58 12.2 t3.9 t2 9.4s 10 13.5 ll Jul. 94.0 76.2 67 66.0 66 52.8 63 tt.4 tt.7 l1 8.8 9 10.4 l0 Aug. 82.7 64.4 65 65.0 65 10.5 60 tl.2 I 1.8 t2 9.1 10 I 1.3 l1 Sept. 40.4 69.2 51 59.0 59 70.2 53 15.1 9.4 I3 rr.2 ll 12.6 12 Oct. 29.4 69.2 47 49.O 49 70.2 45 16.8 t7.t 16 14.2 t4 15.6 15 Nov. 7.8 46.8 26 28.O 28 18.3 29 20.6 18.0 t9 15.5 t7 16.8 n Dec. 50.9 0.1 27 22.0 22 12.7 24 20.3 20.0 2l 16.2 t9 11.9 t9 Apri-Oct 346.7 425.9 395 391.0 391 409.0 337 a Average of21 years, b un"rug" of34 years, c average 125 years, d avetage of20 years, e average of29 years

À -l 48

Anthesis {olt,L The.number of tillers and dry matter production were estimated from a single randomly ^ selected quadrat (0.3 mz) from each plot. The sample was oven-dried to a constant weight at 80 'C and dry matter determined. N concentration was measured at 4 N rates (0, 30, 60,

90 kgN/ha) by the Kjeldahl method.

Maturity

The sampling technique utilised at anthesis was used to measure total above ground dry matter at maturity. Ear number in each quadrat was determined for each sample and grain yield determined after threshing the grain from the sample. Individual kernel weight was estimated from a sample of 100 grains from each quadrat sample. N concentration in grain and straw was determined using the Kjetdahl method. Grain and straw N content (gN/mz) were then calculated by multiplying dry weight and N concentration. Grain yield (t/ha) was determined at the end of experiment by harvesting the remaining plants in the plot using a small plot harvester which harvested eight rows; in 1990 the whole plot was harvested and in 1991 the outside border rows were discarded. Nitrogen harvest index (NHI) was calculated as the ratio of grain N content to total N content of above ground parts at maturity.

3.4.2.2. Nitrogen analysis

Total N concentration was measured using the Kjeldahl method (Appendix 3.1). Dried

material was ground to pass through I mm sieve and I g was used for analysis.

3.4.2.3. Statistical analysis

All data were analysed by analysis of variance. Treatment means rwere compared using the

protected LSD (Steel and Tonie 1960) procedure operating at the 5Vo and lVo level of

significance. In a number of cases, the data was variable, so to better quantify and compare

the response of different cultivars to N, a regression technique was used. Response curves

were determined for grain yield, GNC and dry matter at anthesis for each cultivar using N

rate as the independent variable. Five different models were compared: 49

Linear: y = a+bx

Quadratic: y = â*bx+cx2

Mitscherlich: Y = a-bcx ModifiedMitscherlich: Y= a-bcx-dx

SquareRoot: y = ¿4þ¡ço'5çsx where y is yield (t,/ha), GNC and dry matter at anthesis and x is the nitrogen rate (kgN/ha).

These models are illustrated in Fig. 3.2. of the Review of Literature. The chosen curves generally were those accounting for the highest percentage of variation (R2) and giving the lowest residual mean square (RMS) (Webster and Oliver 1990) (Appendix 3.3). For the sake of simplicity, an equation was chosen even through it may not have been the best model for a particular cultiva¡ if it was the most appropriate for the majority of the cultivars.

Where different models gave similar results, the simplest one was used. The quadratic model fitted the data for grain yield and dry matter production at anthesis most often, so it was decided to use this model for each set of data. The GNC equation was fitted by the linear model most consistently and consequently was used.

Using the fitted curves for dry matter and grain yield responses 4 parameters were calculated to characterise a cultivar's response to N:

(Ð the yield with zero applied N (YO);

(iÐ the yield response at 0 kgN/tra (the slope at N=0);

(üÐ the N required to achieve the maximum yield (Nsp¡), and;

(iv) the maximum yield level (yield at Nspli Yma,Ð.

To assess the relative importance of the different yield components, simple linear correlations were determined using multiple regression. It is recognised that where a relationship is curvilinear, simple linear correlations may not be the most appropriate.

However they allow the most important correlations to be identified.

3.2.5. Response Index (RI)

In malting barley it is desirable to have cultivars that have a high grain yield response and a

low grain N response (as long as GNC is greater then the minimum requirement). To 50 quantify the changes in grain yield and GNC that occurs with applications of N, a Response

Index (RI) was derived. The RI is defined as the ratio of yield response at N=0 to the protein response at N=0. Therefore when the yield response is high relative to the GNC response the RI will be high. The formula for response index is: o, _ Slope of GY response at N=0 kg/ha r\r - Slope of GNC at N=0 kg/ha

3.3. Results 3.3.1. Weather

During the growing season (April-October) Northfield 1991 had highest rainfall and

Northfield 1990 had the lowest. Rainfall in June, July and August at all sites was adequate

(Table 3.2) suggesting responses to N prior to anthesis were unlikely to have been affected an{h¿¡rs ard, by water stress. The amount of rainfall duringngrain filling (September-November) varied

considerably between sites; it was highest at Northfield 1991 (185 mm) and lowest at

Northfield 1990 (78 mm). The mean temperature was lower during winter at Nuriootpa

1990 and Northf,reld 1991 than the other two sites.

3.3.2. Level of production in the different experiments

Vegetative growth, grain yield and yield components were highest at Northfield in 1991, but

the GNC was lowest at this site (Table 3.3). Grain yield was lowest at Nuriootpa in 1990

and Charlick in 1991 while the highest GNC was obtained at Nuriootpa in 1990. There was

a general inverse relationship between site mean grain yield and GNC between different

sites.

3.3.3. Average responses to N fertiliser at each site

The average responses to N at each site (average of 6 cultivars) are shown in Fig. 3.1. Of

the 4 sites used, significant grain yield responses to N occurred at three sites-Northfield

1990, Northfield 1991 and Charlick in 1991 (Fig. 3.1). There was a significant Site x N

interaction because of these differences responses. The experiment at Nuriootpa did not

show a significant response to N in most of the variables measured (Fig. 3.1), nor was there

a Cultivar x N interaction (Table 3.5). The reason for this is not clear, because the rainfall 51 and nitrate at sowing were not very different from the other sites used (Tables 3.1 and 3.2).

The discussions to follow will therefore concentrate on the 3 responsive sites (Northfield

1990, Northfield 1991 and Charlick 1991).

Table 3.3. Average values of production of some factors at the four sites

Site Dry matter Grain Grain N Ears/m2 Kernels/m2 Kernel at anthesis yield concentration (x10-3¡ weight Gtrú) (t/ha) (vo) (mg)

1990 Northfield 459 2.9 r.99 438 6.57 38.4 Nuriootpa 378 2.6 2.52 461 5.86 40.9 1991 Northfield 674 3.3 r.77 52r I 1.89 36.1 Charlick 555 2.5 t.96 465 7.86 39.4

Table 3.4. Summaries of analysis of variance for responses to N in six cultivars of barley

Dry matter Grain Grain N Earslm2 Kernels/m2 Kernel at anthesis yield concentration (x10-3¡ weight GtÍ?) (t/ha) (vo) (mg)

Site *:F* *** *(** * *** ** Cultiva¡ t< *** x*t t<* NS *{<* ** Site x Cultivar * NS NS *{(* NS Nitrogen *x* *** *** {<** *** * {<:F

{<{€{< * * Site X N t<** *x* *** N x Cultivar NS * NS NS NS NS Site x N *t

xxx=P<0.005, xx=p40.01, x-P<0.05, NS=P>0.05 52

(a) (b) 800 o 1 000 C\I o o o E o o.t o) E o T o, 800 .ut I a¡, 600 o = I -c. 5 c o (tt C' E 600 (ú (ú oL Þ 400 ^ o (ú (ú E ^ 400 à E o oà 200 200 020 40 60 80 100 120 0 20 40 60 80 100 120 N rate (kg/ha) N rate (kg/ha)

(c) (d) 4.0 2.7

3.5 G'2.5o\ (ú o ; ! 'Eo ¿-.)^^ 3 .0 g o c '=o cI z.t c 2.5 o E o z 1.9 CI c o I 2.O E CI 1.7 o

1.5 1.5 0 20 40 60 80 100 120 020 40 60 80 100 120 N rate (kg/ha) N rate (kg/ha)

Figure 3.1. (a, b, c, d). Average response to N fertiliser in (a) dry matter at anthesis, (b) dry matter at maturity, (c) grain yield and (d) grain N concentration at 4 sites; Northfield 1990 (O), Nuriootpal990 (A), Northf,reld 1991(O) and Cha¡lick 1991 (f). Each point is the mean of 6 cultiva¡s and lines have been fitted by eye. 53

3.3.4. Varietal responses to N 3.3.4.1. Dry matter at 10 weeks (DMro)

Measurements at 10 weeks were taken to examine the importance of early vigour of crops to the grain yield response to N between cultivars. At both sites where measurements were made there was a large dry matter response to N, although significant responses sometimes occurred only after 15 kgN/ha or 30 kgN/ha were applied (Table 3.5, Fig. 3.2). On average, Weeah produced most dry matter and responded to N up to the highest rate, whereas the response of the other cultivars tended to plateau at N rates above 45 kgN/ha. In

Stirling and Skiff grain yield was positively correlated with early vigour (Table 3.10), while with Weeah at Northfield 1991 there was a negative correlation. Early vigour was not significantly correlated with yield in the remaining cultivars.

3.3.4.2. Dry matter at anthesis

Nitrogen fertiliser significantly increased dry matter production at all sites (Table 3.5, Fig.

at Nuriootpa was sfrgll 3.1a), although the response ,.or/,,,,1 or rttatvtr r'j

The fitted regressions of dry matter at anthesis (DMÐ on N level for the 3 responsive sites,

Northfield 1990, Northfield 1991 and Charlick 1991 (Table 3.8) are shown in Fig. 3.3.

The parameters derived from the regression equations for each cultivar are give in Table 3.6.

Additions of N increased dry matter production, although the responses varied considerably

between sites and cultivars. Clipper in both 1991 experiments and Chebec at Charlick could

not be described by the quadratic polynomial. There \ryere no consistent differences in the

values of the response parameter across the sites (Table 3.6).

3.3.4.3. Dry matter at maturity

At maturity, total dry matter production responded significantly to N at Northfield in 1990

and Charlick in 1991; the responses observed at anthesis at Nuriootpa in 1990 and

Northfield in 1991 were not evident at maturity. There was no interaction between Cultiva¡

and N for plant dry weight at maturity except at Northfield in 1991 (Table 3.5). Significant

differences in dry matter production by the cultivars were found at Northf,reld in both years 54 (Table 3.7,Fig.3.1). At Northfield in 1990, Weeah had the highest dry matter at maturity, but there were no significant differences between the other cultivars. At Northfield in 1991,

Stirling and Skiff had lower dry matter at maturity than the other cultivars, however the other cultivars did not differ in dry matter at maturity (Table 3.7).

Table 3.5. Summaries of analysis of variance of dry matter and tillers/m2 at 10 weeks (DMro, TNro), at anthesis (DMa, TN¿) and dry matter at maturity (DMrn)

DMro TNto Dtvfâ TN¿ DMm

Northfield 1990 Cultivar NS NS * Nitrogen *r<{< ** *** N x Cultivar NS NS NS Nuriootpa 1990 Cultivar NS NS NS Nitrogen **x NS NS N x Cultivar NS NS NS Northfteld 1991 Cultivar ** t< NS ** ** x** Nitrogen *** ** *t(* t<*{< NS N x Cultiva¡ NS NS * * {<* Charlick l99l Cultivar *** t<* NS ** NS Nitrogen ** t< *** {<** *** *tx N x Cultivar NS * * lc* NS

xx*-p<0.005, **=pq0.01, *=pq0.05, NS=P>0.05 55

Clipper Stirling 450 450 (\¡ 400 ô.t 400 E E o) 350 è, 350 C c .o 300 o 300 o o f 250 lrl a= 250 T E o I o- 200 T o- 200 I e a 150 t 150 o Ð (ú (tt E 100 oooo E 100 OOOOo à o o à o o o 50 Õ 50 0 0 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kg/ha)

Weeah Schooner 450 450

(\¡ 400 C\,I 400 E E o) 350 o) 350 I c T C, o 300 .o 300 T C) I o I f 250 J 250 T E a o e 200 o- 200 o. t o o q) o 150 o o 150 o (ú o ct E 100 o o É 100 oooo à à o o o 50 o 50 0 0 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kgiha)

Chebec skiff 450 450 î 400 C\I 400 I I E IIT E Þ 3s0 g) 350 c .i soo .o 300 C) E zso 250 E Eo f zoo o- 200 q o 150 o 150 o o OO (ú (d o E 100 ooo E 100 ooo à o o öso o 50 0 0 0 15 30 45 60 75 90 105 01530456075 90 105 N rate (kg/ha) N rate (kg/ha)

Figure 3.2. The effect of N fertiliser on the dry matter production after ten weeks in six cultivars of barley at Northfield in 1991 (O) and Charlick in 1991(D. 56

Table 3.6. Dry matter response at anthesis to N of six barley cultivars in 1990 and 1991 as estimated by quadratic regression

Cultiva¡ Ye Ymax Nopt. Slope at B2 Signif @trû) GlrrP) (kgN/ha) 0 kgN/ha lgmÐl(kgN/ha¡

Northfield 1990 Clipper 336 537 82 4.89 0.82 * Stirling 367 692 260 2.50 0.83 t< V/eeah 307 543 87 5.44 0.95 ** Schooner 319 888 382 2.98 0.gg *r. Chebec 286 510 97 4.62 0.72 * skiff 266 565 r70 3.51 0.79 t(

Northfield 1991 Clippera 652 -0.13 NS Stirling 413 629 80 5.37 o.72 * V/eeah 619 807 118 3.20 0.51 * Schooner 594 802 55 7.57 0.85 t<* Chebec a 597 t.r2 o.64 * * skiff 504 798 225 2.6t 0.64

Charlick l99l 329 608 19 t.01 NS Clipper * Stirling 372 534 rzt 2.67 0.61 V/eeatt 3t3 677 87 8.35 0.54 {< * Schooner a 539 1.16 0.53 523 0.45 NS Chebeca * skiff 389 63r 79 6.r3 0.52

a Fitted equations did not have a maximum yield and only Yg and slope are shown (see Fig. 3.3) *x-P<0.01, x=P<0.05, NS=P>0.05 57

C pper St rling 000 1 000 N ôl E E ct) o gl 800 800 .9 ol .9, at) Ø o) o -c. o o L C (ú 600 U (ú 600 (ú (d L o o o ! cl 400 (tt E E o à à o o 200 200 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kgiha)

Weeah Schooner

000 1 000 C\¡ o ol E E o o) È, 800 800 .a o (rt o o .g o o oU' ! c o c (ú (5 r (ú a (ú L o (¡) Þ a (ú 400 (d 400 o E E Ð à o o o 200 200 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kg/ha)

Chebec skitf

1 000 1 000 $t (\l E E o) d) 800 o 800 .u, I .9 U, (t, o o (¡) o ! c o (ú (ú 600 OI a G' (ú o o a o o ! (ú 400 o at o E E à à o o 200 200 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kg/ha)

Figure 3.3. The effect of N rate on the dry matter production at anthesis of six cultivars of barley at Northfield in 1990 (O), Northfield in 1991 (O) and Charlick in 1991 (D. 58

Table 3.7. The main effect of Cultivar on dry matter at maturity, earshû, kernels/m2, kernel weight, grain N concentration and grain yield at four sites

Year/Site Clipper Stirling V/eeah Schooner Chebec Skiff LSD (57o)

1990 Dry matter at maturity (gl^2)

Northfield 5s9 515 654 500 560 495 9I

Nuriootpa 535 539 484 52t 481 515 NS

r991

Northfield 946 694 947 1028 915 762 t69 Charlick 730 670 586 762 '159 800 NS

1990 Grain yield (t/ha)

Northfield 2.87 2.96 2.82 2.75 2.96 2.83 NS

Nuriootpa 2.69 2.65 2.45 2.50 2.53 2.65 NS

1991

Northheld 3.r9 3.15 2.84 3.52 3.51 3.56 0.4r Charlick 1.98 2.59 r.'75 2.64 2.75 3.02 0.83

1990 Ears/m2

Northfield 468 428 482 377 45t 422 NS

Nuriootpa 489 466 4t2 470 458 410 NS

1991

Northfield 5t4 497 475 509 557 566 NS Charlick 4r6 422 375 498 520 562 42 r990 Kernels/m2 $r ro-r) Northfield 6.15 6.62 6.84 5.7 t 6.32 7.50 NS Nuriootpa 5.85 5.97 5.53 5.61 5.85 6.35 NS

1991 Northfield 11.41 10.92 11.15 12.31 13.16 t2.t9 NS Cha¡lick 7.O7 8.22 6.02 8.03 7.95 9.86 NS

1990 Kernel weight (mg) Norttrfield 39.9 37.2 38.4 38.5 38.9 37.3 0.7 Nuriootpa 42.5 41.9 40.3 41.0 40.1 40. r 0.9 I99I Northf,reld 39.3 33.7 38.6 37.2 37.4 33.8 t.7 Cha¡lick 41.8 38.5 36.8 41.4 38.8 39.4 NS 1990 Grain N concentration (%) Northfield 1.98 1.99 2.02 2.02 1.95 2.03 NS Nuriootpa 2.53 2.5r 2.52 2.57 2.51 2.51 NS

1991

Northfield t.73 1.89 1.70 1.75 t.'t5 1.78 NS

Cha¡lick 1.95 1.99 2.17 2.rl 2.16 l.8l NS

NS=P>0.05 59 3.3.4.4. Grain yield

There were significant differences in grain yield between cultivars only at Northfield and

Charlick in 1991 (Table 3.7). At Northfield in 1991 Weeah had the lowest mean grain yield, but there was no significant difference between the other cultivars (Table 3.7). At

Charlick in 1991 Weeah also had the lowest mean grain yield and its yield was significantly different from all cultivars except Clipper, while there was no difference between Skiff,

Chebec, Schooner and Stirling.

The fitted regressions for grain yield are shown for Northfield 1990 and 1991, and Charlick

1991 in Fig. 3.4 and the estimated parameters for these curves are given in Table 3.9. The data were often variable and in a number of cases this has resulted in non-significant relationships between N and grain yield. The variability in the data and the consequent poor relationship between N and yield affects comparison between cultivars. Nevertheless, Skiff consistently shows a large response to N at 0 kgN/ha; its grain yield increased by 30-43 kg/kgN which is greater than that of the other cultivars. The maximum predicted yield of

Skiff (Ym¿¡)at Northfield 1991 and Charlick 1991 tended to be greater than the other cultivars and equal to the others at Northfield 1990 (Table 3.9).

There was often a positive relationship between DM¿ and grain yield particularly at

Northfield 1990 (Table 3.10), although there was some variation between sites and cultivars (Fig. 3.5). Yields at Charlick were lower than those at Northfield, 1990, 1991 for

comparable amounts of DMa for all cultivars except Stirling. Therefore, regressions were

fitted to the Northfield data and the Charlick data separately (Fig. 3.5). Grain yields

increased with increases in DM¿ at Northfield. At Charlick there was no significant 'Weeah relationship betweenDM¿ and yield for any cultivar except Skiff. showed relatively

little increase in yield with DM¿, and showed no increase beyond a DM¿ of about 500 g/m2.

In contrast, the yields of Skiff and Stirling increased substantially with improvements in

DMa. The yield responses to DM¿ of Clipper, Schooner and Weeah to DM¿ were generally

similar. 60

Clipper Stirling 4.5 4.5 4.0 4.0 3.5 o 3.5 a (ú ^ct o -Ê. 3.0 5 3'0 E 2.5 o E 2.5 o '=q) 2.O 2.0 L 'Ec o E 1.5 1.5 CI o 1.0 t.o 0.5 0.5 0.0 0.0 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kg/ha)

Weeah Schooner 4.5 4.5 4.O 4.0 o 3.5 3.5 (ú (Il ! 3.0 3.0 Þ 2.5 I 2.5 o '=o) '= 2.O 2.0 .c I 'õc (õ L 1.5 L 1.5 o CI 1,0 1.0 0.5 0.5 0.0 0.0 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kg/ha)

Chebec skff 4.5 4.5 o 4.0 o 4.0 3.5 3.5 ^(ú ^(6 5 3.0 € 3.0 o 2.5 p 2.5 c) '=o '= 2.0 2.0 'Ec 'Ec r.s 1.5 CI " t.o t.o 0.5 0.5 0.0 0.0 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kg/ha)

Figure 3.4. The grain yield responses of six cultivars of barley to N fertiliser at Northfield, 1990 (O), Northfield, 1991 (O), and Cha¡lick, 1991 (f). 61 Table 3.8. Summary of analysis of variance for grain yield and yield component at 3 sites

Parameters Grain Ears/m2 Kernel Kernels/m2 Yield weight

Northfield 199t Cultivar NS NS *** NS Nitrogen *** *** {

Northfield 1991 Cultivar ** NS ** ,< Nitrogen ** *** {

Charlick I99l Cultivar * ** NS NS Nitrogen *** *** * * N x Cultivar NS ** NS NS x*{<=P<0.005, *x=pq0.01, *=P<0.05, NS=P>0.05

Table 3.9. Grain yield response to N fertiliser rate of barley cultivars in 1990 and 1991 as estimated by quadratic regression

Cultiva¡ Ys Ymær Slope at P2 Signif. (lha) (t/ha) 0 kg/Ì.{/ha (kg/kgN)

Northfield 1990 Clipper 2.41 3.1 74 17.5 0.53 NS Stirling )1) 3.3 92 23.6 0.48 NS Weeah 2.37 3.0 74 t7.4 0.87 ** Schooner 2.35 3.2 t87 9.5 0.80 ,< Chebec 2.41 3.5 161 13.6 0.91 ** skiff 1.88 3.1 76 33.r 0.96 t< *( Northfield l99l Clipper 2.89 3.4 60 15.6 0.64 NS Stirling 2.76 3.4 105 11.1 0.73 * V/eeah 2.85 3.0 36 8.4 0.80 * Schooner 3.44 3.6 50 5.8 0,08 NS Chebec 3.11 3.7 77 14.7 0.64 NS skiff 2.57 4.0 66 43.3 0.81 *

Charlick I99l Clipper I 69 2,1 6l 14.4 0.45 NS d< Stirling 1 94 2.9 66 28.6 0.81 V/eeah 1 36 r.9 64 t7.9 0.39 NS Schooner 2 35 2.8 58 5.6 0.40 NS Chebec 2 4t 2.9 80 12.o 0.56 NS skiff 2 23 3.4 75 30.0 0.87 **

x*=P<0.01, *=P<0.05, NS=P>0.05 62

Clipper Stirling 4.0 4.0 Northfield Northfield 3.5 o 3.5 o 3.0 a 3.0 (ú o ! ! o > rà 2.5 o !t E E z.o 'ao) 2.O TI .tf T r.u o 'õc 1.5 I oE o 1.0 t.o 0.5 0.5 0 0 1000 0 200 400 600 800 1000 0 200 400 600 800 (glm2) Dry matter at anthesis (g/m2) Dry matter at anthesis

Weeah Schooner Northfield 4.0 4.O o 3.5 3.5 3.0 o a 3.0 (ú T ! ! I > rà o 2.5 I o t'jr E O T E z.o I '=o 2.0 lrl .g ñ 1.5 Ëtu (J o t,o 1.0 0.5 0.5 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 Dry matter at anthesis (glm2) Dry matter at anthesis (glm2)

Chebec Northfield skiff o Northfield 4.0 o 4.0 oo 3.5 3.5 o Charlick 3.0 (ú 3.0 o -c ! 2.5 orl 2.5 io I '=o 2.0 .9 2.O o 'õ 1.5 'õc 1.5 o 1.0 o 1.0 0.5 0.5 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 Dry matter at anthesis (g/m2) Dry matter at anthesis (glm2)

grain yield ald mgfter production at anthesis Figure 3.5. The relationship between -d¡y foí six cultivars of barley ai three sites in 1990 and 1991. The sites are Northfield in 1990 (O), Northfield in 1991 (O) and Charlick in 1991(D. 63 3.3.4.5. Yield components

A correlation matrix was used to determine any relationship between grain yield and yield components and the simple correlation coefficients are shown in Table 3.10. V/hen sites were combined there was positive relationship between grain yield and ears/m2, kernels/m2 and a negative correlation with kernel weight.

Ear nutnber

Ear number (ENo) responded significantly and positively to added N up to the highest rate of N at the three responsive sites (Northfield in 1990 and 1991 and Charlick in 1991). For many cultivars there was a significant linear correlation between grain yield and ears/m2, although this was influenced by site (Table 3.10, Fig. 3.6).

The number of ears/m2 at Northfield and Charlick in 1991 was higher than at Northfield in

1990 (Table 3.3). Ear number differed between barley cultiva¡s at Northfield in 1990, 1991 and Charlick in 1991 (Table 3.7). At Northfield 1990 and 1991 the cultivars Skifl Stirling,

Chebec and Schooner showed a significant correlation between grain yield and ear/rû. Ãt

Charlick, 1991 only the grain yield of Skiff was correlated with ears/m2 (Table 3.10). The importance of ears/m2 to yield differed between sites and cultivars (Fig. 3.6). Except for

Skiff, ear number had no influence on yield at Charlick. At Northfield yields increased significantly with increases in ears/m2 for all the cultivars except Weeah (Fig. 3.6). The responsiveness of yield to ears/m2 did not differ significantly between cultivars (Appendix

Table 3.6).

Kernel number

There were only relatively small increases in the number of kernels/m2 1trNo; at Nuriootpa

and Northfield (1991) where the effect of N was non-significant (Table 3.7). There was no

significant difference between cultivars in the number of kernels/m2 except at Northfield

1990 when Skiff and Stirling set more kernels/m2 (Table 3.7). There is also a general

positive relationship between dry matter at anthesis and kemel number for all cultivars (Fig.

3.7). The slope of this relationship is greater for Skiff than for the other cultivars, showing

that with Skiff, more kernels/m2 will be produced as vegetative growth ln"r"urJTnppendix / 64 Table 3.5). The relationship between yield and kernels/m2 varied between sites and cultivars. At Charlick increases in kernels/m2 had no significant effect on grain yield (Fig.

3.8), a result that reflected the relationship between yield and earclm2 (Fig. 3,6). At

Northfield in 1990 and 1991 there was a positive response to kernels/m2 in all the cultivars except Weeah. There were linear relationships for Stirling, Chebec and Skiff, while in

Clipper and Schooner the response was curvilinear. The predicated maximum yield for

Clipper occurred at 10,600 kernels/m2 and for Schooner at 14,000 kernels/m2. The responsiveness of Skiff was significantly greater than that of Stirling and Chebec (Appendix

Table 3.6).

Kernel weight

Kernel weight (KWÐ declined as the N rate increased at all sites. Kernel weights were lower at the higher yielding sites (Table 3.3). At all sites and in all cultivars, kernel weight fell with the addition of N fertiliser (Fig. 3.9). Linear regressions described most of the responses and these have been fitted to the data (Table 3.11). Variation in mean kernel weight between sites is clearly evident and this is associated with differences in the number of kernels/m2 lTable 3.3). There were few significant differences between cultivars in the responses in kernel weight to N. At Northfield, 1990 the decline in kernel weight in Stirling was greater than that of the other cultivars, which were not significantly different from one another. In 1991, the kernel weight of Chebec declined less than that of the other cultivars at Northfield, while at Charlick the kernel weight of Skiff was reduced less than that of

Schooner. 65

Table 3.10. The simple linear correlation coefficients between grain yield, yield components and grain N concentration

Cultivar DM10 aTN¿ DN{a ENo. KNo. KV/t. GNC

Northfield 1990 Clipper 0.73* 0.62 0.69 -0.53 0.60 Stirling 0.90** 0.71* o.79* -0.29 0.51 Weeah 0.75x 0.24 0.r2 -0.58 0.27 Schooner 0.84** 0.74x 0.62 -0.91** 0.83* Chebec 0.88r* 0.82x 0.50 -0.64 0.87** skiff 0.77* 0.93** 0.90*x -0.86*x o.64 Norffield l99I Clipper 0.26 0.01 0.20 0.48 -0.28 -0.06 0.16 Stirling 0.81x 0.71* o.67 0.77* 0.06 -0.69 0.58 Weeah -0.71* -0.47 -0.51 -0.48 -0.32 0.61 -0.57 Schooner -0.I7 o.42 0.53 0.30 0.33 -0.02 0.32 Chebec 0.41 0.42 0.50 0.78x 0.63 -0.23 0.39 skiff 0.79* 0.gg*x 0.32 0.91** 0.70* -0.56 -0.15

Chnrlick 1991 Clipper 0.40 0.33 0.43 0.18 0.58 -0.03 0.29 Stirling 0.74* 0.08 0.41 -0.18 0.63 -0.70* 0.32 Weeah 0.40 0.12 0.20 -0.01 0.48 -0.36 0.35 Schooner -0.09 0.40 -0.28 -0.15 0.53 0.54 -0.69 Chebec 0.64 0.50 0.64 0.22 0.34 -0.61 0.39 skiff 0.84x 0.81** o.77** 0.96** o.72* -0.68 -0.01

AII sites Clipper 0.46* 0.66** 0.55** -0.51x -0.2r Stirling 0.70x* 0.51* 0.53** -0.60x* 0.22 Weeah 0.19 0.40 0.53** -0.14 -0.31 Schooner 0.41* 0.29 0.71** -0.03 -0.59** Chebec 0.67x* 0.46* 0.74** -0.71*x -0.20 skiff 0.71** 0.96** 0.83** -0.71x* -0.09

**=P<0.01, x=P<0.05, NS=P>0.05, aTNa=Tiller number at anthesis 66 Clipper Stirling 4.0 Northfield 4.O Northfield 3.5 o 3.5 o o 3.0 (ú 3.0 o (ú .c, o ! 2.5 o 2.5 E E o 2.0 o 2.O o c 1.5 c 1.5 E E (, 1.0 CI 1.0 0.5 0.5 0 0 0 200 400 600 800 0 200 400 600 800 1000 Ear number /m2 Ear number /m2

Schooner Weeah Northfield 4.O 4.0 o 3.5 3.5 o o a a 3.0 a 3.0 -c å3P" Ð 2.5 " > 2Ã E la o OT 2.0 z.o .E I .E .ç 1.5 .s 1.5 (s T s õ r.o (t l.o 0.5 0.5 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 Ear number /m2 Ear number /m2

Northfield Chebec Skiff o 4.0 Northfield 4.0 o 3.5 3.5 3.0 G'3'0 lrl (ú .c. -c Ð 2.5 û 2.5 E 9 E 2.0 o 2.0 Charlick .ç 1.5 c. 1.5 g E o t.o CI 1.0 0.5 0.5 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 Ear number /m2 Ear number /m2

Figure 3.6. The relationship between grain yield and ear number for six cultivars of barley at three sites in 1990 and 1991. The sites are Northfield in 1990 (O), Northfield in 1991 (O) and Charlick in 1991(D. 67

Clipper Stirling 16 16 o Northfield ot 14 Northfield ñ14c. E o È12 o o 12 rb o o o 3lo o o 10 ol (l)^L Charlick q) -oö -o I E E e6 cf 6 a õ 4 o O e4o o Y2 2

0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 (glm2) Dry matter at anthesis (g/m2) Dry matter at anthesis

Weeah Schooner 16 Northfield 16 14 14 o C! ôI ooo E E 12 12 o oo o o 10 I 10 I (¡) o I -o I o c I E I E l c b c 6 o t I õ 4 co) 4 o o \¿ 2 Y 2 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 Dry matter at anthesis (glm2) Dry matter at anthesis (glmz)

Chebec skiff 16 16 Northfield Northfield ñ14 ot 14 oo ç o E 512 ò 12 o O f o Oo a-' 10 o 9lo O T 3B ¡¡o) I E E o o 6 a a c= =6 o)c 4 C4q) o a Y2 Y 2

0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 Dry matter at anthesis (g/m2) Dry matter at anthesis (glm2)

Figure 3.7. The relationship between dry matter production at anthesis and the number of kernels/m2 for six cultivars of barley at Northfield in 1990 (O), Northfreld in 1991 (O) and Charlick in 1991 (f). 68

Glipper Stirling 4.5 4.5 Northfield 4.O Northfield 4.0 o 3.5 o 3.5 ^ct ^G' o 3.0 3.0 o o s a 5 z.s p 2.5 Iq) l- '=c) t+..f '12.0 2.0 L L 'E 1.s a E 1.s o o r.o r.o 0.5 0.5 0 0 15 0 5 10 15 0 510 ('000/m2) Kernel number ('000/m2) Kernel number

Weeah Schooner 4.5 4.5 Northfield 4.O 4.O 3.5 3.5 (ú ^(ú ! 3,0 3.0 o o ç" s I t 9 2.5 rl 8so ã z.s '=o) o) a 2.O 'a 2.0 'õc, Ër c 1.5 I 1.5 o I E 1.0 o r.o 0.5 0.5 0 0 0 5 10 15 0 5 10 15 Kernel number ('000/m2) Kernel number ('000/m2) sk¡ff Chebec Northfield 4.5 Northfield 4.5 4.0 4.0 o 3.5 3.5 o (It a ^(It ! a 3.0 € 3.0 Charlick E 2.5 tfr s 2.5 o) '12.Oo '= 2.O 'õc 'EL 1.5 1.5 (, o 1,0 t.o 0.5 0.5 0 0 0 510 15 0 5 10 15 Kernel number ('000/m2) Kernel number ('000/m2)

Figure 3.8. The relationship between the number of kernels per m2 and grain yield foi six cultivars of barley ãt Northfield in 1990 (O), Northfield in 1991 (O) and Charlick in 1991(D. 69

Clipper Stirling 50.0 50.0

o) 45.0 o) 45.0 E E

! .9 .9 a o 40. o o 40.0 I o 3 = o o õq o c o L o o ¡ Y 35.0 Y 35. 0 oa

30.0 30.0 0 15 30 45 60 75 90 105 0 't5 30 45 60 75 90 105 N rate (kg/ha) N rate (kgiha)

Weeah Schooner 50.0 s0.0

q) o) 45.0 45.0 E I E T -c .9 c o, o 40. o 40.0 a = o)= o E Lc o o 35.0 \¿ 3s.0 Y I 30.0 30.0 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kg/ha)

Chebec skiff 50.0 50.0

o) 45.0 o) 45.0 I E E I ! o) o) 'õ I 'õ 40.0 40.0 3 I õ= õ E c c. I o o o) Y 35.0 v 35.0 o

30.0 30.0 0 15 30 45 60 75 90 '105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kg/ha)

Figure 3.9. The effect of N rate on the kernel weight of six cultivars of barley grówn at Northfield in 1990 (O), Northfield in 1991 (O) and Charlick in 1991 (l). 70 3.3.4.6. Grain nitrogen concentration

Average GNC increased with application of N fertiliser at all sites (Table 3.I2,Fi9.3.1d).

In contrast to the strong influence of N on GNC, there was no significant difference between the GNC of the six cultivars for GNC at all sites (Table 3.7).

The response to N fertiliser in GNC was linear in all cases where significant correlations occurred (Fig. 3.10). The levels of GNC for all cultivars were above l.65Vo N which is in the adequate range for malting quality suggested by Atherton (1989). At Northfield 1990

Chebec had the lowest GNC response but there was no significant differences between the other cultivars. At Charlick 1991 V/eeah and Schooner had a significant higher response than Clipper. Comparison of the slope for grain yield against N at N=0 (Table 3.9) with the

GNC response (Table 3.13) indicates that grain yield and GNC are not well correlated (Fig.

3.1 1).

Table 3.11 The intercept and slope of the regression between kernel weight and N rate at three sites

Cultivar lntercept Slope * se p2 Signif.

Northfield 1990 Clipper 4r.5 -0.032 + 0.019 0.54 * Stirling 40.1 -0.056 + 0.012 0.78 ** * Weeah 40.7 -0.044 + 0.012 0.69 Schooner 40.5 -0.039 + 0.013 0.59 t< Chebec 39.9 -0.020 r 0.009 0.42 NS skiff 39.6 -0.043 + 0.009 0.81 **

Northfield 1991 Clipper 41.3 -0.038 r 0.018 0.43 NS Stirling 36.4 -0.083 + 0.013 0.74 ** V/eeah 41.3 -0.051 + 0.017 0.60 * Schooner 39.7 -0.043 + 0.006 0.90 *r< * Chebec 38.8 -0.021 + 0.007 0.57 skiff 36.9 -0.060 + 0.008 0.89 **

Chnrlick 1991 Clipper 42.7 -0.015 + 0.026 0.05 NS Stirling 39.4 -0.017 + 0.018 0.11 NS Weeah 43.5 -0.088 r 0.036 0.50 NS Schooner 43.4 -0.089 + 0.036 0.51 * Chebec 43.2 -0.035 + 0.028 0.21 NS skiff 41.3 -0.038 + 0.013 0.59 *

x*=P<0.01, x=P<0.05, NS=P>0.05 7T Table 3.12. Summary of analysis of variance for grain N concentration, grain N yield and N harvest index of six barley cultiva¡s at 3 sites

Parameter GNCA GNYA NHIA

Northfield 1990 Cultivar NS NS NS Nitrogen NS *** *** N x Cultivar NS * NS

Nuriootpa 1990 Cultiva¡ NS NS NS Nitrogen * *{< NS * *:ß N x Cultivar NS NS NS

Northfíeld 1991 Cultiva¡ NS NS * Nitrogen ** t< NS t

Charlíck 1991 Cultivar NS NS NS Nitrogen ** t< * :ß** N x Cultivar * NS NS a GNC=Grain N concentration, GNY=Grain N yield, NHl=Nitrogen harvest index *x*=P<0.005, **-P<0.01, *=P<0.05, NS=P>0.05

Table 3.13. Relationship between grain N concentration and N rate for six barley cultivars at 3 sites

Cultivar Intercept(Vo) Slope (b)x10-3 + se p2 Signif.

Norffield 1990 Clipper t.70 5.34 + 0.63 0.93 ** Stirling 1.73 4.97 + O.45 0.96 ** Weeatt 1.81 4.O2 + t.44 0.56 * Schooner t.77 4.79 + 0.76 0.87 ** Chebec r.78 3.21 r 0.86 0.70 *{< skiff 1.75 5.56 + 0.87 0.87 *t€

Northfteld l99l Clipper 1.63 1.88 + 0.35 0.82 ** Stirling 1.81 t.6r + 0.77 0.43 NS Weeah 1.61 t.76 + 0.64 0.55 * Schooner 1.63 2.25 + r.t5 0.40 NS Chebec t.64 t.94 + 092 o.4t NS skiff r.75 0.70 + 1.06 0.07 NS

Charlick 1991 Clipper I 75 3.79 + r.09 0 66 {< Stirling 1 85 2.75 + 4.67 0 06 NS Weeah 1 11.71 + 3.18 0 70 ** 58 ** Schooner 1 72 7.52 + 1.30 0 85 Chebec 1 83 5.99 r 3.95 0 33 NS skiff I 69 2.30 + 3.42 0 07 NS

**=P<0.01, *=P<0.05, NS=P>0.05 72

Clipper Stirling 3.0 3.0 * s c 2.5 u o Ë'r g (ú c c z.o I e.o LI c ()o ()o o z z .c 1.5 1.5 ((l .s L E t CI o 1.0 1.0 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kg/ha)

Weeah Schooner 3.0 3.0

òe * L 2.5 .9 in''u (ú a E c L o 2 .0 z.o co cI o o () o o z o z o 'õc 1 5 .c 1.5 E (, CI

1.0 1.0 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kg/ha)

Chebec skitf 3.0 T 3.0 s s L 2.5 c 2.5 .o .Eo ct T (ú c L o a z.o o 2.0 cI c o o o a z o o z c 5 .c 1.5 'õ (d L L o CI

1.0 1.0 0 15 30 45 60 75 90 105 0 15 30 45 60 75 90 105 N rate (kg/ha) N rate (kg/ha)

Figure 3.10. The effect of N fertiliser rate on the grain N concentration in six cultivars of barley at Northfield in 1990 (O), Northfield in 1991 (O) and Charlick in leel (D. 73

Clipper Stirling 3.0 3.0

E òe c 2.5 c 2.5 .Fo .o (ú a ct o I o L IO c o o I z.o I o o 2.0 C, I c I o o o o () .d. €y z o ?"9. z .s 1.5 c 1 .5 g E o o T 1.0 1.0 1.5 2 2.5 3 3.5 4 4.5 1 1.5 2 2.5 3 3.5 4 4.5 Grain yield (Vha) Grain yield (Uha)

Weeah Schooner 3.0 3.0

òS òS I T 2.5 L 2.5 ll 'Fo .o (tt I (ú T L Ê. q) o 2.0 o 2.0 o o L o oL T o C) o a z z o sg t L c 1 .5 I 'õ 1 .5 o oE o 1.0 1.0 1 1.5 2 2.5 3 3.5 4 4.5 1.5 2 2.5 3 3.5 4 4.5 Grain yield (Vha) Grain yield (Uha)

Chebec skiff 3.0 3.0

-o o\ òS c 2.5 ,Fo ãru at (E E c c o I z.o I z.o ¡ Lo c o o o a ooooo z z 'f,9l .c 1.5 .ç 1.5 (ú g (, (t I

1.0 1.0 1 1.5 2 2.5 3 3.5 4 4.5 1 1.s 2 2.5 3 3.5 4 4.5 Grain yield (Uha) Grain yield (Vha)

Figure 3.11. The relationship between grain N concentration and grain yield for six cultivars of barley at Northfield in 1990 (O), Northfield in 1991 (O) and Charlick in 1991 (I). 74

3.3.4.7. Response index

The Response Index (RI) was derived to integrate the effect of N on the changes in grain yield and GNC. A high RI indicates a higher grain yield response relative to the GNC increase, The values of this ratio indicate/ natthere a¡e differences between cultivars across experiments: Skiff and Stirling have high values, Weeah and Schooner had low values, and

Chebec and Clipper were intermediate (Table 3.I4). The differences in R[ between cultiva¡s tends to be more a reflection of the grain yield response rather than the GNC response.

Table 3.14. Mean responsiveness index at 3 sites

Cultivar F-I(tlhalVo) Northfield Northfield Charlick Mean * se 1990 t99r r99r

Clipper 3.28 8.30 3.79 5.t2 + 1.59 Stirling 4.75 6.89 r0.47 7.35 + t.67 Weeatr 4.33 4.77 1.53 3.54 + 1.05 Schooner 1.98 2.58 0.74 r.77 + 0.05 Chebec 4.24 7.58 2.00 4.6t + r.62 skiff 5.95 61.86 13.04 26.95 + 17.58

3.3.4.8. Nitrogen harvest index

In these experiments nitrogen harvest index (NlÐ declined at higher rates of applied N at all

sites (Table 3.15). Significant Cultivar x N interaction were seen only at Northfield in

199I. NHI was reduced significantly only in Stirling and Chebec but it did not much

change in Schooner, Skiff and Clipper (Appendix Fig. 3.1).

3.3.4.9. Grain N concentration and total shoot N at anthesis

For2 sites, Northfield 1990 and 199t, the amount of N in the crop at flowering and the

GNC were correlated (Fig.3.12, Table 3.16) while at Charlick there was no significant

relationship. However, for the same amount of N in the crop at anthesis, GNC was greater

at Northfield 1990 than at Northfield 1991. This difference possibly can be related to the

amount of rainfall during spring and its subsequent effect on grain filling. The 1990 spring

was considerably drier than that of 1991 with only 77.6 mm of rain compared with 185.2

mm (Table 3.1). 75

Table 3. 1 5. The main effects of Cultiva¡ and N on NHI at different sites

Cultivar Sites Clipper Stirling Weeah schooner Chebec Skiff LSD(57o)

r990 Northfield o.823 0.831 0.772 0.815 0.841 0.861 NS

Nuriootpa 0.822 0.81 1 0.821 0.826 0.817 0.838 NS 1991 Northf,reld o.776 0.750 0.769 0.815 0.801 0.774 NS Charlick 0.718 0.752 0.707 0.7r8 0.772 0.175 NS

Nitrogen (kgN/ha) 015 30 45 60 75 90 r05

1990 Northfield 0.833 0.836 0.842 0.847 0.808 0.840 0.789 0.794 NS Nuriootpa 0.840 0.839 0.833 0.828 0.824 0.817 0.805 0.793 0.027 1991 Northfield 0.823 0.798 0.797 0.818 0.759 0.762 0.770 0.729 0.032 Charlick 0.805 o.769 0.791 0.773 0.722 0.683 0.706 0.682 0.054

xx-P<0.01, x=P<0.05, NS=P>0.05

Table 3.16. The intercept and slope of the correlation between grain N concentration and total shoot N at anthesis for Northfield 1990 and Northfield 1991

Cultivar Intercept Slope p2 Signif. ToGNC (g/m2) t se

Northfíeld 1990 Clipper r.28 0.10 + 0.01 0 97 ** Stirling 1.15 0.12 + 0.05 0 78 t< Weeah 1.68 0.06 + 0.02 0 85 * Schooner t.3r 0.15 + 0.04 0 96 *,C Chebec t.29 0.12 + 0.03 0 92 ** skiff 1.38 0.10 t 0.02 0 94 **

Northfield 1991 Clipper t.26 0.05 r 0.01 0.94 ** Stirling 1.65 0.03 + 0.02 0.47 NS V/eeah r.43 0.04 t0.004 0.97 ** Schooner 0.76 0.11 + 0.03 0.90 ** Chebec r.42 0.04 + 0.04 0.70 NS skiff t.22 0.07 + 0.02 0.81 *

**=P<0.01, x=P<0.05, NS=P>0.05 76

This was reflected in the greater GNC in 1990, even when total N uptake by flowering was similar. The data indicated that some cultivars responded differently tci seasonal conditions and to the increased N taken up by anthesis.

As well as the effect of season, increasing N at anthesis increased GNC. The slope relating

GNC with total shoot N at anthesis was high for Schooner in both years (Table 3.16). The other cultivars were intermediate. Therefore Schooner appears to be a cultivar whose GNC is relatively more sensitive to increased N uptake than other cultivar. The slope of this relationship was not related to the yield responsiveness of the cultivar. For example Skiff and Clipper had similar slopes (Table 3.16) even though they showed quite different yield responses. This result suggests that there may be genetic differences in post anthesis growth and the mobilsation of N to the grain. There was no correlation between grain N yield and total shoot N at anthesis for barley cultivars at different sites.

3.4. Discussion geno\7ic -genetie The aims of these field experiments were to examine the level of variability to N among some current cultivars of malting barley and identify some of the causes of these responses. Of the 4 sites chosen a significant response in grain yield to N was obtained at

3, and the site optimum N varied from about 45 kgN/ha up to about 90 kgN/ha (Fig 3.1b).

The grain yield and GNC levels also varied between sites (Table 3.3). These differences in grain yields, GNC and optimum N rates are typical of fertiliser experiments in southern

Australia (Russell 1968a, 1968b; Xu et al.: 1991, 1992; Jefferies et al. I99O). However, the ability to discriminate yield and GNC responses to N between cultivars depends, in part,

on the general responsiveness of a site. The variability in yield and GNC makes field

evaluation of genetic differences difficult and the effects of seasonal conditions on response

of grain yield and GNC to N in Mediterranean type environments have been described in

Chapter 2 (2.2.4, 2.2.5). 77

Clipper Stirling 3.5 3.5

-o \o ù g.o Ð g.o c c .9 .9 ftl Northtield 90 Þ 2.5 Northfield 90 E z.s C. c o o o () c T c 8 2.0 I 2.0 o z z o L c 'õ Northfield 91 'õ 1.5 L 1.5 CI (5

1.0 1.0 051015 0 51015 Totalshoot N at anthesis (g/m2) Total shoot N at anthesis (g/m2)

Weeah Schooner 3.5 3.5

òe ò9 c 3.0 c 3.0 o I o Northfield 90 (ú Northfield 90 (ú 2.5 c 2.5 o) o co o o o o 2.0 o 2 .0 I z z t 'õc Northfield 91 'õE 1 .5 1 .5 Northfield 91 o CI 1.0 1.0 0 51015 0 5 10 15 Total shoot N at anthesis (gim2) Total shoot N at anthesis (g/m2)

Chebec skiff 3.5 3.5

rO òe Ð g.o 3.0 L É. .9 Northfield 90 .9 Northfield 90 ftl E Þc 2.5 c 2.5 o o () co c I o 8 2.0 o 2.0 z Io o z 'õc o L 1.5 E 5 Northfield 91 (, CI

1.0 1.0 0 51015 0 51015 Total shoot N at anthesis (g/m2) Totalshoot N at anthesis (g/m2)

Figure 3.I2. The relationship between the total amount of N in the shoot at anthesis and the grain N concentration for six varieties of barley at Northfield in 1990 (O), Northfield in 1991 (O) and Charlick in 1991(f) 78

High rates of N applied early in the season (at the time of tillering) resulted in increases in dry matter production (Fig. 3.2). This observation is consistent with other reports with wheat (Spiertz and van der Haar 1978; Ellen and Spiertz 1980). However, the increases in dry matter production early in the season were not correlated with grain yield for most cultivars except in Stirling and Skiff (Table 3.10). For Weeah at Northfield 1991, the large increases in growth were negatively related to yield (Table 3.10). Large amounts of dry matter production early in the growing season may not increase yields, or may even decrease yields if it results in depletion of soil moisture or increased lodging (Fischer 1979). This latter result is in agreement with the report of McDonald (1992).

The data from these experiments often did not strow sTy'nificant differences between // cultivars. Variability was often unacceptably large which obscured some apparent trends.

However, despite the variability in yield and protein responses between sites, there does appear to be some consistent genetic differences. Skiff showed the greatest grain yield response to N of the 6 cultivars used. It also appeared to be more responsive in the relationship between ear number and grain yield (Fig. 3.6), DMa and KNo. (Fig. 3'7) and 'Weeah in the importance of early growth (DMro) to yield (Table 3.10). In contrast, was a

cultivar which showed a much lower yield response (Table 3.9), and whose yield was not

sensitive to changes in ear number, kernel number or DM¿ (Figs. 3.6, 3.8). It tended to

show greater early response to N at the 2 sites where DMto was measured, but on one

occasion this greater response was negatively correlated with yield (Table 3.10). In

individual experiments, differences in yield responses were measured but frequently they

were not consistent across all 3 experiments. For example, Stirling sometimes showed

responses which were similar to these of Skiff (Table 3.9), but in many cases these were not

significantly different to those of the other cultivars because of the variability of the data.

However, the data show that the only reliable genetic differences were in these experiments

between Skiff and Weeah. The differences in responsiveness appear to be related to plant

height because Skiff and Stirling were shorter and more erect than Weeah and Schooner

which were tall cultivars (Plate 3.1). 79

N rolos

fhe allêct of ll on hdrl.y vnd¡llls !¡lh lúo lclcls

\

(

N rnlos

th!.rr!cr ol ll on h¡rl0y !iricr¡.t Y¿il¡ treo lrvclr

Plate 3. 1. Effect of N rate on the plant height of Skiff, Stirling Weeah and Schooner at Northfield in 1991. Yields of SkifT and Stirling were responsive to N while Weeah and Schooner were not responsive. 80 One of the consistent effects of applying N fertiliser was to reduce kernel weight (Fig. 3.9), which is not desirable for malting quality (see Chapter 2, section2.3.4.I). However the average kernel weight of Skiff and Stirling were significantly lower than that of the other cultivars at the 2 of. the 3 responsive sites (Table 3.7). As well, the variation in mean kernel weight between sites for Skiff and Stirling was greater than that of the Clipper, Weeah,

Schooner and Chebec. Across the 4 sites the mean kernel weight of Skiff ranged from 33.8 mg to 40.1 mg (6.3 mg) and Stirling by 8.2 mg, whereas the kernel weight of the other cultivars were only of the order of 3-4 mg. At Northfield in 1990 and 1991 the reduction in kernel weight with added N for Stirling and Skiff tended to be slightly greater than that of the cultivars, although not significantly (Table 3.11). However, these data suggest that the kernel weight of Skiff and Stirling may be lower and more sensitive to environmental conditions than that of the other cultivars examined. The smaller and more va¡iable weight

of the kernels in Skiff and Stirling can not be simply explained by differences in kernels/m2

because there was no relationship between the kernels/m2 produced by individual cultivars

and the kernel weight of the cultivars (Table 3.7). Therefore, the kernel weight

characteristics of Stirling and Skiff may be related to other features of the physiology of the

cultivars during the post anthesis period. Bidinger et aI. (1977) showed that post-anthesis

drought reduced kernel weight in wheat and barley and therefore the reductions in kernel

weight between barley cultivars in this experiment may be due to differences in the degree to

which levels of stress increased around flowering and particularly during grain filling in the

different cultivars. This is investigated further in Chapter 5.

Although there a¡e some genetic differences, the effect of N on kernel weight will be affected

mainly by environmental conditions during grain filling especially the amogút of available ('

moisture. Birch et al. (1993) found in irrigated barley that N applied at rates up to 90

kgN/ha did not reduce kernel weight at sites where grain yield responded to N (see their

Table 2); however, individual grain weight was reduced consistently by N rates greater than

90kgN/tra. They suggested that under irrigation conditions, higher rates of N under low N

conditions could be applied without reducing grain weight. In another experiment Birch and

Long (1990) found that under inigated conditions N application increased individual grain

weight in barley. Gallagher et aI. (1975) suggested that variations in grain weight between 8l barley cultivars were primarily related to the weather during the different growing seasons.

The differences in the kernel weight between the 4 sites could be related in part to the differences in spring rainfall and its interaction with N treatment.

Another undesirable effect of adding N to malting barley is the increase in GNC which can sometimes occur. In these experiments the response in GNC to N showed a linear increase for most cultivars (Fig. 3.10), although the values at 15, 30 and 45 kgN/ha were not often significantly different from that at 0 kgN/ha. However, seasonal conditions had a large influence on GNC and its response to N. The response in GNC observed in these experiments is similar to that reported by Birch and Long (1990) and Birch et al. (1993) who found that GNC in a number of barley cultivars increased but often not significantly until40-45 kgN/ha was applied. At rates lower than 45 kgN/tra they found no increase or decrease in GNC in some barley cultivars. These data suggests that small additions of N will not cause large increases in GNC and therefore would not adversely affect the chance of farmers achieving malting quality.

There was little evidence that there was a large difference in the responsiveness in GNC between cultivars (Table 3.7,3.I4); environmental (i.e. site) effects had a stronger influence on GNC than genotype. In these experiments the lowest GNC occurred at the sites with the highest grain yield, and the change in GNC with N was largely influenced by the yield response. However the data from these experiments show the relationship between grain yield and GNC for these six barley cultivars to be essentially independent (Fig. 3.11),

which is also suggested by Jenner et al. (I99L) for wheat. These results suggest it may be

possible to select cultivars with high grain yield response and low GNC response, if malting

quality is the major objective. NHI was not a useful indicator of GNC. Adding N reduced

NHI at each site (Table 3.16), but the reductions were often small except at the highest N

rates and were not related to GNC. The variations in NHI between the 6 cultivars were

small and generally not significant. GNC therefore depended more on total N uptake than

on the partitioning of N to the grain (Fig. 3.12). 82 The genetic variation in grain yield was more important to the variation in the yield-protein relationship than was the differences in GNC. The responses in yield to N were the result of a complex effect of N on dry matter production and its subsequent influence on kernels/m2 and kernel weight. There were significant genetic differences in some of these relationships. Over the 3 sites, kernel number was associated with dry matter at anthesis in all barley cultivars except Weeah (Fig. 3.7). The relationship for Skiff was also different from that of the other cultivars. The form of this relationship is a measure of the efficiency of production of fertile reproductive sites, and it will be influenced by the source: sink relationship prior to anthesis. Relationships between DM¿ and kernel number also have been reported in wheat by Fischer (1979) and Anderson and Smith (1990). In both cases, short wheats produced 10-13 kernels/m2 per g dry matter and the tall wheats produced 7-8 kernels/g Dl\4/m2. A similar result was found amongst the 6 cultivars of barley at Northfield although the number of grains/gDM was much higher. Skiff, a semidwarf cultivar showed a bigger response (-20 kernels/g DN{/m2¡ than the other, tall cultivars. However, Stirling and Schooner which are not semidwarfs also showed large response of -20 kernels/g

DM/m2. For example, at a DM¿ of 60O glm2, the number of kernels/m2 is estimated as

11,035, 10,031, 9,185, 8,770,8,79L and 8,127 for Skiff, Stirling, Chebec, Clipper,

Schooner and Weeah respectively. Anderson and Smith (1990), found the yield advantage

of a semidwarf wheat was related to both greater kernel numbers per unit area and to a

greater kernel size (see their Figs. 3a and 4b), however, in the present experiments the yield

advantage of semidwarf barley, Skifl was related to kernels/m2, but not to kernels size: in

fact the kernel weight of Skiff was lower than that of the other cultivars with the exception

of Stirling.

Over all sites, high yields also were correlated positively with ear number (Table 3.10).

Skiff, Chebec and Stirling had strong correlations between grain yield and ear number at

Northfield 1990 and 1991, and Schooner in Northfield 1990. Gardener (1972) suggested

with barley the variation in grain yield between cultivars is due to ears/m2 and also

McDonald (1992) with wheat found ears/m2 and kernelslm2hadhigh correlations with grain

yield. 83 I conclude that on basis of the aims of these field experiments there are genetic differences between barley cultivars in response to N fertiliser. In particular, the semidwarf cultivar

Skiff was more responsive than the other, tall cultivars used, and the yield of Weeah was

least responsive to increases in dry matter and ear number. The early response to application

of N was not well correlated with yield. Over all the sites the number of ears/m2 was

increased with N fertiliser and the number of kernels/m2 and ear/m2 was significantly correlated with grain yield. Kernel weight had a negative correlation with grain yield response to N although there were some differences between barley cultivars in kernel

weight. Generally,low rates of N (<45 kgN/ha) will promote significant yield responses at

N deficient sites and GNC may remain low. To look in detail at the genetic differences

between barley cultivars to N, further studies in N uptake and remobilisation by the barley

cultivars will be needed. 84 CHAPTER 4 DIFFERENCES IN NITRATE UPTAKE AND ASSIMILATION AMONG BARLEY CULTIVARS

4. l.Introduction

Results from the field experiments described in Chapter 3 indicated there are differences in grain yield and grain N concentration responses to N between barley cultivars.

Vegetative growth early in the season and the response to N fertiliser differed between some cultivars, although this did not always benefit grain yield. Nevertheless these differences in early growth and responsiveness of barley cultivars could be related in part to differences in the ability of plants to exploit soil N. Nitrate is the most common form of

N taken up by cereals plants growing in the field, and efficient utilisation of soil and fertiliser N is an important and desirable agronomic character in barley. Early in the season, nitrate in the soil tends to be high (see Chapter 2,2.3.1. on nitrate levels in soil) and so it would be desirable to use as much of this nitrate as possible. There are clear indications that active ion abso¡ption by plants is under genetic control (Epstein 1972) and that considerable differences exist between varieties (see Chapter 2,2.4.1). The variation is due to differences in the size and morphology of the roots, demand for mineral elements caused by differences in relative growth rate (Chapin and Bieleski L982), in uptake, in transport (Glass et al.l98l) and in use efficiency. The present study represents an attempt to evaluate the level of genetic variation in nitrate uptake and assimilation in barley at the seedling stage.

4. 2. Preliminary investigations on nÍtrate uptake

4,2.t. Materials and Methods

4.2.1.1. Experiment 1: Short term uptake of nitrate

Eight barley cultivars (Clipper, Schooner, Weeah, Stirling, Skifl Chebec, Franklin,

Galleon) were selected, 6 of which were used in the field experiments described in

Chapter 3. Skiff and Stirling were the most responsive to N in grain yield, Weeah and 85

Schooner were least responsive, whilst Clipper and Chebec were intermediate (Chapter 3).

Galleon and Franklin were included because these cultivars have high potential yields in the held.

Seeds were surface sterilized by immersion in 95Vo ethanol for 10 seconds followed by soaking for 3 minutes in O.2Vo HgCl2 and thoroughly rinsing with sterile water (Herdina

Sccds were sown in square, black plastic pots in sterilised sand and 7 ater and placed in a growth room (2O ! 0.4oC; 16 h light, 8 h dark).

High pressure sodium lamps (Quartz Halogen) provided a photon irradiance of approximately 300 p Einstein m-2 sec-1. Seven days after emergence 18 seedlings were transferred to pots containing 1.5 L of nutrient solution (see Table 4.1) with lmM nitrate concentration and grown subsequently using a hydroponic system in a glasshouse under

natural irradiance and daylight for 13 days (Plate 4.1). The nutrient solution was renewed

every 48 h. Nitrate uptake during this time was not measured. The design was a

randomised complete block with 6 replicates. KOH was used to adjust the pH of the

solution to 7.0. The aerâtion rate was 0.9 L/min. for each pot. At day 13 the plants were

transferred to a nitrate-free solution for 2 days to reduce shoot and root nitrate levels to

negligible concentrations and to allow the plants of each cultivar to reach an identical condition. After this, half the plants in each pot were harvested and the remaining

seedlings were transferred to a 1mM nitrate solution for 34 hours. A sample of solution

(lsml-) was taken at each time and a lml- aliquot of the nutrient solution was diluted to

10mL with de-ionised water and the absorbance measured in a spectrophotometer at 2lO

nm (Cawse t967). A calibration curve was developed using solutions of KNO3 of known

concentrations, and a correction factor (CF) calculated from the slope of the curve to relate

absorbance to concentration. The nitrate concentration of the nutrient solution was

measured at 5 hour intervals over the 34 hours, and nitrate uptake by the plants was

determined as the amount disappearing from the nutrient solution over the time. Nitrate in

the sample was calculated as follows:

xC -V x xCF Nitrate uptake 100 86 where CF=Correction factor from calibration curve, Vt=Initial volume of solution in each pot, V2=final volume of solution in each pot, Ct= Initial concentration of nitrate,

Cz=Final concentration of nitrate.

Table 4.1. Compositions of solutions used for hydroponic studies of nitrate uptake and assimilation

Nitrate solutions 0mM 0.25mM 0.5mM l.OmM

(mg/L) KNO3 8.5 t7.t 34.1 Ca(NO¡)z.4H2O 19.9 38.9 79.9 MgSOa.TH2O 7.6 7.6 7.6 7.6 KH2POa I0.2 t0.2 r0.2 to.2 K2SOa 218.0 210.8 200.1 189.4 CaSOa.2H2O 430.5 416.0 401.5 372.7 Trace Elements (Tn¡a 6.rU 0.36 0.36 0.36 Fe-EDTA 21.8 21.8 2r.8 2r.8

aTE: (MnS0a.7H20 28mglL, Na2Mo04 2H2O 6mgll-, S0+ 5H20 l4mg/I-, NaCl l45mg[L,

ZnSO+.7}J20 1 8mg/L, CuS0+. 5H20 3 I mg, H3BQ 1 1 6mg/L).

Definitíon of units

After the plants were harvested they were partitioned into shoot and root, dried at 80oC for

2 days and the dry weights measured.

(i) The relative growth rate was calculated: p6ft =(lnW2-lnW)l(r2-t1), where'W1

and'W2 are total plant dry weight at t1 and t1 (ä) Nitrate accumulation in was calculated from the different between nitrate

content in root and shoot at H1 and H2. Nitrate concentration the root and shoot was

measured as unreduced nitrate by the E. coli method (McNamara et al. 1971,

Appendix 4.1).

(äi) Nitrate assimilation was estimated by subtracting the nitrate accumulation in

both shoot and root from total nitrate taken up by the plants during the34 hours. 87 (iv) Nitrate assimilation fficiency (NAE) was defined as the dry weight increase per

unit nitrate assimilation.

(v) Nitrate uptake fficiency (NUE) was defined as the total nitrate uptake by plant

per unit nitrate supplied multiplied by 100.

(vr) Percent of nitrate assimlation was derived from the following equation:

7o nitr ateassimilation !4-I4!994PþL x 1 00 = nlrrare uptaKe

4.2.1.2. Experiment 2: Response to different concentrations of nitrate nih"le by Experiment 1 examined the uptake and assimilation ofndifferent barley cultivars at one concentration of nitrate (lmM). To examine the seedling response to different rates of nitrate a smaller number of cultivars were grown at 3 rates of nitrate.

Four barley cultivars (Stirling, Skifl Schooner, Weeah) were selected on the basis of their

dry matter and nitrate uptake responses t€dlitrât€ in the first experiment. Seeds were

prepared and sown as described in Experiment 1. Twenty seedlings of each cultiva¡ were

transferred to a hydroponic system with 3 levels of nitrate (0.25,0.5, 1.OmM (Table 4.1))

in a glasshouse under natural irradiance and daylight in early October. In this experiment,

4 cultivars were grown in a single tray (contain l0 L of nutrient solution) at each nitrate

concentration. The experimental design was a split plot design with N treatment as the

main plots, and with 4 replicates. The nutrient solution was renewed every 2 days. At the

end of experiment (5 leaf stage) the seedlings were harvested and partitioned into shoot

and root and dried at 80oC. Nitrate accumulation in the plant parts rwas measured in the

same way as Experiment 1. Each tray contained all 4 cultivars so measurement of nitrate

uptake for each cultivar was not possible. Nitrate accumulation in the root and shoot was

measured by the E. coli method (McNamaraet aL.1971, Appendix 4.1). 88

ü*

,l \'i,',1_.:,- r _-L ö

t- .t ^t't'À > >.1 Þ" ñ. ç- ! j$û ,.r,-4Â

HvdroPonlc mslhod t ln l.¡ltrate uPtake bêrleY (l.mMoll

llydroD0nlc mghd ll:dr'rlìtrtrit illcllr(l Nlrrurc trpltrle ln [Ì.laY \[rulc trlluÌc ln hrilct l0,SDlloll I l.hù\toll

Plate 4.L The hydroponic system which was used in the glasshouse to identify differences in nitrate uptake between barley cultivars for Experiments land 2. 89

4.3. Results

4.3.1. Experiment L

The eight barley cultivars differed significantly in both root and shoot dry weights at both

harvests (Table 4.2). At the first harvest the root and shoot dry weights of Stirling and

Schooner were significantly lower than the other cultivars. At the second harvest, root and

shoot dry matter was still low in Schooner and Stirling, but the differences with other

cultivars was less and often non-significant. Schooner and Stirling also had the highest

RGRs although statistical significance varied. Schooner's RGR was not significantly

different or higher than Stirling, while only Clipper and Franklin had RGRs significantly

lower than that of Stirling.

Tab\e 4.2. Root, shoot dry and total plant dry weight and relative growth rate of eight cultivars of barley grown in the presence of a lmM nitrate solution. Plants were 20 days old and harvest two (H2) occurred 34h. after harvest one (H1)

Cultivar Root DWt. Shoot DV/t. Total DWt. Relative (mg/plant) (mg/plant) (mg/plant) growth rateb Hra }Iza H1 H2 H1 H2 (h-1 103)

Clipper 76 85 133 189 208 274 8.01 'Weeah 64 81 134 t73 r97 255 8.51 Stirling 50 69 99 t6L t49 230 12.99 Schooner 53 75 105 173 158 248 13.64 Chebec 70 78 147 214 2t7 292 8.44 skiff 70 89 148 201 2t8 290 8.47 Franklin 73 85 t4t t70 214 255 5.46 Galleon 65 84 147 207 212 29r 9.42

LSD(57o) 13 12 26 35 36 44 4.96

aHl=Harvest one, H2=Harvest two bbased on total dry weight

Although different cultivars took up different amounts of nitrate (Table 4.3), there were

only small differences in the rate of nitrate uptake on the basis of root dry weight. Clipper ttïäå*f*"G " rate of nitrate uptake, but there were no signihcant differences between the ^A 90 other cultivars. Therefore, the main cause of the differences in nitrate uptake appeared to be related to the amount of root growth.

The amount of nitrate taken up over the 34h. period differed significantly between cultivars (Table 4.3). Total nitrate uptake per plant did not differ significantly between

Skiff, Franklin, Galleon, Weeah and Chebec. Skiff however had a significantly higher nitrate uptake than the remaining cultivars, while Stirling had the lowest nitrate uptakc

(Table 4.3). Consequently nitrate uptake efficiency (total nitrate uptake as a proportion of nitrate supply) also differed between cultivars at the end of the experiment (Table 4.3).

Skiff, Franklin, Galleon, Weeah and Chebec had significantly higher nitrate uptake efficiencies than Clipper, Schooner and Stirling.

There was no significant relationship between the total dry weight at H2 and the nitrate uptake over the 34h. of the experiment (Fig.4.1a), nor with nitrate accumulation in the shoot (Fig.a.lb). Nitrate assimilation efficiency also differed significantly between cultivars. Stirling and Schooner had significantly higher nitrate use efficiencies than

Franklin, V/eeah and Skiff (Table 4.3).

Weeah and Skiff had a significantly higher root nitrate concentration than Schooner,

Chebec, Ctipper and Stirling (Table 4.3). There was no significant difference in nitrate accumulation in the shoot between cultivars and no overall relationship between plant dry

weight and nitrate accumulation (Table 4.3, Fig. 4.1b). The proportion of nitrate

assimilated by plants was not significantly different between cultivars. At the end of

experiment the proportion of nitrate reduced by all the barley cultivars ranged from 527o

to 59Vo. Table 4.3. Total nitrate uptake, accumulation and assimilation of barley cultivars. All measurments are basis of 34h.(HZ-Hùa

Pa¡ameters Clipper Weeah Stirling Schooner Chebec Skiff Franklin Galleon LSD (5Vo)

Total nitrate uptake 313 349 283 297 333 379 358 350 59 (pmoUplant) Rate of NO3- uptake 2720 3430 3390 3310 3t20 3400 3230 3310 393 (pmoVg root DWday) Root nitrate conc. 63 86 57 66 65 83 77 7l 16 (pmoUplant) Shoot nitrate conc. 62 75 76 76 74 92 82 84 NS $rmoUplanÐ Nifrate accumulation 125 160 134 142 r39 r75 159 155 NS (pmoUplant) Niftate a.ssimilation 188 189 150 156 r93 205 200 195 NS (pmoUplant) Nitrate assimilation 60 53 53 52 58 54 55 55 NS (vo ) Niirate uptake efficiency 38 42 34 36 40 46 43 42 7 (NUE) (7o) Nit¡ate assimilation efficiency 0.39 0.36 0.57 o.57 0.38 0.37 0.22 0.42 0.19 (NAE) (mdttmol)

aHz, Ht-Second and first harvest \o 92

(a) r = +0.62 NS 390 skiff f- (d o o- 370 Franklin .>-- o Galleon o 350 o o E Weeah Chebec o -L 330 o l¿ o (ú 310 o- o Clipper J 290 o o Schooner 270 Stirling E .=z 250 200 220 240 260 280 300 Total dry weight (mg/plant)

(b) c r = +0.29 NS (ú 200 o- 190 o 180 skitf E a f- 170 c Weeah o 160 Franklin Galleon ¡ O (d 150 o E= 140 Chebec ¡ of 130 o schooner C) Stirling a (ú 120 o Clipper g 110 .= 100 z 200 220 240 260 280 300 Total dry weight (mg/plant)

Figure 4.1. Relationship between total plant dry weight an{ (a) nitrate uptake and (b) nitrate accumulation for 6 barley cultivars at H¿. 93

4.3.2. Experiment 2

There was a significant' interaction between Cultivar and Nitrate rates in total plant dry weight but not in the nitrate accumulation (Fig. 4.2). Weeah and Skiff were responsive cultivars, but Stirling and Schooner did not respond to nitrate. Total dry weight in Weeah increased at 0.5mM, but not significantly at the higher nitrate level while Skiff responded at the highest rate only. Nitrate accumulation increased identically with increase level of nitrate for all cultivars (Fig. a.2b).

Table 4.4. Growth responses of four barley cultivars to different levels of nitrate. Weeah Stirling Schooner Skiff LSD (5Vo)

Cultivar effect (mg/plant) Root dry weight. 78.8 59.7 54.1 64.8 7.4 Shoot dry weight. r54.0 111.1 124.3 146.8 t4.4 Total dry weight. 232.7 170.4 t78.5 213.6 r7.7 Root:Shoot ratio 0.53 0.55 0.45 0.45 0.05

Nitrate effect (mg/plant)

0.25 mM 0.5 mM 1.0 mM s 2 Root dry weight 66.2 63.9 62.9 6.qi 'V Shoot dry weight r15.4 r41.4 r45.3 12.5 Total dry weight 181.8 205.3 208.2 15.3 Root:Shoot ratio 0.58 0.46 0.43 0.05

There were significant differences between barley cultivars in total plant dry weight

(Table 4.4, Fig.4.2a). Schooner and Stirling had lower plant dry weights and nitrate accumulation (Fig.4.2c) than Weeah and Skiff, which is consistent with the results of

Experiment l. Also there was a significant difference between nitrate levels in plant dry weight for all cultivars. Root dry weight was not significantly affected by nitrate but shoot and total dry weight increased at 0.5 and 1.0 mM nitrate rates.

The root : shoot ratio could not explain the different response to nitrate of the cultivars.

Weeah and Skiff had high dry matter responses to N but different root : shoot ratios 94 whereas Weeah and Stirling had similar root : shoot ratios but different responses. b¿lt¡c¿r., owlhvqn Therefore, the root growthnappears more important than the root : shoot ratio.

4.4. Discussion

In general, the differences in nitrate uptake between cultivars was positively related to the size and growth rate of the seedlings (Table 4.2 and 4.3). The size and morphology of the roots (Hackett 1968), and differences in relative growth rate (Chapin and Bieleski 1982) have been reported to affect nutrient uptake. The size of the root system appeared to be important because there were fewer differences in uptake per g. root. On the basis of the relationship between nitrate uptake, nitrate accumulation and dry matter accumulation during the experiment (Figs. 4.Ia, b 4.2b), two distinct groups of cultivars could be identified. Stirling and Schooner,ryere small plants, which took up and accumulated less nitrate but had higher relative growth rates than the semi-dwarf cultivars, Skiff and

Franklin and the cultivar W'eeah. The greater nitrate assimilation efficiency of Schooner and Stirling compared with Skiff, Franklin and Weeah suggests that they compensated for their lower uptake with a more efficient utilisation of nitrate (Table 4.3). The other three cultivars, Clipper, Chebec and Galleon, were intermediate. Cultivars which took up less nitrate (Stirling, Schooner and Clipper) utilised the same percentage of the solution nitrate in growth (percentage nitrate assimilation was not significant, Table 4.3), although there was a tendency for the total amount.of nitrate that was assimilated to be lower in these cultiva¡s as well, but it was not statist^fally different. Franklin and Skiff are short cultivars ,l which are related to Triumph whereas Schooner and Stirling have Prior A as a parent.

Prior A is also in the pedigree of Chebec and Clipper. Schooner and Stirling, and to a certain extent Clipper and Chebec, appear to be cultivars with a lower ability to take up

nitrate and accumulate smaller amounts of nitrate in the shoots. On the other hand, Skiff

and Franklin appear to be able to take up larger amounts of nitrate for similar increases in

dry weight (see Fig. 4.1). Whether the different responses in growth, nitrate uptake and

utilization observed in these experiments are related to the different genetic backgrounds

is examined in further experiments. 95

(a)

290 Weeah (! + Þr 270 Þo + Stirling 250 + Schooner 230 bo (.) 2t0 skiff È €L 190 (Ë 170 Io F 150 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 l Nitrate rate (mM)

(b) I É r20 Oi -o 100 ¿ 80 o 60

40 ()Q (€ 20 o 0 z 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 I Nitrate rate (mM)

(c) €80 ê78 ã76 \ù/eeah 174 o ;72 skiff €70 a ? 68 E66 s,lÎing H64 o s62 Schooner S60 z1 60 180 200 220 240 Total dry weight (mg/plant)

Figure 4.2. lnteraction between nitrate levels and cultivar in (a) Plant dry weight and (b) Nitrate accumlation and (c) relationship between nitrate accumulation and total plant dry weight for 4 barley cultivars. 96

4. 5. Effect of genotypes on nitrate uptake and assimilation in barley cultivars

Barley cultivars were found to differ in nitrate uptake, nitrate accumulation in the shoot

and in dry matter production. In Experiments 1 and 2 Stirling and Schooner produced

less dry matter and accumulated less nitrate than Skiff and Weeah (Table 4.2,Fig.4.la,b

Experiment 1).

On the basis of these differences in nitrate uptake and dry matter accumulation, the

cultivars were categorised into two main groups: Stirling and Schooner were small plants,

which took up less nitrate but had higher relative growth rates than the semi-dwarf

cultivars, Skiff and Franklin. It was thought that these differences could be related to the

pedigrees of the cultivars. Therefore, the preliminary experiments were extended to look

at a greater number of cultivars representing four genetic groups.

These experiments differed from the preliminary experiments (Experiments 1 and 2) by

measuring nitrate uptake and assimilation for a longer period of time. Nitrate uptake and

growth of seedlings were measured over 6 days rather than 34h (Experiment 1). The size

of pots was increased to 6 L and the number of plants per pot to 24.

The aim of the present study was to verify the differences in the response of the barley

cultivars to nitrate observed in the preliminary experiments and to extend the results by

looking at nitrate uptake and dry matter production in a larger number of cultivars. The

cultivars chosen represented 4 groups each with a common or similar genetic background.

4.5.1.. Materials and Methods

The study was conducted as 3 separate experiments because there was insufficient space

in the growth room to do all the comparisons at once. In each experiment the semidwarf

group was compared to one of the other groups. 97 4.5.1.1. Plant material

Nineteen improved cultivars of barley were obtained from the collection of the Waite barley breeding program (Table 4.5). Each cultivar is a commercial variety or advanced breeding line. The cultivars were classified into four groups, each with a similar or

coÍlmon pedigree:

(i) Group I: Cultivars derived from semidwarf parents, with high nitrate uptake

cha¡acteristics. Skiff and Franklin were assigned as the leading cultiva¡s.

(ii) Group II: Cultivars which are mostly derived from'Western Australian cultivars and

crosses involved Western Australian cultivars. The leading cultivar of this group, Stirling,

has been found to have low nitrate uptake characteristics in the early stages of growth.

Grimmett is a Queensland cultivar.

(iii) Group III: These cultivars have been derived either from Prior A (originally from

England) or Clipper (a derivative of Prior A). Schooner, the leading cultivar in this group

has low nitrate uptake characteristics.

(iv) Group IV: This group consisted mainly of Victorian cultivars, with the variety

Research as a common parent. The leading cultivar V/eeah has been recognised as having

high nitrate uptake characteristics.

In Experiment 3, Groups I and II were compared, in Experiment 4, Groups I and III were

compared, and in Experiment 5, Groups I and [V were used. WI-2869 from Group I was

used only in Experiment 3.

4.5.1.2. Growth conditions

Seeds were sterilised by immersion in7ÙVo ethanol for 1 minute, soaked for 5 minutes in

17o sodium hypochloride (8ml/l00ml de-ionised water) and thoroughly rinsed with de-

ionised water. The seeds were then sown in square plastic pots containing sterilised sand and watered with de-ionised water in a growth room set at 2O+4oC, and a 12h 98 photoperiod. Fluorescent and incandescent lamps provided an irradiance of 200-300 pEinstein m-l s-I.

Table 4.5. The barley cultivars used in the studies of nitrate uptake

Cultivars Pedigree

Group I (SD and Shannon group)a skiff ((AD x WI-2335) x (CD 28 xWI-2231))/165 Franklin Shannon x Triumph Triumph (H4DM.24566 x Diamant x 1402164) ((ALSA x Abyssinian x St. x Union) Shannon Proctor*4Æthiopian line CI-3208- 1 v/r-2869 (Triumph * G alleon) l7 7 a

Group II Q)b

Dampier//(A 1 4)Prior/Ymer/3iPiroline Olli selection(M9 8)/Research Forrest Atlas 57l(416) Prior/Ymer Grimmett BusseVZephyr wr-2966 (Schooner * Fonest)/55

Group III (T) 'Prior /Proctorgroup'

Schooner Proctor/PriorA.íProcto r I Cl-3 57 6 Clipper ProctorÆriorA Galleon Clipper/Hi pr oly I I 3x Proctor/Cl 3 576 *WI-2468) Chebec ((O/lvf artin*Clipper(2)) -88 / 5 I 612 Prior A Selected Chevalier

Group IV (T) 'Research group'

W'eeah Prior/Research Parwan Plumage Archer/Prior I [Lenta]3 lResearch/Lenta Lara Research/Lenta wr-2728 (WI-2468LHR x Weeah)/7

aShannon mostly derived from semidwarf cultivars, bT=Tall

Eight days after emergence the seedlings were removed from the sand and were

transferred to 6 L round plastic pots with plastic lids (Plate 4.2). Each lid had 25 holes.

Seedlings were supported by inserting their roots through holes in the base of eppendorf

tubes, (7 mL capacity) which were placed in the holes in the lid. Twenty four seedling

were placed in each pot and these were grown using a hydroponic system with a lmM

nitrate solution (Table 4.1) for 26 days (Experiments 1, 2) or 24 days (Experiment 3) at a

pH of 6.0. The remaining hole was used for the supply of air. The aeration rate was 0.9 99 L/min. for each pot. The nutrient solution in each pot was renewed at 10, 15 and 19 days after seedling transfer. At day 20, the plants were transferred to a nitrate-free solution

(Table 4.2) for 24 hours after which 12 plants in each pot were harvested (H1). The remaining plants were transferred to a lmM nitrate solution for 6 days during which time the solutions were changed every 48 hours. At day 26 (Hù the remaining 12 plants were harvested and partitioned into root and shoot and their dry weights measured. The experimental design was a randomised complete block with 5 (Experiment 3) and 4

(Experiments 4, 5) replicates respectively.

4.5.1.3. Measurements

The nitrate concentration of the nutrient solution was measured each time solutions were

renewed before H1, and between H1 and the final harvest. The absorbance of this solution

was measured in a spectrophotometer (model Lambda 5) at2lO nm (Cawse 1967). Nitrate

in the sample was calculated as described in the preliminary Experiment 1.

Nitrate concentrations in the plant parts at H1 and H2 were measured as unreduced nitrate

by the E. coli method (McNamara et a\.1971, Appendix 4.1). RGR, nitrate assimilation, nitrate assimilation efficiency (NAE) and nitrate uptake efficiency (NUE) were

determined as in preliminary Experiment l. Growth rate was calculated from the increase

in total dry weight during 6 days.

4.5.1.4. Data analysis

Analysis of variance were conducted for each experiment on all data parameters. Linear

regressions were calculated for the relationships between (a) the increase in nitrate

between H1 and H2 day period and the increase in dry weight over the same time, and (b)

the total nitrate uptake over 26 days and total dry weight atÞ1 To compare the response

of different groups of cultivar, the slopes and intercepts of the regressions were compared

using Genstat 5. 100

Plate 4.2. Method used to examine differences between genotypes in nitrate uptake and assimilation in barley cultivars in Experiments 3, 4 and 5. 4.5.2. Results

The inclusion of the Group I cultivars in each experiment allows a comparison to the made between the 3 experiments. The growth and nitrate uptake of the semidwarfs was similar

in the first two experiments, but in the third it was a little lower. A possible reason is that

the duration of Experiment 5 was 2 days shorter than Experiments 3 and 4. The general

consistency in nitrate uptake and growth of semidwarf cultivars in these experiments

allows the results of the experiments to be compared.

4.5.2.1. Experiment 3

There were significant differences in root and shoot growth among the cultivars which

were not related to their cultivar groupings (Table 4.6). Shannon and WI-2869 (Group I)

and WI-2966 and Forrest (Group II) produced significantly more root growth than the

other cultivars at both harvests. A similar difference, although not as large, was observed

also with shoot growth. Stirling had the greatest RGR (and the smallest plants), a result

consistent with the results of the preliminary experiments. The growth rates of the 10 cultivars were not significantly different from one another, however, there were

differences in the RGR. Shannon is not a semidwarf but had quite a different RGR.

There were significant differences between cultivars both in the total amount of nitrate

taken up over the 6 days and in the nitrate which accumulated in the plant tissue (Table

4.7). However, as with the dry matter and growth rate data, there was no clear distinction

between the two groups,

There was no significant difference between cultivars in Groups I and II in nitrate uptake per g. root dry weight (Tabte 4.7). Total nitrate uptake per plant did not differ

significantly between Shannon and WI-2869,WI-2966, Forrest but Shannon and WI-

2966, Forrest took up significantly more nitrate than the other cultivars (Skiff, Franklin,

Triumph; Stirling, Dampier, and Grimmett). Stirling and Dampier had the lowest total

nitrate uptake (Table 4.7). r02 Table 4.6. Experiment 3: Effect of lmM nitrate supplied hydroponically on root, shoot and total dry matter production and growth rate of ten cultivars of barley

Cultivar RootDWt. Shoot DV/t. Total DWt. RGRA GRA (mg/plant) (mg/plant) (mg/plant) (mg/ (mg/day) H1 H2 H1 H2 H1 H2 mg/day)

Group I skiff 6t 113 176 400 23',1 513 13.25 46

Franklin 53 t02 t7t 398 223 501 13.39 46

Triumph 66 113 168 419 234 532 t3.r7 50

Shannon 86 142 222 44'l 308 588 t0.77 47 wr-2869 74 t37 188 480 262 618 t4.32 59

Group II

Stirling 38 95 106 311 144 406 17.37 44

Dampier 47 94 t63 407 2t0 501 t4.35 48

Forrest 83 t63 230 464 3t3 627 11.57 52 Grimmett 70 t14 196 370 266 484 9.93 36 wr-2966 78 166 225 513 303 679 13.52 63

LSD(5%) t7 24 39 80 53 96 3.92 NS

Hl=Harvest one, H2=Harvest two, RGR=Relative growth rate, TDM=Total dry matter, GR-growth rate abased on total dry weight

The relationships between the increase in plant dry matter and uptake of nitrate over the 6

days for the 2 groups were not significant (Fig. 4.3a). There were significant linear

relationships between the total nitrate uptake of the plants and the total dry weight at H2

(Fig. 4.3.b), but comparison of the regressions showed the two ä* no, significantly

different from one another. Therefore, the two groups of cultivars showed statistically

similar relationships between dry matter production and nitrate uptake.

Nitrate assimilation efficiency (NAE) did not differ significantly between cultivars except

between Stirling and Grimmett (Fig. 4.4a): Stirling had the highest and Grimmett the

lowest. Nitrate uptake efficiency (NUE) was highest in V/I-2966, Forrest, Shannon and

WI-2869, but there was not difference among these cultivars, whilst Stirling and Dampier 103 had the lowest NUE (Fig. 4.4a). There was no significantly differences in two indices of

N use efficiency (TDN{/nitrate uptake, ÀDÀ¡VÂnitrate) between cultivars (Table 4.7).

Table 4.7. Experiment 3: Total nitrate content, increase in nitrate uptake nitrate uptake per root dry weight, and indices of N use efficiency of 10 cultivars of barley grown with lmM nitrate

Total NO3- uptake ÂNO¡- uptakea Nitrate uptake ÅDM/ÂNO3-b pU¡NO¡- (pmoUplant) (¡tmoUplant) (pmoVmgroot uptake uptake (0-26 day) (20-26 day) Dw/day) (20-26 day) (o-26 day) (20-26 day)

Group I skiff 1 837 9s5 3699 0.300 0.285

Franklin 181 I 937 4003 0.306 0.283

Triumph 1759 845 3lt7 0.337 0.299

Shannon 2234 1 184 3461 0.233 0.263 lVI-2869 2t69 1 180 3695 0.306 o.286

Group II Stirling l45t 828 4101 0.346 0.283 Dampier 1660 858 4023 0.339 0.305

Fonest 2247 1203 3339 0.264 0.279

Grimmett 1919 953 3466 0.230 0.252

Iù./I-2966 245t 1375 3774 0.274 0.277

LSD(570) 286 2t9 NS NS NS

âIncrease in nitrate uptake over the 6 days, blncrease in dry matter over increase in nitrate uptake over the 6 days LO4

(a) y=0.041+2.05x R^2 =0.21 NS Group I 1400 I O/ crl / I Group I a 1300 o / o Group II

o t200 o J¿ cÉ

a 1 100 o= k 1000 y=0.Qll+3.22x o R^2 =0.64 NS Group II q) U) 900 c.) ()¡i o I 800 OOOOOOOOOOÊca\ñr-o\Êc.¡\ñr-o\ ôì c.l cì ôl N ca c.¡ cî¡ ca cî Change in dry weight (mg/plant)

(b) y=-0.021+3.95x 2600 R^2 =0.81* Group I

Ê 2400 Þi o 2200 o .l¿ 2000 È o y=0.346+3.54x (.) R^2 Group [I (€ r800 =0.92'rcr. o ctl 1600 Fo 1400 400 450 500 550 600 650 700 Total dry weight (mg/plant)

Figure 4.3. Experiment 3: Relationship between increase in dry weight and nitrate uptake over 6 days (a), and (b) total dry weight and total nitrate uptake over 26 days for 10 barley cultiva¡s. ã È='a llioa (Vo) ä'õ É Nitrate uptake effrciency (7o) Nitrate uptake efficiency (Vo) Nitrate uptake efficiency tssd 9äs tslJ(rÞ(JtO\\¡OO OOOOOOOOO-l.JlrÀ(.¡¡O\\¡æ HIJ(r5L^O\\¡Oo OOOOOOOOO = Franklin -r ='5 Franklin 5ã 3 Franklin ã'õ. Triumph Èu¡dB Triumph tsz,z^ê (D Triumph Shannon >e{ Shannon o go É v/I-2869 fo$., Shannon Schooner 9oaå Þ) Stirling 'H< ÞFt rt) t¡¡tsp Weeah (, ()¡^(Diri Galleon Dampier -'ão'Þ ã7f Lara Clipper Forrest ÞÞo'o wI-2728 Prior A Grimmet <ÞôÈ oå- wr-2966 ¡-Þl= '-t Parwan Chebec ä.6', o9P99P ÞÞ OP999P OÊl.JN) Øu)4Ø o9999PP9 OF¿HN)bJ (JtOUTO(Jt (JrOLr¡O(J¡O(JrOl.)tJ(rtJ) (JrOLhO(.,r ÀþØ= dx' Nitrate assimilation efficiency (mg/pmol) Nitrate assimilation efficiency (mg/pmol) Nitrate assimilation efficiency (mg/pmol) EÞ I z z O tr lrl (¡ 106

4.5.2.2. Experiment 4

There were significant differences between the nine barley cultivars in root, shoot and total plant dry weights at both harvests (Table 4.8). At the first harvest Shannon and

Galleon had significantly higher root dry weights than the other cultivars. The remaining cultivars did not differ significantly. At harvest 2 root dry weights of Prior A, Clipper,

Chebec, Franklin, Schooner, Triumph and Skiff were not significantly different but those of Galleon and Shannon were again greater. Shoot dry weight was higher in Galleon,

Shannon, Skiff and Franklin than in the others at first harvest. It was higher in Galleon and Shannon and the lowest in Chebec, Schooner and PriorA at the second ha¡vest.

Table 4.8. Experiment 4: Effect of lmM nitrate supplied hydroponically on root, shoot and total dry matter production and growth rate of nine cultivars of barley

Cultivar Root DV/t. Shoot DWt. Total DWt. RGRa GRA (mg/plant) (mg/plant) (mg/plant) (mg/ (mg/day) lmglday H1 H2 H1 H2 H1 H2

Group I skiff 87 t67 191 395 279 562 11.79 47

Franklin 86 t39 189 383 2t5 s22 10.71 4t

Triumph 69 151 169 386 239 537 13.61 50

Shannon 129 t97 241 467 371 663 9.75 49

Group III

Schooner 7I r47 147 351 2t8 497 t4.lt 47

Galleon 122 234 268 538 390 773 TT.44 64

Clipper 78 t27 157 318 235 445 10.88 35

Prior A 66 122 149 295 216 416 11.63 33

Chebec 65 139 t46 337 210 475 t3.64 44

LSD(57o) 29 50 53 91 79 130 2.73 13

H1=Harvest one, H2=Harvest two, RGR=Relative growth rate, TDM=Total dry matter,GR=growth rate abased on total dry weight r07 In Experiment 4, as in Experiment 3, the differences in growth were between cultivars rather than between the 2 groups. RGR was significantly different between cultivars and ranged from 14.1 d-l in Schooner to9.7 d-lin Shannon. V/ithin Group I cultivars, the

RGR of Shannon was significantly lower than Triumph (Table 4.8). The high RGR for

Schooner is consistent with the results of the preliminary experiment. The growth rates of the semidwarf cultivars in Experiment 2 were similar to those in Experiment 1. The significant difference in growth rate is largely due to the value for Galleon; there is no significant difference in the growth rates of the other cultivars.

Total nitrate uptake per plant did differ between cultivars but nitrate uptake per g. root dry weight did not differ between cultivars. Galleon had a significantly higher nitrate uptake

than the other cultivars, while Prior A had the lowest uptake (Table 4.9). The nitrate uptake over 6 days was significantly related to the growth of the plants in Group III but not in Group I (Fig. 4.5a). However, there were positive relationships between total nitrate content and total plant growth for cultivars in both groups (I and III) (Fig. 4.5b).

Galleon was different from the other Group III cultivars because of its high dry matter production and its high nitrate uptake. However, comparisons of the regressions in Fig.

4.4b found that the slopes and intercepts were not statistically different. Therefore, the

relationships between nitrate uptake and plant dry matter for Groups I and III are similar.

Nitrate assimilation efficiency did not differ significantly between Chebec, Schooner,

Triumph, Galleon and Skiff (Fig. a.ab), but the differences between Chebec and the

remaining cultivars (Shannon, Clipper, Franklin and PriorA) were significant. The

differences were significant between Galleon and the others cultivar except Shannon (Fig.

4.4b). There were no significant difference between cultivars for either (TDlWnitrate

uptake) or (ÂDlv{/Ânitrate) (Table 4.9). 108

(a) y=0.069+1.93x 1600 R^2 =0.11 NS Group I d Oi 1500 I Group I o 1400 O Group lll

.vc) 1300

(b) y=0.87+3.32x R^2 =0.91{c Group I 2600

(€ a 2400

J. 2200 c) ,\1 ñl )È 2000 o ¡rctt 1800 y4.29+2.82x R^2 Group III d =0.99** Fo 1600

1400 400 500 600 700 800 900 Total dry weight (mglplant)

Figure 4.5. Experiment 4: Relationship between increase in dry weight and nitrate uptake over 6 days (a), and (b) total dry weight and total nitrate uptake over 26 days for 9 barley cultivars. 109

Table 4.9. Experiment 4: Total nitrate content, increase in nitrate uptake and nitrate uptake per root dry weight

Total NO3- uptake ÂNO¡- uptakea Nitrate uptake ÂDlv{/ÀNo3-b mu¡No:- (pmoVplant) (pmoVplant) (pmoVmgroot uptake uptake (0-26 day) (20-26 day) Dw/day) (20-26 day) (0-26 day) (20-26 day)

Group I skiff r993 1248 327t 0.228 0.283

Franklin 1871 tt45 3372 0.218 0.289

Triumph t77t I 109 3360 0.269 0.303

Shannon 2282 l4t3 2920 0.208 0.29r

Group III

Schooner 1654 to47 3292 0.265 0.300

Galleon 2485 t487 2836 0.259 0.311

Clipper 1563 953 3 108 0.225 0.289

Prior A 1511 869 3062 0.244 0.275

Chebec 1589 1031 3368 0.262 0.310

LSD(57o) 480 297 NS NS NS

âIncrease in nitrate uptake, bln"r"ur" in dry matter over increase in nitrate uptake over the 6 days

4.5.2,3. Experiment 5

The 8 barley cultivars differed significantly in dry matter production at both harvests

(Table 4.10). Root dry matter production was significantly higher than the other cultivars

in Shannon at the first harvest. Weeah, Lan,WI-2728 and Franklin had the lowest root

dry weight at the first harvest. At the second harvest root dry weight was higher in

Shannon, Parwan and Skiff and low in Franklin,WI-2728 and Lara. Shoot dry weight did

not differ between Shannon, Skifl Parwan and Triumph but was significantly higher for

these than the others at the first harvest. There were also no differences between Shannon,

Weeah, Parwan and Skiff in shoot growth at the second harvest, but shoot dry matter was

low in Triumph, Lara, V/I-2728 and Franklin (Table 4.10). Total plant dry weight at H1

was not different between Shannon, Skifl Parwan and Triumph but it was low in Lara,

'Weeah, WI-2728 and Franklin. At the second harvest plant dry weight did not differ 110 between Shannon, Weeah, Skiff and Parwan but was lower in Lara, WI-2728, Triumph and Franklin (Table 4.10).

The period of Experiment 5 was shorter than the other two experiments and the semi dwarf cultivars showed lower growth rates and nitrate uptake than in the other two experiments. Shannon and Skiff had higher nitrate uptake than Triumph and Franklin which is consistent with Experiments 3 and 4.

Table 4.10. Experiment 5: Effect of lmM nitrate supplied hydroponically on root, shoot and total dry matter production and growth rate of eight cultivars of barley

Cultivar Root D\ù/t. ShootDWt. TotalDWt. RGRA GRA (mg/plant) (mg/plant) (mg/plant) (mel (mg/day) mg/day) H1 H2 H1 H2 H1 H2

Group I skiff 84 160 185 366 269. 527 l l.61 43 Franklin 62 105 153 290 216 394 10.31 30 Triumph 80 131 t66 290 246 420 8.t2 29 Shannon 110 r72 192 4to 303 s90 lt.3't 48

Group N

Weeah 5s 145 t29 400 r84 545 19.77 60

Lara 54 125 130 304 184 429 14.r3 4l wt-2728 59 tt7 t34 310 193 427 t2.70 39

Parwan 83 t72 t74 368 257 540 lt.9'7 47

LSD(7o5) 23 24 30 82 43 9t 6.08 t6

Hl=Harvest one, H2=Harvest two, RGR=Relative growth rate, TDM=Total dry matter, GR- growth rate abased on total dry weight

'Weeah; RGR was significantly different between cultivars due to the high RGR of there were few significant differences (Table 4.10). There were also significant differences between cultivars in growth rate. Weeah had a much higher growth rate than other cultivars (Table 4.lI). Skifl Lara,WI-2728, Franklin and Triumph did not differ in their growth rates. Triumph and Franklin were significantly lower than Shannon, Parwan and

Weeah. 111

J) æs There v.rr€ísignificant difference in the nitrate uptake per g. root dry weight between / cultiva¡s (Table 4.ll). WI-2728 had a significantly lower uptake than the other cultivars.

There were no differences among Franklin, Skifl Shannon, Lara and Weeah which were all high. Nitrate uptake per plant also differed between the barley cultivars. Shannon had a signif,rcantly higher uptake than all the other cultivars except Weeah. Uptake was low in

Lara, Triumph, Franklin and WI-2728. The linea¡ correlation between the change in plant dry weight and uptake of nitrate over the 6 days was significant for the Group I but not for group IV (Fig. 4.6a). There were significant linear correlations between the total nitrate uptake and the total dry weight at H2 Gig. a.6b). A comparison of the regression indicated the 2 Groups, I and IV, had the same slopes but different intercepts, ie, the lines were parallel. Therefore, nitrate uptake by the cultivars in Group IV was lower for the same amount of dry matter production, or conversely, dry matter production was higher when similar amounts of nitrate are taken up. This trend was seen in the indices of N use efficiency (ADÌWÀnitrate and TDN{/nitrate), where the values for the Group IV cultivars tended to be higher than those of Group I, although the differences were not significant.

NAE differed between the cultivars. It ranged between 30.8 mgDM/¡rmol nitrate in 'Weeah and I3.7 mgDN{/pmol in Triumph (Fig. 4.4c) and tended to be greater for Group

IV cultivars. NUE did not differ between Shannon and Weeah, but the differences were significant between Shannon and the other cultivars (Fig. 4.4b). tt2

y=0.12+4.39x (a) R^2 =0.96**, Groupl 1400 (! I Group I Þ. 1300 o O Group fV É t200 =- c) 1 100 o áã 1000 o o ./ liñl 900 o./ 800 /y=o.M+2.31x .rtc.) R^2 =0.51 NS GroupIV (€ 700 lro O o Ê 600 \OæOC.l+\OoOON+\OoOOooooooooooooo Ê tr C\l õì C.¡ õì ô¡ cA ca cA Cf¡ C.¡ T Change in dry weight (mg/plant)

(b) y=-0.043+4.32x 2200 R^2 -0.99** Group I

(! a 2000 o Ê ¿ 1800 o) J¿ GI È 1600 Ðc¡ (Ë k 1400 y=-0.052+4.34x R^2 -0.923** Group IV (Ë 1200 o Fo 1000 350 400 450 500 550 600 Total dry weight (mg/plant)

Figure 4.6. Experiment 5: Relationship between increase in dry weight and nitrate uptake over 6 days (a), and (b)total dry weight and total nitrate uptake over 26 days for 8 barley cultivars. 113

Table 4.11. Experiment 5: Total nitrate content, increase nitrate uptake and nitrate uptake per root dry weight

Total NO3- uptake ÅNOg- uptakea Nitrate uptake ÂDm/ANO3-a TDlf/NO¡- (¡rmoVplant) (pmoVplant) (pmoUmgroot uptake uptake (0-26 day) (20-26 day) Dw/day) (20-26day) (0-26 day) (20-26 day)

Group I skiff 1806 1089 2991 0.240 0.297

Franklin r286 757 2940 0.250 0323

Triumph l3'77 827 2545 0.t92 0.303

Shannon 2t42 1307 3053 0222 0.277

Group IV

Weeah 1882 1008 3340 0.460 0.293

Lara 1439 863 3151 0.290 0.301 w1-2737 1242 638 2230 0.404 0.335 Parwan r786 to47 2659 0.266 0,315

LSD(57o) 262 162 591 NS NS

âIncrease in nitrate uptake, bin"t"use in dry matter over increase in nitrate uptake over the 6 days

4.6. Discussion

Overall, the present study found that in seedlings of barley cultivars, lmM nitrate nutrient

solution promoted differences in dry weight of the root and shoot, in nitrate uptake, in

nitrate and assimilation as well as in the nitrate assimilation efficiency and nitrate uptake

efficiency.

The major aim of the these experiments was to examine whether there were any genetic

differences in nitrate uptake in a range of barley cultivars. The result of the preliminary

experiments, Experiments 1 and 2, suggested that at least two groups of cultivars could be

identified; one group, which included Schooner and Stirling, produced small seedlings and

took up small amounts of nitrate and the semidwarf cultivars, including Skiff and

Franklin, which took up greater amount of nitrate. Lt4

The trend of increased nitrate uptake with greater plant dry weight over 6 days in

Experiment 3 was not statistically different between the two groups (Fig. 4.3a), which suggests that the differences between cultivars were related to the size of the plant. There were differences in dry matter production which were reflected in differences in nitrate uptake. In both the preliminary experiments and Experiment 3 there is a consistent result between cultivars, namely that Skiff had high nitrate uptake and dry weight, but Stirling had low nitrate uptake and dry weight (Tables 4.2,4.3,4.6,4.7).

In the fourth experiment, cultivars of Schooner's group, with the exception of Galleon, showed lower plant dry weight and nitrate uptake (Fig. a.5). In contrast to Schooner group's, the semidwarf cultivars of Group I had higher plant dry weight and nitrate uptake.

PriorA which is a grand parent of Schooner had a low nitrate uptake and low dry weight.

This may be a characteritic of cultivars which have Prior A as a parent.

In all experiments nitrate uptake over the 26 days was consistently related to the size of the plant and there were few differences in uptake per g. root. The genetic differences could be described by growth rate and the size of seedlings. However, in the fifth

experiment, there were differences between Group I and Weeah's group (Group IV) which

were not related to the size of the plants dry weight alone. The correlation between the

increase in nitrate uptake and the increse in dry weight between day 20 and day 26 is

significant for semidwarf cultivars but not for W'eeatr group's (Fig. 4.6a). However, all 4

cultivars of Group fV were below the line of Group I indicating they took up less nitrate

for comparable increases in growth over the 6 days. A similar result was found with total

dry weight and total nitrate uptake (Fig. a.6b) but in this instance the difference between

the 2 groups over the 26 days of the experiment was statistically significant.

The comparisons of Group I and Group IV show that cultivars from Group IV took up less

nitrate but produced larger seedlings. Therefore they were apparently more efficient at

using nitrate but tended to not take up a lot. In this group Weeah produced a relatively

large amount of dry matter, which was consistent'with the two preliminary experiments. 115 Experiment 2 also showed that seedlings of V/eeah to be responsive to N. These results are consistent with field observations of early season growth (Fig. 3.2) which showed that

Weeah often produced more dry matter and was the most responsive of the 6 cultivars at

Northfield and Charlick in 1991 (Chapter 3). Thus, if Weeah is representative of the

Group IV cultivars it appears that they may to be able to produce more dry matter for a given nitrate supply them many other cultivars, which can result in greater early dry matter production. However this early response to N is not necessarily related to increased grain yield per se (see Chapter 3).

There were other consistent results across the experiments. In the preliminary experiments 2 different groups were distinguished: Skiff and Franklin (semidwarf) and

Weeah had high nitrate uptake and dry weight, while Stirling and Schooner had low nitrate uptake and dry weight. In the 3 experiments which compared different groups, the nitrate uptake and plant dry weight of Skifl Franklin and Weeah was again higher than Stirling and Schooner. The differences between cultivars may be related to their respective pedigrees because Triumph which is parent of Skiff and Franklin, also had

higher nitrate uptake and dry weight than Prior A which is a parent of Stirling and

Schooner.

Scno\pic The g€a€tie differences in nitrate uptake appeared largely to be due to the differences in

the size of the plants particularly roots, rather than differences in the ability of different

cultivars to assimilate nitrate. There were few differences in the rate of nitrate uptake per

g root, suggesting that differences in uptake rüere caused by the size of the root system.

Perby and Jensen (1933) also found the differences in N uptake of barley cultivar was ; /. Á. related to root size. In addition to root size, differences among cultivar of barley in net ion

uptake may also due to different flux rates into and out of the roots (Schimansky and

Marschner I97l; Glass et al. 1980; Glass and Perley 1980).

Over the 3 experiments cultivars which appeared to grow more vigorously, took up more

nitrate. These results agree with those of Hackett (1968) for barley and Reed and

Hageman (1980) for maize who found that high nitrate uptake was associated with a more 116 extensive root system. Results from field experiments (Chapter 3) show that at Northfield in 1991 dry matter production at 10 weeks of Weeah and Skiff was higher than that of

Stirling and Schooner (Table 4.12). Northfield 1991 was the most N-responsive site and also the site where dry matter production was lowest and most affected yields. Therefore, there was a different response to N in early growth between cultivars. The responses in early growth in Skiff and Weeah in this field experiment are consistent with the responses observed in the hydroponic studies, however, only Skiff showed a response in grain yield to N; Weeah was not responsive. Therefore, different levels of nitrate uptake in early growth are not always a benefit to grain yield. The yield responsiveness of cultivars will also depend on the characteristics of their response to environmental conditions during later stages of growth, particularly to and temperature stress. "vater

Table 4.12. Interaction between Cultivar and N in plant dry weight at l0 weeks after sowing at 0 and 45 kgN/ha in field experiment (Chapter 3)

Cultivar Northfield 1991 Charlick 1991 045 Relative 045 Relative response response (keN/ha) (vo) (kgN/ha) (vo)

Clipper 85 104 +22 182 310 +70 Stirling 65 9I +40 205 319 +56 Weeah 101 122 +21 202 255 +26 Schooner 74 102 +37 t96 296 +51 Chebec 96 92 0 225 368 +64 skiff 68 113 +66 203 304 +50

Mean 82 104 31 202 308 53

LSD(57o) 2I 56

6. 7. General conclusions

Results from these experiments in a hydroponic system show genetic differences in nitrate uptake between barley cultivars. These differences can be explained mainly by differences in plant growth. According to the results of these 5 experiments the lt7 differences in growth and nitrate uptake cha¡acteristic could be related to the pedigree of these cultivar, mainly through the size of the seedling. However, the Group IV cultivars do appear to have physiological differences from the other 3 Groups. There were some consistencies in the results from the hydroponic studies and measurements of early vegetative growth in a field experiment (Chapter 3) although the differences in vegetative growth may not necessarily be linked to yield. Skiff, Weeah, Stirling and Schooner all had high growth rates and nitrate uptake, but in the field experiments (Chapter 3) the grain yield of Skiff and Stirling were responsive to N in grain yield whereas the yield of Weeah and Schooner did not have or had a low response to N in grain yield. Therefore, high vegetative growth early in the growing season, although promoting uptake of nitrate from the soils, is not always related to the final grain yield response. However, it is possible to have a cultivar such as Skiff that can take up nitrate efficiently and which may be more responsive to increased nitrate in the soil. 118 CHAPTER 5 REMOBILISATION OF DRY MATTER AND NITROGEN DURING GRAIN FILLING IN BARLEY CULTIVARS

5.1. Introduction

In areas where cereals are grown in southern Australian it is common for available soil N and moisture availability to be low during the grain-filling period. Under such conditions the N requirement of the developing grain is largely satisfied by the mobilisation of N from vegetative tissues (Dalling et al.1976). In general, cereals have completed 70-100Vo of.their total acquisition of reduced N by anthesis. Remobilisation of this N from vegetative tissues is the principal source of N for the developing grain, although some studies have reported N uptake during grain filling accounting for as much as 50Vo of grain N at maturity

(Gregory et al. L981; Spiertz and Ellen 1978). As well, the mobilisation of dry matter formed during the pre-anthesis period makes a significant contribution to final yield, especially when post-anthesis photosynthesis is reduced by stress.

The influence of the environment on postanthesis N uptake has also been noted (Gregory et al. 1981) although few estimates of genotype x environment interactions have been made.

The importance of remobilised N for barley growing under Mediterranean conditions is noteworthy because conditions after anthesis are usually hot and dry, leading to problems of water stress and photosynthate limitation (Sofield et al. 1977; Papakota and Gagianas 1991).

Consequently, yield and grain protein concentration may depend largely on the translocation of pre-anthesis assimilates to the grain. There are reports that the N translocated from vegetative pafts to the developing grain after anthesis is under genetic control (van Stanford and Mackown 1987; Halloran 1981). Due to the likely importance of remobilised dry matter and N in cereal growing regions in South Australia a glasshouse study was conducted to examine the effect of N on the mobilisation of dry matter and N from vegetative organs after

anthesis in different cultiva¡s of barley under moisture stress and non-stress conditions. 119 The objectives of the present study were to:

i) evaluate differences in post-anthesis N and dry matter mobilisation among barley

cultivars; and

ii) determine the effects of post-anthesis water stress on the grain yield and GNC

response to N among barley cultivars.

5.2. Materials and Methods 5.2.1. Plant culture

The experiment was conducted in the glasshouse. Six barley cultivars, Clipper, Stirling,

Weeah, Schooner, Chebec and Skiff were grown at two levels of N and either kept well watered or stressed after anthesis as described under treatments.

The plants were grown in 15 cm diameter pots lined with plastic bags containing2.4 kg of steam sterilised John Innes potting mix (1:1 coarse sand: loam mix) to which no N was added, but which was adequate for the other plant nutrients. The potting mix had a pH of about 6.5. The amount of mineral N (ammonium-N, nitrate-N) in the soil was determined by extraction with 2N KCI and measured using an automated distillation procedure 2 t1

(Rayment and Higginson 1992). The concentration of inorganic N was 9.4 ttglg.

Four seeds per pot were sown on 19 May 1993 and seedlings thinned to 2 plants per pot after emergence. The experiment was set up as a factorial in a randomised complete block design with 4 replicates. Aphids when present were controlled using a pyrethrum spray.

5.2.2. Treatments 5.2.2.I. Nitrogen

Nitrogen was applied as a solution of ammonium nitrate (34 VoN) at rates €qËi+€t€at+e5o (LN) (¡¡¡) erdJee+gNr+È 264nand 530 mg¡f ammonium nitrate per pot respectively in 3 split

applications - at the two leaf stage, at tillering and at stem elongation. r20 5.2.2.2. Water

Plants were watered regularly until ear emergence to minimise the possibility of pre-anthesis

moisture stress. Two watering treatments (well-watered and water-stressed) were then

imposed between ear emergence and maturity, based on the water holding capacity of the

soil. The treatments were selected as follows: four pots containing soil, but with no plastic bag liners, were saturated with water, covered with black plastic and allowed to drain to a constant weight. The pots were then weighed and the moisture content of the soil measured.

It was found to be 17.3 Vo (wlw) and his was assumed to represent the field capacity of the t / soil in the pots. All pots were brought to field capacity at ear emergence and the water stress

treatment imposed after 3 days. Pots in the well watered treatment were maintained at field

capacity by frequent weighing and the addition of water to return them to field capacity,

while pots in the water stress treatment received half the quantity of water supplied to the

well watered pots at each watering.

5.2.2.3. Plant harvest

Harvests were made at 10 days after anthesis on the main stem (H1) and at maturity (H2). (grotit" + strrar,.r) Shõot, root and grain from each replicate were oven dried at 80'C for 48 h, weighed and the

N concentration determined by Kjeldahl N analysis (Appendix 3.1). The number of tillers

and ears per pot were also counted. Dry matter remobilised (DMR) and total N remobilised

(NR) between Harvest 1 and Ha¡vest 2 were derived from the following formulae:

DMR = non grain dry matter (H1)-non grain dry matter (H2)

NR = non grain total N (H1)-non grain total N (Hz)

where non grain dm - (total dry matter-grain dry weight) and non grain total ]r[ = (total N- grainN). At ll,, onl"¡ hølrr-Èhesse¡t eta^ts u{ú4¿ 1*çy6lad,.

5.3. Results 5.3.1. Growth at 10 days after anthesis (H1) 5.3.1..L. Dry matter production

There were no significant interactions between N and Cultivar in root or shoot dry weight

(Table 5.1). Root and shoot dry weight was higher in all cultivars at the higher rate of N. t2l No significant differences were found between cultivars in root or shoot dry weight (Table s.l).

5.3.I.2. Grain dry weight

There was a significant interaction between Cultivar and N for grain dry weight. Skiff showed a significant response to N at H1 whereas the yield of all the other cultivars was not affected by N treatments (Table 5.1).

5.3.1.3. Tiller number

There was a significant interaction between N and Cultivar in tiller number. At the low rate of N there were no differences between cultivars but at the high rate of N Skifl Schooner and Clipper produced significantly more tillers than the other cultivars. All cultivars responsed significantly to N (Table 5.1).

5.3.1.4. Plant N Root and shoot N

Root N concentration (RNC) was highest in Schooner, lowest in'Weeah and increased at the higher rate of N (Table 5.2). There were no significant differences between cultivars for root N content, but it was higher at the higher level of N. There was no interaction between

N and Cultivar for RNC or root N content.

Shoot N concentration (SNC) was significantly higher in Schooner and Clipper than in the other cultivars, while Chebec and Weeah had the lowest shoot N concentrations (Table 5.2).

There was no interaction between N and Cultivar in SNC and shoot N content (Table 5.2). Table 5.1. The influence of two levels of N on root, shoot dry weight, grain dry weight and tiller number of six barley cultivars 10 days after anthesis. Cultivars differences over all treatments are indicated by the me

Cultivar Root drv sht Shoot drv weisht Grain drv weisht Tiller number LNa HNa Mean LN HN Mean LN HN Mean LN HN Mean (g/pot) (g/pot) (g/poÐ (tillersþot)

Clipper 3.63 4.66 4.15 9.43 12.30 10.87 0.42 o.49 +# o.r+ç 10.3 16.0 13.2 Stirling 3.51 3.83 3.67 10.42 t4.41 12.45 . 0.76 0.58 0.67 9.0 13.8 11.0 Weeah 4.Ot 4.05 4.03 8.17 13.38 11.08 0.44 o.76 0.60 8.8 13.0 10.9 Schooner 2.85 4.3t 3.58 8.13 t3.o4 10.59 0.33 0.66 0.50 8.5 16.3 t2.4 Chebec 3.78 4.24 4.01 9.98 12.85 11.42 0.6r 0.67 o.64 10.0 13.8 tt.9 skiff 2.83 4.17 3.50 9.43 13.73 I 1.58 0.66 1.28 0.97 8.8 16.0 12.4

Mean 3.44 4.2t 3.82 9.36 13.30 11.33 0.54 0.74 ufrr o-b4 9.2 r4.8 11.97

LSD (57o) Cultivar NS NS o.25 0.98 N o.62 0.82 o.20 0.80 Cultivar x N NS NS 0.35 1.96 aLN=Low F) h¡û.t¿¡ N, HN=High N i rnQox, o+ o.ll- h¿a"f¡^ AwE -

N) N) 123 Grain N concentration

There was no interaction between N and Cultivar for GNC. There were no significant differences between cultivars but the higher rate of N gave an increase in GNC in all cultivars (Table 5.2).

5.3,2. Growth at maturity (Hz) 5.3.2.1. Dry matter production

There was no significant interaction between Cultivar, N and Water treatment in root dry weight (Table 5.4). Root dry weight increased significantly by adding N only in Clipper,

Skiff and V/eeah (Table 5.3). The high level of N increased root dry weight significantly under water stress conditions.

Shoot dry weight increased significantly with N in all cultivars (Table 5.3) but the greater increase occurred under non-stressed conditions (Table 5.4). The intéraction between

Cultivar, N and Water stress was not significant for shoot dry weight at H2 (Table 5.4).

Table 5.3. The influpnce of two levels of N on root and shoot dry weight of six barley cultivars at maturity. AvAroqaL over sù'€ssc¿ rixol. 7r¡¡-shessc.d- treatt"¿n*s.

Cultivar Root drv weisht Shoot dry weight b LNa HNa Mean LN HN fvfean (9pot) (g/pot)

Clipper 2.48 4.33 3.41 13.78 19.38 16.58 Stirling 3.65 3.81 3.7r 13.58 18.48 16.03 Weeah 3.18 4.t2 3.65 14.60 19.89 17.25 Schooner 2.76 2.6r 2.69 T2.47 18.t2 t5.27 Chebec 3.43 3.45 3.44 13.70 r8.59 16.15 skiff 2.27 3.67 2.98 12.27 17.39 14.83

Mean 2.96 3.67 3.3r 13.39 18.64 16.02 LSD (s%) Cultivar 0.61 r.57 N 0.35 0.90 Cultivar x N 0.86 NS b aLN=Low inc.t¡r,d¿S N, HN=High N Srafrr" Table 5.2. The influence of two levels of N on root N concentration, root N content, shoot N concentration, shoot N content and GNC of six barley cultivars 10 days after anthesis

Cultivar RNCb Root N content SNCb Shoot N content GNCC LNa HNa Mean LN HN Mean LN HN Mean LN HN Mean LN HN Mean (vo) (mg/poÐ (7o) (mg/poÐ (vo)

Clipper o.67 0.93 0.80 2.43 4.35 3.39 t.r2 r.43 1.27 9.66 16.88 12.37 1.86 2.t5 2.01 Stirling 0.42 0.99 o.82 2.t9 3.84 3.10 t.o7 1.28 1.18 II.4L t8.37 13.66 t.7I 2.O8 1.89 Weeah 0.61 0.92 o.76 2.43 3.69 3.10 1.08 r.t2 l. l0 8.98 15.45 11.06 1.79 1.96 1.88 Schooner 0.74 1.06 0.90 2.tt 4.49 3.30 1.28 1.52 1.40 9.47 20.r2 13.81 1.99 1.93 1.97 Chebec 0.63 0.9r 0.77 2.38 3.89 3.r3 0.95 1.15 1.05 9.41 14.55 1o.77 r.69 2.O5 1.87 skiff 0.69 0.84 o.77 1.95 3.49 2.12 1.18 t.2t 1.19 tt.2l 18.75 12.80 1.98 2.24 2.rl

Mean 0.66 0.94 0.80 2.25 3.96 3.12 1.11 1.28 1.19 10.02 17.35 t2.4r r.84 2.07 r.96

LSD (57o) Cultiva¡ 0.08 NS 0.14 1.89 NS N 0.05 0.53 0.08 1.09 0.12 Cultiva¡ x N NS NS NS NS NS aLN=Low N, HN=High N, bRNC=Root N concentration, SNC=Shoot N concentration, CGNC= Grain N concentration

N) 5 t25

Table 5.4. The effect of N and water stress on root and shoot dry weight of six barley cultivars at maturity. Water stress treatments were imposed after anthesis

Water stressed Well watered Cultivar LNa HNa Mean LN HN Mean (g/pot)

Root dry weight Clipper 2.32 4.32 3.32 2.65 4.33 3.49 Stirling 2.50 3.45 2.98 4.79 4.t7 4.48 Weeah 2.75 3.43 3.09 3.60 4.82 4.21 Schooner 1.80 2.13 1.97 3.72 3.10 3.41 Chebec 2.67 3.22 2.95 4.t9 3.68 3.94 skiff 1.57 3.82 2.70 2.97 3.52 3.25

Mean 2.27 3.39 2.83 3.66 3.94 3.67

LSD (57o) Water x N r.28 Cultivar x N 0.86 Cultivar x'Water NS CultivarxNxWater NS

Shoot dry weightÞ

Clipper 1 1.87 t6.42 T4.T5 15.69 22.33 19.01 Stirling t2.4r t4.49 13.45 r4.75 22.48 18.62 Weeatr 12.93 16.29 14.61 t6.28 23.50 19.89 Schooner IT,7I 14.64 r3.06 13.1 3 2r.60 t7.37 Chebec 11.88 14.21 13.05 15.52 22.96 19.24 skiff I1.40 t4.92 13.16 13.13 19.86 16.49

Mean 12.03 15.16 13.60 14.t5 22.2t 18.69

LSD (57o) Water x N 0.50 Cultivar x N NS Cultiva¡ x'Water NS CultivarxNxWater NS

aLN=Low b N, HN=High N h.rt-rr^ 3l'4t^ t26 5.3.2.2. Grain yield

There was a significant interaction between Cultivar, N and'Water stress in grain yield at H2

(Table 5.5). Under water stress conditions and the low rate of N there were no significant differences between cultivars except for Skiff which had the lowest grain yield. Under well watered conditions and low N there was no differences between Weeah, Chebec and Stirling in grain yield, but Skiff and Schooner had significantly lower grain yields (Table 5.6).

Table 5.5. The summary of analyses of variance of grain yield and yield components

Grain Ea¡ Kernel Kernel GNC SDMRa SNRa yield No No. weight

Cultivar * * NS NS NS * t(*t< N * *r< *** ** ** ,

Table 5.6. The influence of N and water on grain yield of six barley cultivars at maturity. Water stress treatrnents were imposed after anthesis

Water stressed Well watered Relative response to N Cultiva¡ LNa HNa Mean LN HN Mean at well watered (g/pot) (vo)

Clipper 5.39 6.26 5.83 5.86 8.09 6.98 +38 Stirling 6.32 3.56 4.94 5.98 10.35 8.17 +73 Weeah 4.88 4.39 4.64 7.54 7.25 7.40 -4 Schooner 5.38 4.87 5.13 5.02 9.54 7.28 +90 Chebec 5.r7 3.21 4.r9 7.r5 10.38 8.77 +45 skiff 2.98 4.57 3.78 4.26 9.45 6.86 +122

Mean 5.02 4.48 4.75 5.97 9.18 7.58 +60

LSD (57o) N x water 0.68 Cultivar x'Water 1.18 CultivarxNxwater r.67

aLN=Low N, HN=High N r27 Under water stress conditions the cultivars Clipper, Weeah, Schooner and Skiff showed no significant response to N while the yield of Stirling and Chebec fell. In contrast there was a large and positive response to N under well watered conditions in all cultivars except in 'Weeah. The highest response to N under well watered conditions was observed for Skiff,

Schooner and Stirling. Chebec and Clipper were intermediate in their response to N.

5.3.2.3. Yield components Ear number

There was no significant interaction between Cultivar, N and'Water for ear number at maturity (Table 5.7). There was a significant interaction between N and Cultivar for ear number at maturity (Table 5.5,5.7). Ear number was significantly higher at the high N rate for all cultivars except Weeah. Skiff showed the highest response to N (Table 5.7). Ear number was reduced by postanthesis water stress equally at both levels of N (Water x N interaction non-significant, Table 5.5); overall ear number was reduced from 10.0/pot to

8.8/pot by postanthesis stress.

Table 5.7. The influence of two levels of N on ear number per pot of six barley cultivars at maturity

Water stressed Well watered Cultivar LNa HNa Mean LN HN Mean Cultivar mean

Clipper 7.75 10.00 8.88 8.75 10.75 9.75 9.06 Stirling 7.70 8.25 7.88 7.00 11.50 9.25 8.56 Weeah 7.75 7.50 7.63 8.75 10.00 9.38 8.50 Schooner 7.75 ro.25 9.00 8.75 11.25 10.00 9.50 Chebec 7.70 11.50 9.50 8.50 12.50 10.50 10.00 skiff 7.00 13.00 10.00 8.50 14.25 I 1.38 10.69

Mean 7.54 10.08 8.82 8.38 tt.7r 10.04 9.39

LSD (57o) Cultivar 1.5 Water 0.9 Water x N NS Cultiva¡ x N 2.1 Cultivar x Water NS CultivarxNxW'ater NS

aLN=Low N, HN=High N r28 Table 5.8. The effect of N and water stress on kernel number per pot of six barley cultivars at maturity. Water stress treatments were imposed after anthesis 'Well Water stressed watered Cultivar LNa HNa Mean LN HN Mean Cultiva¡ mean

Clipper II7 131 r24 139 183 161 143 Stirling 151 83 TL7 t54 239 r97 r57 V/eeah 97 111 t04 159 194 177 r40 Schooner tr7 103 110 145 202 173 r42 Chebec 103 94 99 r57 225 19l r45 skiff 69 100 85 99 202 151 118

Mean 109 r04 r0l r42 208 r75 I4I

LSD (57o) Cultivar NS 'Water t7.5 'Water x N 24.3 Cultivar x N NS Cultivar x Water NS CultivarxNxWater NS aLN=Low N, HN=High N

Table 5.9. The effect of N and water stress on kernel weight of six barley cultivars at maturity. Water stress treatments were imposed after anthesis

Water stressed Well watered Cultivar LNa HNa Mean LN HN Mean (mg)

Clipper 44.r 43.4 43.7 47.3 46.O 46.7 Stirling 45.5 35.0 40.3 50.9 47.4 49.2 Weeah 45.8 45.9 45.9 47.3 48.1 47.7 Schooner 42.9 4t.4 42.2 46.t 50.2 48.2 Chebec 47.0 29.r 38.1 49.3 45.6 47.5 skiff 45.7 36.7 4r.2 43.r 5r.4 47.3

Mean 45.2 38.6 4r.9 47 .3 48.1 47 .7

LSD (57o) Water 1,7 Water x N 2.4 Cultivar x Water NS CultivarxNxWater 5.8

aLN=Low N, HN=High N r29 Kernel numher

Water stress reduced kernel number significantly. Also applying a higher rate of N

increased kernel number per pot under well watered condition, but there was no effect of N

on kernel number with post-anthesis stress (Table 5.8).

Kernel weight

There was a significant Cultivar x N x Water interaction for kemel weight at maturity (Table

5.5). Postanthesis water stress caused significant reductions in kernel weight at the higher

level of N in Stirling, Chebec and Skiff, whereas at the low level of N water stress had no

effect on kernel weight.

Under well watered conditions, N level had no effect on kernel weight in any cultiva¡ except

Skiff, in which there was a significant increase, while with postanthesis stress, kernel

weight was lower in the higher N treatment in Skiff, Stirling and Chebec (Table 5.9).

5.3.2.4. Plant N at maturity (H2) Root and shoot N

At maturity the concentration and content of N in the roots was lower than at H1. Root N

concentration was significantly lower for Skiff compared with Clipper and Schooner at the

low level of N only (Table 5.10) and RNC did not differ significantly between the other

cultivars. Over all cultivars at both water treatments the higher level of N significantly

increased root N content (Table 5.11).

There was no interaction between Cultivar and N in SNC and shoot N content at maturity

(Tables 5.10, 5.12). There were no significant interaction between Cultivars, N and Water

in SNC and shoot N content (Table 5.I2,5.I3). Under water stressed conditions the higher

level of N increased shoot N concentration while under well watered conditions SNC

decreased significantly (Table 5.12). Table 5.10. The influence of two levels of N on root N concentration, root N content, shoot N concentration and shoot N content of six barley cultivars at maturity

Cultivar RNCb Root N content SNCb Shoot N content LNa HNa Mean LN HN Mean LN HN Mean LN HN Mean (vo) (mglpoÐ (vo) (mglpoÐ

Clipper 0.56 0.60 0.58 1.38 2.51 1.97 o.7l 0.71 0.71 12.8t t6.76 14.78 Stirling 0.47 0.66 o.57 1.72 2.50 2.lt 0.66 0.80 0.73 10.96 15.95 13.45 Weeah 0.49 0.73 0.61 1.62 2.95 2.29 0.68 0.65 0.66 12.80 14.57 13.69 Schooner 0.54 0.74 o.64 1.48 1.85 r.67 o.77 0.69 o.73 14.13 14.31 14.22 Chebec 0.45 0.76 0.61 1.53 2.59 2.06 o.69 0.13 o.7l tI.39 t5.97 13.68 skiff 0.43 0.62 o.52 0.96 2.24 1.60 o.67 0.66 0.61 11.81 13.r2 12.46

Mean o.49 0.68 0.59 r.45 2.45 1.95 0.70 0.7t 0.70 12.32 15.11 13.71

LSD (57o) Cultivar 0 01 0.44 NS NS N 0 04 0.26 NS 1 .25 Cultivar x N 0 09 NS NS NS aLN=Low N, HN=High N, bRNC=Root N concentration, SNC=Shoot N concentration

ou) 131 Table 5.1 1. The effect of N and water stress on root N content of six barley cultivars at maturity. Water stress treatments were imposed after anthesis

Water stressed Wellwatered Cultivar LNa HNa Mean LN HN Mean Cultiva¡ (mg/pot) mean

Clipper 1.40 2.75 2.08 1.36 2.39 1.88 1.98 Stirling r.23 2.76 2.00 2.20 2.23 2.22 2.tt Weeah r.49 2.88 2.19 t.76 3.02 2.39 2.29 Schooner 1.00 1.91 r.46 r.97 t.79 1.88 r.61 Chebec t.24 2.9r 2.08 1.83 2.27 2.05 2.06 skiff 0.72 2.55 t.64 r.20 t.94 t.75 1.60

Mean 1.18 2.63 1.91 t.72 2.27 2.OO 1.95

LSD (57o) Cultivar 0.44 Water 0.26 Cultivar x N NS Cultivar x'Water NS Water x N 0.36 Cultivar x N x'Water NS aLN=Low N, HN=High N

Table 5.12. The effect of N and water stress on shoot N concentration of six barley 'Water cultivars at maturity. stress treatments were imposed after anthesis

W'ater stressed Well watered Cultivar LNa HNa Mean LN HN Mean Cultivar (vo) mean

Clipper o.74 0.80 o.77 0.68 o.62 0.65 0.71 Stirling 0.67 0.99 0.83 0.66 0.60 0.63 0.73 V/eeatr o.74 0.78 0.76 0.61 0.53 0.57 0.67 Schooner 0.81 0.82 0.82 0.73 0.55 0.64 0.73 Chebec 0,80 0.91 0.86 0.59 0.60 0.60 0.73 skiff 0.68 0.74 o.7t 0.67 0.58 0.63 0.67

Mean o.74 0.84 0.79 0.66 0.57 0.62 0.70

LSD (57o) Cultivar NS Vy'ater 0.05 Cultiva¡ x N NS Cultivar x Water NS Water x N 0 .o7 CultivarxNxWater NS

aLN=Low N, HN=High N r32

Table 5 . I 3. The effect of N and water stress on shoot N content (grain + straw) of six barley cultivars at maturity. Water stress treatments were imposed after anthesis

Water stressed Well watered Cultivar LNa HNa Mean LN HN Mean Cultivar (mg/pot) mean

Clipper tL.34 16.92 14.t3 14,27 15.60 t4.94 14.78 Stirling 9.81 t7.97 13.89 t2.r0 13.93 t3.02 13.45

V/eeatr 13.79 t4.49 r4.t4 1 1.81 14.66 13.24 13.69 Schooner 13.60 15.51 t4.56 14.66 13.11 13.89 14.22 Chebec 12.77 16.07 14.42 10.01 15.87 12.94 13.68 skiff 10.58 12.07 11.33 13.04 14.16 13.60 t2.47

Mean 11.98 15.50 t3.74 12.65 14.72 13.69 13.71

LSD (57o) Cultiva¡ NS N 1 .25 Water NS Cultiva¡ x N NS Cultivar x Water NS V/ater x N NS CultivarxNxWater NS aLN=Low N, HN=High N

With post anthesis stress, the shoot N content increased at the high N level although not significantly so (Table 5.13).

Shoot N content increased with higher N but was not significantly affected by water stress.

The amount of total N in the plant at maturity was not different from that at H1 (Appendix

Table 5.1).

Grain N concentration

On average GNC did not differ significantly between cultivars at maturity (Table 5.14).

Averaged over all cultivars water stress increased GNC at the high level of N (1.367o cf.

2.27Vo ) but not significantly at the low level of N. Adding extra N increased GNC both

under stress and non-stress conditions (Table 5.14), but the increase was greater with

postanthesis stress. This response was the same in all cultivars. 133 Graín N yield

There was a significant interaction between Cultivar, N and V/ater for grain N yield per plant

(Table 5.15). At the low level of N, under water stress conditions, grain N yield was significantly lower for Skiff than in all the other cultivars except Chebec. Grain N yield increased significantly in Clipper, Weeah and Skiff with increased N application. At the low level of N under well watered conditions Skiff had the lowest grain N yield while Weeah gave the highest. At the high N rate however Vy'eeah had the lowest grain N yield while

Schooner had the highest (Table 5.15).

Cultivars differed in their increase in grain N yield at maturity in responses to applied N

(Table 5.15). Skiff had the highest response to N, it was lowest in Weeah, Stirling and

Chebec and moderate in Schooner and Clipper. In the well watered treatment the mean over all cultivars for grain N yield was higher than in water stressed treatment. ]4L K

5.3.3. Dry matter and N remobilisation 5.3.3.1. Root dry matter and N remobilisation There was no significant interaction between Cultivar, N and Water in dry matter remobilisation from the root (Table 5.16). Overall, water stress increased root dry matter remobilisation at both levels of N but there was no effect of N or cultivar (Table 5.16).

There was no significant interaction between Cultivar, N and V/ater stress in root N remobilisation (Table 5.17). Water stress reduced root N remobilisation at the higher level of N. There was no significant difference between cultivars in the amount of N remobilised. 134 Table 5.14. The effect of two levels of N and water stress on GNC of six cultivars of barley at maturity. Water stress treatments were imposed after anthesis

Water stressed Well watered Cultivar LNa HNa Mean LN HN Mean Cultivar (vo) mean

Clipper 1.26 2.00 r.63 1,.25 1.48 t.37 1.50 Stirling r.23 2.30 r.77 t.t4 1.23 1.69 1.48 V/eeah 1.50 2.42 r.96 1,13 1.49 1.31 t.64 Schooner r.44 2.t0 r.77 t.34 1.39 r.37 1.57 Chebec t.22 2.55 1.89 1.15 r.26 t.2r 1.55 skiff r.42 2.25 1.84 1.39 1.32 r.36 1.60

Mean 1.35 2.27 1.81 1.23 r.36 1.30 1.55

LSD (57o) Cultiva¡ NS N 0 .09 'Water 0.92 Cultivar x N NS Cultiva¡ x Water NS Water x N 0.13 CultivarxNxWater NS

aLN=Low N, HN=High N

Table 5.15. The effect of two levels of N and water stress on grain N yield of six cultivars of barley at maturity. Vy'ater stress treatments were imposed after anthesis

Water stressed Well watered Cultivar LNa HNa Mean LN HN Mean Cultivar (mg/pot) mean

Clipper 6.77 11.60 9.t9 7.32 11.11 9.22 9.20 Stirling 7.79 8.20 7.99 6.84 r2.76 9.80 8.89 V/eeatr 7.39 r0.59 8.99 8.56 t0.79 9.68 9.33 Schooner 7.76 9.97 8.87 6.78 13.23 10.01 9.44 Chebec 6.29 8.04 7.17 8.10 13.04 10.57 8.87 skiff 4.22 10.33 7.26 5.92 12.53 9.23 8.25

Mean 6.70 9.79 8.25 7.25 12.24 9.75 8.99

LSD (57o) Cultiva¡ NS Vy'ater o.67 Cultivar x N r.63 Water x N 0.94 Cultiva¡ x'Water NS CultivarxNxWater 2.29

aLN=Low N, HN=High N 135

Table 5.16. The effect of two levels of N and water stress on dry matter remobilisation from the root of six cultivars of barley. Water stress treatments were imposed after anthesis

Water stressed Well watered Cultivar LNa HNa Mean LN HN Mean Cultivar (g/pot) mean

Clipper 1.31 0.33 0.82 0.98 0.33 0.66 o.74 Stirling 1.01 0.39 0.70 -t.28 -0.34 -0.81 -0.06 V/eeah t.25 0.62 0.94 0.40 -0.77 0.59 0.77 Schooner 1.06 2.17 t.62 -0.86 t.2r 1.04 r.33 Chebec I.I2 1.01 1.07 -0.4r 0.56 0.49 0.78 skiff r.25 0.36 0.81 -0.15 0.65 0.40 0.6r

Mean r.t7 0.81 0.99 -o.22 0.28 0.03 0.51

LSD (57o) Cultivar NS Water 0.56 Water x N NS Cultivar x N NS Cultiva¡ x'Water NS Cultiva¡xNxwater NS

aLN=Low N, HN=High N

Table 5.17. The effect of two levels of N and water stress on root N remobilisation of six cultivars of barley. Water stress treatments were imposed after anthesis

Water stressed Well watererl Cultivar LNa HNa Mean LN HN Mean Cultivar (mg/pot) mean

Clipper 1.04 1.60 1.32 t.07 r.96 r.52 t.42 Stirling 1.15 1.07 1.1r -0.01 1.61 0.80 0.96 V/eeah o.94 0.81 0.88 0.67 0.68 0.67 o.76 Schooner 1.11 2.58 1.85 0.14 2.70 1.42 r.64 Chebec t.r4 0.99 1.07 0.54 1.62 1.08 r.08 skiff 1.23 0.94 1.09 0.75 1.56 1.16 1. 13

Mean 1 . 10 t.33 r.22 0.52 1.69 1.11 r.r7

LSD (57o) Cultiva¡ NS 'Water NS Water x N 0 .61 Cultiva¡ x N NS Cultivar x'Water NS Cultiva¡xNxwater NS

aLN=Low N, HN=High N 136 5.3.3.2. Shoot dry matter and N remobilisation There was no significant Cultivar x N x Water interaction for Shoot dry matter

remobilisation per pot (Tables 5.5, 5.18), nor was there a significant interaction between N

and Cultivar in shoot dry matter remobilisation per pot. There were however significant

differences between cultivars in shoot dry matter remobilisation per pot (Table 5.18). Shoot

dry matter remobilisation was lowest in Weeah and highest in Stirling. Shoot dry matter

remobilisation increased under water stressed conditions. Over all cultivars and N

treatments shoot dry matter remobilisation under water stressed conditions was 1.89 g/pot

while under well watered conditions there \ryas no remobilisation.

There was no significant interaction between Cultivar, N and Water in N remobilisation

(Tables 5.5, 5.19) but there was between Cultivar and N (Table 5.20). At the low N rate,

the N remobilisation from the straw did not differ significantly between cultivars. At the . higher N rate Skiff, Schooner and Stirling remobilised more N than the other cultivars (Table 5.20). Post-anthesis water stress decreased slightly N remobilisation, but ruvrrobu lüso.fi ar^., uns significantly more under well watered conditions (8.9 gN cf 9.1 gN).

^e

5.3.4. Contribution of shoot dry matter remobilisation to grain yield and N remobilisation to grain N yield

There was no significant Cultivar, N and Water interaction in the estimated contribution of

shoot dry matter mobilisation to grain yietd (Table 5.21). V/eeah and Clipper had little shoot

dry matter mobilisation and therefore its contribution to grain yield was low in these

cultiva¡s. Water stress increased the contribution of shoot dry matter remobilisation to grain

yield in all cultivars (Table 5.21), but N had no effect (Table 5.23). There was no

correlation between shoot dry matter mobilisation and grain yield (Fig. 5.1). The increase in

shoot dry matter mobilisation at the higher level of N was significantly correlated (r=0.95**)

with the increase in grain yield with N only under well watered conditions (Fig. 5.2). The

cultivars whose yield was most responsive to N, Skiff, Schooner and Stirling, had high

contribution of shoot dry matter remobilisation to grain, while the unresponsive cultivars 'Weeah and Clipper had relatively little shoot dry matter remobilisation. 137

N remobilisation as a proportion of grain N yield differed between cultivars (Table 5.23). It was highest in Skiff, Stirling and Schooner which also had the higher amounts of N remobilisation (Table 5.20). There was no significant interaction between Cultivar, N and

Water in the contribution of N remobilisation to grain N yield (Table 5.22). Grain N yield was not correlated with N remobilisation (Fig. 5.1).

Table 5.18. The effect of two levels of N and water stress on shoot dry matter remobilisation of six cultivars of barley, Water stress treatments were imposed after anthesis

Water stressed Well watered Cultivar LNa HNa Mean LN HN Mean Cultivar (g/pot) mean

Clipper 2.52 1.65 2.09 -0.83 -1.68 -r.26 0.42 Stirling 3.56 2.97 3.27 0.89 1.70 1.30 2.28 Weeah 0.27 0.72 0.50 -0.41 -2.77 -1.59 -0.55 Schooner 1.46 2.6r 2.04 -0.32 1.13 0.41 r.22 Chebec 3.09 r.43 2.26 1.01 -0.38 0.32 r.29 skiff 0.35 1.93 I.I4 -0.25 1.87 0.81 0.98

Mean 1.88 r.89 1.89 o.02 -0.02 0.0 0.94

LSD (57o) Cultiva¡ 1.53 Water 0.89 Cultiva¡ x N NS Cultivar x Water NS Water x N NS Cultivar x N x'Water NS

aLN=Low N, HN=High N 138 Table 5.19. The effect of two levels of N and water stress on remobilisation of N in six cultivars of barley at maturity. Water stress treatrnents were imposed after anthesis

Vy'ater stressed Well watered Cultivar LNa HNa Mean LN HN Mean Cultivar (mg/pot) mean

Clipper 6.73 9.70 8.22 5.87 1 1.09 8.48 8.35

Stirling 8.76 1 1.59 10.18 8.10 t3.39 r0.75 10.46 Weeah 5.96 9.3r 7.64 5.60 9.32 7.46 7.55 Schooner 6.29 t3.75 t0.02 5.82 14.20 10.01 to.02 Chebec 6.84 6.97 6.9r 6.09 7.72 6.9r 6.9r skiff 8.19 t2.43 10.31 7.92 13.65 r0.79 10.55

Mean 7.13 10.62 8.88 6.57 11.56 9.07 8.97

LSD (57o) Cultiva¡ r.34 'Water 0.77 Cultivar x Water NS Cultivar x N 1.89 Water x N NS Cultivar x N xWater NS aLN=Low N, HN=High N s\æf (c¡nò (sNQ) Table 5.20. The effect of cultivar on^dry matter^and N remobilisatioqof six barley cultivar. Negative value is mean amount of'öry mater'iemobilised and positive sign is mean no remobilisation of dry matter

SDMRb SNRb Cultivar LNa HNa Mean LN HN Mean (g/pot) (mg/pot)

Clipper 0.85 -0.02 0.42 6.30 10.39 8.35 Stirling 2.23 2.34 2.28 8.43 12.49 10.46 Weeah -0.07 -t.02 -0.55 5.78 9.32 7.55 Schooner o.57 1.87 t.22 6.06 13.98 10.02 Chebec 2.05 0.53 1.29 6.46 7.34 6.90 skiff 0.05 1.90 0.98 8.05 13.04 10.5s

Mean 0.9s -0.93 0.94 6.85 11.09 8.97

LSD (57o) Cultivar 1 .53 t.34 N NS 4.25 Cultivar x N NS 1.89 aLN=Low N, HN=High N, b SDMR=Shoot dry matter remobilisation, SNR= Shoot N remobilisation 139 Table 5.2t. The effect of two levels of N and water stress on the contribution of shoot dry matter remobilisation to grain yield of six cultivars of barley at maturity. Water stress treatments were imposed after anthesis

'Wafer stressed Well watered Cultivar LNa HNa Mean LN HN Mean Cultivar (vo) mean

Clipper 49.5 30.5 40.0 -r3.2 -36.1 -24.7 7.7 Stirling 74.9 56.6 65.8 10.1 35.2 22.6 44.2 Weeatt 9.5 r7.4 13.4 -r5.7 -40.2 -27.9 -7.3 Schooner 31,0 48.8 39.9 -5.1 29.9 t2.4 26.2 Chebec 53.8 34.9 44.4 26.0 -3.6 rt.2 27.8 skiff t0.7 60.5 35.6 -5.7 23.0 8.7 22.1

Mean 38.3 4t.5 39.9 -0.6 r.4 0.4 20.r

LSD (57o) Cultivar 30.3 Water t7.r Cultivar x N NS Cultivar x Water NS N x water NS CultivarxNxwater NS aLN=Low N, HN=High N

TabIe 5.22. The effect of two levels of N and water stress on the contribution of N remobilisation to grain N yield of six cultivars of barley at maturity. Water stress treatments were imposed after anthesis

Water stressed Well watered Cultivar LNa HNa Mean LN HN Mean Cultiva¡ (vo) mean

Clipper 83.2 107.3 93.3 61.5 r03.7 82.6 88.9

Stirling 122.8 r07.4 1 15.1 99.9 t53.3 126.6 120.9 V/eeatr 61.5 t16.t 89.1 65.0 r23.6 94.3 9r.7 Schooner 68.s t52.9 110.9 57.9 225.7 141.8 126.3 Chebec 73.3 81.6 77.5 98.5 98.2 98.4 87.9 skiff 134.0 282.5 208.3 87.4 165.6 126.5 t67.4

Mean 90.6 14r.4 116.0 78.4 145.0 111.0 rr3.9

LSD (57o) Cultivar 41.8 Water NS Cultivar x N 58.9 Cultivar x Water NS N x water NS Cultiva¡xNxwater NS

aLN=Low N, HN=High N 140 Well watered

(a) (b) 9.0 r=+0.35 NS 10.6 a r=-0.44 NS o Chebec 10.4 ^ 8.5 Chebec o o o o- 10.2 Stirling è, a Ëro lJ 10.0 Schooner õ /.c o 9.8 '= o Stirling' vrlàean z 9.6 we3arr .= 7.0 a Schooner c Clipper o .E E skiff 9.4 o o.s (t o o 9.2 cl per skiff 6 9 -2-1012 6 7 8 I 10 11 Shoot dry matter remobilisation (g) N remobilisation (mg)

Water stressed

(c) (d) 6.0 9.5 r=+0.35 NS o r=-0.69 NS Clipper o ^ 5.5 8 g.o o Clipper o o- Weeah o o ct) Schooner Schooner o Ëu.o Stirling e.s a Io Ë o.u '= o z 8.0 o .c_ 4.0 Chebec t- Stirling g o ru o 3.s skitf E skitf . 3 7 0.0 1.0 2.0 3.0 4.0 6 7 I I 10 11 Shoot dry matter remobilisation (g) N remobilisation (mg)

Figure 5.1. Relationship between shoot dry matter remobilisation and grain yield (ar c), N-remobilisation and grãin N yield (b, d) in well watered and water stressed conditions for 6 baley cultiva¡s. t4l Well watered

7.O 6.0 l=+0.95*' r=+0.51 NS skitf o a a o a Schooner €. s.o Schooner o- 6,0 ¡ Skiff o) Stirling -o o t9 +.0 Stirling E 5.0 o Þ .fÐ Chebec a 4.0 9. g.o z O .E Chebec c, Clipper a (ú 3.0 g 2.0 Clipper o) o a U' 2.0 H r.o (ú Weeah E E() 1.0 Ë 0.0 ¡Weeah -1 0 -3-2-10123 0123456789 lncrease N remobilisation (mg) lncrease shoot dry matter remobilisation (g)

Water stressed

(c) (d)

2.O r=+0.57 NS 7.O r=+0.52 NS o 1'5 o a skiÍÍ o- 6.0 o ê l.o o o, skitf 3 o.s Clipper Þ 5.0 o -9. .o) 0.0 4.0 Clipper .s -0,5 o o z Weeah 'õc o 8, -r.o Schooner 3.0 o) Weeah E -l.s o o (ú Ø 2.0 (ú -2.0 o (t) Schooner E L E Chebec () 1.0 -2.5 Stirltng L Stirling -3 0 -2-1 0'l 2 012 3 4 5 6 7 8 lncrease shoot dry matter remobilisation (g) lncrease N remonilisation (mg)

Figure 5.2. Relationship between incre¿se in grain yigld 9n{ ¡ltoo-t dry.matter reñrobilsation (a, c), inciease N remobilisation and grain N yield (b' d) in well watered and water stressed conditions of six barley cultivars r42

Table 5.23. The relative contribution of shoot dry matter remobilisation to grain yield and Nremobilisation to grain N yield during grain hlling in six barley cultivars bsovvcy(zoi bNnrcNy(ør) Cultivar LNa HNa Mean LN HN fvfean

Clipper 18.1 -2.8 7.7 72.3 105.5 88.9 Stirling 42.5 45.9 44.2 111.3 r 30.3 120.8 Weeah -3.r -r1.4 -7.3 63.3 r20.2 91.7 Schooner t2.9 39.3 26.t 63.5 189.3 126.4 Chebec 39.9 15.7 27.8 85.9 89.9 87.9 skiff 2.5 41.8 22.2 rt0.7 224.1 167.4

Mean 18.8 21.4 20.r 84.5 143.2 113.9 LSD (s%) Cultivar 30.3 4 r.8 N NS 24.r Cultivar x N NS 58.9

aLN=Low N, HN=High N, bSplrrlUCY =The contribution of shoot dry matter remobilisation to grain yield, bNR/GNY=The contribution of N remobilisation to grain N yield

5.4. Discussion

The aims of this experiment were to examine differences in post-anthesis N and dry matter

remobilisation between barley cultivars and to determine the effects of post-anthesis water

stress on grain yield and GNC response to N. Uptake of N by wheat and barley after

anthesis may vary markedly depending upon growth conditions, nutrient supply and

genotype (Austin and Jones 1975; Ellen and Spiertz 1980; Schorring et al. 1989). N was

supplied to the plants prior to anthesis, and the accumulation of N by the grains was

dependent mainly on N mobilised from vegetative tissues (Appendix 5.1). The total N

measured in the root and shoot at maturity was not different from that at 10 day after

anthesis. Therefore during the post-anthesis period there was a re-allocation of N from root

to shoot (Appendix Table 5. t).

At the first harvest, all cultivars showed a similar increase in shoot dry weight witþ an $¡a-ut"^ application of N. However, even at this early stage (10 days after anthesis) the mean Yield r43 of Skiff was greater than that of the other cultivars and it was the only cultivar to show a significant yield increase with N (Table 5.1). Therefore the greater yield response found at maturity in Skiff was already apparent at the early stages of grain growth. The increase in the number of ears/pot was greater in Skiff than in the other cultivars (Table 5.7) and similarly the increase in kernel number at the higher level of N was greater. Although these increases may help explain the greater response to N of Skiff at the first harvest they do not explain why its yield was much greater. A possible reason for Skiffs higher yield is that not only did it produce more fertile tillers in response to N, but these tillers flowered and started to produce grain shortly after the main stem flowered. Harvest one was based an anthesis in the main stem and so the yield at this time was affected by the development of the tillers relative to the main stem. Tiller development in Skiff was more synchronous than in the other cultivars, so the contribution of tillers to total grain yield at H1 was greater than for the other cultivars. Therefore, the greater responsiveness of Skiff at H1, which carried through to maturity, ñây be related to both its greater tillering capacity as well as the synchrony of ear emergence and grain set.

The results from this pot experiment were generally similar to these observed in the field experiments presented in Chapter 3. Skiff was the most responsive cultivar and Weeah was the least responsive. Stirling also showed a large response to N, but unlike the field experiments, Schooner was responsive to N in the glasshouse. As in the field, the responsiveness of these cultivars was related to the number of ears produced and to the number of kernels/pot. As already discussed, Skiff showed a large response in both ears/pot and kernels/pot, while'Weeah produced fewer ears/pot and did not respond to the higher level of N.

The greater response observed in Skiff in this experiment was associated with its poor yield at low N rather than a high yield at the higher level of N. Under both levels of watering

Skiff was the lowest yielding cultivar at the low level of N; adding N substantially increased its yield, but its yield at high N level was often not significantly different to that of the other cultivars. A similar response was observed in the field experiments. This result suggests r44 that Skiff may be poorly adapted to low fertility conditions and requires relatively high fertility to achieve high yields.

Postanthesis water stress had a large effect on the yield response of the cultivars. When watering was reduced after anthesis to induce stress, the response to N was lower, and in some cases there was a decline in yield (Table 5.6). The genetic differences in responsiveness were most clearly seen when the plants were kept well watered after anthesis

(Table 5.6). Therefore the ability to distinguish responsive and non-responsive cultivars will be greater when postanthesis stress is low.

Malting quality is influenced greatly by GNC. Several studies have found that protein, as a percentage of grain dry matter, increases with drought (Salter and Goode 1967; Brooks et al.

L982). Results from this study also demonstrated the strong effects of postanthesis stress on GNC but there was also an influence of cultivar on this responses (Table 5.14). GNC increased with water stress at the high level of N (the increase, at the low N level was just non-significant), but the grain N yield decreased with stress (Table 5.15). In other words,

despite less N being transported to the grain, the proportion of N in the grain was increased

with stress. The important factor determining GNC in these barley cultivars is therefore the

amount of carbohydrate deposited in the grain during grain filling. This result is consistent

with previous studies with wheat. Nicolas et al. (1985) found GNC was significantly

higher under drought than under well-watered conditions. However, the N content per grain

and the grain N yield per ear were lower under drought, indicating that the higher percentage

N of grains was due to their smaller size. Drought also reduced the sink size of the grains,

as measured by the starch granules and the number of endosperm cells (Nicolas er a/.

1985a). As well, high protein cultivars of wheat were found to have a lower kemel weights

than a low protein cultivar (Donovan et al. 1977). Jenner et aI. (1991) suggested the

response in the wheat grain to developing water stress is similar to the response to elevated

temperature. They consider that deposition of starch is more sensitive than is the deposition

of protein hence stress will affect carbohydrate levels more than protein levels in the grain.

There has been considerable work with wheat, but relatively less with barley on the effects

of stress on grain filling. In a comparison of wheat (cv Sun 9E) and barley (cv Clipper) t45 Brooks et al. (1982) found that water stress had no significant effect on the absolute protein content of wheat or barley grains but protein as a percentage of dry matter was higher in

stressed grains of both cultivars. However in barley, water stress reduced kernel weight by only 3-5 mg, whereas in wheat the reduction was about 10 mg. Clipper was also used in the present pot experiment and it also showed relatively little change in kernel weight in response to N and water stress (Table 5.9). However, Skiffs kernel weight was quite

sensitive to postanthesis stress, which was largely due to change in non-protein dry matter.

The comparison between Sun 9E and Clipper reported by Brooks et al. (1982) and between

Clipper and Skiff in the present study indicate that there may be differences between wheat

and barley in response of kernel weight to stress as well as differences between cultivars of

barley. The apparent greater dependence of GNC on the carbohydrate content of the grain,

rather than on the grain N content, suggests that cultivars like Skiff and Stirling (and

possibly Chebec) whose kernel weight is apparently more sensitive to postanthesis stress,

may have a greater potential for high GNC. This is a disadvantage of growing Skiff for

malting barley in semi-arid areas despite its higher yield potential in f,ield trials.

There were significant differences between cultivars in the amount of dry matter and N

remobilised from the shoot during grain frlling (Table 5.2O), and their contributions to grain

yield and grain N yield respectively (Tables 5.2I, 5.22). The estimates of the average

contribution of remobilised dry matter to final yield ranged from a small decrease of 0.4Vo

under non-stressed conditions up to approximately 4O7o under stressed conditions, although

there was.considerable variation between cultivars. However, the values are within the

range quoted for wheat and barley under field conditions (Gallagher et al. 1975; Bidinger er

al. 1977;Palta et al. t994; Schnyder 1993). The average amount of N remobilised from the

straw of the different cultivars was equivalent to more than 87Vo of the N in the grain at

maturity which is comparable to the SlVo of total grain N quoted by Palta et aI. (1994).

Therefore the levels of remobilised dry matter and N and their estimated contribution to the

grain yield and grain N yield are realistic values and reflect values expected in the field. In

the 3 most responsive cultivars, Skifl Schooner and Stirling, there was also a loss of N

from the shoot which was not recovered in the grain (Table 5.21). r46 Despite the significant differences between cultivars in shoot dry matter remobilisation and N remobilisation there is little evidence to show that these can help explain genotypic differences in responses to N. There was no significant correlation between the average amounts of shoot dry matter remobilisation and grain yield or between the average N remobilisation and grain N yield (Fig. 5.1). The increase in grain yield was significantly correlated with the increase in shoot dry matter remobilisation when there was no postanthesis stress (Fig. 5.2a); that is, the non-significant yield increase with N in Weeah, was associated with a small change in the amount of shoot dry matter remobilisation postanthesis, while Skiff and Stirling, which had large yield responses to N also had high shoot dry matter remobilisation. However, it is difficult to say whether this a cause-and- effect relationship or a reflection of the greater demand for carbon during grain filling because the higher yieldings cultivars also set more kernels. Under stressed con¿itionffíi /--, no relationships between increased remobilisation of dry matter or N and the yield or GNY responses (Fig. 5.2 a-d). Under dryland conditions, postanthesis water stress is to be expected and situations where postanthesis stress is very low will be the exception.

Therefore, the responses in the stressed treatment are probably more representative of the field environment. From these relationships, it can be concluded that among the 6 cultivars used in this study the genetic differences in remobilisation of dry matter and N are probability not strongly associated with the responses in grain yield or grain N yield. Also in this experiment, postanthesis dry matter remobilisation was not important to kernel weight because Skifl which was responsive cultivar have a variable kernel weight.

Kernel weight is an important factor determining malting quality. This study showed that there is a strong effect of both N and postanthesis water stress on the kernel weight of the different cultivars. The kernel weight of Skiff and Stirling were greatly reduced by postanthesis stress at the higher level of N. This sensitivity in kernel weight to stress is similar to the differences observed in the field experiments (Chapter 3). However, in the glasshouse, Chebec's kernel weight was also greatly reduced by postanthesis stress, which contrasts with the field results. In wheat, semidwarf cultivars often are more sensitive to stress than tall cultivars (eg Gale and Youssefian 1985; Anderson and Smith 1990) and kernel weight is sometimes reduced more. For example, Anderson and Smith (1990) found r47 that the kernel weight of the semidwarf wheat Aroona fell from about 42 mg to 35 mg with stress induced by late sowing while the decline in the kernel weight of the tall cultivar

Gamenya was from 38 mg to 34 mg. Nedel et aI. (1993) with 2-row semidwarf barley found kernel weight of semidwarf cultivars was lower and more variable than tall cultivars.

Skiff is a semidwarf cultivar while in this experiment Stirling was relatively short; their greater sensitivity to postanthesis stress may therefore be related to their reduced height.

Remobilisation of stored assimilate from the stems and leaves contributes to the growth of the grain, especially under stressed conditions when photosynthesis is greatly reduced

(Bidinger et aI. 1977; Gallagher et al. 1975; Pheloung and Siddique 1991). It has been

suggested that the ability to remobilise large amounts of assimilate and translocate it to the

grain is a desirable trend for cereals in dryland environments (Gale and Youssefian 1985;

Anderson and Smith 1990), and chemical desiccation techniques have been proposed to

select for stability in kernel weight and grain yield (Turner and Nicolas 1987; Hossain et al.

1990). Such studies have shown that there is considerable variation between varieties in

grain yield and kernel weight stability (eg Hossain et aI. l99O). It is argued that the greater

kernel weight stability observed in some cultivars is due to greater remobilisation of dry

matter, but often there are no estimates of dry matter remobilisation to support this claim. In

the present experiment, the results do not agree with these claims: dry matter remobilisation

was not correlated with kernel weight or with grain yield. The kernel weight of Skiff,

Stirling and Chebec were most sensitive to postanthesis stress (Table 5.9) even though the

amount of dry matter remobilisation was high (Table 5.18). Hossain et al. (199O) also did

not find a strong correlation between the loss of dry matter from the stem and the reduction

in yield or kernel weight in a group of 10 winter wheat varieties. Therefore although this

and other studies have shown that there is genetic variation in kernel weight and kernel

weight stability, this does not seem to be strongly related to the ability to remobilise dry

matter from the stem and leaves.

This study confirmed some of the results found in the field experiments. Postanthesis water

stress affected the response of grain yield and GNC to N and caused an increase in GNC,

which is not desirable for malting quality. There were differences in the effect of water 148 stress on grain yield in the different cultivars and averaged over all cultivars water stress reduced grain yield. The grain yield responses to N were partly due to an increase in ear number and kernel number. Under well watered conditions and high N grain yield response to N was high in all cultivars except Weeah, a result similar to that found in the field study

(Chapter 3). There were genetic differences in remobilisation of dry matter and N during grain filling, but there was no clearcut association with the yield or GNC responsiveness of the different cultivars. At the lower level of N, postanthesis water stress reduced yields less

(Table 5.6) and did not significantly affect kernel weight (Table 5.9) or GNC (Table 5.14), compared with barley grown at the higher level of N. These responses perhaps help explain the conservative approach of farmers to applying N to malting barley in environments where postanthesis water stress occurs. This study shows that the supply of N to malting barley cultivars should be tailored both to environmental conditions (especially rainfall) and to agronomic practices which will reduce the effects of water stress during grain filling. To avoid water stress agronornic practices such as time of sowing, rate of sowing and control weeds are important. t49 CHAPTER 6 GENETIC VARIATION IN THE RESPONSE OF BARLEY CULTIVARS TO NITROGEN

6.1. Introduction

Malting barley is an important, high value crop in southern Australia. Many farmers who grow barley do not apply nitrogenous fertiliser because of the risk of increased grain protein levels. These farmers may suffer a yield disadvantage that in many cases is not compensated by the premium paid for malting grade. The long term aim of research into N responsiveness is to breed cultivars that will give maximum grain yield responses but which will not have a high grain protein responses. The work described so far.in this¿T+sis has used a relatively small number of cultivars, although even within these,^differences in responses have been identified. However, it would be useful to examine a wider range of genetic material and also assess the practicability of screening for N responses in the field.

6.2. Materials and Methods 6.2.1. Experiment 1: Northfield

An experiment was conducted at Northfield, South Australia in 1991. Seventy eight barley cultiva¡s of diverse genetic background (Appendix 6.1) were grown at two levels of nitrogen

(0, 50 kgN/ha as urea applied immediately before sowing time). The barley was sown on

25 July I99l at a rate of 55 kg/tra in 4 row plots, 4.2 m in length. The size of each plot was 2.52 Íp. A conventional row spacing of 15 cm was used. A basal dressing of triple

superphosphate (20 kgP/ha) was drilled with the seed at sowing. The experimental design

was a split plot with 3 replicates; main plots were levels of N and subplots were cultivars.

6.2.1.1. Measurements

Agronomic data were collected during the season. They included tiller number, plant height,

flag leaf length, general growth, chlorosis of leaves, scores for boron toxicity, time of

flowering and temperature differential as follows: 150

(t) Tiller number per plant was counted at 5 locations within the plot with 5 plants in each

location (25 samplesþlot) and the average calculated. lyr¿n:wr¿m¿rnb r^leÉ laþr- 6t a6'tl.¿sûs.

(ä) Plant height was measured after anthesis from ground level to the top of the head in

each plot using a metre rule with 10 values per plot.

(äi) FIag leaf length was measured with a ruler at the end of anthesis on l0 leaves in each

plot.

(iv) General growing was determined on the basis of score from 1 to 10 with high scores

indicating good growth. The basis of this measurement was green appearance and

lack of disease.

(v) Chlorosis score of leaves ranged from 1 to 5 with a high score indicating the leaves

were very chlorotic.

(vi) Boron toxicity in some cultivars was evident. Therefore the plots were scored for

toxicity symptoms using a scale of boron toxicity of 0 to 9 with higher toxicity

showing a higher value.

(vii) The time of flowering for each cultivar was recording when 507o of. the plants in the plot had completed anthesis (69 on the Zadok's scale). Data is expressed as the

number of days after 1 October.

(väi) Temperature dffirential was measured at the end of anthesis (29 October) with an

infrared thermometer, Model 510B, with a 4" field of vision. The infrared

thermometer was held at a declination angle of 30o from the horizontal, resulting in a

spot size of about 0.25 m2. Measurements were made on two replicates in the middle

of a sunny day (l I a.m.-2 p.m.). The measurement is the difference between canopy

temperature and air temperature. Canopy temperature detects water stress in plants

based on the assumption that transpired water evaporates and cools the leaves below 151

the temperature of the air (Jackson L982). As water becomes limiting, transpiration is

reduced and the leaf temperature increases. Under stressed conditions canopy

temperature can be higher than air temperature and the difference will therefore be high

and positive: T¿=f"-1¿ where T6 is temperature differential, Ts is canopy temperature

and T¿ is air temperature.

At maturity the plots were harvested with a small harvester. A sample of grain was taken from each plot and grain protein concentration (GPC) was determined by the Kjeldahl method (Appendix 3.1).

6.6.2. Experiment 2z Charlick

The experiment was carried out at Charlick Experiment Station, South Australia in 1993.

The same 78 cultivars of barley used in Experiment I were grown at two levels of N. Seed of the barley cultivars was planted on 22 July 1993 at a rate of 62 kghain eight-row plots,

4.2minlength. Thesizeof eachplotwas 5.04m2. Aconventionalrowspacingof 15cm was used.

Nitrogen was applied as urea at two stages: with the seed at sowing at a rate of 20 kgN/ha and broadcast at the mid-tillering stage (23 on Zadok's scale) at a rate of 25 kgN/ha. The experimental design was a split plot with 5 replicates; main plots were cultivars and subplots were N treatments. A different experimental design to that at Northfield 1991 was used to improve the ability to recognise different responses of individual cultivar to N.

6.2.2.1. Measurements

Measurements of growth at anthesis (at Zadok's growth stage 69), were based on a quadrat size of 2 row x 0.5 m at 2 locations from 3 replicates. Samples were cut at ground level in each plot, oven-dried at 80oC for 2 days, subsampled and ground to pass through a 2 mm sieve. N concentration at anthesis, time of flowering, plant height, general growing and flag leaf length were determined by the same methods as Experiment 1. At Charlick boron toxicity was not evident, so cultivars were not scored for damage. Grain protein concentration in the flour was determined by near infrared reflectance (NIR) (Technicon 152 InfraAlyzer ru 400 R) after the NIR had been calibrated with Kjeldahl method (Bran and

Luebbe; U.K. Software Training; V 5.10).

6.2.2.2. Measures of responsiveness

The responsiveness of each cultivar was measured using 3 indices, designated RI1, RI2,

NEI. RI1 and R[2 are Response Indices based on yield and protein responses to N, and NEI is a N efficiency index based only on grain yield. These indices are calculated as follows:

(Ð Rlr-(et##a)t(qffie) (ii)Rrz-(ffi)r(æ) (iii)NEI=æ where GPC59 and GPC6 are GPC at 50 (or 45) and 0 kgNÆra, GYso and GYo are grain yield at 50 (or 45) and 0 kgN/fia.

The 3 responsive indices were compared to find which index better explained different responses of cultivars to N. RI1 is the inverse of the response index described in Chapter 3 because in many cases the change in GPC is small oÍ zeÍo, which may lead to very high values if the change in GPC remained as the denominator.

6.2.2.3. Statistical analysis An initial analysis of variance was performed to verify genetic variation in the traits measured and to examine the effect of N on these traits. A correlation matrix was used to select variables that were significantly related to yield and protein. Due to the large number of cultivars and variables measured there were many sources of variation in data. These variables may have overlapping effects on each other, so it was important to distinguish between sources of variation among data using multiva¡iate analysis. A principal component analysis (PCA) based on the covariance of the measurements was used to simplify the complex data set into a few independent components (Software; Genstat 5, 1987). Grain yield and GPC were not included in the PCA. The PCA was conducted on standardised variables ( H) because of the wide range in the values of the different data sets. 153

PCA is a data reduction technique used to identify a small set of variables which account for a large proportion of the variation in the original data. In practice, the method of PCA is used to find linear combinations of variables (known as principal components) which explain much of the variation in the data. 21, the first principal component (PC) is a normalised linear combination of the original variables explaining mærimum variance of the data. The second PC (Zù explains as much of the residual unexplained variance as possible. If the first and second PCs explain a reasonable amount of the variation, it is possible to reduce p variables to 2 principal components. The PCs (ZÐ are formed as linear combinations of the original variables. If the latter are denoted by Xi, then Zl-L ali Xi where i=l,...p, and generally Z¡=la¡i Xi.

The coefficients a1i,....aki are referred to as PC loadings, which are the eigenvectors of the covariance matrix of the X variables. The value and the sign of these eigenvectors indicate the relative importance of each variable in the PC. The proportion of the total variance explained by successive PCs is calculated as the proportion of each eigenvalue of the total.

PCA is useful where the original variables are correlated, positively or negatively.

PCA was used to derive a small set of variables which could then be used to examine yield and protein responses to N. Once the important PCs were derived, grain yield, GPC, yield response (GYSo-GY 6) and GPC response (GPC56-GPC.g), Rlt, RI2 and NEI were then regressed against PC1, PC2, or PC3. In the case of the responses of grain yield, GPC, RI and NEI, the PCs at 0 kgN/ha were used in order to assess what characteristics of cultivars not receiving N contributed to their responsiveness.

6.3. Results 6.3.1. Weather

The amount of rainfall at the two sites is presented in Table 6.1. Total amount and seasonal rainfall at Northfield in 1991 was higher than at Charlick in 1993, and the seasonal distribution was different at each site. At Charlick rainfall early in the season was adequate but less than at Northfield. At Charlick rainfall during the growing season was more evenly distributed and it increased at the end of season. At Northfield the amount of rainfall in June r54 was high which was adequate for establishment of seedlings, but it later decreased, and was very low in November. The amount of rainfall during grain filling (September-November) was slightly higher at Northfield (185 mm) than at Charlick (171 mm).

6.3.2. Level of production 5-i^ Generally the average level of production was higher at Charlick in 1993 compared with

Northfield (Table 6.2) andthe GPC at Charlick was lower.

6.3.3. General responses to N

The mean grain yields at 0 kgN/tra and 50 kgN/ha at Northfield were 2.I2 and2.53 tlhaand at Charlick the mean yields at 0 and 45 kgN/ha were 2.61 and 3.41 tlha. Examples of responses to N at 2 sites are presented in Plate 6.1. The grain yield response to N between cultivars was significant at both sites (Tables 6.3,6.4) and ranged from -0.37 t/ha to 1.53 t/ha and 0.03 to 1.54 üha at Northfield and Cha¡lick respectively. The GPC response for the

2 sites varied from -0.56Vo to 4.93Vo and -0.07Vo to l.98%o respectively. There were significant differences between all cultiva¡s in the total grain N at both sites, but a significant

N x Cultivar interaction was observed only at Charlick 1993. Although there were significant cultivar differences in response to N at both sites, the grain yield responses of the cultivars at the 2 sites were not significantly correlated with each other (Fig 6.1).

6.3.4. Relationship between grain yield and GPC

There was a significant negative relationship between the mean grain yield and GPC for the

78 cultivars at the 2 sites (Fig 6.2 a, d). However, there was no significant correlation at either site between the absolute response of grain yield and GPC or between the relative responses grain yield and GPC (Flg 6.2 b, c, e, f). It seems that there are some barley cultivars which have high grain yield responses and low GPC responses as well as others which had high yield and GPC responses (Fig 6.1). At Northfield 199I, cultivars such as

Indian Dwarf (cultivar number 38, Appendix Table 6.1), Galleon (31) and WI-2621(53) and at Charlick 1993 Skiff (48), WI-2583 (25), WA-79S/42O (63), V/I-2805 (55) and WI-2582

(20) had high absolute and relative responses in grain yield but small responses in GPC. 155

Plate 6.1. A general view of the field experiment at Charlick in 1993 showing responses to N 156

The six barley cultivars which have been used over all experiments are identified in Fig. 6.2.

Generally the location of these cultivars in the relationships between mean grain yield and

GPC and response to N differed at both sites. At Northfield, mean grain yield and GPC of

Skiff and Schooner were high, Clipper and'Weeah had low mean grain yield and high GPC,

while Chebec had a high grain yield but a low GPC. Grain yield and GPC were low in

Stirling. In terms of absolute and relative responses of grain yield and GPC responses,

were high in Stirling and Schooner but low in Clipper and Skiff. Chebec and Weeah had

low yield responses and high GPC responses (Fig. 6.2b, c). At Charlick'Weeah, Chebec

and Schooner had low mean grain yield and GPC, Stirling had high grain yield and GPC,

while Skiff had high yield and low GPC. Clipper also had a high mean GPC but low mean

yield. In terms of absolute and relative responses Skifl Stirling and Chebec had high yield

and low GPC but Weeah showed low grain yield and GPC responses, while Schooner had a

high grain yield and GPC response. Clipper also had a high GPC response but a low yield

responses (Fig. 6.2 e, f).

6.3.5. Correlations between grain yield and other factors

At Northfield in 1991 at 0 kgN/tra there was a significant negative correlation between grain

yield and GPC and date of flowering, and a significant but positive correlation with total

grain N, plant height and tiller number (Table 6.5). At 50 kgN/ha GPC, date of flowering

and temperature differential had a significant negative correlation with grain yield, while the

correlation between total grain N and plant height was positive (Table 6.5). At Charlick

1993 with.both rates of N, there was a negative relationship between GPC, plant height and

grain yield, but a positive correlation between grain N yield and grain yield (Table 6.6).

Time of flowering was not significantly correlated with yield. The correlations among all the

variable measured at 2 levels of N are presented in Appendix Tables 6.2 and 6.3. 157

Table 6.1. The monthly distribution of rainfall at Northfield in 1991 and Charlick in 1993

Month Northfield 1991 Charlick 1993 (mm/month)

January 15.4 18.2 February 0.0 9.3 Ma¡ch 9.7 13.0 April 32.2 0.0 May lt.2 23.8 June 103.5 31.1 July 76.2 79.6 August 64.4 49.4 September 69.2 80.6 October 69.2 26.8 November 46.8 64.0 December 0.1 64.0

April-October 426 324 September-November 185 L7I Total 498 456

Table 6.2. Average values of production and measurements of some factors at two sites

Parameter Northfield 1991 Charlick 1993

Grain yield (t/ha) 2.32 3.01 Grain protein concentratign (Vo) 13.3 8.3 Grain N yield (gktÊ) 4.94 3.99 Plant height (cm) 68 6t Tiller number (per plant) 5.8 Flag leaf length (cm) 8.1 4.2 Boron toxicity (0-9) 3.7 Chlorosis (1-5) 2.8 General growing (1-10) 5.2 5.8 Temperature differential f C) -0.70 Date of flowering (d. after I Oct.) 22.0 12 Dry matter at anthesis (Elrf,) _a 489 N concentration at anthesis (7o) 1.1 I Total N at anthesis (gt^2) 5.38

aNot measured 158

Table 6.3. Summary of analyses of variance of data at Northfield in 1991

Parameter Nitrogen Cultivar Nitrogen x Cultivar

Grain yield (t/ha) **{c t<* NS Grain protein concentration (Vo) *** *** NS Grain N yield (glrû) *x {

**t(-P<0.005, x*-t.0.01, *=p40.05, NS=Þ0.05

Table 6.4. Summary of analyses of variance of data at Charlick in 1993

Pa¡ameter Nitrogen Cultivar Nitrogen x Cultivar

Grain yield (t/ha) *{€* *** x*r< Grain protein concentration (Vo) *** *t

x*x-P<0.005, xx=p40.01, *=P<0.05, NS=P>0.05 159

2.0 (a)

(d -c r=+0.11NS o E 1 5 o) o Ë o o o o a z 1 0 o (ú o o oo o o to U'c ttta o ¡ a o 0.5 a a o oU' O o o a E a O o rtO o loot '= 0 o oa c o o 'õ o o o a (5 a -0.5 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.O Grain yield response at Charlick (Vha)

(b) G'150o\ E o r=+0.14 NS o Ëo o z- 100 (d o o o o o o Co *50 o o o ao E o oa '=o o o a Fo a o) (¡)

(d P -so 010203040506070 Relative grain yield response at Charlick (%)

Figure 6.1. The correlations between Charlick and Northfield for (a) absolute yield response and (b) relative yield response. 160 Northfield Charlick (a) Y=15.6-0.99 x (d) Y=12.0-1.23 x 6 * 11 l=-0.56 * r= -0.56 Clipper òS òs a c o Stirling c 1 0 o .o 1 5 o (ú o ot I s a skiff c o ro{-mff L o 1 4 a . o I a C)c o cC) o o o o o o Weeah o .g 1 3 .E I t o a t o o e a o Schooner o o- Clipper o- aao c 12 O .c 7 I (ú E Chebec - CI Stirling o Chebec a 11 6 1.0 1.5 2.O 2.5 3.0 3.s 4.0 1.5 2.O 2.5 3.0 ' 3.5 4.0 4.5 Grain yield (Vha) Grain yield (Vha)

(b) r=0.19 NS (e) 5.0 r=0.13 NS o o Schooner 2.5 s 4.5 t a q) 4.O d 8 ,.0 o schooner .D o o at ' oL 3.5 fo. c. o o- g_ 1.s Clipper o U' 3.0 o a o O Ø a a a E o o o 2.5 a o 1'o o o fL I ¡ fL a o skiff CI 2.0 OO I t o OO o o o o 1.5 a g0'5 f ï a . o a õ c pper J al, 1.0 Stirling õ a J -o skiff o oo 0.5 o 80 , rling Chebec 0 -0. 5 Iôl qcgfqqqc\l Iq -0 0.0 0.5 1.0 1.5 2.O ggOOOOoFFF Absolute yield response (Uha) Absolute yield response (Vha)

(c) r=0.17 NS (f) r=0.04 NS 45 25 Chebec o goo a o 8ro o o oo 835

Figure 6.2. The relationship between grain yield and grain protein concentration in mean (a), absolute (b) and relative response (c) at Northfield 1991 and Charlick 1993 (d, e, f¡ for 78 barley cultivars. 161

Table 6.5. Simple linear correlations between measured attributes and grain yield at Northfield 1991, n=78

Parameter 0 kgN/ha 50 kgN/ha

Grain protein concentration -0.619** -0.504*x Grain N yield 0.973** 0.972x* Date of flowering -0.567*x -0.416x* Tiller number 0.444*x 0.030 Plant height 0.354** 0.258* Temperature differential -0.t42 -0.335** Flag leaf length -0.207 -0.106 Boron toxicity 0.045 0.1 t8 Chlorosis -0.126 -0.065 General growing 0.089 0.111 r (SVo) + 0.223 r (|Vo) ! 0.290

Table 6.6. Simple linear correlations between measured attributes and grain yield at Charlick in 1993, n=78

Pa¡ameter 0 kgN/ha 50 kgN/ha

Grain protein concentration -0.447** -0.549xx Grain N yield 0.615** 0.730x* Date of flowering 0.067 0.r47 Dry matter at anthesis 0.151 -0.053 Shoot N concentration at anthesis 0.104 0.201 Shoot total N at anthesis 0.162 0.087 Plant height -0.258* -0.417x* General growing -0.015 0.027 Flag leaf length -0.079 -0.054

r (5Vo) + 0.223 r (lVo) + 0.290 162 6.3.3. Principal component analysis 6.3.3.1. Northfield I99l

Principal component analysis at 0 kgN/ha

At OkgN/tra the cumulative variation of the first 3 PCs accounted for 72.4 Vo of the total variation (Table 6.7). PC1, which accounted for 4I.2Vo of total variation, showed date of flowering, flag leaf length, plant height and temperature differential were more important than the other attributes, although date of flowering and flag leaf length were the major components (Table 6.7). PC1 was high when date of flowering, flag leaf length and temperature differential were high and when plant height was low. Temperature differential and flowering were positively correlated, indicating that the temperature differential (and therefore stress) increased as flowering was delayed. Long flag leaves and short stature were also positively correlated with temperature differential. PC2, which accounted for

17 .6Vo of total variation indicated temperature differential, boron toxicity and flag leaf length were more important than the other variables. A large temperature differential, long flag leaf and greater symptoms of boron toxicity contributed to a high PC2. The symptoms of boron toxicity were positively correlated with temperature differential. The third principal component which accounted for 13.6Vo of total variation showed flag leaf length and temperature differential to be important. PC3 was greater with a long flag leaf and low temperature differential.

Principal component analysis at 50 kgN/ha

At 50 kgNha, PC1 which accounted for 42.5Vo of total variation showed date of flowering to be the most important variable, while chlorosis, plant height, Boron toxicity and temperature differential were relativly less important (Table 6.8). Boron toxicity scores varied inversely with flowering date. PC2 which accounted for IS.lVo of variation, indicated flag leaf length and chlorosis were more important than the other variables. PC3 which accounted for

13.OVo of variation showed flag leaf length, temperature differential and Boron toxicity had high variability. Table 6.7 . Eigenvectors of eight principal components from Principal Component Analysis of some attributes of barley cultivars grown with 0 kgN/ha at Northfield 1991

Latent vectors (Loadings) Va¡iable 1 2 3 45 6 7 8

Date of flowering 0.733 -0.136 0.3t7 0.091 0.518 -0.075 0.243 0.053 Tiller number -0.246 0.011 0.004 -0.097 o.146 -0.384 0.162 0.858 Plant height -0.389 -0.119 -0.036 -0.564 0.485 -0.051 0.391 -0.352 Temperature differential 0.382 0.587 -o.641 -0.291 -0.002 -0.057 0.101 0.028 Flag leaf length 0.730 0.458 0.615 -0.492 -0.263 -0.059 -0.118 -0.003 Boron toxicity -o.267 0.515 0.112 0.292 0.539 0.478 -0.188 0.106 Chlorosis -0.168 0.376 0.111 0.488 -0.015 -0.615 0.304 -0.329 General growing 0.019 -0.070 -0.087 -0.118 0.338 -0.481 -0.781 -0.r32

Eigenvalues 96.5 41.2 31.8 28.3 16.9 9.9 6.3 3.7 Percent of variance 4r.2 r7.6 t3.6 12.0 7.2 4.2 2.7 1.6 Cumulative variance%o 4t.2 58.8 72.4 84.3 91.5 95.7 98.4 100.0

(})Oì Table 6.8. Eigenvectors of eight principal components from Principal Component Analysis of some attributes of barley cultivars grown with 50 kgN/tra at Northfield l99l

Latent vectors (Loadings) Variable I 2 3 45 6 1 8

Date of flowering 0.709 0.169 0.066 0.365 0.082 -0.280 0.049 0.494 Tiller number -0.r48 0.089 0.140 -0.146 0.128 -0.899 -0.216 -0.239 Plant height -0.334 -o.255 -0.159 -o.376 0.271 -0.o37 -0.049 0.761 Temperature differential 0.308 0.166 -0.545 -0.677 0.123 o.o23 0.024 -0.004 FIag leaf length -0.071 0.671 0.598 -0.078 0.34r o.239 0.052 -0.076 Boron toxicity -0.329 o.t57 -0.507 o.459 0.625 -0.022 0.085 0.023 Chlorosis -0.386 0.538 -0.t92 0.151 -0.613 -0.099 -0.065 0.335 General growing -0.086 -0.025 -0.049 -0.079 -0.080 -o.207 0.966 -0.0r5

Eigenvalues 90.1 32.O 21.7 23.8 17.4 9.1 7.2 4.9 Percent of variance 42.5 15.1 13.0 tt.2 8.2 4.3 3.4 2.3 Cumulative varianceVo 42.5 57.6 70.6 81.8 90.0 94.3 97.7 100.0

èo\ 165

Principal component analysis of response to N

The change in the measured parameters between 0 and 50 kgN/ha were also analysed by

PCA. PC1 which accounted for 3l.2%o of the total variation, showed changes in the temperature differential was the most important variable (Table 6.9). PCr was high when temperature differential increased. PC2 which accounted for 18.9Vo of variation indicated the change in boron toxicity score, flag leaf length and plant height were most important.

Date of flowering did not change with different N treatments, and so its contribution to the overall variation was low.

Correlations between principal components and yield and protein responses

The correlation between PC1 and PC2 (at 0 kgN/ha) and grain yield response were positive

(Table 6.10, Fig. 6.3). There was no significant correlation between yield or protein response and PC3. The positive correlation between PC1 in which the most important parameters were time of flowering, flag leaf and height suggests that cultivars which were late flowering and short had greater grain yield responses, than early, tall cultivars. A high temperature differential also contributed to a high response in grain yield. The correlation between PC2 and grain yield increase was also significantly positive (Fig. 6.3). The most important variables in PC2 were temperature differential, boron toxicity and flag leaf length which suggests that plants showing greatest symptoms of boron toxicity score and had a long flag leaf and which had a high canopy temperature were most responsive to N.

In the correlations between PC1, PC2 and the increase in grain yield (Fig. 6.3) some cultivars showed responses well above or below the regression line. Azuma (cultivar 13), Cytris (16), RV/CF42R (21), Indian Dwarf (38), 2EBYT23 (17), Schooner (30),

Moondyne (49),WI-2815 (64) and Minerva (1) show above average responses whereas

WA-75Si329 (61), Resibee (3), Mazurka (10), Compana (19) and Ketch (44)bad lower than average responses. There were no significant correlations between grain protein

concentration and the first 2 PCs (Table 6.10), but the correlation between grain N yield and

PC2, was signif,rcant and positive (Table 6.10). Table 6.9. Eigenvectors of eight principal components from Principal Component Analysis of the increase in the values of some attributes of barley cultivars when 50 kgN/tra was applied at Northfield 1991

Latent vectors (Loadings) Variable I ) 3 45 6 1 8

Date of flowering difference 0.o79 o.25r -o.2t9 -0.276 o.o4l -0.412 o.657 -0.450 Tiller number difference -0.037 0.160 0.035 o.122 0.199 0.424 0.ltr 0.482 Plant height difference -o.tr1 -0.4t7 o.399 0.560 -o.u2 0.006 0.168 -0.555 Temp. differential difference o.974 0.013 0.017 0.175 -o.ttz 0.890 0.011 -0.014 Flag leaf length difference o.t24 -0.528 o.399 -o.73r 0.007 o.072 0.003 -0.o79 Boron toxicity difference o.o21 -0.674 -0.686 0.084 -0.029 -0.24 -0.033 0.082 Chlorosis difference -0.082 0.015 -o.174 -0.026 0.876 0.269 -0.183 -0.298 General growing difference 0.087 -0.060 0.363 o.r47 o.419 -o.7tl -0.011 0.393

Eigenvalues 73.4 44.3 33.6 21.3 21.6 16.6 t0.2 8.7 Percent of va¡iance 3t.2 18.8 14.3 11.6 9.2 7.1 4.4 3.7 Cumulative variance%o 3r.2 50.0 &.3 75.1 84.9 91.9 96.3 100.0

a o\ r67

Northfield Charlick (a) (c) 300 1 (ú (ú 000 .c, 200 13 16 900 è, 21 è, t¿ .Y 800 48 0) 100 c) 700 1 an U) 3625 10 (ú (ú c) 0 600 64 L 0) t 77 o o 500 ,9'uar"t L 1 00 .= 400 t.!,u- õ E' 22 .9 -200 .9 300 ¿.95 olrltff 671 3oe"'1îe 200 'õc c u -300 GI "xA L L 100 (5 61 (, -400 0 -4 -3 -2 -1 0 1 2 3 4 -4 -3 -2 -1 0 1 23 PC1 PC1

(d) 300 (b) 1 000 (õ (ú 16 13 ! 900 è, 200 è, l¿ 2'l l¿ 800 4g 38 z'!st 1 o 100 610 o 700 36 (t, 27 (t 25 10 (ú zg (ú 63 o) 600 o 0 L 11 o 67 o 500 62 57 .c 212 .E 1 00 11 78 400 t35 o 41. d0 og g 2v.¡ c) .0) -200 3' 300 2[,20 25 6s s&6 ¿q eït \,.- I 200 ¿o c ro3 44s c 2W 'õ -300 'õ 42 L 100 41 L 61 o -400 CI 0 -2-1.5 -1 -0.5 0 0.5 1 1.5 2 -4 -3 2 -1 0 12 3 PC2 PC2

Figure 6.3. The correlation between lCI,PC2 ar¡d g¡aiq.yield increase at Nõrthfield (a, b) and Charlick (c, d). Numbers in the figure refer to the cultivars number in appendix Table 6.1. 168

Table 6.10. The intercept and coefficient of regression analysis of responses in grain yield, GPC and grain N yield against PC at Northfield in 1991

PCla PCta Dependent Intercept b r Signif. Intercept b r Signif,

Grain yield increase -87.3 28.3 0.26 * -87.3 37.6 0.23 {É GPC increaseb 3.I4 -0.14 0.18 NS 3.14 -0.03 o.o2 NS GNY 0.84 0.19 0.09 NS 0.84 0.78 increaseb o.23 '< aPCt=Principal Component 1, PC2=p¡i¡¡cipal Component 2 bGPC=Grain Protein Concentration, GNY=Grain N Yield x=P<0.05, NS=P>0.05

6.3.6.2. Charlick 1993

Principal component analysis at 0 kgN/ln

Cumulative variation of the first 3 PCs at OkgN/tra was 82.4Vo (Table 6.11). PC1, which accounted for 42.lVo of total variation showed date of flowering and shoot total N at anthesis to be the most important parameters although flowering date was the more important. PC1 would be high when flowering date was late and when total N at anthesis was low (Table 6.11).

PC2, which accounted for 2l.6Vo of cumulative variation, showed height and shoot N concentration at anthesis to be most important although there was also an effect of flowering

time. Greater height and reduced shoot N concentration at anthesis and earlier flowering will

decrease PC2. The third PC, which accounted for only I8.7Vo of total variation, showed

shoot N concentration, height and total N in shoot at anthesis to be the major variables.

When shoot N concentration is low the PC3 is high (Table 6.11).

Principal component analysis at 45kgN/ln

At 45 kgN/ha, PC1 which accounted for 40.6Vo of variation, showed date of flowering,

shoot total N and DMa at anthesis to be the most important factors. PC2 indicated plant

height, shoot N and DMa were important variables (Table 6.12). Values of PC2 will be high

when growth of the plants was high. Table 6.1l. Eigenvectors of seven principal components from Principal Component Analysis of some attributes of barley cultivars grown with 0 kgN/ha at Charlick 1993

Latent vectors (Loadings) Variable I 2 3 45 6 7

Date of flowering 0.822 -0.327 -0.162 0.037 0.376 -0.2r9 -0.000 Dry matter at anthesis -0.231 0.256 -0.118 -0.091 0.605 -0.159 -0.682 Shoot N concentration at anthesis -0.185 -0.518 -0.723 -0.006 -o.278 0.135 -0.282 Shoot total N at anthesis -0.327 0.021 -0.412 -0.078 o.496 -0.r23 0.675 Plant height 0.212 0.653 -0.465 0.380 -0.291 -0.282 0.003 General growing 0.225 0.275 -0.169 0.014 0.196 0.898 0.017 Flag leaf length 0.177 0.238 -0.150 -0.916 -0.205 -0.087 0.013

Eigenvalues 91.8 50.1 43.4 18.9 14.8 7.1 0.2 Percent of variance 42.1 21.6 18.7 8.1 6.4 3.r 0.1 Cumulative variance%o 42.1 63.7 82.4 90.1 96.9 99.9 r00.0

o\ \o Table 6.12. Eigenvectors of seven principal components from Principal component analysis of some attributes of barley cultivars grown with 45 kgN/ha at Charlick 1993

Latent vectors (Loadings) Variable I 2 3 45 6 7

Date of flowering 0.623 0.306 -0.559 0.208 -0.378 -0.138 -0.005 Dry matter at anthesis -0.3M o.347 0.201 0.498 -0302 0.003 -0.6r8 Shoot N concentration at anthesis -0.302 o.2t2 -0.615 -0.398 0.385 -0.035 -0.42r Shoot total N at anthesis -0.509 0.460 -o.206 0.208 -0.o44 -0.033 0.663 Plant height o.L42 0.595 o.433 -0.597 -0.167 -0.232 -0.004 General growing 0.160 o.276 o.o32 -0.049 0.068 o.943 0.009 Flag leaf length 0312 o.314 0.192 0.386 o.764 -0.186 0.003

Eigenvalues r12.6 64.0 46.8 25.2 20.9 7.7 0.3 Percent of variance 40.6 23.1 16.9 9.1 1.5 2.8 0.1 Cumulative vanance%o 40.6 63.1 80.6 89.6 91.r 99.9 100.0

\ì O t7r

Principal component analysis of response

When the increases in the values of the parameters between 0 and 45 kgN/ha were analysed by PCA, the change in shoot N concentration and content were the most variable factors in

PC1 $able 6.13). PC2 which accounted for 2l.IVo of variation showed changes in DMa and shoot N concentration were important between 2 rates of N. The score for general growth was the only factor contributing to PC3, which accounted for I9.5Vo of the variance.

Correlations betvveen principal components andyíeld and protein responses

Yield response was not significantly correlated with PC1. There was however, a significant negative relationship between PC2 and grain yield response to N (Table 6.14). This negative relation between PC2 and grain yield response indicates tall cultivars which had a low N content in the shoot at anthesis and which were late flowering tended to be less responsive. In the relationships between PC¡PC2 and increase grain yield (Fig. 6.3) the

cultivars Skiff (48), Minerva (1), Mazurka (10), WI-2583 (25), Bandulla (36), WA-

79Sl42O (63), WI-2815 (64), WI-2805 (55) and CH35l3l7 (11) had responses well above

average while Clipper (34), Resibee (3), Prior A (41), MC 90(42), Betzes (34), Weeah

(29), CI-4226 (23) andWI-2728 had low responses. There was no significant correlation

between GPC response and the first 2 components (Table 6.14). There was a significant

negative correlation only between grain N yield response and PC2, which is the same trend

as grain yield (Fig. 6.3 c, d). PC3 was not correlated with either yield or protein response.

Table 6.14. The intercept and coefficient of regression analysis of responses in grain yield, GPC and grain N yield against PC at Charlick 1993

APCI apCz Dependent Intercept b r Signif. Intercept b r Signif.

Grain vield increase 402.8 TI.7O 0.09 NS 402.8 -64.00 0.35 :F* GPC increaseb 0.55 t -0.02 0.05 NS 0.551 0.05 0.08 NS GNY increaseb 6.58 -0.11 0.05 NS 6.58 -0.72 0.27 *

aPCt-Principal Component 1, PC2=p¡incipal Component 2 bGPC=Grain Protein Concentration, GNY=Grain N Yield x*=P<0.01, *=P<0.05, NS=P>0.05 Table 6.13. Eigenvectors of seven principal components from Principal Component Analysis of the increase in the value of some attributes of barley cultivars when 45 kgN/ha was applied at Charlick 1993

Latent vectors (Loadings) Variable I 2 34 5 6 7

Date of flowering difference 0.003 -o.029 0.018 o.072 o.r79 -0.833 -0.518 Dry matter at anthesis difference -0.281 -0.675 0.017 -o.04 0.000 0.379 -0.565 Shoot N conc. at anthesis difference -o.706 0.623 -0.014 -0.043 -o.029 0.150 -0.296 Shoot total N at anthesis difference -o.&9 -0.384 0.071 0.021 0.007 -0.333 o.562 Plant height difference 0.018 -0.033 o.029 -o.r12 -o.975 -0.163 -0.086 General growing difference 0.040 0.058 0.986 0.145 0.006 0.045 -0.018 Flag leaf length difference 0.008 -0.055 0.146 -0.978 o.127 -0.033 -0.031

Eigenvalues 92.7 33.7 16.8 10.1 3.8 r.4 1.1 Percent of variance 58.1 2t.l 10.5 6.3 2.4 0.9 0.7 Cumulative variance%o 58.1 19.3 89.8 96.r 98.4 99.3 100.0

-ll.J t73 6.3,4. Responsive Indices

The summary of the analyses of variance and the averages of the 3 responsive indices is presented in Table 6.15. RI1 was the most variable of the 3 indices and there was no significant differences between cultivars at both sites (Fig. 6.4 a, b). RI1 was also ambiguous because negative values could occur with either a decline in yield or in GPC with added N. It was therefore decided to disregard RI1. RIz was less variable and was significant at the IÙVo Ievel at Northfield and the 5Vo level at Charlick (Table 6.I5,Fig. 6.4 c, d). NEI, which only indicates the grain yield response, showed the greatest significance to N (Fi. 6.4 e, f). Overall there are consistent results between RI2 and NEI for each cultivar at each site (Fig. 6.4).

Table 6.15. Summary of analysis of variance and averages of three responsive indices at two sites

Northfield 1991 Cultivar effect Mean Range Coefficient variabtlity (Vo) RIT NS -0.73 -40.2-26.4 1050.4 RIZ .r-â 0.74 0.39-r.67 46.5 NEI *X 0.63 0.55-2.34 42.2

Charlick 1993 RIr NS 0.60 -4.80-19.6 662.6 RIz + 1.24 0.97-1.9 18.3 NEI {< t< L32 r.o2-r.9 15.5

xx- a+ = P<0.10, P<0.01

6.3.4.1. Correlation between principal components and responses indices

Northfield 1991

There was a significant positive correlation between PC1, PC2 (at 0 kgN/ha) and RI2 (Fig.

6.5 a, b). There is no correlation between PC1 and NEI, but there was significant positive

relationship between PC2 and NEI (Table 6.16, Fig. 6.5 c, d). Later maturity and low

height were corelated with high RI2 (Table 6.7). The positive correlation between R[2 or

NEI and PC2 (Fig. 6.5 b, d) indicated that when temperature differential and boron toxicity

scores were high , R[2 and NEI were high. 174

Response index 2 (%) Response lndex 2 (%) I JIJJJ l\) OOO<)TJJJJJ\)'Nsb,bbiusbrbob @ oluÞo@ b wr-269v7 AZUMA VADA v/r-2583 9) SKIFF q CYTRIUS MAZI.JRKA IDWARF z MINERVA o MOONDYNE o FRANKLIN g¡ MINERVA = wr-2582 :a RWCF42R o= wA-79S/420 õ' GPROMISE wr-2805 x TMP*GMTI2I/1 d cH35l317 cr-4226 TRIUMPH PROCTOR IDWARF v/A-75S/323 PARWAN SCHOONER wl-2231 t- GIMPEL t- wr-2597 U) wr-269l/7 CN wA-8ls/474 o v/A-8ls/470 o CAMBRINUS o 2EBYT23 o Ìl FORREST KORU oq òe BETZES s É GPROMISE H STIRLING wr-2728 o DIAMANT wI-2805 Or CYTRIUS GALLEON F 2EBYT23 v/A-8ls/474 MOONDYNE wA-8ls/473 È wI-2785 NOYEP o SCHOONER wI-2808 Þ ZEPHYR TRIUMPH NOYEP ct-3576 Fto YMER wA-8ts/467 F0 wA-775/409 wA-79S/420 oqo PROCTOR MC90 wt-2737 CLIPPER o+) BANDULLA wl-2621 ñ KLAGES wI-2585 wt-2477 wl-2736 N) wA-815/467 FRANKLIN P wr-2736 NORDAL o SHANNON o wA-8ls/465 d c YAGAN c PARWAN Þ) wt-2727113 wr-2785 wA-755t323 KLAGES È g) wA75S/329 Ê) wA-815/469 z õ cPI-18197 wA-77S/409 [n DAMPIER FORREST COMPANA cPI-18197 p PIROLINE PIROLINE wA-815/470 wl-2597 È wr-2585 MAYTHORPE wA-8ts/473 wt-2477 ä'Ft wA-8ls/465 cH35t3t1 cr-3576 YAGAN oo-l NORDAL YMER d KETCH SHANNON ÞFl VADA STIRLING o LARA LARA FUJINIJO v/A-815/46ó o v/A-8ls/466 wl-2582 É GRIMMET PRIORA wt-2621 ZEPHYR WEEAH GRIMMET FtÊÐ GIMPEL BANDULLA (t) MAYTHORPE KANIERE êÊ0 OCONNOR wt-2231 l'.J KANIERE wt-2737 U) TMP*GMTI21/1 wr-2727n3 wA-815/469 wr-2583 v)o RV/CF42R KETCH wI-2808 SKIFF AZUMA OCONNOR wt-2728 WEEAH MC90 DAMPIER RESIBEE DIAMANT BETZES CAMBRINUS GALLEON FUJINIJO cr-4226 RESIBEE KORU COMPANA PRIORA MAZURKA CLIPPER u/A-75S/329 n5

Nitrogen response index Nitrogen responsive index o fu lu 9:^-lulu i¡ o('to(¡ (¡o(¡ool wt-269y7 AZUMA CYTRIUS MAZURKA VADA c) MINERVA o- MOONDYNE z SKIFF IDWARF o wr-2583 o MINERVA w^-7951420 =q) =J I RWCF42R BANDT]LLA GPROMISE 9-= TRIUMPH õ' o- v/I-2582 - cl-4226 cH35l3l7 SCHOONER FRANKLIN wA-75S/323 wI-2805 2EBYT23 CYTRIUS PROCTOR GIMPEL PARWAN wr-269u7 rwA-775/409 Ø Ø IDWARF wA-815/470 o o (rl CAMBRINUS ('l TMP*GMTI2I/1 wt-2231 wr-2728 s wA-75S/323 òS wA-8ls/473 KLAGES GALLEON GPROMISE wA-8ls/467 YMER NORDAL PROCTOR wr-2808 wA-75S/329 TRIUMPH SCHOONER \r/A-815/465 FORREST cr-3576 wI-2585 NOYEP 2EBYT23 wI-2736 wA-8 st47 4 KORU NOYEP MC90 wA-79S/420 PIROLINE rwA-8l5/474 MOONDYNE LARA PARWAN cPI-18197 KLAGES TMP*GMT12I/1 wI-2585 wr-2597 wA-77S/409 wA-815/465 wr-2785 SHANNON FRANKLIN o DIAMANT O wl-280s c STIRLING C CLIPPER wr-2736 wA-8ls/473 =wA-Bls/469S cPr-18r97 A) RWCF42R H cn3st3n U'- cl-3576 Ø wt-2477 AZUMA BETZES wI-2785 wl-2597 YAGAN wt-2621 KANIERE LARA DAMPIER FORREST GIMPEL STIRLING wt-2737 YMER wr-2727113 PRIORA wt:2477 wt-2737 wA-8ls/467 wA-8ls/466 GRIMMET YAGAN ZEPHYR MAYTHORPE wl-262r wt-2727113 VADA wl-2231 OCONNOR wI-2582 NORDAL wr-2583 COMPANA PIROLINE wA-8ls/466 SHANNON FUJINJO BANDULLA wI-2808 DAMPIER wA-8ls/470 ZEPHYR KETCH KANIERE MAYTHORPE WEEAH wA-815/469 KETCH cl-4226 GRIMMET wt-2728 CAMBRINUS KORU DIAMANT WEEAH SKIFF GALLEON OCONNOR BETZES FUJINIJO MC9O MAZURKA PRIORA RESIBEE RESIBEE wA-75S/329 CLIPPER COMPANA 176

The trends of RI2 and NEI were the same as the grain yield increase and indicated cultivars such as Azuma (13), Cytris (16), Indian Dwarf (38), RWCF42R (21) and Vada (22) had values of RI2 and NEI higher than the general trend while cultivars such as Resibee (3),

Compana (19), Mazurka (10) and WA-7551329 (61) had lower RI2 and NEI (Figs 6.3,

6.s).

Charlick 1993

RI2 and NEI were not correlated with PC1, but there was a negative correlation between

RI2, NEI and PC2 (Table 6.16, Fig. 6.6 b,d). Date of flowering and shoot total N at anthesis contributed negatively to PC2 (Table 6.11) which indicates that late flowering cultivars had high RI2 and NEI and responsiveness was high when the amount of shoot total

N at anthesis was high. Generally the trend between RI2, NEI with PC2 is the same as increase in grain yield (Fig. 6.6 b,d). Cultivars such as V/I-2815 (64), Skiff (48), Mazurka

(10), V/I-2583 (25) and Minerva (1) with high RI2 and NEI showed high increase in grain 'Weeah yield and cultivars such as Clipper (25), Resibee (3), (29), MC90 (42),Prior A (41) and Betzes (34) showed low RI2 and NEI response and low grain yield responses (Fig.

6.6).

Table 6.16. The intercept and regression coefficient of regression analysis of Principal Component Analysis with response indices at 2 sites

APCT APCz Dependent Intercept b r Signifi. Intercept b r Signif.

Norffield 1991 RI2b 0.74 0.05 o.24 * 0.74 0.08 o.23 * NEIb 0.92 0.04 0.16 NS 092 0.10 0.25 *

Charlick 1993 * RI2b t.24 0.01 0.06 NS r.24 -0.03 o.26 NEIb r.32 0.01 0.01 NS t.32 -0.03 0.27 * aPCt-Principal Component l, PC2=p¡i¡cipal Component 2 bRll=p".nonse Index 1, R[2=Bs5ponse Index 2, NEl=Nitrogen Response Index *=P<0.05, NS=P>0.05 177

2.5 (a) 2.5 (c) 13 x c) N 2.O E 2.0 X oo) o 13 .5 ls"82oo. 1.5 22 .c 21 6 16 q) ,Eo 'õ 21 ¿g38 U) 7 18 AA27 E 46t L 23 27 o 1 0 o 1.0 18 o- 77 c. U' o 58 77 c) o) 4l fE 0.5 0.5 I 1 z= 61 0.0 0.0 -4 -3 -2 -1 0 1 2 3 4 -4 -3 -2 -1 0 1 2 3 4 PC1 PCl

(b) (d) 2.5 2.5 13 X C\¡ 2.0 x ,o o Ë E 13 1 .5 16 z4g .g 16 22 Ê 1.s 382't 40 29 o) 38 49r 'õo U' 21 eg 40 27 L zr23 o 1 .0 64 1.0 67 67 o- Ë 69 U) c c) 69 24 6i cc 0.5 3 19 8, O.S ..9 0.0 z o.o -2-1.5 -1 -0.5 0 0.5 1 1.5 2 -4-3-2-10 12 PC2 PC2

Figure 6.5. The conelation between I C1 and PC2 and RI2 (a,b), *t (9' d)-æ Northfield. Numbers in the figure refer to the cultivars number in appendix Table 6.1. 178

(a) (c) 2.0 2.0 64 64 x o o 0 C\¡ c,- x '- t.c 11 o 48t 77 1 5 totr., () o 63'Ì0 77 c 22 .= 38 o 6r q) 3 .n 22 E r.o c (¡) o 21 3sz o- 1 .0 23 41j U' c q) 9, O.S tÍ. o z= 0.5 0.0 -3-2-10123 -4-3-2-101 23 PC1 PC1

(d) 2.O (b) 2.0 64 64 x ìoo C\,I x ,c 0) 25 48 10 o 10 õ 1 5 c 11 7720 (¡) .c 35 q) C) zs .62 1 .5 lóô U' .Ê¡= 11 c 68t o) 62 o 21 35 CL 1 91 67 c .0 za fi o) 13 oU' o) fr o s67 ìð\ z.Ë Åv 42 0.5 1 .0 -3-2-10123 -4-3-2-101 23 PC2 PC2

Figure 6.6. The conelation between PCL,PC2 and RI2 (a, b) and NEI (9' d)-at Charlick. Numbers in the figure refer to the cultivars number in appendix Table 6.1. 179

6'4' Discussion hov¿ 3eno\Prc Results of these experimentsnestablished genetie differences in grain yield and GPC responses to N in barley cultivars. However, there was a large seasonal effect on the differences in the responses of cultivars and over the 2 experiments and there was no significant correlation between the yield response at the two sites (Fig. 6.2). Yields at

Northfield were lower than at Charlick which can be related to a later sowing and also perhaps to boron toxicity . One of the implications of these differences in response to N for screening and selecting cultivars is that seasonal conditions are very important in determining differences in grain yield responses between cultivars. The lack of a correlation between sites suggests that responsive cultivars will need to be adapted to other environmental limitations (such as soil boron levels, length of the growing season) and for the barley growing areas of South Australia there will need to be a number of cultivars suited to the range of environments where barley is grown.

The PCA showed that some of attributes measured in this experiment affected grain yield and grain yield response (Tables 6.10, 6.14) but there was no correlation between PCs and

GPC response. Although there were differences between cultivars in GPC response it was not as great as yield response (Fig. 6.2 c, f), and the variation in yield/protein relationship was related more to variation in the yield response.

There was a significant negative relationship between mean grain yield and mean GPC (Fig.

6.2 a, b) but not between the responses, either absolute and relative responses (Figs. 6.3 b,

e and 6.2 c, f¡. These data demonstrated that grain yield and GPC response were essentially

independent variables (Fig 6.2). This result agrees with data presented by V/heeler et al.

(1937) from Arthurton which also showed variation in response of yield and GPC to N were

independent. Therefore it is possible to have cultivars which have high grain yield with low

GPC response suited for malting barley, as well as high yield and high protein response

which would be suitable for feed barley. The responses of the 6 barley cultiva¡s which were

used in the fietd experiments (Chapter 3) varied between experiments, although Skiff, 180

Stirling and Schooner had relatively higher grain yield responses than V/eeah, Clipper and

Chebec but variable GPC responses.

The results of PCA analysis indicates that tall cultivars were less responsive in both years.

Therefore the experiments demonstrated plant height to be an important characteristic to N responsiveness. This result agrees with and extends the results obtained from the field experiments (Chapter 3) and the glasshouse study (Chapter 5) which showed the short cultivar, Skiff was more responsive than the tall cultivar V/eeah. In the two experiments at

Northfield and Charlick, lodging was not a problem, so the results suggest that there is some other factor associated with height which affected responsiveness. Apart from genetic differences, height is also reduced with N deficiency, so the negative effect that height had on yield response may indicate that cultivars better able to grow at low N levels (and hence are taller) may, as a consequence, be less responsive. Although lodging was not a problem in these experiments, tall cultivars may be more prone to lodging in high yielding situations which will also reduce yield response.

In the two experiments at Northfield and Charlick, cultivars which were late flowering tended to have lower yields at both levels of N, but in terms of the response to N, late flowering contributed to a higher yield response. The data collected in the two experiments do not enable a clear explanation to be given for this. Results of this experiment shows there is a significantly interaction between N and cultivar for date of flowering time at least at

Charlick. Therefore flowering time is an important characteristic affecting responsiveness to N.

At Northfield the PCA suggested that cultivars which showed more severe symptoms of boron toxicity at OkgN/ha tended to be more responsive to N. Gupta (1919) considers that

N is very important in affecting boron uptake by plants. He cites a number of examples

where, under high levels of boron, addition of N decreased boron uptake and concentrations

in plants. On average, at Northfield adding N reduced the boron score from 3.8 to 3.6.

Adding N at Northfield may therefore have been associated with alleviating boron toxicity. 181

The effect would have been greatest in the most sensitive cultivars, which may be an explanation of the positive contribution of boron score to N responsiveness.

There was considerable scatter of cultivars around the regression lines in Fig. 6.3; there were some cultivars which were well above or well below the lines, indicating them to be more responsive and less responsive respectively. Of the 6 cultivars which were examined in field experiments (Chapter 3) Skiff (48) was above the line at Charlick 1993 while Weeah

(29) and Clipper (26) were below the regression line, the other cultivars (Stirling (43),

Schooner (30) and Chebec(59)) were intermediate but the trend for these cultivar was consistent at Northfield. Comparison of the Triumph and Weeah groups, which were used

in the hydroponic experiments (Chapter 4), shows the Triumph group (Skiff, Franklin (77)

and Shannon (32)) were more responsive than the Weeah group (Weeah (29),Parwan (14),

WI-2728 (54)) except for Lara (66). Resibee (3), which is a single plant selection from

Research, was also well below the regression line. Therefore, all the cultivars in the

Research group, with the exception of Lara, showed poor responses to N in the field, even

though in the hydroponic studies (Chapter 4) they showed high early vigour. This result is

consistent with the poor responses with V/eeah in Chapter 3.

Assessment of the 3 responsiúe indices showed R[1 was variable and not useful, but RI2

and NEI can be useful as selection criteria because their trend was generally the same as the

increase grain yield response. The ranking of RI2 and NEI were not the same at the 2 sites

(Fig. 6.a) which is possibly a reflection of the different yield response to N in 2 experiments

and perhaps the importance of boron toxicity at Northfield. At each site, the correlation of

RI1 and NEI with PC1 or PC2 showed a similar distribution of cultivars (Figs. 6.5, 6.6);

that is, cultivars which had values of RI2 well above or below the regressions also had

values of NEI above or below the regressions. Therefore, both these indices can be useful

to identify responsive cultivars. However, the measurement of NEI is simpler because it

only involves yield.

This experiment has shown that it is possible to screen a large number of cultivars at 2 N

rates in the fîeld to help identify desirable cultivars. However, it also showed that there is a 182 large influence of the environment on the responsiveness of barley cultivar to N, which means that such screening needs to be conducted over a number of sites and years. The importance of interactions with other nutritional problems to N responsiveness was seen at

Northfield where boron toxicity was an important influence on response to N. 183 CHAPTER 7 GENERAL DISCUSSION

This general discussion seeks to interpret the main findings and identify gaps in knowledge for future resea¡ch. The results from each experiment have already been partly reviewed in separate discussions for each chapter. In southern Australia, the areas where malting barley can be grown successfully year after year are relatively small. Nitrogen defrciency is becoming a major factor limiting both grain yield and grain protein in winter cereals grown in South Australia as the intensity of cropping increases and soil fertility, in general, declines

(McDonald 19S9). Sometimes farmers in South Ausralia who wish to grow malting barley do not apply nitrogenous fertiliser because of the risk of increased grain protein levels.

Maltsters require plump grain which is relatively low in protein (up to approximately ll.8

|o),highin sta¡ch and which has a high and even rate of germination. The variety of barley has a major role in determining grain quality and therefore the malting grade of barley

(Sparrow t972; Smith 1990). However, little is known of the yield and protein responses to N of the main South Australian cultivars and recent advanced lines.

The study was a preliminary examination of the level of genotypic variability to N among some current cultivars of malting barley and examined patterns of response to N in the field, dry matter and N remobilisation, genetic differences in nitrate uptake and assimilation and genetic variability among 78 barley cultivars. The studies conducted in this progranìme found that consistent genotypic variation exists among barley cultivars in response to N fertiliser in field experiments (Chapters 3 and 6), in nitrate uptake (Experiments 1 to 5,

Chapter 4), and in remobilisation (Chapter 5).

Genetic differences in N responses in the field with barley has been observed by Wheeler et

al. (1987) in South Australia and by Birch and Long (1990) and Birch et al. (1993)in variability in eueensland. In the experiments described in Chapter 3, there was considerable

yield and protein responses between sites, but despite this there does appear to be some

consistent genetic differences between cultivars. The late-maturing semidwarf cultivar Skiff

showed tho greatest gr¿in yield response to N of the 6 cultivars used (Fig. 3.4). In contrast

'Weeah, a tall early-maturing cultivar, showed a much lower yield response (Table 3.9). 184

Stirling sometimes showed a response to N which was similar to that of Skiff (Tablc 3.9)' but this did not occur consistently at all sites. Therefore, the results show that the only reliable genetic difference was between the semidwarf cultiva¡ Skiff and the tall cultiva¡

Weeah. In wheat, the reduction in height caused by the introduction of the Rht genes resulted in an increase in yield potential and a greater ability to respond to N (Riggs et al.

1981; Gale and Youssefian 1935). The greater responsiveness of Skiff to N seen in Chapter

3 and Chapter 5 and the importance of short stature to yield responsiveness in Chapter 6 is consistent with this effect. However, Nedel et al. (1993) compared tall and semidwarf isotypes of 2-row and 6 row barley (produced by mutagenesis) and found no advantage of reduced height to responsiveness to yietd. Notwithstanding these results, the results presented in this thesis suggest that a reduction in height may improve yield responsiveness, although the importance of height to yield and yield responsiveness to N should be investigated further.

Apart from plant height, the time of ear emergence (Chapter 5) and the pattern of development (Chapters 3 and 4) may also contribute to the responsiveness of a variety to N.

Time of ear emergence affects the level of stress to which plants are exposed near anthesis and during grain filling and therefore it can influence both yield and quality- The PCA described in Chapter 5 cultivars suggested that flowering time contribute to yield responsiveness, but the importance of maturity is influenced greatly by seasonal rainfall. At

Northf,reld in 1991 late flowering \4,as an advantage but at Charlick in 1992, when rainfall

was slightly lower, early flowering contributed positively to N responsiveness. However,

the results highlight the importance of matching maturity to the expected seasonal rainfall in a

region to improve the chance of obtaining positive responses to N.

The differences observed between Skiff and Weeah in all experiements suggest that the

pattern of development may also contribute to the reponsiveness of a cultivar to N. The

reponse in the number of ears/m2 was an important factor in the yield reponses observed in

a more Chapter 3. Tiller production of Skiff was greater compared with Weeah, and it had

prosrrate gowth habit which suggest that its early development was quite different from that

of Weeah. Moreover, there was a high degree of synchrony in tiller development in Skiff 185

(Chapter 4) it was concluded ttrat both tiller compafed with Weeah. In the glasshouse study to the greatef production and the relative rates of tiller development contributed has an erect growth habit responsiveness of skiff. In contrast, vy'eeah is earlier maturing, tillering capacity and possibly and has an asychronous tillering habit which would reduce will also influence the the responsiveness to N. The synchrony of tiller development elongation and therefore may also competition for nutrients and photosynthate during stem this However' as noted above' these affect a cultiva¡'s sensitivity to stress during time' The relative importance of cultivars also differ in height which influences N responsiveness' development - needs to be examined further' these two characteristics - height andpattern of are not the only things In malting barley production grain yield and yield responsiveness Therefore strategies to improve important to the farmers; grain quality is also important' jeopafdise malting quatity' In this thesis' kernel yield responsiveness to fertiliser should not quality, although there are other weight and GNC (or GPC) were used as measures of malt extract, diastatic power' ø-amylase important measurements of malting quality such as kernel weight are 2 important factors and and p-amylase (Chap tet 2). However GNC and (Fig Adding N decreased kernel weight GPC has a srrong effect on malting quality 2.1). consistently(Fig.3.8)althoughtheextentofthedeclinevariedamongcultivars.The some cultivars the effect was exacerbated glasshouse experiment (chapter 5) showed that for the kernel weight was va¡iable and by postanthesis stress. Skiff was a cultivar in which greatlyreducedathighlevelsofNandwithpostanthesisstress.Inwheat,areductionin in kernel weight (Gale and Youssefian plant height less frequently resulted in a reduction competition for assimilate because of the 1985), possibly due, in part, to the increase in kernel weight (which will affect grain increase in kernels/m2. Therefore, in terms of Reducing height to increase yield plumpness) it appears that there is a trade off. smaller and more variable grain size' responsiveness may result in cultiva¡s with a the kernel weight of semidwarf wheat Anderson and Smith (1990) for example, found that that the wheat, while Nedel et at' (1993) found to be more sensitive to stress than that of tatl to be more sensitive to N than the kernel kernel weight of the semidwarf isotypes of barley of quality to ttre malting barley production' weight of the tall isotypes. Given the importance and their variability needs to be investigated the effect of reduced height on the size, GPC 186

Starch is the major component of the grain and the lower sta¡ch content at high rates of N was reflected in lower kernel weights. It has been suggested that greater mobilisation of assimilate stored in the stem and translocated to the grain will increase the stability of kemel weight by increasing rhe supply of sucrose to the grain (Brooks et al. 1982). However, although genetic differences in remobilisation of dry matter were measured (Chapter 5), there was little evidence to show that this was important in maintaining kernel weight. Skiff and Stirling were two cultiva¡s that showed the greatest variability in kernel weight, yet these cultivars remobilised the greatest rimount of dry matter during grain frlling (Iable 5.18). The lack of an apparent relationship between remobilisation and kernel weight stability is at variance with desiccation studies in wheat (Blum et al.l983a; Hoassain et al.1990; Regan et at. L993). Further work which compares remobilisation in wheat and barley and its importånce to yield should be undertaken.

The other important effect of N was on GNC. Consistently, adding N increased GNC

(Chapters 3, 6), although the effect depended on seasonal conditions and postanthesis stress

(Chapters 3, 5 and 6) as well as cultiva¡ (Chapters 5, 6). A major cause of the effect of N

on GNC was its effect on kernel weight. The amount of ca¡bohydrate deposited in the grain

in these barley cultivars had an important affect on GNC. Drought has more effect on

accumulation of sta¡ch than on protein deposition in the grain (Brook's et al. L982; lenner et

al. l99l). The partial dependence of GNC on the starch content of the grain suggests that

cultivars like Skiff and Stirling in which kernel weight is sensitive to postanthesis stness may

also have more variable GNC, which is not desirable for consistent production of barley of

malting quality. The results of these experiments have highlighted a problem associated with

attempts to improve N responsiveness in malting barley. Cultivars like Skiff, which have

been shown to be high yielding and responsive. However, it was found from an

examination of yield and protein responses of a diverse range of cultivars that responses in

yield and GNC to N were independent (Fig 6.2), which suggests that it may be possible to

select for high yield responsiveness and low GNC response.

(Chapter Despite the risk of increasing GNC by adding N fertiliser, the field studies 3)

suggest it is possible to add small amounts of N without causing a large increasc in GNC, 187 which may adversely affect malting quality. Depending on seasonal conditions and residual soil N, rates of 15-45 kgNlha appear to be suff,rcient to improve yields without greatly increasing GNC. The importance of postanthesis stress to GNC and yield (Chapter 5) suggests that other management practices, such as sowing time and disease control, which reduce stress, will atso maximise the chances of getting yield responses with acceptable

GNC and kernel weight for malting. The results also suggest that sensitive cultivars like

Skiff should only be grown in the higher rainfall areas, if malting barley production is the principal aim.

In most situations, N is applied early in the season. As well, mineralisation of organic N results in a flush of nitrate in the soil soon after sowing (see Chapter 2, section 2.3-2). To minimise the loss of this N requires cultivars which have an ability to take up nitrate during the seedling stage. The studies reported in Chapter 4 indicated that significant genetic differences in growth and nitrate uptake exist. Results of Experiments 1 and 2 in Chapter 4, suggested that Skiff and Franklin consistently produced large seedlings and took up large amount of nitrate but Stirling and Schooner produced smaller seedlings and had lower nitrate uptake. The difference in nitrate uptake among cultivars was related to the size and $owth rate of the seedlings (Table 4.2 and4.3). Differences in $owth rate (Chapin and Bieleski

lg82) and in the size of the root (Hackett 1968) have been reported to affect nutrient uptake.

There were some consistencies in the results from the hydroponic studies and measurements of early vegetative gowth in the field (Chapter 3). Over 3 experiments (Chapter 4,

Experiments 3, 4 and 5) cultivars which glew more vigorously, took up more nitrate. Data from the field experiments (Chapter 3) showed that at Northfield 1991, dry matter

production at 10 weeks in Skiff and Weeah was higher than that of Stirling and Schooner

(Table 4.12). The responses in early growth in Skiff and Weeah in Northfield 1991 a¡e

consistent with the response observed in the hydroponic studies, but only Skiff showed a

response in grain yield to N; Weeah was not responsive. Therefore, high vegetative $owth

early in the growing season may not always be related to the gfain yietd response and so the

hydroponic screening needs to be done in conjunction with field screening'

t 188

The experiments demonstrated the strong effect that environmental conditions havc on yield and protein responses in barley. However, the studies in Chapter 6 also indicated the¡e we¡e significant genotypic differences in grain yield and GPC response to N in a large number of barley cultivars although the grain yield responses of the different cultivars at both sites differed, which showed the importance of seasonal conditions on grain yield and GPC response to N. PCA analysis showed date of flowering, ptant height and growth were important parameters related to grain yield responses, but not in GNC. Tall cultiva¡s did not response much to N in grain yield. The PCs had a greater influence on yield and yield responsiveness than on GNC or response in GNC. Also the results of the Northfield experiment indicates that sensitivity to high soil boron gfeatly may affect responsiveness to

N. Despite the large genotype x environment interaction, the screening of a large number of cultivars at 2 N rates in the field may allow useful estimations of the responses of yield and

GNC to N to be made. However, the results of the preliminary field evaluation (Chapter 6) which showed a strong environmental effect on responsiveness, suggest that screening should be done in the tatget environment in order to identify and develop genotypes that are highly and consistently responsive to N, especially in environments where available moisture during grain filling is variable.

Three indices of responsiveness were examined, but onlry 2; RIz and NEI were found to be potentially useful as selection criteria for screening a large number of lines in the field

(Chapter 6). As the ranking of yield response of cultivars generally was essentially the same

for both RI2 and NEI, it may be preferable to use only NEI as responses indices for

screening for N responsiveness. However, screening in the field is variable and achieving a

yield response depends on the fertility of the soil, seasonal weather and rates of N

application. Another screening technique used was the measurement of growth and nitrate

uptake (Chapter 4). This method identifred cultivars which have high early vigour, although

as discussed previously, high vegetative growth early in the growing season is not always

related to the grain yield response; however, it may be possible to select cultivars which are

more efficient and can take up more nitrate from the soil. and still have a high yield

response. 189

The practical implications as a result of the studies reported in this thesis are summarised as follows:

Grainyíeld ospects

(i) Moderate levels of N (<45 kgN/ha) appear to be suitable to increase grain yield

without greatly reducing the chance of increasing in GNC. However, management

strategies should aim to reduce postanthesis stress (see below).

(ii) Semidwarf barley cultivars such as Skiff may yield more than tall tyPes, but such

cultiva¡s may need more fertile conditions and adequate soil moisture and low moisture

stress to fulfil their potential. Such cultiva¡s therefore should only be grown in higher

rainfall areas and should never be sown later in the seasons than the optimum planting

time.

Qwhty aspects (iii) There is evidence from the studies described in this thesis and from other published

results to suggest that kernel weight of semidwarf cultivars may be more sensitive to

envi¡onmental stress than that of tall cultiva¡s, which is not desirable for malting quality.

Therefore practical management such as time of sowing and rate of sowing are necessary

to maintain grain size.

(iv) Cultivars which show a high yield response to N may also have more variable kernel

weight and GNC, which would result in a reduction in the consistency of quality.

(v) GPC is affected by environmental and genetic factors. Small applications of N may

not increase GNC but may result in significant yield increases. Other management

strategies which can be considered include choice of cultivar and time of sowing to

reduce postanthesis stress. 190

Breedíng crspects

(vÐ It may be possible to undertake a preliminary screening of a large number of lines (eg

1,000) using a large scale glasshouse hydroponic system to identify cultiva¡s with

vigorous early growth, but in addition it would be useful to do a more intensive screening

of lower number of cultivars delected from the first screen with measurements of growth

and nitrate uprake (if possible) of the seedlings to identify parents for funher studies.

(vii) It is possible to deveþ field a screening technique for a large number of genot¡pes

with 2 N rates for grain yield and GPC response to N, although considerable va¡iation

from year to year and site to site is to be expected. Screening should probably to targeted

at specifrc environments.

(viii) It is possible to select for high grain yield and low GPC response to N in barley

cultivars for malting quality, because of ttre independence of responses in yield and GPC.

Future work

The suggested future research as a result of this study is as follow:

(i) The results from the field experiments showed short stature to be an important

cha¡acteristic for responsive cultivars, but the semidwarf cultiva¡ Skiff showed a low and

variable kernel weight and a tendency for high GPC. Therefore the importance of height

to quality and stability of quality needs to be examined. This should be done with a

number of rates of N to def,rne the N-responses in yield and quality. Additionally it

should be possible to select for semidwarf barley with larger and more stable gfain size.

(ii) Time of flowering was related to N responsiveness, although the form of the

relationship differed in the two years with dissimila¡ rainfall. As well, based largely on

the differences between Skiff and Weeah, there was some evidence that the pattern of

development may influence N responsiveness through its effect on tiller production, the

synchrony of tiller development and the level of competition for photosynthate and N within the plant. Further studies on the importance of plant development to N

responsiveness should be undertaken. 191

(iii) Further study on the physiology of grain filling in responsive and non-responsive

cultivars is needed, to help understand the inter-relationship between grain protein

deposition and sta¡ch deposition in responsive and non-responsive cultivars.

(iv) In this thesis, quality was defined in terms of kernel weight and GNC (or GPC). It

is necessary to look at quality in more detail, eg the effects of N and genotype on malt

extract, diastatic power, cr-amylase and p-amylase activity over a range of envi¡onments.

(v) Screening for responsiveness both in the field and in hydroponic system was found to

be feasible. It would be useful to use these methods to undertake genetic studies to

examine the genetic basis of N uptake and yield/protein responsiveness. The Weeah-

group of cultiva¡ (non-responsive) and the TriumpVskiff group (responsive) could form

the basis of populations that could be used for further genetic studies.

(vi) The importance of boron toxicity in influencing responses to N among the 78

cultivars at Northfield in 1991 highlighted the importance of other nutrient deficiencies

and toxicities to N responsiveness. Further work on the interaction between N and other

nutrients, palticularly boron, manganese and zinc in South Australia, is needed

The research described in this thesis hightighted the difficulty in selecting for high grain yield response to N and low GNC in the field because of the effect of seasonal weather and field management practices on grain N. The results of this thesis suggest that there a¡e genetic differences in nitrate uptake between barley cultivars at the seedling stage, which were related to the root system, although early vigour to yield in the field was not always beneficial to yield. This research has provided some understanding of the limitations to grain yield and GPC in barley cultiva¡s and suggested management strategies for a limited range of environments. Further work over a more extensive set of environments and using more new barley cultivars is needed to extend these principles' t92 APPENDICES

Appendix 3.1. Measurement of total N concentration using the Kjeldahl method oC The sample (1g) was digested with 15 mL of concentrated sulphuric acid at 390 which converts the N in the sample into ammonium sulphate ((NlI¿)zSO¿). A catalyst tablet of 3.5 g (se) was added to promote the oxidation of the organic matter. The catalyst tablet also contain potassium sulphate which raises the temperature of the digest and thereby increases the rate of the reaction. After digestion, 50 mL H2O is added and the ammonium is released from the digest by steam distillation with 35 mL alkali (NaOH). Ammonium was collected in a27o boric acid solution which is titrated with 0.1 M HCI acid. The boric acid solution contain l%o ofmethyl red/ bromocresol green indicator solution (Bremner 1965).

Appendix Table 3.1. Origin and characteristics of 6 barley cultivars used in field experiments

Cultivars Height Maturity Origin Pedigree

Clipper Med.tall Early-medium SA Proctor/PriorA skiff Semidwarf Medium-late SA ((AD x WI-2335) x (CD 28 x WI- 2231))lL6s

Stirling Med.tall Early WA Dampier/(A 1 4)Prior/Ymer/3Æiroline rWe€att Tall Early-medium Vic. Prior/Resea¡ch

Chebec Med.tall Early-medium SA ((Offartin*Clipper(2)*Wl-2468)- 88lsl6n

Schooner Med.tall Medium SA hoctor/PriorA//Proctor/Cl-3576 193 Appendix Table 3.2. The residual mean squares of models chosen to fit dry matter at anthesis against N

Cultivar auadratb Linear Mitscherlich ModiFred Squarc Mitcherlich Root

Nortlfield 19X) Clipper 1609 2594 834 458 598 Stirling r469 r256 t477 t470 1418 Weealt 2391 3055 2777 t453 3453 Schooner r523 1290 t523 1902 1548 Chebec 369r 3882 3423 4293 381 1 skiff 4ro9 3576 4t36 4153 4?ß7

Northfield 1991 Clipper 1055 8926 TzIM t1435 9503 Stirling 3756 4715 7379 4æ9 4212 Weeatt 5488 4836 8&9 6850 5527 Schooner 2&6 8937 6759 3041 3923 Chebec 5300 4912 19170 6286 54/,6 skiff 42tt 3558 8398 4980 4@6

Chslick 1991 Clipper 1278 r3/.r 12972 15947 r385 Stirling 267t 2397 2549 3019 2,48',1 V/e€alt t4v+ 1510 15676 r3796 1739 Schooner 5965 5038 5031 4481 ffi5 Chebec 9698 82t4I 6796 7944 9613 skiff 4850 617t 5600 5327 614t

Appendix Table 3.3. The residual mean squ¿ues of models chosen to fit grain yield against N

Cultiva¡ Quadratic Linear Mitscherlich Modified Square Mitcherlich Root

Norrhfield 1990 Clipper 0.066 0.075 0.061 0.076 0.059 Stirling 0.242 0.225 0.254 0.2u 0.268 V/e€att 0.011 0.028 0.011 0.012 0.011 Schooner 0.022 0.019 0.021 0.012 0.014 Chebec 0.017 0.017 0.017 0.01ó 0.019 skiff 0.013 0.07ó 0.012 0.012 0.022

Nortffield 1991 Clipper 0.020 0.041 0.035 0.017 0.030 Stirling 0.024 0.024 0.019 0.008 0.013 AND Weeatt 0.013 0.030 0.012 0.010 Schooner 0.063 0.058 aND 0.073 0.059 Chebec 0.032 0.039 0.048 0.039 0.033 skiff 0.080 0.2r9 0.28r 0.072 0.188

Clørlick l99l Clipper 0.039 0.052 0.023 0.020 0.019 Stirling 0.034 0.095 0.053 0.039 0.053 Weeat¡ 0.082 0.097 0.080 0.103 0.078 Schooner 0.054 0.07r 0.037 0.033 0.029 Chebec 0.032 0.035 0.032 0.039 0.035 skiff 0.032 0.084 0.028 0.035 0.024

âND=Not done 194

Appendix Table 3.4. The residual mean squares (x10-3¡ of models chosen for frned grain N concentration against N

Cultiva¡ Quaeatic Linear Mischerlich Modiñed Square Mitcherlich Root

Nortlfield l9N Clipper 3.6 3.94 3.61 4.49 4.63 Stirling 1.04 1.56 0.96 1.09 t.73 Y/e€att 14.7 20.03 18.30 t2.94 r0.98 Schooner 6.22 5.33 6.25 6.63 6.33 Chebec 6.ll 7.02 6.10 7.t6 5.72 skiff 4.il 7.22 5.27 5.26 s.57

Northfield 1991 Clipper 1.38 r.29 t.42 1.09 r.10 Stirling 6.51 5.47 6.4t 7.44 6.54 V/eeah 4.67 4.07 4.62 5.73 4.7r Schooner 12.50 12.43 8.60 10.69 9.16 Chebec 9.42 8.31 9.28 Il.13 9.67 skiff 11.83 10.54 6.2r 7.08 8.03

Charlick 1991 Clipper 11.43 11.70 12.2 13.48 13.99 Stirling 24.78 20.70 24;n 30.32 24.50 Y/eeatt I 1.17 93.88 r1.16 13.77 11.10 Schooner 17.97 15.Ø 18.15 21.t6 17.35 Chebec tt.2t 94.99 rt.32 r0.52 92.76 skiff 13.07 r0.96 13.t2 12.r9 11.34

Appendix Table 3.5. Coeff,rcient of the fitted equations for the relationship between kernel number and dry matter at anthesis, and between kernel weight and kernel number at Northfield 1990, 1991, Charlick 1991 (n=22)

Cultivar a b1 xl0-3 + se ¡2 Signifi.

Kernel number vs DM¿ l. Clipper 148 t4.4 + 2.6 0.58 rl 869 r5.3 r.4.9 0.31 Stirling * Weeatt 227t 9.8 + 2.8 0.35 a Schooner -2345 18.6 + 3.5 0.56 rl t787 t2.3 t2.3 0.57 Chebec rl skiff 679 r7.3 t 2.9 0.63

Kernel weight vs Kernel number Clipper 43.r -3.4 ! r.7 0.15 NS Stirling 42.0 -6.5 + 2.r 0.30 NS We€alt 38.9 -0.3 ! 2.5 0.01 NS -9.2 + 0.01 NS Schooner 38.4 1.8 * 43.1 -4.4 + 0.22 Chebce 1.7 :l skiff 44.r -7.3 + 1.7 0.46 195

Appendix Table 3.6. Coefficient of fitted equations for relationships between yield components at Northfield 1990, 1991 (n=6)

Cultivar a b1 x10-3 * se P2 Signifi.

Grain yield vs Ea¡ number Clipper 1.60 2.gI + 9.1 x10-4 0.42 *** Stirling 1.44 3.51 +10.5x10-4 0.45 *:lr WeÊah 3.05 0.00 t10.5x10-4 0.02 NS Schooner 1.30 3.92 t 5.6x10-4 0.78 *** *** Chebec 0.84 4.98 + 9.9x10-4 0.64 skiff 0.87 4.53 t 5.5x10-4 0.83 :l** Kemel numbervs DM¿ Clipper -1.01 16.9 + 3.6 x 10 -3 0.62 *** Stirling -1.06 18.8 I 6.0 x 10 -3 0.41 ** Weeah t.32 r2.8 + 2.7 x l0 -3 0.62 *** Schooner -3.38 2t.4 t 3.6 x 10 -3 0.72 *** **,1. Chebce t.25 t4.l + 2.3 x t0 -3 0.72 skiff -0.61 20.2 t 3.2xl}-3 0.74 r¡ rl. {.

Grain yield vs Kernel number 2.47 63.7 0.37 ** Clipper tO.02 * Stirling 2.30 85.9 r 0.04 0.29 rWeeah 2.84 -1.70 r 0.02 0.00 NS + {.* Schooner 2.21 99.9 0.02 0.77 *** Chebce 2.26 t02.9 r 0.02 0.61 skiff 1.60 163.0 + 0.03 0.75 **

Kernel weight vs Kernel number 41.8 -245 + 0.16 NS Clipper 0.15 * Stirling 40.8 -611 + 0.22 0.36 Weeah 37.5 108 t 0.21 0.02 NS Schooner 39.4 -150 t 0.11 0.t2 NS -301 + 0.49 ** Chebce 40.9 0.08 ** skiff 42.6 -720 + 0.11 0.76

Grain yieldvs DM¿ 1.54 0.26 o.47 ** Clipper 2.14 t *¡t* Stirling t.23 3.50 t 0.44 0.56 NS Weealt 2.69 o.22 t o.28 0.03 *** Schooner 1.58 2.il t 0.19 0.83 + **,¡ Chebce 2.17 1.82 0.19 0.69 *** skiff t.52 3.24 + 0.43 0.54 196

Appendix Table 3.7. Coefhcient of fitted equations for kernel weight and kernel number against N at Northfreld 1990, 1991, Charlick 1991 (n=6)

Cultivar a bt x10a t se xl64 p2 Signifi.

Northfíeld 1990 Clipper 45 7 -9.5 t 4.3 0 45 NS Stirling 40 9 -5.6 t 3.9 0 26 NS 43 -7.3 + 7.8 0 I3 NS Weeah 4 * Schooner 51 6 -24.4 + 9.4 0 53 40 4 -2.4 + 3.2 0 08 NS Chebec * skiff 4L 8 -6.0 t 1.1 0 85

Northfíeld 1991 Clipper 54.6 -r3.4 + 5.2 0.s3 * Stirling 28.6 4.7 + 6.8 0.07 NS Weealt 28.9 8.8 t 6.2 0.26 NS Schooner 36.7 0.6 + 5.7 0.02 NS Chebce 46.7 -7.8 t 4.3 0.36 NS skiff 44.7 -8.9 I 4.2 0.43 NS

Charlick I99l Clipper 41.9 -0.2 t 8.6 0.0 NS Stirling 42.4 -4.8 t 3.9 0.20 NS V/eeah 46.4 -t2.4 + r6.8 0.08 NS Schooner 26.7 15.2 t 7.8 0.39 NS 0.02 NS Chebce 45.5 -5.2 t 14.7 * skiff 47.7 -8.4 t 2.6 0.63 t97

Appendix Table 3.8. Coefficient of fiaed equations for the relationship between yield component agáinst N at Charlick 1991 (n=6)

Cultivar a b1 x10-3fse p2 Signift.

Grain yield vs Kernel number Clipper t.2 111.8 t 6.5 x 10-2 0.33 NS Stirling 1.6 122.9 t 6.1 x 10-2 0.40 NS Weeah 0.9 140.0 t 10.5 x 10-2 0.23 NS Schooner 1.2 146.8 t 9.5 x l0-2 0.29 NS Chebec 1.9 101.0 + 11.4 x l0-2 0.01 NS skiff 1.3 Itg.4+ 7.2xl}-2 0.51 *

Grain yield vs Ear number Clipper 1.7 0.7+ 1.2 x 10-3 0.03 NS Stirling 3.0 0.9+ 1.9x 10-3 0.03 NS Weeah 1.8 -0.0+ l.5x 10-3 0.00 NS Schooner 2.9 -0.9+ 2.4x10-3 o.o2 NS Chebce 2.4 0.6 t L.2x lO-3 0.05 NS ¡¡** skiff 0.6 4.4+ 0.5 x 10-3 0.93

Kernel numbervs DM¿ Clipper 3.3 7.lt 1.8x 10-3 0.73 *** Stirling 6.3 4.1 t10.3x10-3 0.03 NS V/eeah 5.7 6.0t 2.6xlO-3 0.01 NS Schooner 3.8 63+ 7.6x10-3 0.1 1 NS Chebec 8.1 -0.0+ 3.1 x l0-3 0.00 NS skiff 5.8 7.3t 5.9x 10-3 o.20 NS

Grain yieldvs DMa Clipper 1.61 0.7+ 0.6x 10-3 0.19 NS Stirling 1.64 2.Ot 1.9 x 10-3 o.L7 NS Weeah 1.55 3.7 t 7.4x10-3 0.04 NS Schooner 3.36 -1.6+ 2.1x10-3 0.08 NS Chebec 1.89 1.51 7.1 x 10-3 0.41 NS * skiff 1.25 3.2t 1.1 x 10-3 0.60 198 Appendix Table 3.9. Mean of parameters measurement at Northfield in 1990 Cultiva¡ Nraæ Yield DMa TNa DMm ENo KNo K'Wt GNC GI{Y RDM

Clipper 0 2.3t 295 341 419 362 4.80 4t.61 l.ó9 3.28 t29 l5 2.81 469 473 408 5.48 39.41 1.89 4.06 tr7 30 2.82 439 440 509 426 5.73 4t.56 1.82 4.31 139 45 2.84 5r5 608 485 6.80 41.06 1.87 5.22 t& 60 3.12 5l I 479 5s3 443 s. l8 40.72 1.99 4.23 151 -t5 3.27 50'l 638 543 7.72 37.65 2.rl 6.12 185 90 2.60 528 627 613 523 5.'79 38.59 2. l8 4.9r 198 105 3.15 541 660 519 7.69 38.24 2.31 6.77 2t5 Stirling 0 2.73 40ó 368 356 323 3.34 40.24 t.73 2.27 50 l5 1.65 339 443 346 4.22 37.37 1.86 2.92 r51 30 3.01 453 413 521 395 6.38 40.22 1.87 4.84 128 45 2.99 455 533 429 6.22 37.56 L9l 4.43 r62 60 3.44 525 463 527 429 8.77 37.04 1.99 6.45 129 75 3.49 567 638 523 9.61 36.4t 2.06 7.24 190 90 3.r6 539 562 563 503 7.10 33.94 2.17 5.2t 156 105 3.25 572 543 474 7.32 34.39 2.3t 5.78 t42 Weeah 0 2.33 344 336 554 418 7.52 40.05 1.99 5.95 187 l5 2.70 358 549 418 6.90 39.35 1.84 4.96 172 30 2.7 t 388 428 613 422 4.84 40.42 1.88 3.69 2t7 45 2.95 508 67r 412 7.09 40.81 t.73 5. l4 228 60 3.02 s6't 558 681 402 6.52 37.33 2.lt 5.t4 194 75 3.06 545 690 510 7.96 36.22 2.09 5.91 2U 90 2.82 634 43'l 715 373 6.41 36.7t 2.26 5.34 223 105 2.98 534 761 409 7.32 36.36 2.27 5.93 265 Schooner 0 2.19 331 3M 393 300 4.65 41.85 r.75 3.4 100 l5 2.63 324 493 375 5.54 38.38 1.88 4.t2 155 30 2.'t7 454 428 426 34t 5.00 38.21 1.86 3.59 97 45 2.70 4M 494 346 5. l9 38.98 1.93 4.43 l4l 60 2.78 496 4'17 5t'l 385 4.86 39.76 2.t7 4.26 14l 75 2.76 547 561 420 6.23 38.38 2.07 5. l0 128 90 2.98 526 482 618 455 5.75 36.06 2.27 4.69 l8l 105 3.r6 598 496 397 5.86 36.38 2.21 4.73 96 Chebec 0 2.45 287 283 477 363 5.94 41.00 1.69 4.tt t& l5 2.60 318 504 416 6.24 38.35 l.89 4.43 r67 30 2.62 4ll 410 573 M8 6.l0 39.12 1.90 4.47 212 45 3.05 49t 554 443 6.88 39.7 1.88 5.t2 136 60 3.15 535 484 593 468 7.31 37.5t 2.t0 2.75 157 75 3.05 407 500 460 5.31 39.1 8 1.98 4.t4 l4l 90 3.42 47t 451 736 553 9.69 38.25 2.13 7.86 252 105 3.32 555 545 453 5.59 37.72 2.04 4.34 t26 1) skiff 0 1.84 292 335 299 221 3. l6 40.28 1.82 2.31 15 2.56 299 407 366 4.71 38.05 1.89 3.43 12l 30 2;79 365 353 497 420 6.41 38.56 1.80 4.44 140 45 2.93 429 535 463 7.01 37.61 l.9l 5.03 l4r 60 3.t7 456 518 559 531 8.97 37.4 2.07 6.95 163 75 3. l9 419 497 486 8.16 35.7t 2.09 6.74 t42 90 3.t7 593 620 6't3 6t2 I 1.07 34.69 2.33 8.81 2t2 105 3.03 484 495 577 9.92 36.17 2.36 8.56 t24

Cultivar NS NS NS NS NS NS 't+* N * Cultivar x N NS NS NS + * NS NS NS Appendix Table 3.10. Mean of parameærs measurement at Northfield in 1991 r99 Cultivar Nraùe Yield DMa TNa DMm ENo KNo KWt GNC GI.TY RDM

Clipper 0 2.89 701 560 918 410 I1.59 39.85 t.67 7.79 210 l5 3.18 648 492 893 468 10.50 43.22 1.63 7.40 t92 30 3. l9 592 483 1004 500 I 1.98 38.75 t.& 7.59 52 45 3.l s 570 508 838 472 10.06 40.58 t.76 7.l l 138 60 3.34 779 578 843 52r r 0.61 37.92 1.72 6.77 341 75 3.56 682 523 995 573 I 1.58 38.48 1.71 - 7.93 134 90 3.23 849 580 963 567 tr.20 39.90 l.8l 8.03 330 105 3.02 &7 521 tn2 600 t3;76 36.03 l.83 8.99 l5 Stirling 0 2.59 4r8 384 690 378 10. l3 35.58 r.19 6.52 105 t5 3.19 429 423 476 426 tt.22 35.00 1.87 7.r2 12 30 3.09 654 5ó3 176 480 12.26 35.00 1.85 7.'74 286 45 3.09 548 392 919 494 t2.86 36.08 1.90 8.66 66 60 3.r6 583 503 579 430 9.21 33.28 1.99 5.98 324 '75 3.28 623 502 883 565 9.66 33.88 1.78 6.00 19 90 3.40 653 615 476 555 lt;79 30.50 1.98 7.06 533 105 3.37 602 60ó 484 &4 10.18 30. l0 2.02 6.0r 465 V/eeâh 0 2.78 618 471 916 405 lt.29 39.08 r.63 7.tt t37 l5 3.0'7 640 496 982 457 t2.03 41.30 1.58 7.81 r55 30 3.0r 767 508 952 s03 n.97 4t.25 1.72 8.42 305 45 2.89 733 493 r072 518 12.40 41.r7 1.67 8.47 172 60 3.08 695 431 758 391 8.30 36.35 t.72 5.27 294 15 2.75 743 480 848 457 l1.00 37.48 1.63 7.20 34 90 2.60 9r6 583 1052 560 12.70 36.97 1.83 9.04 358 105 2.52 754 491 995 508 11.20 35.38 1.80 8.03 2t'l Schooner 0 3.3',7 585 486 l0l8 461 14.01 39.03 1.50 8.21 105 l5 3.58 706 556 898 505 r 1.56 39.83 1.70 7.84 267 30 3.58 736 6t2 1050 s68 13.7 | 38.55 t.76 9.42 2t8 45 3.69 807 558 1078 589 t4.34 37.60 1.87 9.98 282 60 3.'7 t 874 632 l09t 572 14.09 37.20 1.83 9.54 321 75 3.08 701 540 961 541 l 1.46 36.63 1.70 7.09 215 90 3.74 694 576 980 586 12.t4 34.92 l.7l 7.23 178 105 3.42 659 630 I 150 638 13.94 35.58 1.93 9.52 87 't.46 Chebec 0 3.08 586 479 813 388 tt.41 38.35 t.66. 208 t5 3.39 706 610 867 470 r2.t3 37.55 1.72 7 .79 292 30 3.28 558 508 813 450 I l.l3 37;78 l.6l ó.78 r67 45 3.88 623 510 967 513 12.53 38.63 1.13 8.43 t42 60 3.48 788 g8 987 558 12.89 36.00 1.88 8.66 268 75 3.67 808 615 919 530 t2.07 36.73 1.65 7.36 320 90 3.6s 888 688 9t4 579 t2.67 36.10 t.82 8.19 432 105 3.63 881 't06 1027 583 13.59 36.43 1.90 9.48 357 skiff 0 2.80 496 4't7 673 4t4 10.14 36.00 l.6l 5.94 l8l l5 3.04 489 480 897 513 11.64 37.25 1.88 8. l0 24 30 3.13 695 6r-l 6s8 445 9.97 34.90 l.89 5.80 343 45 3.92 609 649 707 583 l 1.35 34.55 l.7l 6.l0 261 60 4.32 578 740 810 692 13.79 33.47 1.73 8.21 178 75 4.09 651 '7 19 871 6'77 t4.24 32.30 |.74 8.34 114 90 3.73 7Q2 684 7M 580 t2.34 30.55 l.8l 6.90 339 105 3.45 728 673 777 623 14.08 31.32 l.8l 8.08 331

Cultivar NS NS NS NS +*)i *,| * N NS NS *** NS Cultivar x N NS NS NS NS 200 Appendix Table 3.1 l. Mean of parameærs measurement at Charlick in 1991 Culrivar Nrare -Yield DMa TNa DMm ENo KNo Kwt GNC GI.ry RDM

Ctipper 0 1.53 342 27'l 522 352 5.52 44.00 1.79 4.3'l 62 t5 l.93 353 362 583 36s 6.26 39.93 r.66 4.08 28 30 2.26 612 412 848 363 8.62 43.55 r.88 7.23 t42 45 2.13 547 418 748 440 7.15 44.30 1.99 6.56 t22 60 2.10 457 400 633 4t2 5.'t7 39.60 2.06 4.69 5l 15 l .88 755 582 863 513 8.07 38.00 2.17 6.72 205 90 r.89 527 450 757 408 7.06 43.80 1.98 6.2t 7t r05 2.08 59'7 433 875 413 8.1 I 4t.57 2.tr 7.13 60 Stirling 0 r.93 358 362 608 398 6.46 40.73 l.98 5.12 r6 t5 2.34 400 360 4'75 385 6.02 4t.35 r.75 4.24 341 l0 2.'l I 480 43'l 593 392 9.69 37.90 1.84 5.91 200 45 2.53 522 490 563 423 7.t5 35.75 r.80 4.55 2ll ó0 3.02 405 423 887 410 l 1.5 36.85 2.81 12.23 59 75 2.75 535 397 700 385 8. l0 37.80 t.34 4.07 r39 90 2.91 505 383 665 3't8 7.5t 37.75 2.19 5.95 t26 105 2.44 550 562 868 600 9.30 40.00 2.23 7.'t9 55 Weeâh 0 1.26 373 322 520 348 6.06 39.55 |.49 3.72 99 l5 1.83 405 328 582 4t7 6.1l 45.50 1.55 3.59 63 30 r.70 420 322 s05 342 5.71 43.33 2.38 5.03 t32 45 1.73 537 3ó0 430 362 4.85 42.42 2.20 3.26 276 60 l.8l 817 552 86"Ì 525 7.58 36.28 1.82 6.t3 297 '15 2.37 582 408 6s2 352 7.06 32.32 2.60 6.86 195 90 l.55 782 510 470 428 4.36 34.85 2.85 4.26 459 r05 t.77 595 255 697 230 6.42 37.00 2.63 6.72 t37 Schooner 0 2.t6 527 458 608 427 7.48 39.25 l.77 5.33 228 l5 2.7 | 525 432 730 372 7.50 45.50 r.85 6.t7 137 30 2.84 618 472 987 598 10.88 43.33 r.95 8;79 96 45 2.79 692 572 8ó8 550 10.18 42.42 1.92 8.25 257 60 2.92 578 497 680 497 7.t8 36.28 2.02 5.15 r69 15 2.39 670 573 678 583 6.59 32.32 2.43 5.06 374 90 l.55 593 510 615 510 5.t2 34.88 2.54 4.58 160 105 t.77 810 255 930 623 8.71 36.38 2.41 7.6t 207 Cheþc 0 2.43 543 440 122 467 8.06 42.75 2.t3 7.42 r62 l5 2.47 463 397 685 422 7.U 43.38 t.72 5.06 82 30 2.93 580 388 833 427 9.22 42.58 t.92 7.16 138 45 2.56 558 447 735 478 7.45 44.60 1.73 5.52 153 60 3.05 707 583 630 438 7.t2 38.05 2.t4 5.57 349 75 2.83 5't2 &5 823 635 8.68 36.30 3.03 10.05 65 90 2.84 512 442 845 580 8.52 42.40 2.30 8.s3 32 r05 2.88 772 658 795 533 8. r3 40.78 2.26 7.37 302 skiff 0 2.O9 405 407 648 370 8. l3 41.67 2.00 6.99 95 l5 2.85 48'7 463 &0 512 7.95 40.55 r.33 4.73 178 30 3.08 482 483 790 s78 9.77 38.98 1.63 6.7 5 76 45 2.93 545 485 735 510 8.51 4t.47 t.7 5 6.t4 162 -733 60 3.25 723 748 597 9.06 38.58 2.33 't.68 320 75 3.43 568 't33 975 650 t2.t2 37.53 l.5l 7.09 5l 90 3.42 680 662 943 642 l 1.65 39.53 2.O2 9.49 r89 r05 3. l3 575 570 923 &o l r.66 36.53 1.89 8.40 100

Cultivar NS NS NS NS NS NS NS N * NS Cultivar x N NS :} NS NS NS NS NS 201

Y 0.83-0.00152 X Clipper r= -0.530 NS Stirling = 0.85 0.85 r= -0.816' o s p o X 0 80 o o ;0.80 o o o) Þ E o .c oo c 0.75 ; 0.75 U) .n o o (¡) ¿ o ¿ o cct 0.70 .E o.zo c c o q) o o) o) o 0.65 e 0.65 .= z z o 0.60 0.60 0 20 40 60 80 't00 120 0 20 40 60 80 100 120 N applied (kgN/ha) N applied (kgN/ha)

Weeah Schooner Y= 0.85-0.00052 x 085 o r =-0.587 NS 0.90 OO r = -0.733' òS a o x 0 .80 0.85 o x ! o o .ç Þ 0.80 o 0 .75 .c U) o o o o

Chebec y= 0.824-o.ooo32 x skiff 0.85 0.85 r = -0.290 NS r = -0.810' òS o àS x 0.80 ;0.80 OO o o o E Ec o o .g 0.7s 0.75 ;U' U)o o ¿ o o (!¿ c 0.70 E o.zo c c o) o o, o) o 0.65 E o.6s z= z 0.60 0.60 0 20 40 60 80 100 120 0 20 40 60 80 100 120 N applied (kgN/ha) N applied (kgN/ha)

Appendix Figure 3.1. The interaction between Cultivars and N for N harvest index of 6 barley cultivars at Northfield 1991. 202

Appendix 4.1. Nitrate determination in plant tissue

Nitrate accumulation in the dried plant material was determined by enzymatic reduction

(Escherichia coli nitrate reductase) of nitrate to nitrite and the measurement of the latter with a colorimetric reagent (McNamaraet al.l97l).

General procedure

Shoot and root material of barley were ground to pass a 40 mesh screen. 60 mg of dried, ground, plant material was placed in a 100oC water bath for 30 min. with 40 mL double distilled water. After cooling, the extract was frltered through Whatnan No. 1 filter paper.

Use of dissimilatory nitrate reductase from Esclvrichia colí

The assay mixture contained 0.5 mL of 0.1 mM K phosphate buffer (pH 7.5), 0.5 mL of

0.4 M sodium formate,0.lml. E. coli extract (provided by Dr. V/allace of the Plant Science

Depatment at the Waite Institute), 0.1 mL plant extract containing less than 100 nmol nitrate and H2O in a final volume of 2 mL. After incubation at 45oC for 4 h the reaction \ryas stopped by the addition of 1 mL of O.\lVo (w/v) N-(1-naphthyl) ethylenediamine dihydrochloride. After l0 min, 2 firL water was added and the mixture centrifuged for 15 min. The absorbance of the pink diazo dye compound was determined at 540 nm and the reading converted to amounts of nitrate using the calibration graph. 203

Appendix Table 5.1. Balance between root and shoot total N of 6 barley cultiva¡s

10 days after anthesis Mamrity Cultivar Root Shoot Total Root Shoot Total Vo (mg/pot)

Clipper 3.39 t2.37 t5.76 1.98 15.2t t7.19 109 Stirling 3.10 13.66 t6.76 2.ll 13.46 15.57 93 Weeah 3.10 11.06 14.t6 2.29 13.69 15.98 113 Schooner 3.30 13.81 L7.LI r.67 t4.26 15.93 93 Chebec 3.13 r0.77 13.90 2.07 13.68 t5.75 113 skiff 2.72 12.80 15.52 r.70 t2.47 T4.17 9l 2M Appendix 6.1. Waite Accessions and pedigrees of lines used in the N response trials at Northfield in 1991 and at Cha¡lick in 1993

I Wl-2605=Minerva 53 ¡¡¡-26! l=(G/Promise* Bx 7 2l 161 5)l L3 2 WI-1127=NoyeD 54 WI-2728=(WI-2468lhr*We€åt¡)t 3 tilI-2.060--Resibee 55 WI-2805=(WI -2339*Y¡1-2ffi) I 18 * O Y¡1-2J{=Ymer 56 WI -2 808=((Cl ip* CPD/ 14 ZEBY Tn) f25 * 5 Wl-2068=Cambrinus 57 WI-27 85 =((Clip 2295 )* (G ertÈ 223 r)) I 5 L 6 Wl-26lGNordal antlT-139 58 WI-273GSch BX 8814 l4lr G3 7 Wl-zl4l=Dampier 59 WI-2737{hebec 8 V/I-2477=(I,ll-2231*(A.Xþba* Clipper))D2 60 v/A-75Sß23 9 M-2l78=Diamant 61 v/A-75S/329 10 ltJl-?4ïÈMaz:r:ka 62 wA-755/409 l1 v¿I-26rÈcH35ßn 63 wA_79S/420 12 \ilI-355=Pi¡oline & WI-28 I 5=((Sch*206)*(G/P*239Ðn 13 Wl-?ßgÈ.{zumaGolden 65 1ry'¡-2¡ 17=((ubam*Clip(2))*Sch)/13 14 Wl-2637=Parwan 6 Iåra 15 WI-2231=(Proctor*CI-3576)8525 67 Koru 16 Wl-2632=Cytris 68 (Triumph * G t'nmett)/ l2l | | l7 W.-?-&5=2F,BYT23 69 wA-815/46ó 18 Wl-2709=Gimpel 70 v/A-8ls/465 19 Wl-575=Compana 'lI rwA-8ls/469 20 Wl-2582=(Chppr*2231)l2l '72 wA-815/470 2l \ilI-2646=R\ilCPt2R 73 wA-815/473 22 rWI-27lGVada 74 lWA_8t5/474 23 'Wl-775=C14226 75 rù,/A_8l5/467 24 WI-2377=Fuji Nrþ 76 Klages 25 Wl-2583=(Clippri2295)/5 77 Franklin 26 \Ml-2200=Clipper 78 Yagan 27 W-357=Proctor 28 Wl-35ÈMaythorpe 29 Wl-22l4=Weeatr 30 Wl-ZóþSchooner 31 Wl-2580{alleon 32 Wl-2599=Shannon 33 WI-768=CI-3576 34 Wl-2l9l=Betzes 35 IVI-260GForrest 36 \ryI-2633=Bandulla 37 ÌWI-958=CPI-18197 38 Vr'I-Zf=Indian Dwarf 39 Wl-2ó35{rimmeu 40 Wl-%59=Goldenhomise 4L WI-1195=Prior A 42 rtr/I-2539=MC 90 43 \ilI-Zí38=Stirling 44 \VI-2137=Ketch 45 Wl-2577=Triumph 46 WI-2ó76=O'Connor 47 W-216?=7*phyr 48 \ilI-2584=Skiff 49 Mmndyne 50 WI-2585=(øæphyËKetch)*rM-2335)153 51 Wl-2ó39--Kaniere 52 WI-2597=((Clipper*CPI-18I97)*WI'2476)l& Appendix Table 6.2a. Simple linear co¡relations between some parameters of barley grown at 0 kgN/ha at Northfield l99l ni78, r (5vo) = fl .223; r (lvo) = fl .290

Grain yield 1.00

GPCA -0.62** 1.00

GNYA 0.97++ -0.43** l.00

Date of flowering -0.5J++ 0.57r* -0.49*+ 1.00

J0.6 l'< + Tiller number 0.44++ -0.39+* 0.40** I 1.00

Plant height 0.35*x -0.17 0.36** -0.56*x 0.64** 1.00

Temp. differential -0.14 0.18 -0.11 -0.57*# -0.35r* -0.32+* 1.00

Flag leaf length -0.2r 0.28* -0.17 -0.13 -0.04 -0.01 0.1 r 1.00 B toxicity 0.05 -0.22 -0.02 -0.38*x 0.31** 0.22 -0.07 0.r2 1.00 ** Chlorosis -0.l3 -0.18 -0.20 -0.3 I 0.30** -0.08 -0.10 0.07 0.55** 1.00

General growing 0.09 0.2r 0.08 0.1 I 0. r3 0.17 0.08 -0.t2 -0.10 -0.14 1.00

Grain GPCA GNYA Date of Tiller PIant Temperature Flag leaf B toxicity Chlorosis General yield flowering number height differential length growrng aGPC = Grain Protein concentration, GNY = Grain N Yield

l.) O Ur at 50 kgN/ha at Northfield l99l between some palameters of barley grown Apoendix Table 6.2b . Simple linear correl arions ;:iã:; (a'i"i = mtzzt: r (t%) = +f.'zeo

Grain yield 1.00 1.00 GPCA -0.50x* -0.30** 1.00 GNYA 0.97++ 0.42+* -0.42** 1.00 Date of flowering -0.42*+ 0.03 -0.33** 1.00 Tiller number 0.03 -0.r2 0.26* -0.72** 0.38** 1.00 Plant height 0.26* -o.2lt -0.42*+ 0.37** -0.14 -0.31** 1.00 Temp. differential -0.34** 0.08 1.00 -0.11 -0.02 0.24* 0.05 0.04 leaf length -0.11 0.02 Flag -0.00 1.00 0.r2 -0.41** 0.15 0.22 -0.19 B toxicitY o.r2 -0.16 -0.r7 0.22 0.39+* 1.00 -0.08 -0.07 -0.51** 0.24* 0.14 Chlorosis -0.07 -0.03 -0.04 0.l6 1.00 0.11 -0.21* 0.16 0.24* -0.14 General growing 0.11 -0.23* Chlorosis General Tiller PIant Temperature Flag leaf B toxicity Grain GNYA Date of grow¡ng GPCA height differential length yield flowering number

GNY Grain N Yield acpc = Grain Protein concentration, =

Oò'J o\ Appendix Table 6.3a. Simple linear_c_onelations between some of barley grown at 0 kgN/tra parameters at Charlick 1993 n=78, r (5vo) = fl .223; r (r7o) = fl290

Grain yield 1.00

GPCA -0.45** 1.00

GNYA 0.62*" 0.42t* 1.00

Date of flowering 0.07 -0.57*x -0.41*x 1.00 DMa 0. r5 0.32** 0.42** -0.56*r r.00

SNC 0.10 -0.06 0.40** -0.06 -0.05 1.00 STN 0.16 0.27* 0.08 -0.52** 0.81** 0.54** 1.00

Plant height -0.26t 0.05 -0.23* 0.16 0.12 -0.13 -0.00 1.00

General growing -0.02 -0.34rr -0.30** 0.41** -0.01 -0.23* -0.t] 0.35** 1.00

Flag leaf length -0.08 -0.22 -0.26* 0.23* -0.03 -0.t2 -0.12 0.29** 0.35** 1.00

Grain GPCA GNYA Date of DMaa SNCa STNa Plant General Fìag leaf yield flowering height growrng length

aGPC=GrainProteinconcentration,GNY=GrainNYield,DMa=Drymatteratanthesis,SNC=ShootNatanthesis,STN=ShoottotalN

N) {O berween some orbarlev srown at 45 kgN/tra parameters at charlick 1ee3 fJtr:l1ä; îo]_"Å;.trBr,rJtì#; !'fff îî*lations

Grain yield r.00

GPCA -0.55** 1.00

GNYA 0.73** 0.16 r.00 1.00 Date of flowering 0.15 -0.53** -o.25* DMa -0.05 0.16 0.06 -0.38** 1.00 0.01 1.00 SNC 0.20 -0.21 0.07 -0.38** 0.65*+ 1.00 STN 0.09 0.01 0.10 -0.35*+ 0.81** 0.15 -0.11 0.05 1.00 Plant height -0.42** 0.09 -0.43x* 0.16 -0.04 -0.08 -0.07 0.49** 1.00 General growing 0.03 -0.35** -0.26* 0.44** -0.10 -0.26* -o.23* 0.32** 0.47** 1.00 Flag leaf length 0.05 -o.2'7* -o.27* 0.39**

General Flag leaf Grain GPCA GNYA Date of DMaa SNCA STNA Plant yield flowering height growlng length

aGpC=Grainproteinconcentration,GNY=GrainNYield,DMa=Drymatteratanthesis,SNC=ShootNatanthesis,STN=ShoottotalN

l.J æo 2t9 Appendix. List of papers published from part of this thesis.

Fathi, G., Lance, R.C.M. and McDonald, G.K. (1992). Effect of nitrogen fertiliser on the yield and grain nitrogen concentration of malting barley. Proc. 6th Aust. Agron. Conf.,

Armidale, N.S.V/. p. 579.

Fathi, G., McDonald, G.K. and Lance, R.C.M. (1993). Cultivar differences in res¡ronses to N fertiliser in malting barley. Proc. 7th Aust. Agron. Conf., Adelaide, p. 360.

Fathi, G., McDonald, G.K. and Lance, R.C.M. (1993). Differences in nitrate uptake and assimilation among barley cultivars. Proc. 6th Aust. Barley Tech. Symposium. Launceston,

Tasmania. 7th- 19th September pp. 94-97 .

Fathi, G., McDonald, G.K. and Lance, R.C.M. (1993). The grain yield and protein responses to nitrogen fertiliser in six cultivars of barley. Proc. 44th Aust. Cereal Chem.

Conf., Ballarat, Victoria, (In press). 2to

REFERENCES

ABARE, (1990). Commodity Statistical Bulletin. Australian Bureau of Agricultural Economic and Resource Economics, Canberra.

Agboola, A.^. (1972). The relationship between the yields of. eight va¡ieties of Nigerian - muze and content of nitrogen, phosphorous and potassium in the leaf at flowering stage. J. Agric. Sci.79: 39L'396. Aktan, S. (1976). Nifate-nitrogen accumulation and distribution in the soil pr-ofile {*ing a fà[ow grain rotationãs influenced by different levels of soil profile moisture. M.S. Thesis, Oregon State University, Corvallis.

Alessi,. J., Power, J.F. and Sibbit, L.P. (1979). Yield, quality and nitrogen fertiliser recovery of standard and semi-dwarf spring wheat ás affected by sowing data and fertiliser rate. J. Agttc. Sci. Camb. 93:87-93.

Alston, A.M. and Lungley, D.R. (1975). Nirogen time of application and depth of placement. pp. 81-93. CSIRO Australian Land Reso No.1.

Anderson, E.L., Kamprath, E.J. and Moll, R.H. (1984). _Nitrogen-fertility effects on accumuÍation,'remobilisation and partitioning of N and dry matter in corn genotypes differing in prolificacy. Agron. 1.76:397-4M.

Anderson, W.K. (1935). Differences in response of winter cereal va¡ieties to applied nitrogen in the field. II. Some factols associated with differences in response. Field Crops Res. 11: 369-385.

Anderson, W.K. and Smith, V/.R. (1990). Yield advantage of two semi-dwarf_compared with two tall wheatsdepends on sowing time. Ausl. J. Agric. Res. 41: 8LL-26. 'W.K., (1991). for differences Anderson, Seymour, M. and D'Antuono, M.F. -Evidence between cultíva¡s in responsiveness of wheat to applied nitrogen. Aust. J. Agric. Res. 42: 363-377. Arends, Macleod, L., Lance, R., V/allace-r'W., Evans, E', , Gilmour, R., Tatr, A., Weining, S. andHenry, R. astatic po\¡/er of Australian barlys. Tenth Aust. Plant

Arnon, I. (1975). Physiolo gical principles of dryland crop production.- In: Physiological Aspects of Dryland Éatr"itg. (Îld: U.S. öupta), pp-. 3-t+s. (Oxford and IBH, New Delhi)

Arnon,I.' (1979). Optimising yields and,water use in Mediterranean_a.g.nc{ty¡e in soil. In: Meditenaneãn type-ilimates and their yield potential. 14th Coll. Int. Poøsh lnstitute, Sevilla. Spain. Bern, pp.311-348.

Aschmann, H. (1973). Di n ec_osystems. In: Mediterranean type .di Castri and H.A. Moony), pp. 11-19. 2ll Aspinall,- D., Nicholls, P.B. and May, C.H. (1964). The effect of soil moisture stress on the $owth of barley. I. Vegetative development and grain yield. Aust. J. Agric. Res. 15:729-45

Atherton, M.J. (1989). Quality requirements: Malting barley. In: Cereal Production fd: E.M. Gallagher). Próceedings of the Second International Summer School in Agriculture, pp. 120- 126. (Butterworths I-ondon).

Atkins, R.8., Stanford, G. and Dumenil, L. (1955). Effects of nitrogen and phosphorus fertilisers on yield and malting quality of barley. Agric. Food Chem. 3: 609-614.

Austin, R.B. (1989). Maximising crop production in water-limited environments. In: Drought Resistance in cereats. lEd: Baker.), pp. 13-25. (CAB. International IGSU, U.K.).

Austin, R.B. and Jones, H.C. (1975). The physiology of wheat. In: Annual Report of the Plant Breeding Institute, Cambridge 1974, pp. 30-73.

Austin, R.B., Bingham, J., Blackwell, R.D., Evans, L.T., Ford, M.4., Morgan, C.L and Taylor, M: (1980). Genetic improvments in winter ryhgt! yields since 1900 and ass-ociated physiological changes. J. Agric. Sci. 94: 675-689.

Austin, R.8., Ford, M.4., Edrich, J.A. and Blackwell, R.D. (1977). The nitrogen economy of winter wheat. J. Agric. Sci. 88: 159-t67.

Baldridge, D.8., Brann, D.E., Ferguson, 4.H., Henry, J.I. and Thompson, R.K (1985). Cultural practices. In: Barléy (Ed: D.C. Rasmusson),p.457-482. Agron. Monogr. 26. (Amèrican Society of Agrronomy, Madison, rü/isc.).

Barham, R.W., and Rasmusson. (1981). Inheritance of photoperiod response in barley. Crop Sci. 2l:454-456.

Barley, K.P. and Naidu, N.A. (1964). The performance of three Australian wheat va¡ieties at high levels of nitrogen supply. Aust. J. Exp. Agric. Anim. Husb. 4:39-48.

Bauer, A. Garica, R., Kanemasu, E.T., Blad, 8.L., Hatfield, J.L., Major, D.I1'-Rggginato, R.J., and.Hubbard, K.G. (1988). Effect of latitude on phenology of 'Colt' winter wheat. Agric.Forest Meteor. 44 I3l-140.

Beauchamp, E.G., Kannenberg, L.W. and Hunter, R.B, (L976). Nitrogen accumulation and uanslocation in coñr genotypes following silking. Agron. 1.68 4L8-422. Bechtel, D.B., Gains, R.L. and Pomeranz, Y. (1982). Protein secretion in wheat endosperm formation of the matrix protein. Cereal Chem. 59 336-43.

Bell, F.C. (1979). Perception. In: Arid-land Ecosystems. (Eds: D.W. Goodall and R.A. Perry). I. Cambriãge University Press. (Cambridee UK).

wheat in a subhumi$ Benbi,' DK. (1990). Efficiency of nitrogen use by dryland 5egio-n !n relati'on to optimising itre amouni of availãble water. J. Agric. Sci.(Camb.). 115: 7- 10.

Bendelow, V.M. (1964). Inheritance of free beta-amylase in barley. Can. J. Plant. Sci. 44: 550-554.

Benzian, 8., Darby, R.J., Lane, P., Widdowson, F.V. and Verstraeten, L.M.J. (1983). Relaiionshíp between N concentration of grain and,grain yielÉ-in recent winter wheat e*periments in England and Belgium, some wiih large yields. J. Sci. Food Agric.34: 212

Benzian, B.M. and Lanc, P. (1979). Some relationships between yield and grain protein of wheat experiments in south-east England and comparisons with such relationships elsewhere. J. Sci. Food Agric. 30: 59-70.

Bhatia, C.R. and Rabson, R. (1976). Bioenergetic considerations in cereal breeding for protein improvement. Science 194: 1.418-1421.

Bhullar, S.S. and Jenner, C.F. (1985). Differential responses to high temperatures of starch and nitrogen accumulation in the grain of four cultivars of wheat. Aust. J. Plant Physiol. L2: 363-7 5.

Bhullar, S.S. and Jenner, C.F. (1986). Effects of temperature on the conversion of sucrose to strach in the developing wheat endosperm. Aust. J. Plant Physiol. 13: 605-15.

Bidinger, F., Musgrave, R.B. and Fischer, R.A. (L977). Contribution of stored preanthesis assimilate to grain yield in wheat and barley, Nature 270:43L-433.

Birch, C.J. and Long, K.E. (1990). Effect of nitrogen on the growth, yield and grain protein content of barley (Hordeum vulgare). Aust. J. Exp. Agric .30:237-242.

Birch, C.J., Fukai, S. and Broad, I.J. (1993). The effect of nitrogen on the yield and selected aspects of quality of two varieties of malting barley grown under contrasting õonditions-of wâter availability. In Proceedings of the 7th Australian Agronomy

Bishop, J.R. (1930a). The nifogen content and quality of barley. J. Inst. Brew. 36:352- 363.

Bloom, A.J. and Chapin, F.S. (1981). Differences in steady-state net ammonium and nitrate influx by cold- and warm-adapted barley varieties. Plant Physiol. 68: 106/.- to67. Blum, A. and Dnuel, Y. (1990). Physiological attributes associated with drought resistance of wheat cultivars in a Mediterranean environment. Aust. J. Agric. Res. 14:799-810.

Blum, 4., Poia¡kova, H., Golan, G. and May :r, I. (1983a). Chemical desiccation of wheat plants as a simulator of post-anthesis stress. I. Effects on translocation and kernel growth. Field Crops Res. 6: 51-58.

Bolton, F.E. (19S1). water and nitrogen through soil and crop management. gen in Medilenagean-type E-nvironments. (Eds:1. Mon 231-247. (Martinus NijhofflDr W Junk Publishers: The Hag

Bowerman, P. and Haris, P.B. (1974). The rate and time of application of nitrogen on continuous spring barley. Exp. Husb . 27 : 45-49.

Bray, C.M. (1983). Nitrogen Metabolism in Plants. (Longman, London).

Bremner, J.M. (1965). Total nitrogen. In: Methods of Soil Analysis (Ed: C.A.Black), pp. 1149-1178. (Amer. Soc. Agron. Madison, rWisc.).

Broadbent, F.E., DeDatta, S.K. and Laureles, E.V. (1987). Measurement of nitrogen utilisation efficiency in rice genotypes. Agron. J.79:786-791.

Brooks, A., Jenner, C.F. and Aspinal, D. (l )82). Effects of water deficit on endolPerm starch granules and on þain physiology of wheat and barley. Aust. J. Plant. Physiol. 9:423-36. 213 Brown, P.L. (1971). Water use and soil water depletion by dryland winter wheat as affected by nitrogen fertilisation. Agron. 1.63: 43-46.

Brown, S.C., Keatinge, J.D.H., Gregory, P.J. and Cooper, P.J.M. (1987). Effects of fertiliser, varièty and location on barley production under rainfed conditions in Northem Syna. l. Root and shoot $owth. Field Crops Res.16: 53-66. Bruckner, P.L: and Morey, D.D. (1988). Nitrogen effects on soft red winter wheat yield, agronomic characteristics and quality. Crop Sci. 28: L52-157.

Bruetsch, T.F. and Estes, G.O. (1976). Genotype variation in nutrient uptake efficiency in corn. Agron. J.68: 521-523.

Buddenhagen, I.W. (1990). Legumes in rmer systems in Mediterranean climates. In: the systems of the Mediterranean areas. (g4i Thé role of legumes-Ibrabim in farming 4.n. Osman M.H. and M.A. Jônes), pp. 3-29. (Kluwer Academic Publishers Netherlands).

Burger, W.C. and LaBerge, D.E. (1985). Matting and Brewing Quality_., In: Barley (Ed: - D. C. Rasmusson), pp. 367-398. (Amer. Soc. Agron. Madisom, Wisc.).

Cacco, G., Saccomani, M. and Ferrari, G. (1983). Changes in the uptake and assimilation efficiency for sulfate and nitrate in maize hybrids selected during the period 1930 to 1975. Physiol. Plant. 58: L7L-L74.

Calder, F.W. and Macleod, L.B. (L974). Effects of soil pH and NPK fenilisation on yield and quality of two barley cultivars. Can. J. Soil. Sci. 54: 1-6.

Camberato, J.J. and Bock, B.R. (1990a). Spring wheat responses to enhanced ammonium supply. I. Dry matter and nitrogen Content. Agron. I. 82: 463-467.

Camberato, J.J. and Bock, B.R. (1990b). Spring wheat responses to enhanced ammonium supply. tr. Tillering. Agron. J. 82: 467-473. Campbell,^ C.4., Davidson F.G. (1977). Effects of fertiliser N and soil moisture on yield, protein coltent and N accumulation on the aboveground-parts an. J. Soil. Sci. 57: 3ll-327.

Cannell, R.Q. (1969a). The tillering pattern in barley varieties. I. Production, survival and contribution to yield by component tillers. J. Agric. Sci. 72: 405-422. Cannell, R.Q. (1969b). The tillering pattern in barley varieties. -II. The effect of the remperature, light intensity anä daylength on the ire_quengy .o_logggrrence of colebptile no¿iand second tillers iri Uartéy. J. Agric. Sci. 72: 423-435.

Cawse, P.A. (1967). The determination of nitrate in soil solutions by ultraviolet spectrophotometry. Analyst 92: 3L1-3 15. Ceccarelli, S., Acevedo, E. and Grando, S. (1991). Breeding for field stability in uttpredictable environments: Single trai¡s, _interactioñ between traits, and architecture of genotypes. Euphytica 56: 169-185.

Chandler, R.F. (1970). Overcoming physiological barriers to higher yields through plant breeding. Potassium Symp. 9: 42L-434.

Chapin, F.S. and Bieleski, R.L. (1982). Mild phosphor-us stress in.barley qttd a related ' low-phosphorus-adapted barley-grassl Ph-osplorus fractions and phosphate absofotiori in relationìo growth. Physiol. Plant. 54:309-317.

Chevalier, P. and Sch¡ader, L.E. (1977). Genotypic differences in nitrate absorption and partirioning of N among plant parts in maiáe. crop. Sci. 17: 897-901. 214 Cla¡k, R.B. (1982). Plant genotype di ation, accumulation and use of mineral elementi.- In: eral Nutrition of Plants (Ed. M. R. Saric), pp.4l-55 of Sciences and Arts, Belgrade).

Clark, R.B. (1983). Plant genotype differences in the uptake, translocation. accumulation, and use of mineralélemèñts required for plant growth. Plant Soil 72: L75-L96.

Clark, R.B. (1990). Physiology of cereals for mineral nutrient uptake, gry,_m! efficiency. In: Crops as Enhanceriof Nutrient Use. @ds: V.C. Baligar and R.R. Duncan), pp. 131-183. (Academic Press, San Diego, California).

Clark, R.B. and Brown, J.C. (1980). Role of as related to breeding and genetics. In: Moving up nd Obstacles (Eds: L.S. Murphy, E. Doll and L.F. Soc. Agron. Madison, Wisc

Cooke, G.W. (1932). Fertilising for Maximum Yield. (Granada, London).

Cooper, P.J.M. (1983). Crop management in rainfed agriculure with special reference to water use efficienry. In: Proc. 17th Coll. lnt. Potash Institute, Rabat and Ma¡rakech, Morocco, 1983, I.P.I. Bem, pp,63-79.

Cornish, E.A. (1950). The infl.uence of rainfall on the yield of wheat in South Australia. Aust. J. Sci. Res. (Series B) 3: 178-218.

Cox,. M.C., Qualset, C.O. and Rains, D.W. (1935). Genetic variation for nitrogen assimiìation and translocation in wheat. II. Nitrogen assimilation in relation to grain yield and protein. Crop Sci. 25: 435-M0.

Craswell, E.T. and Godwin, D.C. (1984). The efficiency of nitrogen fertiliser applied to cereals in different climates. Adv. Plant. Nut. 1: 1-55.

Craswell, E.T. a Isotopic studies of the nitrogen balance in a cracking ery in Þlant and soil in relation to the depth of fertilisei t. J. Soil Res. 14: 75-83.

Craufurd, P.Q. and Cartwright, P.M. (1989). Effect of photoperiod and chlo_rmequat on apical=development ãnd growttr in a'spring wheai (Triticum aestivum L.) cultivar. Ann Bot, 63: 515-525.

Cutforth, H.rüy'., W., Clarke (1988). Growth onents and t of two high-yie DT367, co ultivars, Neepawa and'Wakooma

Dabney, S.M. and Bouldin, D.R. (1985). Fluxes of ammonia over an alfalfa field. Agron. J.77:572-578. is of nitrogen redi Dalling,-filling M.J. (19S5). in cere thysiolog-ical and Enhañce Crop HarPer, L.E. Schra pp.55

Dalling, M.J., Boland, G. and Wilson, J.H. (197.6)., Relation between acid proteinase "ãctivity and reáistribution of N during E¡ain development in wheat. Aust. J. Plant Physiol. 3:721-730.

M.J., Halloran, G.M. and Wilson, J.H._ (1975). The relation between nitrate Dalling,-reductase activity and grain nirogen productivity in wheat. Aust. J. Agric. Res. 26: l-10. 215 Darwinkle, A. (1983). Ea¡ formation and grain yield of winter wheats as affected by time of nitrogen supply. Neth. J. Agric. Sci. 3l: 2Ll-225.

Davidson, J.L., Christian, K.R., Jones, D.8., and Bremner, P.M. (1985). Responses of wheat to vernalisation and photoperiod. Aust. J. Agric. Res. 36: 347-59.

Dawson, I.A. and'Wardlaw, I.F. (1939). The tolerance of wheat to high temperature during reproductive growth, trI. Booting to anthesis. Aust. J.Agric. Res. 40: 965-980.

De Datta, S.K., Obeemea, W.N. and Jana, R.K. (1972). Protein content of rice grain as affected by nitrogen fertiliser and some triazines and substituted ureas. Agron. J. 64:785-788.

Decka¡d, E.L., Tsai, C.Y. and Tucker, T.C. (1986) Nutrition on Quality of Agronomic Crops. In: Nitrogen in Crop .D. Hauck.)' pp.60l- 615.-Amer. Soc. Agron., Crop Sci. Soc. A Amer.

DeDatta, S.K. and Broadbent, F.E. (1990). Nirogen-use efficiency of 24 nce genotypes on an N-deficient soil. Field Crops Res. 23:81-92.

Desai, R.M. and Bhatia, C.R. (1976). Nirogen uptake and nitrogen harvest index in durum wheat cultiva¡s varying in thei¡ þain protein concentration. Euphytica 27: 561-566.

Doddema, H. and Telkamp, G.P. (1979). Uptake of nitrate by mutants of Arabidopsis thaliana, disturbed in uptake or reduc[ion of nitrate. II. Kinetics. Physiol. Plant. 45:332-338.

Donald, C.M. (1968). The breeding of crop ideotypes. Euphytica. L7:385-403.

Donovan, G.R. and Lee, J.W. and Hill, R.D. (1977). Composition changes in the developing grain of high and low protein wheats. I. Chemical composition. Cereal Chem. 54: 638-645.

Doran, J.W. (1980). Soil microbial changes associated with reduced tillage. Soil Sci. Soc. Amer. 1.44:765-771.

Dougherty, C.J., Love, B.G. and Mountier, N.S. (1979). Response surfaces of Kopara wheat for seeding rate and levels and times of applicaiion of nitrogen fertiliser. N.Z. J. Agric. Res.22: 47-54.

Doyle, A.D. and Kingston, R.W. (1992). Effect of sorring rate on grain yield, kernel ' weight, and giain protein percentage of barley (Hordeum vulgar L.) in northern New South Wales. Aust. J. Exp. Agric.32:465-471. Doyle,' A.D. and Shapland, R.A. (1991). Effect of split nitrogen agnllgationson the yield and protein cõnteni of @land wheat in northern New South Wales. Aust. J. Exp. Agric. 3L:85-92. Doyle,A.D., .8. (19?-3). d quali South V/al Y Tech Tasmania, DuBois, J.B. and Fossati, A. (1981). In luence of nitrogen uptake and partitioning êfficiency on grain yiètA aiA grain protein conõentration of twelve wheat genotype! (Tritiõum aéstivum L.),2. Pflãnzenzucht. 86: 4t-49. Dyson,- P.W. (1977). An investigtion into the relations between some growth parameters and yield of barley. Ann. Appl. Biol. 87:471-483. 2t6 Easson, D.L. (1984). The timing of nitrogen application for spring barley. J. Agric. Sci. 1O2:673-678.

Eck, HV. (1983). V/inter wheat response to nitrogen and irrigation. Agron. J. 80: 902-8.

Ellen, J. and Spiertz, J.H.J. (1975). The influence of nitrogen and Benlate on leaf area duratioñ, grain growth and pattern of N-, P- and K-_ uptake of winter wheat (Triticwn aestivwt).2. Ãcker Pflanzenbau 141: 231-239. grain Ellen, J. and Spiertz, J.H.J. (1980). Effect of rate and timing of nirogen dre ssing _o^n yield fõrmation of winter wheat (Triticwn aestivum L.). Fert. Res. 1: 177-190.

Ellington, A. and Reeves, T.G. (1990). Regulation of soil_nitrogen t_e1.qt^"_f9r. wheat, by - direct drilling following clover pasture. Soil and Tillage Res. 17: L25-142.

Elliot, D.E., Growden, B. and Reuter, D.J. (1985). prediction and co¡rection. Termination rePort Research Committee of South Australia. (S.4. Ellis,' R.L., Collinson, S.T., Hudson, D. and Patefield, W.M. (L992). The analysis-of reciprocal transfer experiments to estimate the duration of photoperiod-sensitive and photoperiod-insensitive phases of plant development: an example in soyabean. Ann. Bot. 65:87-92.

Engel, R.E. (1991). Simulated growing season precipitation and nitrogen effects on winter wheat. Agron. J. 83: 180-185.

Epstein, E. (L972). Mineral Nutrition of Plants: Principles and Perspective. (John Wiley and Sons, New York),pp.325-3$.

Evans, L.T. (ed) (1975). Crop Physiology: Some Case Histories. (Cambridge University Press, London).

Evans, L.T. and Wardlaw, I.F. (1976). Aspects of comparative physiology of grain yield in cereals. Adv. Agron. 28: 301-359. Evans,' L.T., Wardlaw, L.F. and Fisher, R.A. (1975). Wheat. In: Crop Physio-logy: some casehistories. @d: L.T. Evans), pp. 101-149. (Cambridge Univ. Press, London).

Fairey, D.T.L., Hunt, A.and Stoskopf, N.C. (1975). !ay-!e1gth 1{l9e-1c¡^on-reproductive '' development and tillering in'Ferglls' barley. Can. J. Bot. 53: L770-2775-.

Farquhar, G.8., Firth, P.M., Wetselaar, R. and Weif, B. (1930). On the gaseous exchange ^ oî ammonia between leaves and the environment. Determination of the ammonia compensation point. Plant. Physiol. 6:710-714.

Fischer,.physioiogicil R.A. (1979). Growth and water limitations to dryla¡rd^yle.at yield in Australia, a framework. J. Aust. Inst. Agnc. Sci. 45: 83-94.

Fischer, R.A. (1981). Optimising the use of water and nitrogen through the breeding of crops. Plant and Soil. 58: 249-278.

Fischer, R.A. (1984). Wheat. In: 'symposium on Potential_Productivity of-fleld_props ïnder piffetént Environment-s'. (Ed: \V.H. Smith and S.J. Banta). pp. 129-L53.

Fischer, R.A. (1985). Number of kernels in wheat crops and the influence of sola¡ rediation and temperature. J. Agnc. Sci. 105: 447-461.

Fischer, R.A. and'Wall, P.C. (1976). Wheat breeding in Mexico and yield increases. J. Aust. Inst. Agnc. Sci. 42: 139-148. 217 Fjell,- D.L., Paulsen, G. M., Walter, T.L. and Lawless, J. R. (1984). Relationship among- nitrogen and phosphorous contents of vegetable parts and agronomic traits of normãl-and high-prbtein wheats. J. Plant Nutr. 7: 1093-1102.

Fletcher, G.M. and Dale, J.H. (1974). Growth of tiller buds in barley: effects of shade treatment and mineral nutrition. Ann. Bot. 38:63-76.

Flood, R.G. and Halloran, G.M. (1984a). Basic development rate in spring wheat. Agron. J.76:260-264.

Food and Agriculture Organisation of the United Nations (FAO) (1981). FAO Fertiliser Programme: 20 yðars increasing crop yields 1961-1981. (FAO, Rome, Italy), 6Op.

Franca, G.E. (1981). Differences in dry-matter yield and the uptake, distribution, and use of nitrogen by sorghum genotyies. Ph.D. Thesis, university of Nebraska, Lincoln. @iss. Abstr. 41, 4018b).

Frank, A.B. and Bauer, A. (1984). Cultivar, N and soil water effects on apex development in spring wheat. Agron. J.76:567-660.

Frank, 4.B., Bauer, A. and Black, A.L. (1987). Effects of air temperatue and water stress on apex development in spring wheat. Crop Sci. 27: ll3lL5.

French, R.J. and Schultz, J.E. (1934). W'ater use efficiency of wheat production in a Mediterranean-type environment. I. The relationship between yield, water use and climate. Aust. J. Agnc.Res. 35: 743-764.

Gale, M.D. and Youssefian, S. (1985). Dwarfing genes in wheat. In: Progress in plant breeding. (Eds: G.E. Russell), p. 1-35. (Butterworths, London). Gallagher,- J.N., Biscoe, P.V. and Scott, R.K. (1975). _larley and its environment. V. Stability of grain weight. J. Appl. Ecol. 12: 319'36.

Gallagher, J.N.,and Biscoe, P.V. (1978). Radiation absoption, growth and yield of cereals. J.Agn.Sci.91: 47-60.

Gardener, CJ. (1972). The response of barley genotypes to agronomic-manipulation in a lrieditenaneán type of climate. Ph.b. Theiiì, Univ-ersity of Adelaide, South Australia.

Gardener, C.J. and Rathjen, A.J. (1975). The differential response of barley genoly,pe^s_ !o trít ogen applicaiion in a Mediterranean-type climate. Aust. J. Agric. Res. 26: 219- 230.

Garwood, E.A. and Williams, T.E. (1967). Growth, water use and nutrient uptake from the subsoil by grass swards. J. Agric. Sci. 69: 125-L30.

Gasser, J.K.R. and Iordanou, I.G. (1967). Effects of ammonium sulphate and calcium nitrate on the growth, yield and nitrogen uptake by barley, wheat and oats. J. Agric. Sci. 68: 307-316.

sqage of applicatiotlgl and Gately,'-quality F.T.F. (1971). Effect of form of nitrogen an4 lle.yield of málting barley (cultivarProctor). Ir. J. Agric. Res. 10: 173-184.

Genstat 5 Committe (1987). Genstat 5. Reference Manual Clarendon Press Oxford. p. 749.

Gentilli, J. (Editor). (1971). Climates of Australia and New Zealand. World Survey of Climatology. Vol. 1 3. Elsevier, Amsterdam, 405p' 218 George, B.J. and Skinner, R.J. (1984). Determination of the N requirement for winter wheat using an experiment of simple design. In: Nitrogen requirements of cereals. References Book, 385, Ministry of Agriculture, Fisheries and food, HMSO, pp.183

Gerloff, G.C. (1976). Plant efficiencies in the use of nitrogen, phosphorus and potassium. In: Proceedings of Workshop on plant Adaptation to Mineral stress in Problem Soils (Ed: M.J. V/right), pp. 16l-173. National Agricultural Library, Beltswille,

Gibbon, D. (1981). Rained farming systems in the Mediterranean region. In: Soil Vy'ater and Nitrogen in Mediterranean-type Environments. (Eds: J. Monteith and C. Webb), pp. 59-80. (The Hague: Ma¡tinus Nijhoff/Dr W Junk).

Glass, A.D.M. and Perely, J.E. (1980). Varietal differences in potassium uptake by barley. Plant. Physiol. 65: 160-164.

Glass, A.D.M., Siddiqui, M.Y. and Giles, K.I. (1981). Correlation between potassium uptake and hydrogen efflux in barley varieties. A potential screening method for the isolation of nutrient efficient lines. Plant Physiol. 68: 457 -459.

Gonzalez Ponce, J., Salas, R.M.L. and Mason, S.C. (1993). Nitrogen use efficiency by winter barley under different climatic conditions. J. Plant Nutri. 16: 1249-L261.

Goodman, I.J. (1977). Selection for nitrogen responses in Lolium. Ann. Bot. 4l: 243-256.

Greenland, D.J. (1971). Changes in the ni rogen status and physical condition of soils under pastures, with special referencè to the maintenance of the fertility of Australian soils used for growing wheat. Soils and Fert. 34: 237-51.

Gregory, P.J. Marshall, B. and Biscoe, P.V. (1981). Nutrient relations of winter wheat. 3. Nirogen uptake, photosynthesis of flag leaves and translocation of nitrogen to grain. J. Agric. Sci. (Cambr.). 96: 539-547.

Gregory,-growth P.J., Shepherd, K.D. and Cooper, P.J.M. (1984). Effects of fertiliser on root - and water use of barley in northern Syria. J. Agric. Sci. (Cambr.). 103: 429-438.

Grieve, C.M., Lesch, S.M., Francois, L.E. snd Mass, E.V. (1992). Analysis of main-spike yield components in salt-stressed wheat. Crop Sci. 32:679'7O3.

Gupta, U.C. (1979). Boron nutrtion of crops. Adv. Agron. 31: 273'307.

Hackert, C. (1968). A study of the root system of barley. I. Effects of nutrition on two varieties. New Phytol. 67 : 287 -3O0. Hagem .(tTf)'oïLiil: 31. Natl. Acad. Sci. Washin

Halloran, G.M. (1981). Grain yield and protein relationships in a wheat cross. Crop Sci. 2l 699-701. Halloran, G.M. and Lee, J. W. (1979). Plant nitrogen distribution in wheat cultiva¡s. Aust. J. Agnc.Res. 30: 779-789.

Halloran, G.M. and Pennel, A.L. (1982). Duration and rate of developmental phases in wheat in two environemnts. Ann. Bot. 49: ll5-12L.

Halse, N.J. and Weir, R.N. (1970). Effects of vernalisation, photoperiod, and temperature on phenological development and spikelet unmber õf Australian wheat. Aust. J. Agric.Res.21: 383-93. 2r9 Hamblin, 4., Tennant, D. and. Perry, M.W. (1987). Management of soil water fo¡ wheat production in Western Australia. Soil Use Manag. 3:63-69.

Harmsen, K. (1984). Nitrogen fertiliser use in rainfed agriculture. Fert. Res. 5: 371-382.

Harmsen, K., Shepherd, K.D. and Allan, A.Y. (1983). Crop response to nitro_gen_ a-nd phosphoruì in rainfed agriculture. In: Nutrient Balances and the Need for Fertilisers in Semi-arid and Arid Regions. IPI, Bern.

Harper, L.4., Sharpe, R.R., Langdale, G.W. and Giddens, J.E. (1987).-¡lgogçl cycling in a wheat cróp. Soil, plant and aerial nitrogen transport. Agron. 1.79:965-973.

Harper, V.R., Catchpoole, R., Davis, and Weier, K.L. (1983). Ammonia volatilisation: ' Soil, plant, aird microclimate effects on diurnal and seasional functions. Agron. J. 75:212-218.

Hay, R.E., Early, E.B. and DeTurk, E.E. (1953). Concentration and translocation of nirogen compounds in the corn plant (Zea mays) during grain development. Plant Physiol. 28:606-621. Hay,- R.K.M. and Walker, A.J. (1989). An Inuoduction to the Physiology of Crop Yield. Chapter 6, Temperate Cereals. pp.159-187. (Longman Scientific and Technical, London.)

Hay, R.K.M., and Kirby, E.J.M. (1991). Convergelce and synchr.ony -_919v]ew of the co- ordination of dévelopment in wheat. Aust. J. Agric. Res. 42: 661-700.

Haynes, R.J. (1986). Uptake and assimilation of mineral nitrogen by-Plqls: In: Mineral ' Nitrogen in the-Plant-Soil System. (Ed: R.J. Haynes), pp. 303-378. (Acadmic Press Inc.l,ondon).

Haynes, R.J. and Goh, K.M. (197S). Ammonium and nitrate nutrition of plants. Biol Rev. 53:465-510. Hegeman, R.H. (1979). Integration of nitrogen a_ssimilation in relation to yiglq. -ll: Nítrogen àssimilation õf Plants. (Eds: E.J. Hewitt and C.V. Cuting), pp. 591-612. (Academic Press, London).

Henry, R.J. (1989). Factors influencing the rate of modification of barley during malting. J. Cereal Sci. 10: 5l-59. (1-3),(t-9-P-Fl"cqlase isoenzymes by high Henry,' . R.J. (1990). Quantitative analysis of performance liquid chromatogiaphy. J. Cereal Sci. 12: L87-192.

Herdina, and Silsbury, J.H. (1989). of faba bean (Vicia faba) *d pru (pisúm satiium i.) bium, nitrate supply, and growth temperatufe. Aust. J

Hibberd, D.E. (1978). Sorghum nutrition and soils. In: Proceedings Grain Sorghum 'Workshop, Warwick, 197 8. pp.1 4-94.

Hoener, I.R. and DeTurk, E.E. (1938). The absorption and utilisation of nitrate nitrogen d"ri"g vegetative'growth by Iilinois high pirctein and Illinois low protein corn- J. Soc. Agron . 30: 232-243. Sheet. Holden,.Department K. and Jefferies, S.P. (1988). Growing bettelmalting barley. Fact of Agriculture, South Australia . FS. 27188

Holmes, D.P.(1983). Inflorescence development of semi-dwarf and standard height w¡9tt ôultivars in different photoperiodind nirogen treatments. Can. J. Bot. 51: 941- 956. 220 Hossain, A.B.S., Sea¡s, R.G., Cox, T.S. and Paulsen, G.M. (1990). Desiccation tolerance and its relationship to assimilate partitioning in winter wheat. Crop Sci. 3O: 622- 627.

Huffaker, R.C. and Rains, D.rù/. (1978). Factors influencing nitrate acquisition by plants; assimilation and fate of reduced nitrogen. In: Nitrogen in the Environment. (Eds: D.R. Nielson, and J.G McDonald), pp.1-43. (Academic Press, New York).

Isfan, D. (1990). Nitrogen physiological efficiency index in some selected spring barley cultiva¡s. J. Plant Nutr. 13: 907-914.

Isfan, D. (1993). Genotypic variability for physiological efficiency index of nitrogen in oats. Plant and Soil 154: 53-59.

Isfan, D., Cserni, I and Tabi, M. (1991). Genetic variation of the physiological efficiency index of nirogen in triticale. J. Plant Nutr. 14: 1381-1390.

Jackson, R.D. (1982). Canopy temperature and crop water stress. Adv. Inig. 1: 43-85.

Jackson, T.L. (1977). Fertiliser needs for developing countries. Proc. International Symposium on Rainfed Agriculture in Semi-arid Regions. pp. 541-551. (University of California, Riverside, California).

Jansson, S.L. and Persson, J. (1982). Mineralisation and immobilisation. In: Nitrogen in agdcultural soils. (Ed: F.J. Stevenson),pp:229-252. (Am. Soc. Agron., Madison, Wisc.).

Jefferies, S.P. (1990). Wheat and Barley N Response Trials. Report of South Australian Field Crop Evaluation. Department of Agriculture, South Australia. Jefferies, S.P., Summerton, B. and Wheeler, P. (1990). Wheat and Barley Variety Evaluation and Agronomy. Department of Agriculture, South Australia.

Jenner, C.F., Ugalde, T.D. and Aspinall, D. (1991). The physiology of starch and protein deposition in the endosperm of wheat. Aust. J. Plant Physiol. 18: 2lL-26.

Johnson, C.B., Whittington, W.J. and Blackwood, G.C. (1976). Nitrate reductase as a possible predictive test of crop yield. Nature 262: I33-I34.

Johnston,G.F.S. and Jeffcoat, B. (L977). Effects of some growth regulators on tiller bud elongation in cereals. New Phytol. 7 9 : 239 -245.

Jones, H.G. and Kirby, E.L.M. (1977). Effects of manipulation of number of tillers ánd lvater supply on grain yield in barley. J.Agric. Sci. 88: 391-397.

Jung, P.E., Peterson, L.A. and Schrader, L.E. (1972). Response of irrigated corn to time, rate and source of applied N on sandy soils. Agron. J. 64: 668-670.

Kanemasu, E.T., Stainer, J.L., Biere, AW., Worman, E.D. and Stone, J.F. (1983). Irrigation in Great Plains. Agric.Water Manage 7:157-I78.

Kassam, A.H. (1981). Climate, soil and land resources in North Africa and West Asia. Plant and Soil58: pp.1-29.

Keeney, D.R. (1982a). Nitrogen management for maximum effrciency and minimum pollution. In: Nitrogen in Agricultural Soils. (Ed: F.J. Stevenson),pp.7ll-733. (Am. Soc. Agron. Madison, Wisc.).

Kerntos, V.D. and Orphanos, P.I. (1979). Nitogen and phosphorus fertilisers for wheat and barley in a semi-arid region. J. Agric. Sci. Camb. 93:.7lI-7I7 . 22r Kirby, E.J.M. (1969). The effects of day length upon the development and growth of wheat, barley and oats. Field Crops Abstr. 22: l-7.

Kirby, E.J.M. (1977). The growth of the shoot apex and the apical dome of barley during ear initiation. Ann.Bot . 4l: 1297 -1 308.

Kirby, E.J.M. and Appleya¡d, M. (1984). Cereal Development Guide. 2nd ed. Arable Unit, National Agricultural Centre, Coventry.

Kirby, E.J.M. and Jones,H.G. (L977). The relation between the main shoot and tillers in barley plants J.Agric. Sci. 88: 381-389.

Kirby, E.J.M. and Riggs, T.J. (1978). Development consequences of two-row and six-row ear type in spring barley. J. Agric. Sci. 91: 207-216.

Kirby, E.J.M., Appleyard, M. (1980a). Effects of photoperiod on the relation between development and yield per plant of a range of spring barley varieties. Z. Pflanzenzucht. 85: 226-239.

Kirby,E.J.M. (1985).Significant stage of ear development in winter wheat. In: Wheat Growth and Modelling. @d: W.Day and R.K.Atkin), pp. 7-25. (NATO ASI Series. Plenum Press, New York and I-ondon.)

Kitchen, B.M. and Rasmusson, D.C. (1983). Duration and inheritance of leaf initiation, spike initiation, and spike growth in barley. Crop Sci. 23: 939-943.

Koranteng, G.O. and Matthews, S. (1982). Modif,rcations of the development of spring barley by early applications of CCC and GA3 and the subsequent effects on yield components and yield.. In: Chemical Manipulation of Crop Growth pp.343-357. @d: J.S. Mclar Kowalenko, C.G. (1978). Organic nitrogen, phosphorus and sulfur in soils. In: Soil organic matter. (Eds: M. Schnitzer and S.U. Kahn), p.95-136. (Elsevier North- Holland, Inc. New York).

Ktamer, T. (1978). Environmental and genetic variation for protein content in the winter wheat (Triticum aestivum L.). Euphytica 28: 209-218.

Lal, P., Reddy, G.G. and Modi, M.S. (1978). Accumulation and redistribution of dry matter and N in triticale and wheat varieties under stress condition. Agron. J. 70: 623-626.

Lance, R.C.M. and Macleod, L.C. (1993). Diastase: A breeding perspective and sources of variation. Proceedings of the 6th Australian Barley Technical Symposium. Launceston, Tasmania. 7th-19th September pp. 157-160.

Langer, R.H.M. (1979). The dynamics of wheat yield. N.Z. rWheat Rev. 14:32-4O.

Langer, R.H.M. and Hanif, M. (1973). A study of floret development in wheat (Triticum asetivum L.). Ann. Bot. (N.S.) 37:743-75L.

Langer, R.H.M. and Hanif, M.(1970). A study of floret development in wheat (Triticum aestivum L.) Ann. Bot.743-51.

Langer, R.H.M. and Liew, F.K. (1973). Effect of varying nitrogen supply at different stages of the reproduction phase on spikelet and grain production and on grain nitrogen of wheat. Aust. J. Agnc.Res.24: 647-656.

Lauer, J.G. and Partridge, J.R. (1990). Planting date and nitrogen rate effects on spring malting barley. Agron. J. 82: 1083-1088. 222

Lawlor, , A.E.,Legg, B.L. and Pa¡kinson, K.J. (1981). Growth of ught: Crop development, photosynthesis, dry matter content. J.Agric. Sci. 96: L67-186.

Leeper, G.W. (editor). (1970). The Australian Environmenr. CSIRO/ÌVÍUP, Melbourne, 4th Edition. 163p.

Leeuwrik, D.M. (1974). Farming systems in the Mediterranean region. International Workshop on farming systems. Nov. 18-21. International Crop Research Institute, Begumpet Hyderabad India. pp. 303-309.

Littler, J:W. (!963). Effect of time and rate of application of urea to wheat. Qld. J. Agric. Sci.20: 323-3M.

Littler, J.W., Kelso, W.T. and Marley, J.M. (1969). Effects of nitrogen fertilisers on the yield and malting quality of barley on the southern Darling Downs, Queensland. Qld. J. Agnc. Anim. Sci. 26: 309-317.

Lockhard, D.A.S. (1989). Cereal farming and harvesting. In: Cereal Science and Technology. (Ed: Palmer G.H.), pp. 15-31. (Aberdeen University press, Great Britain).

Loffler, C.M., Rauch, T.L. and Busch, R.H. (1985). Grain and plant protein relationships in hard red spring wheat. Crop Sci. 25:521-524.

Loffler, G.M. and Busch, R.H. (1982). Selection for grain protein yield and nitrogen panitioning efficiency in hard red spring wheat. Crop Sci. 22:59L-595.

MacKown, C.T. and van Sanford, D.A. (1986). Post anthesis nitrate assimilation in winter wheat. Plant Physiol. 8l:17-21.

Macleod, L.8., Sterling, J.D.E. and Macleod, J.A. (1975). Effects of rate, source, and method of application of N, P and K on glowth, nutrient uptake, and yield of barley. Can. J. Soil. Sci. 55:6L-67.

Macleod, L.C., Wallwork, M.A.B., Dowling, M.A. and Giles, L.C. (1993). Malt beta- glucan-variation, modification and prediction. In: Proceedings of the 6th Australian Barley Technical Symposium. Launceston, Tasmania. 7th-l9th September pp.20-23.

Mason, M.G. (1968). Sources of nitrogen for cereals-their characteristics and costs. J. Exp. Agric.9: 556-558.

Mason, M.G. and Rowland, I.C. (1990). Nitrogen fertiliser response of wheat in lupin- wheat, subterranean clover-wheat and continuous wheat rotations. Aust. J.E*p. Agric. 3O:231-236.

McDonald, G.K. (1989). The contribution of nitrogen fertiliser to the nitrogen nutrition of rainfed wheat crops in Ausralia: a review. Aust. J. Exp. Agric.29: 455-481.

McDonald, G.K. (L992). Effects of nitrogenous fertiliser on the growth, grain yield and grain protein concentration of wheat. Aust. J. Agric. Res. 43: 949-67.

McGuire, C. F., Hockett, E.A. and Wesenberg, D.M. (L979). Response of agronomic and barley quality traits to nitrogen fertiliser. Can. J. Plant. Sci. 59: 831-837.

McKeel, H.S. (1962). Nitrogen metabolism in plants. Cla¡endon Press, Oxford.

McNamara, 4.L., Meeker, G.8., Shaw, P.D. and Hageman, R.H. (1971). Use of assimilator nitrate reductase from Escherichia coli and formate as a reductive system for nitrate assays. J. Agric. Food Chem. L9:229-231. 223 McNeal, F.4., Berg, M.A. and Watson, C.A. (1966). Nitrogen and dry matter in five spring wheat varieties at successive stages of development. Agron. J. 58: 605-608.

McNeal, F.H., Berg, M.4., Brown, P.L. and McGuire, C.F. (1971). Productivity and quality response of five spring wheat genotypes, Triticum aestivum L. to nitrogen fertilisation. Agron. J. 63: 908-910.

McNeal, F.H., Boatright, G.O., Berg, M.A. and Watson, C.A. (1968). Ninogen in plant parts of seven spring wheat varieties at successive stages of development. Clop Sci.8: 535-537.

McNeal, R.H., Berg, M.4., McGurie, C.F., Stewart, V.R. and Baldridge, D.E. (1972). Grain and plant nitrogen relationship in eight spring wheat crosses, Triticum aestivum L.- Crop. Sci. 12: 566-601.

Mengel, K. (1983). Responses of various crop species and cultivars to fertiliser application. Plant Soil 72: 305-319. Mikesell, M.E. and Paulsen, G.M. (1971). Nitrogen translocation and the role of individual leaves in protein accumulation in wheat grain. Crop Sci. lI:919-922.

Miller, H.F., Kavanaugh, J. and Thomas, G.W. (1975). Time of N application and yields of corn in wet, alluvial soils. Agon. J. 67: 401-404.

Morrison, M.J., McVetty, P.B.E. and Shaykkewich, C.E. (1989). The determination and verification of a baseline temperature for the growth of Wester summer rape. Can. J. of Plant Sci. 69: 455-464.

Muruli, B.I. and Paulsen, G.M. (1981). Improvement of nitrogen use efficiency and its relationship to other traits in maize. Maydica 26: 63-75.

Myers, R.J.K. and Paul, E.A. (1971). Plant uptake and immobilisation of l5N-labelled ammonium nitrate in a field experiment with wheat. In: Nitrogen-l5 in Soil-Plant Studies. Proc. of a research Co-ordination meeting. (Vienna: International Atomic Ener

Nedel, J.L., Ullrich, S.E., Clancy, J.A. and Pan, W.L. (1993). Barley semidwarf and standard isotype yield and malting quality response to nitrogen. Crop Sci. 33: 258- 263.

Needham, P. (1982). The role of nitrogen in wheat production: Response, interaction and prediction of nitrogen requirement in the U.K. The Fertiliser Society Proceedings, London zlL,125-147.

Nicholls, P.B.and Muy, L.H. (1963). Studies on the growth of the barley apex. I.Interrelationsgips between primordium formation, apex length, and spikelet development. Aust. J.Biol.Sci. 16: 561-571.

Nicolas, M.E., Gleadow, R.M. and Dalling, M.J. (1985a). Effect of post-anthesis drought on cell division and starch accumulation in developing wheat grains. Ann. Bot. 54:433-444.

Nicolas, M.E., Simpson, R.J., Lambers, H. and Dalling, M. (1985). Effects of drought on partitioning of nitrogen in two wheat va¡ieties differing in drought-tolerance. Ann. Bot.55:743-754.

Nix, E.L. (1975). The Ausralian climate and its effects on grain yield and quality. In: Australian Field Crops, Vol. 1: Wheat and other temperate cereals. (Eds: A. Lazenly and E.M. Maiheson), p.183-226. (Angus and Robertson: Sydney). 224 Novoa, R. and Loomis, R.S. (1981). Nitrogen and plant production. Plant Soil. 58: 177- 204.

Nuttall, W.F. (1973). Yield response of Conquest barley as affected by nitrogen fertiliser and soil tests for ammonium and nitrate nitrogen. Can. J. Soil Sci. 53: 163-168.

Oades, J.M. (1989). An introduction to organic matter in mineral soils. In: Minerals in Soil Environments. (Eds: J.B.Dixon and S.B. Weed), pp. 89-159. Soil Sci. Soc. Amer. Madison, V/isc.

Olson, R. 4., Frank, K.D., Deibert, E.J., Dreier, 4.F., Sander, D.H. and Johnson, V.A. (1976). Impact of residual mineral N in soil on grain protein yields of winter wheat and corn. Agron. J.68:769-772.

Olson, R.A. and Kurtz, L.T. (1982). Crop nitrogen requirements, utilisation, and fertilisation. In: Nitrogen in Agricultural soils. (Ed: F.J. Stevenson), PP. 567-604. (Amer. Soc. Agron. Madison, Wisc.).

Olson, R.V. (1980). Fate of tagged fertiliser applied to irrigated corn. Soil. Sci. Soc. Amer. J.44: 514-517.

Oosterhuis, D.M.and Cartwright , P.M.(1983). Spike differentiation and floret survival in semi dwarf spring wheat as affected by water stress and photoperiod. Crop Sci. 13: 7tt-717.

Paccaud, F.X., Fossati, A. and Cao, H.S. (1985). Breeding for yield and quality in winter wheat: Consequences for nitrogen uptake and partitioning efficiency. Z. Pflanzenzuecht. 94: 89-100.

Palta, J.4., Kobata, T., Turner, N.C. and Fillery, I.R. (1994). Remobilisation of Carbon and Nirogen in wheat as influenced by postanthesis water deficits. Crop Sci. 34: tl8-124.

Pan, W.L., Teyker, R.H., Jackson, V/.4. and Moll, R.H. (1987). Diurnal variation in nitrate, potassium, and phosphalg uptake inrnaiz-e-see{ling¡-Çonsiderations in screening genotypes for uptake efficiency. J. Plant Nutr. 10: 1819-1833.

Papakosta, Despo. K. and Gagianas, A.A. (1991). Nitrogen and dry matter accumulation, remobilisation and losses for Meditenanean wheat during grain filling. Agron. J. 83:864-870.

Parsons, J.rW. and Tinsley, J. (1975). Nitrogenous substance. In: Soil components, Vol. 1. Organic components.(Ed: J.E. Gieseking), p. 263-304. (Springer-Verlag New York).

Parton, W.J., Mogan, J.4., Altenhofen, J.M. and Harper, L.A. (1988). Ammonia volatilisation from spring wheat plants. Agron. J. 80: 1423-1436.

Pate, J.S. (1971). Movement of nitrogenous solutes in plants. In: Nitrogen-15 in soil plant studies. Atomic Energy Agency STL/PUB/278. Vienna. p.165-187.

Pate, J.S. andLayzell, D.B. (1931). Ca¡bon and nitrogen partitioning in the whofe plan¡A thesis based on empirical modelling. In: Nirogen and Carbon Metabolism. (Ed: J.D. Bewley), pp.94-134. (Nijhoff, the Hague, Netherlands).

Pate, J.S., Layzell, D.B. and McNeil, D.L. (1979). Modelling the transport and utilisation of carbon and nitrogen in a nodulated legume. Plant Physiol. 63: 730-737 .

Pearman, I., Thomas, S.M. and Thorne, G.N. (L977). Effects of nirogen fertiliser on growth and yield of spring wheat. Ann. Bot. (N.S.). 41: 93-108. 225 Pearson, C.J., and Muirhead, W.A. (1984). Nitrogen uptake. In: Control of Crop Productivity. (Ed: C.J. Pearson), pp. 73-88. (Academic Press, New york).

Penning de Vries, F.W.T., Brunsting, A.H.M. and Van Laar, H.H. (1974). Products, requirements, and efficiency of biosynthesis: a quantitative approach. J. Theor. Biol. 45: 339-377. Penny, . (1974). Results from experiments with winter wheat, ;:#3,il1åijiåå'-:#fi '$i,ii:illî1,îf:åtiff ,ilö'Jå alon

Perby, H. and Jensen, P. (1983). Varietal differences in uptake and utilisation of nitrogen and other macro-elements in seedlings of barley, Hordeum vulgare. Physiol. Plant. 58:223-230.

Perby, H. and Jensen, P. (1984). Net uptake and partitioning of nitrogen and potassium in cultivars of barley during ageing. Physiol. Plant. 61:559-565.

Perby, H. and Jensen, P. (1986). Variation in growth and accumulation of N, K+, Ca2+, and Mg2+ among barley cultivars exposed to various nutrient regimes and roolshoot temperatures. Physiol. Plant. 67 : 166-172.

Perby, H. and Jensen, P. (1987). Vegetative adaptation to N stress regimes in two barley cultiva¡s with different N requirement. In: Genetic Aspects of Plant Mineral Nutrition. (Eds: W.H. Gabelman and B.C. Lougman), pp.36l-367. (Nijhoff, Dordrecht, Ne

Perry, M. and Hillman, B. (eds) (1991). The wheat book. A technical manual for wheat producers. Department of Agriculture, Western Australia. Bulletin 4L96. p.182.

Peterson, G.A. and Foster, A.E. (1973). Malting barley in the United States. Adv. Agron. 25:328-378.

Pheloung, P.C. and Siddique, K.H.M. (1991). Contribution of stem dry matter to grain yield in wheat cultiva¡s. Aust. J. Plant Physiol. 18: 5306/..

Pilbeam, D.J. and Kirkby, E.A. (1990). The Physiology of Nitrate Uptake. In: Nitrogen in Higher Plants (Ed: Y.P Abrol), pp.39-61. Indian Agricultural Research Institute. (New Delhi, India).

Pinthus, M.J. and Nerson, H. (1984). Effect of photoperiod at different grcwth stages on the initiation of spikelet primordium in wheat. Aust. J. Plant Physiol. LL: l7-22.

Power, J.F. and Alessi, J. (1978). Tiller development and yield of standard and semi- dwarf spring wheat varieties as affected by nitrogen fertiliser. J. Agric. Sci. Camb. 90:97-108.

Prescott, J.A. and Thomas, J.A. (1949). The length of the growing season in Australia as determined by the effectiveness of the rainfall. Proc. Roy. Geog. Soc. South Aust. 50:42-46.

Pushman, F.M. and Bingham J. (1976). The effect of a granular niEogen fertiliser and a spray of urea on-the yield and bread making qualify of ten winter wheats. J. Agric. Sci.87:281-292.

Rahman, M.S. (1930). Effect of photoperiod and vernalisation on the rate of development and spikelet number per-ear in 30 varieties of wheat. J Aust. Inst. Agric. Sci. 46: 68-70. 226

Rahman, M.S. Wilson, J.H. (L977). Determination of spiklet number in wheat. 1. Effect of varying photoperiod on ear development. Auit. J. Agric. Res. 28: 565-74.

Rahman, M.S. Wilson, J.H. and Aithen, Y. (L977). Determination of spikelet number in wheat, II. Effect of varying light'level on ear development. Aust. J. Agtic. Res. 28: 575-81.

Randall, P.J., Freney, J.R., Smith, C.J., Moss, H:J., ìMrigley, Ç.W. pd- Galbally' I..E. (tggO). nfJebt of additions of nitrogen and_ sulphui to irrigated whea!_at heading on grarn yreld composition and milling and baking quality. Aust. J. Exp. Agric. 30:

Rasmusson, D.C. and Cannall, R.Q. (1970). Selection for grain yield and components of yield in barely. Crop Sci. 10: 51-54.

Rawson, H.M. (1992). Plant responses to temperature under conditions of elevated CO2. Aust. J. Bot.40: 473-490.

Rawson, H.M. and Bagga, A.K. (1979). Influence of temperature between floral initiation and flag leaf eñiergence on grain number in wheat. Aust. J. Plant Physiol. 6: 391- 400.

Rayment, G.E. and Higginson, F.R. (1992): Nitr_oge-n. In: Australian Laboratory ' Handbook of SoiIãnd Water Chemical Methods- (Inkata Press, Melbourne).

Reed, A.J. and Hageman, R. H. (1980). Relationship between nitrate_uptake, flux, an_d reduction ai¿ the accumdtatiori of reduced nirogen in maize (Zea mays L.). I. Genotypic variation. Plant Physiol. 66: 1179-1183.

Regan, K.L., V/han, B.R. and Turner, N.C. (1993). Fvaluation of chemical desiccation as a selection iechnique for drought resistance in a dryland wheat breeding program. Aust. J. Agric.Res. 44: 1683-1691.

Remy, J.C. and.Viaux, Ph. (1982). The use_of-nitrogen fertilisers in intensive wheat growing in France. Proc.-Fert. Soc. 21L: 67-92. future nutrition research strategijs for Reuter,'South D.J. (1989). Philosophical ^Discussionpriorities for AusÉatian field crops. paper prepared for the Wheat and Barley Research Committees for S.A. August 1989.

Riggs, T.J., Strat gan,-C.!. and-Ford, M.A. (1981). Cgryp1lso¡t ""-' of .pting England and Wales between 1880 and 1980. J. Agric. Sc

in soil-plant-atmosphere_ sys_tep-. In: Soil Water Ritchie,'and J.T. (1981). Water dynamics the Ñitrogen in Mediterranean-type eÑiro_nnpnts. (EAs: J. Monteith, and C. Webb), ppl St-lO. (Martinus Nijhòff/DrV/. Junk: The Hague).

Roberts, E.J.R., Summerfield, R.J. (1987). Measurement and predication of floweTng-in ânnual crops. In: Manipulation ôf Flowering. (Ed: J.G. Atherton). pp. 17-50. (Butterworths, l,ondon.)

Rodriguez-p, M.S. (1977). Varietal differences in maize in the uptake of nitrogen and its - tranìlocation to the grain. Diss. Abstr. 38: 5690-8. T. (1976). The internal nitrogen cycle Rosswall, -vegetation, and sòil. ln: Nitrogen, phosphórus änd P-E }tp' Z' Éõórógiòaf guil. no.ã2, p.tSt:\e1. (Swed ch Council, Royal Swedish 227 Russell, J.S. (1968a). Nitrogen fertiliser and wheat in a semi-arid environment. 2. Climatic factors affecting response. Aust. J. Exp. Agric. Anim. Husb. 8:22L-231.

Russell, J.S. (1968b). Nitrogen fertiliser and wheat in a semi-arid envi¡onment. 3. Soil and cultural factors affecting response. Aust. J. Exp. Agric. Anim. Husb. 8: 340- 348.

Russell, J.S. (1980). Nitrogen in dryland agriculture. International Congress on Dryland Farming Proceedings, Adelaide 1980, pp.202-2o7. (South Australian Department of A griculture: Adelaide. ).

Russell, R.S. (1977). Plant root systems: Their Function and Interaction with the soil. (Maidenhead, England: McGraw-Hill).

Saini,H.S; and Aspinall. (1982). Abnormal sporogensis in wheat (Triticum aestivum L.) induced by short periods of high temperature. Ann of Bot. 49: 835-46.

Salter, P.J. and Goode, J.E. (1967). Crop responses to water at different stages of growth. Commonw. Bur. Hortic. Plant. Crops Res. Rev. No. 2.

Sanchez, P.AS., Ramirez, A.E. and de Calderon, M.V. (1973a). Rice responses to nitrogen under high solar radiation and intermittent flooding in Peru. Agron. l. 65: 523- 529.

Sander, D.H., Allaway, W.H. and Olson, R.A. (1987). Modification of nutritional quality by environment and production practices. In: Nutritional Quality of Cereal Grains Genetic and Agronomic Improvement. (Eds: R.A. Olson and K.J. Frey), pp. 45- 82. Am

Schimansky, C. and Ma¡schner, H. (1971). Suitability of 86Rb as a tracer for potassium in studies relating to potassium uptake by maize, sugarbeet and four varieties of barley.- Z. Pflanzenernahr. 129 : l4l - L47 .

Schjorring, J.K., Nielsen, N.E., Jensen, H.E. and Gottschau, A. (1989). Nitrogen losses from field-grown barley plants as affected by rate of nitrogen application. Plant and Soil. Ll6:167-175.

Schnyder, H. (1993). The role of carbohyd¡ate storage and redistribution in the source- sink relations of wheat and barley during grain filling-a review. New Phytol. 123: 233-245.

Scott, W.R., Appleyard, M., Fellows, G. and Kirby, E.J.M. (1983). Effect of genotype and positioñ-in the ear on carpel and grain gro,wth and mature grain weight of spring barley. J. Agric. Sci. 100: 383-391.

ShadlR. and Dale, J.E. (1980). Growth-regulating substances and the g¡owth of tiller buds in barley; effects of IAA and GA3. J.Exp. Bot. 31: 1'L9L-LL97.

Siddique, K.H.M., Belford, R.K., Perry, M.W., and Tennant, D. (1989). Growth, development and light interception of old and modern wheat cultivars in a Mediterranean-type environment. Aust. J. Agric. Res. 40: 473-487.

Sillanpaa, M. (1982). Micronutrients and the nutrient status of soils: A global study. FAO Soils Bulletin 48, FAO, Rome.

Simmons, S.R., Rasmusson, D.C. and V/iersma. (1982). Tillering in barley.: genotype, rowspacing and seeding rate effects. Crop Sci.22:801-805.

Simpson, R.J., Lambers, H. and Dalling, M.J. (1983). Nitrogen redistribution.during grain $owth in wheat (Triticum aestivum L.). fV.Development of a quantitative model óf ttre translocation of nitrogen to the grain. Plant Phystol. T l: 7 -14. 228 Sinclair, T.R. and de Wit, C.T. (1975). Comparative analysis of photosynthate and nitrogen requirements in the production of seed by various crops. Science 89: 565- 567.

Singh, N.T., Vig, 4.C., Singh, R. and Chaudhury, M.R. (L979). Influence of different levels of irigation and nitrogen on yield and nutrient uptake by wheat. Agron. J. 7l:40L-4O4.

Singh, R. and Priha¡, S.S. (1978). Fertiliser use in rainfed areas. Proc. IndiaÆAO/1.{orway Seminar on the development of the complementriry use of mineral fertilisers and organic materials in India. pp. 63-66.

Singh, R., Singh, Y, Prihar, S. S. and Singh, P. (1975). Effect of N fertilisation on yield water use efficiency of dryland wheat as affected by stored water and rainfall. Agron. J.67:599-603.

Single, W.V. (1975). Frost injury. In: Australian Field Crops. Vol. l: Wheat and other temperate cereals. (Eds: A. Lazenby, E.M. Matheson), pp. 364-383. (Angus and Robertson, Sydney).

Smith, C.J., Freney, J.R., Chapman, S.L. and Galbally, I.E. (1989a). Fate to urea applied to irrigated wheat at heading. Aust. J. Agric.Res. 40: 951-963.

Smith, C.J., Whitfield, D.M., Gyles, 4.O., and Wright, G.C. (1989b). Nitrogen fertiliser balance of irrigated wheat grolvn on a red-brown earth in south-eastern Ausnalia. Field Crop Res. 2l:269-275.

Smith, D.B. (1990). Barley seed protein and its effects on malting and brewing quality. Plant Varieties and Seeds 3: 63-80.

Smith, R.C.G. and Harris, H.C. (1981). Environmental resources and restraints to agricultural production in a Mediterranean-type envi¡onment. In: Soil water and Nitrogen in Mediterranean-type Environments. (Eds: Monteith J and V/ebb C ), pp.31-57. (Ma

Sofield, I., Evans, L.T., Cook, M.G. and Wardlaw, I.F. (1977>. Factors influencing the rate and duration of grain frlling in wheat. Aust. J. Plant Physiol. 4: 785-797.

Sparrow, D.H.B. (1972). A study of genotypic differences in the malting quality of barley. Ph.D. Thesis, University of Adelaide, South Australia.

Sparrow, P.E. (1979). The comparison of five response curyes for reptesenting the relationship between annual dry matter yield of grass herbage and fertiliser nitrogen. J. Agric. Sci. 93: 5L3-520.

Spiertz, J.H.J. and Ellen, J. ( .978). Effects of nitrogen on crop development and grain growth of winter wheat in relation to assimilation and utilisation of assimilates and nutrients. Neth. J. Agric. Sci. 25: 210-231.

Spiertz,- J.H.J. and van der Haar, H. (1978). Differences in grain -growth, c-rop photosynthesis and distribution of assimilates between semi-dwarf and standa¡d cultivars of winter wheat. Neth. J. Agric. Sci. 26: 233-249.

Spratt, E.D. and Gasser, J.K.R. (1970). Effects of fertiliser nitrogen- and water su^pply on the distribution of dry matter and nitrogen between the different parts of wheat. Can. J. Plant Sci. 50: 613-625.

Stanford, G., and Hunter, A.S. (1973). Nitrogen requirements of winter wheat (Triticum aestivum L.) varieties Btueboy and Redcoat. Agron. I. 65: 424-M7. 229 Steel, R.G.D. and Torrie, J.H. (1960). Principles and procedures of statistics. (MacGraw- Hill Book. Co., Inc. New York) p.813.

Stevenson, F.J. (1982). Nitrogen in Agricultural soils. Agron. Monogr. No. 22. (A4er. Soc.Agron., Crop Sci. Soc. Amer., Soil Sci. Soci. Amer. Madison, Wisc). p 940

Stoddard, F.L. and Marshall, D.R. (1990). Variability in grain protein in Australia hexaploid wheats. Aust. J. ASnc.Res. 41: 277-88.

Storrier, R.R. (1962). The availability of mineral nitrogen in a wheat soil from southern New South Wales. Aust. J. Exp. Agric. Anim. Husb. 2: L85-L92.

Storrier, R.R. (1975). Wheat nutrition. In: Australian Field Crops Volume. 1: Wheat and other Temperate Cereals. (Eds: A. Lazenby and E.M. Matheson.), pp. 256-287. (Aguns and Robertson : Sydney.).

Tanner, C.B. and Sinclair, T.R. (1983). Efficient water use in crop production: Research or research? In: Limitations to efficient water use in crop production. (Eds: H.M. Taylor, \M.R. Jordan and T.R. Sinclair), pp. l-27. (Amer. Soc. Agron. Madison, tü/is

Taylor, 4.C., Storrier, R.R., and Glimour, A.R. (197 8). Nitrogen needs of wheat. Il. Grain yield response to nitrogenous fertiliser. Aust. J. Exp. Agric.Anim. Husb. 18:118-120.

Tew, T.L. and Rasmusson, D.C. (1978). Scoring photoperiod sensitivity in barley. Crop Sci. 18:897-898.

Thomas, S.M., Thorne, G.N. and Pearman, I. (1978). Effect of nitrogen on growth, yield and photorespiratory activity in spring wheat. Ann. Bot. (N.S.) 42:826-837.

Tinker, P.B. and Widdowson, F.V. (1982). Maximising wheat yields and some causes of yield variation. Proc. Fert. Soc. 2ll: L49-184.

Tinker, P.B.H. (1978). Uptake and consumption of soil nitrogen in relation to aglongmic practice. In: Nitrogen Assimilation õf Plants. (Eds: E.J. Hewitt and C.V. Cutting), pp.L0l-122. (Academic Press, New York).

Turner, N.C. and Nicolas, M.E. (1987). Drought resistance of wheat for light-textured soils in a Mediterranean climate. In: Drought tolerance in winter cereals. (Eds: J.P. Srivastava, E. Porceddu, E. Acevedo and S. Varma), 203-217. (John V/iley New York

Van Keulen, H. (1977). Nitrogen requirements of rice with special reference to Java. Gontr. Centr. Res. Inst. Agnc. Bogor. Indonesia No. 30. 67p.

Van Keulen, H. (1981). Modelling the interaction of water and nitrogen: In: Soil water and Nitrogen in Mediterranean-type Environments. (Eds: J. Monteith, and C. Webb), pp.ZOS-220. (Ma¡tinus NijhofflDr V/ Junk Publishers: The Hague).

Van Oosterom, E.J., Ceccarelli, S. and peacock, J.M. (1993). Yield response o_f barle-y t_o rainfall and temperature in Mediterranean environments. J. Agric Sci. Camb. 121: 307-313. van Sanford, D.A. and Mackown, C.T. (1987). Cultivar differences in nitrogen remobilisation during grain filt in soft red winter wheat. Crop. Sci. 27: 295-3W.

Vlek, P.L.G., Fillery, I.R.P. and Burford, J.R. (19S1 tion, and loss of nitrogeñ in soils of the arid region Nitr_ogen_in Meditenãnean-type Environments. @ds: J. pp. 133-173. (Martinu 230 Walpole,- P.R. and Morgan, D.G. (19710. A quantitative st_udy of grain frlling in th¡ee cultiva¡s of Hordeum vulgare L. Ann.Bot. 35: 301-310.

Warington, I.J., Dunstone, R.L, and Green, L.M. (1977). Temperature effects at thre€ - developmental stages on the yield of wheat ea¡. Aust.J. Agttc. Res. 28: ll-27.

Webster, R. and Oliver, M.A. (1990). Statistical Methods in Soil and Land Resource. Survey. Oxford University Press, New York.

Wetselaar, R. and Farquhar, G.D. (1980). Nitrogen losses from tops of plants. Adv. Agron. 33:263-302.

Wheeler, R.D., Lance, R.C.M. and Sparrow, D.H.B. (1987). Variation in responses of barleys to applied nitrogen. In: Proceedings of the 4th Australian Agronomy Conference. (Ed: T.G. Reeves), p. 253. Melbourne.

V/hiteaker, G., Gerloff, G.G., Gabelman, W.H. and Lindgern, D. (1976). Intraspecific differences in growth of beans at stress levels of phosphorus. J.Am. Soc. Hortic. Sci. 101: 472-475.

Whiteman, P.C. and Wilson, G.L. (1965). The effects of water stress on the reproductive development of Sorghum vulgare. Pres. Queesland Univ. Papers. (Dept. of Botany). 4:233-239. (cited by Salter, 1969). rWiddowson, F.V., Penny, A. and Williams, R.J.B. (1961). Applying nitrogen fertilisers for spring barley. J. Agric. Sci. 56: 89-47.

V/illiams, C.H. and Raupach, M. (1933). Plant nutrients in Australian soils. In: Soils an Australian Viewpoint. (Division of Soils CSIRO/Academic Press).

Wood, J. (1980). The mathematical expression of crop response to inputs. Proc. Colloq. Int. Potash Inst. 15: 263-27L.

'Woodend, J.J., Glass, A.D.M. and Preson, C.O. (19S6). Intraspecific va¡iation for nitrate uptake and nitrogen utilisation in wheat (Triticum aestivum L.) grown under nitrogen stress. J. Plant Nutr. 9: 1213-1225.

V/oodruff, D.R. (1972). Cultivar variation in nitrogen uptake and distribution in wheat. Aust. J. Agric. Res. Anim. Husb t2:5ll-6.

Xa,Z.H.,Elliott, D.E and Winn, D. (1992). Influence of optimum rate of N fertiliser applied on whe¿t Proceedings of V/orkshop on Tillage Systems, Ro Root Xu,2.H.,' Elliott, D.E., Hannam, R.J. and \üy'inn, D. (1991). Yield and quality-responsesof wheat and barley to application of nitrogen fertiliser in South Australia. Proceedings of Workshôp on Tillage Syitems, Rotyations, Nutrition and Associated Root Di

Young R.4., Ozbun, J.L., Bauer, A. and Vasey, E.H. (1967). Yield.response of spring - wheat and barely io nitrogen fertiliser iri relation to soil and climatic factors. Soil Sci. Soc Am Proc. 31':407-410.

Zadoks, J.C., Chang, T.T. and Konzak, C.F. (L974). A decimal code for the growth stages of cereals. WeedRes. 14: 4L5-421.