See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/262564743

The Effect of Heat Stress on Fruit-Set and Fruit Yield of Groundnut

(Arachis hypogaea L.).

Thesis · April 1999

CITATIONS READS
5 1,483

1 author:

P. V. Vara Prasad
Kansas State University
496 PUBLICATIONS 24,647 CITATIONS

SEE PROFILE

All content following this page was uploaded by P. V. Vara Prasad on 24 May 2014.

The user has requested enhancement of the downloaded file.

 THE UNIVERSITY OF READING

THE EFFECT OF HEAT STRESS ON FRUIT-SET AND

FRUIT YIELD OF GROUNDNUT (Arachis hypogaea L.)

By

Pagadala Venkata Vara Prasad

A thesis submitted to The University of Reading for the degree of

Doctor of Philosophy

The University of Reading

Department of Agriculture

Apri11999
 ABSTRACT

Groundnuts (Arachis hypogaea L.) grown in the semi-arid tropics are commonly exposed to

air and soil temperatures> 35°C. This research in controlled environments has shown that

continuous exposure of plants to hot air (day/night, 38°/22°C) and/or hot soil (38°/30°C)

temperatures during the reproductive phase (from flower bud appearance until reproductive
.
maturity) significantly reduces total dry matter production, the partitioning of dry matter to

pods and seed yields. The effects of hot air and hot soil temperature were additive and without

interaction. Hot air temperature had no effect on flower production but significantlyreduced

the proportion of flowers setting pegs (fruit-set) and hence the number of fruits. In contrast,

hot soils significantly reduced flower production, the proportion of pegs forming pods and 100

seed weight. There was no evidence that plants dependent on symbiotic N2 fixation were more

susceptible to heat stress than those dependent on inorganic N. Sensitivity to short (6 d)

episodes of hot air temperature (38°/22°C) was acute during the period between 6 d before

until 15 d after first flowering (OAF). The magnitude of that sensitivity depended on the

number of floral buds exposed to heat stress before anthesis. In the Spanish cv. ICGV 86015,

daytime air temperatures ~34°C imposed for only 6 d beginning at 9 DAF, significantly

reduced flower number, pollen production and viability, fruit-set and seed yield. Fruit-set was

most sensitive to heat stress during the first 6 h of the daylight period (AM). Warmer nights

(28° cl 22°C) had no effect on flower numbers, but significantly reduced both pollen

production and viability, and hence fruit-set. There were negative quantitative relations

between flower number and day temperatures between 28° and 48°C. In contrast, reductions in

fruit-set were quantitatively related to AM temperature >37.3°C. Pollen production and

viability were also linearly reduced when day temperature was >34°C. These data will help

plant breeders to screen germplasm and identify heat-tolerant cultivars, and should also

improve simulation models of groundnut crops in the semi-arid tropics.

 11

ACKNOWLEDGEMENTS

I express my deep sense of gratitude to the Felix Foundation for its generous financial

support without which I could not have taken up my PhD programme.

I thank my supervisors Dr P.Q. Craufurd and Professor R.I. Summerfield for their patience,

continuous support, inspiring guidance and excellent advice in designing the research and

preparation of the manuscript. The constructive criticism and thought-provoking remarks

have helped me to improve my scientific knowledge and writing skills.

I am thankful to Drs T.R. Wheeler and Aiming Qi for their suggestions and discussions. I am

grateful to Mr K. Chivers for his untiring technical and engineering support and Messers

S. Gill, H. Dorji, A. Pilgrim and Ms C. Hadley and 1. Redman for their assistance during the

experiment.

I thank my friends Gopal, Mayank, Abhijit, Kanthi, Sudhanshu, Jitender, Elizabeth, Renuka

and Chandrika for their support and help during my research. Thanks are due especially to

Mrs Sue Redman and Mr Alistair Robinson and Mr Ramesh Babu and their families for their

kind concern and constant encouragement.

Finally, I express my gratitude to my parents, my wife, my teacher Dr V. Satyanarayana and

my friends Sudhakar, Ravi, Yashoda and Vijaya for their inspiration and unconditional

support throughout my studies.

 iii

TABLE OF CONTENTS

ABSTRACT i

ACKN"OWLEDGEl\fENTS ii

TABLE OF CONTENTS : iii

LIST OF ABBREVIATIONS AND SYMBOLS •••••••••.••..••••••.••••.••.•.••••.•••••

x

LIST OF TABLES xii

LIST OF FIGURES xv

CHAPTER 1 GENERAL INTRODUCTION 1

1.1 ORIGIN AND DISTRIBUTION 2

l.2 TAXONOMY AND CLASSIFICATION 3

l.3 GROUNDNUT PRODUCTION AND PRODUCTMTY WORLD-WIDE .. 6

1.3.1 Asia 6

1.3.2 Africa 9

1.3.3 North America 9

1.3.4 South America 10

1.3.5 Europe 10

1.4 UTILIZATION 10

l.5 CONSTRAINTS TO PRODUCTIVITY IN THE SAT 11

CHAPTER 2 GROWTH, DEVELOPMENT, NITROGEN NUTRITION

AND SEED YI"ELD OF GROUNDNUT 14

 iv

2.1 GROWTH AND DEVELOP11ENT 14

2.1.1 Growth stages.................................................................................... 14

2.1.2 Seedling and vegetative growth 16

2.1.3 Reproductive growth and maturity 17

2.2 NODULE FORMATION AND NITROGEN FIXATION . 20

2.2.1 Root colonisation and infection 20

2.2.2 Nodule development , 20

2.2.3 Nodule function 21

CHAPTER 3 EFFECT OF TEMPERATURE ON GRO\VTH,

DEVELOPMENT, NITROGEN NUTRITION AND SEED YIELD .....••23

3.1 EFFECT OF TEMPERATURE ON GROWTH, DEVELOP11ENT AND

SEED YIELD 23

3.1.1 Seed germination and seedling emergence 23

3.1.2 Vegetative growth , 25

3.1.3 Reproductive growth 27

3.1.3.1 Bud development 27

3.1.3.2 Flowering 28

3.1.3.3 Pegging 30

3.1.3.4 Podding 31

3.1.3.5 Podfilltng and yteld 31

3.2 EFFECT OF TEMPERATURE ON NODULATION AND NITROGEN

FIXATION 34
 v

CHAPTER 4 OBJECTIVES OF CURRENT RESEARCH ••••.•••••••••••••.••

37

CHAPTER 5 GENERAL MATERIALS AND l\fETHODS .•.•.••.••.•.••••••••

40

5.1 PROGRAMME OF EXPERIMENTS 40

5.2 CULTIVARS RESEARCHED 41

5.3 CONTROLLED ENVIRONMENT FACILITIES 41

5.3.1 Polyethylene tunnels 42

5.3.2 Blackout facility 42

5.3.3 Soil temperature controI.. 45

5.3.4 Growth cabinets 47

5.4 STRUCTURE AND PRESENTATION OF RESULTS AND DISCUSSION

................................................................................................................... 47

CHAPTER 6 EFFECTS OF CONTINUOUS HOT Am AND HOT SOIL

TEl\fPERATURE ON DRY MATTER PRODUCTION,

PARTITIONING AND POD YIELDS 49

6.1 INTRODUCTION 49

6.2 MATERIALS AND ~rnODS 50

6.2.1 Cultivars and temperature treatments 51

6.2.1.1 Experiment 1 (1996) 51

6.2.1.2 Experiment 2 (1997) 52

6.2.2 Plant husbandry 53

6.2.3 Observations and data analysis 54

 VI

6.3 RESULTS 55

6.3.1 Experiment 1 (1996) 55

6.3.1.1 Cultivar responses 55

6.3.1.2 Response to air temperature 58

6.3.2 Experiment 2 (1997) 61

6.3.2.1 Response of cultivars to hot air temperature or hot soil

temperature 61

6.4 DISCUSSION 64

CHAPTER 7 SENSITIVITY OF GROUNDNUT TO TIMING OF HEAT

STRESS DURING REPRODUCTIVE DEVELOPMENT •••••••......•....••••

70

7.1 ABSTRACT 70

7.2 INTRODUCTION 71

7.3 MATERIALS AND ~THODS 73

7.3.1 Environmental conditions 73

7.3.2 Cultivars and plant husbandry 74

7.3.3 Reciprocal transfertreatments 75

7.3.4 Observations and data analysis 75

7.4 RESULTS 77

7.4.1 Cultivar responses 77

7.4.2 Effect of temperature 78

7.4.3 Effect of reciprocal transfers 80

7.5 DISCUSSION 83
 vu

CHAPTER 8 EFFECTS OF SHORT EPISODES OF HEAT STRESS ON

FLO\VER PRODUCTION AND FRUIT-SET OF GROUNDNUT ••••••.•.

89

8.1 ABSTRACT 89

8.2 INTRODUCTION 90

8.3 MA.TERIALS AND l\1ETIlODS 91

8.3.1 Temperature treatments 93

8.3.1.1 Experiment 1. Duration of exposure to hot temperature 93

8.3.1.2 Experiment 2. Timing of exposure to hot temperature 94

8.3.2 Cultivar and plant husbandry 94

8.3.3 Observations 95

8.3.4 Data analysis 96

8.4 RESULTS AND DISCUSSION 97

8.4.1 Experiment 1. Duration of exposure to hot temperature 97

8.4.2 Experiment 2. Timing of exposure to hot temperature 100

8.4.3 Relations between FN, fiuit-set, RNt and temperature 102

CHAPTER 9 FRUIT NUMBER IN RELATION TO POLLEN

PRODUCTION AND VIABILITY IN GROUNDNUT EXPOSED TO

SHORT EPISODES OF HEAT STRESS ••••••.•.••••.•••••••••••.••••.•••.•.••••••.•.••

108

9.1 ABSTRACT 108

9.2 INTRODUCTION 109

9.3 MATERIALS AND ~rnODS II0

9.3.1 Temperature treatments 112

 viii

9.3.2 CuItivar and plant husbandry 113

9.3.3 Observations 114

9.3.4 Data analysis 115

9.4 RESULTS 116

9.4.1 Number of flowers and fruit-set.. 116

9.4.2 Number of pegs and pods 116

9.4.3 Pollen production and viability 120

9.5 DISCUSSION 123

CHAPTER 10 EFFECT OF HOT AIR AND SOIL TEMPERATURE ON

POD YIELD OF GROUNDNUTS DEPENDENT ON SYMBIOTIC OR

INORGANIC NITROGEN 128

10.1 ABSTRACT 128

10.2 INTRODUCTION 129

10.3 MATERIALS AND 11ETHODS 131

10.3.1 Temperature and nitrogen treatments 132

10.3.2 Cultivar and plant husbandry 133

10.3.3 Observations and data analysis 133

10.4 RESULTS 135

10.4.1 Effects of nitrogen sources : 135

10.4.2 Effects of air and soil temperature 137

10.4.3 Interaction between temperature regimes and nitrogen sources 140

10.5 DISCUSSION 142

 IX

CHAPTER 11 GENERAL DISCUSSION 147

11.1 FUTURE RESEARCH 165

REFERENCES 168
 x

LIST OF ABBREVIATIONS AND SYMBOLS

% percent
AM Ante Meridiem (before noon)
°C degree Celsius
Cd thermal time in Centigrade days
cf compare
cm centimetre
CO2 Carbon dioxide
cv. cultivar
d day
df degree of freedom
DAF Days after first flowering
DAP Days after planting
DAS Days after sowing
DBA Days Before Anthesis
e.g. for example
Ed. Editor
Eds Editors
Eq. Equation
Exp. Experiment
FAO Food and Agricultural Organisation
Fig. Figure
FN Number of flowers plant" opening in the 6 d stress period
g gram
h hour
ha hectare
ID Hot Temperature
i.e. that is
ICRISAT International Crops Research Institute for the Semi-Arid Tropics
kg kilogram
kPa kilopascal
L litre
lat. latitude
long. longitude
Ltd Limited company
m meter
mm minute
N Nitrogen (inorganic)
number of variables
Dinitrogen (symbiotically fixed nitrogen)
Optimum Temperature
probability
 Xl

PAR Photosynthetically Active Radiation

PEL Plant Environment Laboratory
PM Post Meridiem (after noon)
PPFD Photosynthetic Photon Flux Density
ppm parts per million
R Reproductive stages of groundnut
~ Coefficient of determination
RN Reproductive Number (number of pegs and pods) plant"
RUE Radiation Use Efficiency
s second
SAT Semi-Arid Tropics
SD Standard Deviation
SED Standard Error of Difference
SLA Specific Leaf Area
SLN Specific Leaf Nitrogen
SLW Specific Leaf Weight
sp., spp. species
T Temperature
t tonne
Tb Base temperature
TIN Total Flower Number plant"
Tm Maximum temperature
To Optimum temperature
UK United Kingdom
USA United States of America
Var. Variety
W watt
WD Whole Day
YBP years before the present
 XlI

LIST OF TABLES

Table 1.1 Harvest area, yield and production of groundnuts (in shell) in different parts
of the World (FAO, 1998) 7

Table 2.1 Growth stage descriptors for groundnut (Boote, 1982) 14

Table 3.1 Optimum temperature- for vegetative and reproductive growth and
development of groundnuts. .. 24

Table 5.1 Origin, botanical type and relative heat tolerance of the four cultivars used
.. vanous contro 11ed envi
In .
environment expenments .41

Table 6.1 Cumulative total flower number plant", proportion of flowers setting pegs
and pods (fruit-set, angular transformed), total dry matter, pod harvest index and
pod yield of two Spanish (ICGV 86015 and 796) and two Virginia (ICGV
87282 and 47-16) cultivars. Data are the means of two temperature treatments
and five replicates 56

Table 6.2 Number of pegs and pods (reproductive number plant"), total dry matter
plant", pod harvest index and root-to-shoot ratio of two Spanish (ICGV 86015
and 796) and two Virginia (ICGV 87282 and 47-16) cultivars grown at mean
day/night optimum (25°C) or hot (30°C) air temperatures from 21 to 90 DAS.
Data are the means of five replicates 60

Table 6.3 Proportion of flowers setting pegs and pods (fruit-set, angular transformed),
number of pegs and pods planfl(reproductive numbers), total dry matter plant",
pod harvest index, root to shoot ratio and 100 seed weight of groundnuts grown
at mean day/night optimum air (25°C) or hot air (30°C) or at mean day/night
ambient soil (25°C) or hot soil temperature (34°C). Data are the mean of
cultivars and five replicates 62

Table 7.1 Total dry weight, pod harvest index, total flower number, proportion of
flowers setting pegs (fruit-set), and the number of pods in the Spanish cv. ICGV
86015 and Virginia cv. ICGV 87282. Data are means of temperature and
transfer treatments 78

Table 7.2 Total dry weight, pod harvest index, total flower number, proportion of
flowers setting pegs (fruit-set), proportion of pegs forming pods, and the
number of pods in the optimum (28°122°C) and hot temperature (38°/22°C)
control treatments. Data are means of cultivars and transfer treatments ......... 80
 X111

Table 7.3 Number of flowers produced during the 6 d 'stress' period and in the 6 d
following the 'stress' period in optimum (28°/22°C) and hot (38°/22°C)
temperature control and reciprocal transfer treatments. Data are means of
cultivars and transfer treatments 83

Table 8.1 Mean day and night air temperatures eC) in the poly-tunnel from sowing
to 37 d after sowing (DAS) and from 44 to 52 DAS, and in the four 6 d
temperature treatments imposed in the growth cabinets between 38 to 43 DAS .
........ 92

Table 8.2 Effect of2, 4, or 6 d duration of exposure to hot days of 34°, 42° or 48°C
on the numbers of pegs and pods produced during the 6 d treatment period (RNt
plant"). Control plants were maintained at 28°C for 6 d 98

Table 8.3 Effect of2, 4, or 6 d duration of exposure to hot days of 34°, 42° or 48°C
on the proportion of flowers producing fruits (fruit-set, angular transformed)
during the 6 d treatment period. Control plants were maintained at 28°C for 6 d.
....................................................................... 100

Table 9.1 Mean day and night air temperatures eC) in the poly-tunnel from planting
to 36 d after sowing (DAS) and from 43 to 53 DAS, and in the eight 6 d
temperature treatments imposed in the growth cabinets between 37 to 42 DAS .
........ 111

Table 9.2 Effect of night temperature on the cumulative number of flowers opened
during the 6 d stress period (FN), the proportion of those flowers which set pegs
or pods (fruit-set, angular transformed), and the number of pegs and pods (RNt),
and pollen production flower" and proportion of viable pollen (angular
transformed) on day 6 of the stress period in groundnut. Data are the means of
four day temperature treatments and four replicates. Standard errors are given in
parentheses 119

Table 10.1 Total per plant values of dry matter production, pod yield, pod harvest
index, root-to-shoot ratio, specific leaf nitrogen (SLN), nodule number and
nodule dry weight at reproductive maturity (90 DAS) in different N-source
treatments. Data are the means of air and soil temperatures and replicates .
........ 136

Table 10.2 Total per plant values of dry matter production, pod yield, pod harvest
index, root-to-shoot ratio, specific leaf nitrogen (SLN) and 100 seed weight at
reproductive maturity (90 DAS) in different mean (day/night) air and soil
temperature treatments. Data are the means of N-sources and replicates within
each temperature regime 138
 XlV

Table 10.3 Number of nodules and nodule dry weight per plant at reproductive
maturity (90 DAS) in different mean (day/night) air and soil temperature
treatments. Data are the means of N-sources and replicates within each
temperature regime 139
 xv

LIST OF FIGURES

Fig. 1.1 Global distribution of the semi-arid tropics (SAT) 8

Fig. 2.1 Reproductive stages of groundnut showing (a) appearance oftirst flower, RI;
(b) pegging, R2; (c) podding, R3; (d) full pod, R4; (e) start of seeding, R5;
(f) full seed, R6; (g) mature seed, R7; and (h) harvest maturity, R8 15

Fig. 5.1 Photographs of (a) the polyethylene covered tunnel structure and
(b) groundnut plants growing in the pots inside the poly-tunnel. 43

Fig. 5.2 Schematic diagram showing the blackout facility in the polyethylene covered
tunnel structure. . 44

Fig. 5.3 Photographs showing the soil temperature control facility constructed inside
the polyethylene covered tunnel structure , 46

Fig. 6.1 Temporal sequences of the numbers of flowers produced per plant dol in
(a) the Spanish cv. ICGV 86015; (b) the Spanish cv. 796; (c) the Virginia cv.
ICGV 87282; and (d) the Virginia cv. 47-16, when each were grown in mean
day/night optimum (25°C, • ) or hot (30°C,O) air temperatures. The stages R2,
R3 and R5 show the appearance of first peg, pod and seed at optimum
temperature, respectively 57

Fig. 6.2 Numbers of flowers planrl(a), the proportion of flowers setting pegs and
pods (fruit-set, angular transformed) (b), and pod yield plant" (c) in Spanish and
Virginia botanical type cvs grown in mean day/night optimum (25°C, open bars)
or in hot (30°C, solid bars) air temperatures. Vertical bars are the SED for
comparing the two temperatures. . 59

Fig. 6.3 Numbers of flowers planrl(a), the proportion of pegs forming pods (peg to
pod ratio, angular transformed) (b), and pod yield plant" (c) in the Spanish cv.
ICGV 86015 and the Virginia cv. ICGV 87282 grown in mean day/night
optimum air (25°C, open bars) or in hot air (30°C, solid bars) and at mean
day/night ambient soil (25°C, open bars) or hot soil (34°C, solid bars)
temperature. Vertical bars are the SED for comparing the two temperatures .
..................................................................................................................... 63

Fig. 6.4 Relative difference (%) in total dry matter and pod yield plant" at mean
day/night hot air (30°C) or hot soil (34°C) or hot air and hot soil combined
temperature treatments in relation to optimum air and ambient soil temperature
control treatment. 67
 XVI

Fig. 7.1 Number of flowers produced per plant dol over time (d) from first flowering in
(a) the Spanish cv. ICGV 86015; and (b) the Virginia cv. ICGV 87282, were
grown continuously at 28°/22°C (.) and 38°/22°C (0). Vertical, bars are the
SED for comparing the two temperatures. The stages R2, R3 and R5 show the
appearance offirst peg, pod and seed respectively at 28°/22°C 79

Fig. 7.2 Effect of transferring plants of (a) the Spanish cv. ICGV 86015 and (b) the
Virginia cv. ICGV 87282 from 28°/22°C (OT) to 38°/22°C (HT) (.) and from
38°/22°C to 28°/22°C (0) at different times relative to flowering on the number
of pegs and reproductive dry weights. Horizontal dotted lines show the 95%
confidence intervals for comparing transfers with control (Con) means. Vertical
bars show SEDs for comparing transfer means. The stages RI, R2 and R3 show
the appearance of first flower, peg, and pod respectively at 28°/22°C 81

Fig. 7 .3 Variance of regressions of number of pegs relative to control values against

the cumulative number of flowers opening in each 6 d period relative to start of
temperature treatment in plants transferred (a) from 28°/22°C (OT) to 38°/22°C
(HT) and (b) from 38°/22°C to 28°122°C 86

Fig. 7.4 Relations between (a) total number of pegs relative to 28°122°C (OT) control
values and the number of flowers produced in the 6 d period following an
episode of 38°/22°C (HT) and (b) total number of pegs relative to 38°/22°C
control values and the number of flower produced in the 6 d period following an
episode of28°/22°C 87

Fig. 8.1 Relation between the number of pegs and pods (RNt plant") and the number
of flowers (FN plant") in plants exposed to day temperatures of 28° to 48°C for
2, 4 or 6 d. Fitted line: RNt=-6,41(±1.57) + 0.80(±O.07) FN, ~=0.94, n=IO,
P<O.OOI 99

Fig. 8.2 Effect of timing of exposure to hot temperature on (a) number of flowers (FN
plant"); (b) proportion of flowers setting fruits (fruit-set, angular transformed);
and (c) number of pegs and pods (RNt plant") produced during a 6 d treatment
period when plants were exposed to day temperatures of 28°, 34°, 42° or 48°C
for the whole 12 h day (0800 to 2000 h; WO, 0) or for only during the first 6 h
(0800 to 1400 h; AM, A), or during the second 6 h (1400 to 2000 h; PM, 0) of
the 12 h day. Vertical bars denote the SED fortreatments l0l

Fig. 8.3 Relations between (a) number of flowers (FN plant") and mean day (0800 to
2000 h) temperature; (b) percentage fruit-set (angular transformed) and mean
AM (0800 to 1400 h) temperature; and (c) number of pegs and pods (RNt
plant") and mean day (0800 to 2000 h) temperature during a 6 d treatment
period when plants were exposed to temperatures of 28°, 34°,42° or 48°C for
2, 4 or 6 d periods (0), or exposed only during the first 6 h (0800 to 1400 h;
 XV11

AM, A} or the second 6 h (1400 to 2000 h; PM,D) of a 12 h light period. The

solid symbols in (c) represent values where AM temperature was >37.3°C, while
the open symbols represent the values where AM temperature was S 37.3°C.
Fitted line in (a): y=68.8(±5.35)-1.3(±O.15)x; n=16, r= 0.84; P<O.OOl and in
(b): y=2.29 (±O.024)-0.048 (±O.0055)x; n=16, r=0.95; P<O.OOI. 103

Fig. 9.1 Relation between mean day temperature eC) and (a) number of flowers
opened during the 6 d stress period (FN); (b) proportion of those flowers which
set pegs or pods (fruit-set, angular transformed); and (c) number of pegs and
pods (RN.) of groundnut cv. ICGV 86015 grown at night temperatures of 22°
(e) or 28°C (0). Fitted lines (a) y=58.29 (±5.16)-1.07(±O.133x, r=0.93,
P<O.Ol; (b) at 22°C: e, y=137.8(±14.2)-2.81(±O.36)x, r=0.93, P<O.Ol and at
28°C: 0, y=123.5(±14.3)-2.81(±O.36)x, ~=0.92, P<O.Ol; and (c) at 22°C: .,
y=40.21(±1.81}-0.856(±O.046}x, r=0.98, P<O.OI and at 28°C: 0, y=36.42
(±O.66)-0.856(±O.046)x, ~=0.98, P<O.Ol. Vertical bars denote standard errors
and are shown where they exceed the size of the symbol. 117

Fig. 9.2 Relation between fruit-set (angular transformed) and (a) pollen production
flower" and (b) pollen viability (angular transformed) at a night temperature of
22° (e) or 28°C (0). Fitted lines: pollen production y= -25.1(±5.2}+0.01O
(±O.OOl) x, r=0.95, n=8, P<O.OI; pollen viability y= -69.l(±9.73} + 1.73(±O.17}
x, r=0.94, n=8, P<O.OI 121

Fig. 9.3 Effect of day temperatures of 28° (e), 34° (O), 42° (A) and 48°C (A) on
(a) pollen production flower"; and (b) pollen viability (angular transformed)
over time during the 6 d stress period. The value of pollen viability at 4 dafter
the start of the temperature treatment at 48°C day temperature is from one
replicate only. Vertical bars denote the standard errors and are shown where
they exceed the size of the symbol. : 122

Fig. 9.4 Relation between pollen production flower" and cumulative day temperature
(>34°C) after the start of the 6 d temperature treatment at a night temperature
of 22° (e) and 28°C (0). Fitted line: y= 9055(±490) -164 (±14.2) X, r=0.88,
n=18, P<O.OOl 125

Fig. 10.1 Relation between pod yield (g plant") and specific leaf nitrogen (SLN, g N
o2
m ) when plants dependent on symbiotic N2 (squares), symbiotic N2 plus 20 N
(circles) and inorganic N (triangles) were grown at optimum air and ambient soil
(0 ,0 ,A ), hot air ( IJ , () , A ), hot soil ( [I , () , 4.) or hot air and hot
soil ( • , • , A ) temperatures from first flower appearance (28 DAS) to
reproductive maturity (90 DAS) 141
 XVlll

Fig. 11.1 Relative difference (%) in pod yield at mean day/night hot air (30°C) or hot
soil (34°C) or hot air and hot soil combined temperature treatments (30o/34°C)
in relation to values at an optimum temperature of 25°C, when temperature
treatments were imposed from flower bud initiation (Exp. 1), or from first
flower appearance (Exp. 2) or from pod initiation (Exp. 3) until reproductive
maturity at 90 DAS 158

Fig. 11.2 Relative difference (%) in pod yield when plants were exposed to 6 d of
mean day/night air temperature of 30°C at different times relative to the onset of
flowering in relation to a mean day/night air temperature of 25°C, as recorded in
the current controlled environment studies (.) and as simulated by the
PNUTGRO model (0) 163
 1

CHAPTER 1 GENERAL INTRODUCTION

The groundnut or peanut (Arachis hypogaea L.) is one of the important grain legume

crops predominantly grown in the tropical and the semi-arid tropical countries as a

principal source of edible oil and vegetable protein (Bunting et al., 1985).

Groundnuts are cultivated between 400N to 40° S of the equator although yields vary

enormously from about 3000 kg pods ha" in the USA, to 1500 kg ha" in South

America and Europe and <800 kg ha"l in the developing world of Asia and Africa

(Summerfield and Roberts, 1985). Temperature extremes, especially hot air and hot

soil temperatures, together with water deficits are the major environmental factors

influencing the productivity of groundnuts. Ironically, research on heat stress on

groundnut growth, development, nitrogen nutrition and yield have received less

attention and are not well understood (Williams and Boote, 1995).

Groundnut crops grown in the semi-arid tropics are often exposed to soil and

air temperatures as hot as 40°C and >35°C, respectively (ICRISAT, 1994). Further,

with the present trend of global warming, temperatures are likely to become hotter,

perhaps by as much as 3° to 4°C (Houghton et al., 1990). This predicted increase in

air temperature would further reduce the productivity of major food crops (parry et

al., 1990; Warrick, 1990). For groundnut, it has been predicted that increases in

current mean temperatures by 2° to 3°C would reduce the seed yield by 23-36 %

(Hundal and Kaur, 1996).

Hot temperatures are a major constraint to crop adaptation and productivity,

especially when temperature extremes coincide with the critical stages of plant

development (rvicWiIliam, 1980). It is also evident that many plant species have
 2

evolved capabilities to protect themselves against these extreme events, either by

metabolic alterations or physiological changes. Some plant species and cultivars

within crops have better heat tolerance than others (Alexandrov, 1964; Berry and

Bjorkman, 1980). Thus, it is of prime importance to identify those cultivars and to

use them as parents in breeding programmes to develop heat-tolerant cuItivars for

hostile climates of the semi-arid tropics.

In order to identify and screen cuItivars for heat tolerance and/or to develop

suitable agronomic management technologies to reduce the effects of heat stress, it is

essential:

1. To understand the effects of heat stress on crop growth and development;

2. To identify the stage of crop development most sensitive to heat stress; and

3. To determine the method (i.e., intensity, duration and timing) of imposing heat

stress in any quantified screening method.

This research seeks to understand and quantify the effects of heat stress on

reproductive growth and development in groundnut. It also looks to define the

reproductive growth phase most sensitive to heat stress and to develop a method that

can be used for screening and evaluating cuItivars for heat-tolerance under controlled

environment conditions.

1.1 ORIGIN AND DISTRIBUTION

The cultivated groundnut originated in South America (Krapovickas, 1969; Gregory

et al., 1980). The term Arachis is derived from the Greek word "arachos", meaning a

weed, and hypogaea, meaning underground chamber, i.e. in botanical terms, a weed

with fruits produced below the soil surface (Stalker and Simpson, 1995).
 3

The earliest archaeological records of groundnuts in cultivation are from

Peru, dated 3750-3900 years before the present (YBP) (Hammons, 1994).

Groundnuts were widely dispersed through South and Central America by the time

Europeans reached the continent, probably by the Arawak Indians, and there is

archaeological evidence from Mexi,?o,dated 1300-2200 YBP (Krapovickas, 1969).

After European contact, groundnuts were dispersed world-wide. The

Peruvian runner type was taken to the Western Pacific, China, Southeast Asia and

Madagascar. The Spanish probably introduced the Virginia type to Mexico, via the

Philippines, in the sixteenth century. The Portuguese then took it to Africa, and later

to India, via Brazil. Virginia types apparently reached the Southeast USA with the

slave trade. Gibbons et al. (1972) noted substantial secondary diversity in Africa and

Asia; the types they found and their locations supported these various conjectures

regarding dispersal.

1.2 TAXONOMY AND CLASSIFICATION

The genus Arachis belongs to family Fabaceae, subfamily Papilionaceae, tribe

Aeschynomeneae, subtribe Stylosanthinae (Rudd, 1981). This genus is

morphologically well defined and distinguished from other genera by having a peg

and geocarpic reproductive growth. The genus Arachis has more than 70 wild

species, of which only Arachis hypogaea L. is domesticated and commonly cultivated

(Ramanatha Rao, 1988).

The taxonomy of the genus Arachis has been described by Gregory and

Gregory (1976), Gregory et al. (1980) and by Smartt (1990). It includes 37 named

species and a number of undescribed species. The genus has been divided to nine
 4

sections (Krapovickas and Gregory, 1994), i.e., Arachis, Caulorrhizae, Erectoides,

Extranervosae, Heteranthae, Procumbentes, Rhizomatosae, Trierectoides and

Triseminalae. The section Arachis comprises annual and perennial diploid (2n = 20)
and two annual tetraploids (2n = 4x = 40). The leaves of Arachis hypogaea L. are

tetrafoliolate, and plants are typically erect or decumbent and pegs penetrate the soil

at an angle of approximately 45°.

Most of the earlier classifications of Arachis hypogaea L. were based on

growth habit, presence or absence of seed dormancy and relative time to maturity

(Ramanatha Rao, 1988). In later classifications, characters such as branching pattern

and location of reproductive branches have been included (Gregory et al., 1951;

Bunting, 1955, 1958; Krapovickas and Rigoni, 1960; and Smartt, 1961).

The classification of Gregory et al. (1951) was based on a comprehensive

study in which the cultivated groundnuts were divided into two large botanical

groups, Virginia and Spanish-Valencia, on the basis of branching pattern. Bunting

(1955) named two basic types of branching as "alternate" and "sequential" and

suggested that the cultivar groups within the two branching patterns should be

considered as subspecies.

In the Virginia group, the main stem does not have reproductive axes.

Alternating pairs of vegetative and reproductive axes are borne on the cotyledonary

laterals and on other n+ 1 branches (where 'n' is the main axis, and primary,

secondary and tertiary branches are n+l, n+2 and n+3, respectively). This system was

termed the 'alternate branching pattern' by Bunting (1955). The first two branches on

the n+ 1 laterals are always vegetative and the alternate branching pattern is repeated

in the higher order branches.

 5

In the Spanish-Valencia group, reproductive branches are borne in a

continuous series on successive nodes of the cotyledonary and other lateral branches,

on which the first branch is always reproductive (termed the 'sequential branching

pattern' by Bunting, 1955). Reproductive branches are also borne directly on the

main axis at higher nodes. Most n+~ and n+ 3 nodes are reproductive.

Krapovickas and Rigoni (1960) proposed the subdivision of Arachis

hypogaea L. into two subspecies: A. hypogaea subsp. hypogaea and A. hypogaea

subsp. fastigiata. Subspecies hypogaea has a central axis that never bears

inflorescences and has lateral branches where vegetative branches alternate regularly

with reproductive branches. The inflorescence is simple, seeds show dormancy and

plants are late maturing (120 to 150 d depending primarily on temperature and crop

density). In general these types branch profusely, have a runner or spreading bunch

habit. The USA market types Virginia and Virginia Runner (var. hypogaea), and the

distinct variety hirsuta belong to this group.

Arachis hypogaea subsp. fastigiata comprises plants that are always erect,

with inflorescence on the central axis, and without a regular pattern in the sequence

of reproductive and vegetative branches. The inflorescence is simple (var. fastigiatdy

or compound (var. vulgaris), pods are concentrated around the central axis, and

seeds do not show dormancy and the plants are early maturing (90 to 120 d). In

general these types are sparsely branched and have an erect bunch habit. The USA

market types Valencia and Spanish belong to this group.

 6

1.3 GROUNDNUT PRODUCTION AND PRODUCTIVITY WORLD-WIDE

The world groundnut (in shell) harvested area in 1997 was 23.5 million ha with a

total production of30 million mt (FAO, 1998) (Table 1.1). Total harvested area in

1997 increased by 3.7 million ha when compared to that of 1990, while production

increased by 7 million mt. The world's average productivity was about 1274 kg ha".

It is cultivated in close to 90 countries and requires tropical, subtropical, or warm

temperate climates for optimum production. Groundnut is therefore an oilseed crop

on a global scale. It grows best in regions, which lie between 400N and 400S

(Virmani and Singh, 1986). The most productive soils are light, fiiable, well-drained

sandy loams (for ease of harvesting), but the crop will also grow in heavier soils.

Groundnuts are predominantly grown in the developing countries of Asia and Africa.

About 90% of the total world production come from this region and about 60% of

production come from the semi-arid tropics (SAT) (Fig. 1.1). Roughly two-thirds of

this is used for oil, making it the second most important source of vegetable oil after

soyabean (Glycine max (L.) Merr.; Bunting et al., 1985).

1.3.1 Asia

Asian countries have the largest area of groundnuts in the world (58%) contributing

67% of the total production. India has the largest acreage (8.1 million ha) followed

by China (3.8 million ha), Indonesia, Myanmar, Thailand, Taiwan, Japan, Pakistan

and Korea. There has been an increase in the harvested area by 28% in Asia, with the

East Asia sub-region (comprising China, Hong Kong, Japan, Korea and Taiwan)

contributing more than 50% to this increase. More than 34% of the groundnut area
 7

Table 1.1 Harvest area, yield and production of groundnuts (in shell) in
different parts of the \Vorld (FAO, 1998).

Country/Continent Harvest area Yield Production

(1000 ha) (kg ha-l) (1000 mt)

..
\Vorld 23520 1274 29970

Asia 13620 1483 20190

India 8100 988 8000
China 3750 2590 9720
Iricionesia 630 1560 980

Africa 8710 841 7320

Nigeria 2250 1124 2531
Sudan 1380 761 1051
Senegal 800 632 506

N. America
USA 570 2809 1604

S. America 450 1368 616

Argentina 298 1351 403
Brazil 88 1598 140
Paraguay 24 1291 31

Europe 11 1229 14
Bulgaria 10 1000 10
Greece 0.7 4126 3

Yugoslavia 0.2 1250 0.25

 8

/
/
-.--
 9

harvested of the world is in India followed by 16% in China. The average

productivity of this region is 1483 kg ha".

In India the important groundnut growing states are Gujarat, Andhra Pradesh,

Tamil Nadu, Karnataka, Maharashtra, Madhya Pradesh, Uttar Pradesh and Rajasthan.

It is grown in all three seasons - rainy, post-rainy, and during summer months. It is

mostly grown under rainfed conditions, and only about 10-15% of the area is

irrigated.

1.3.2 Africa

In Africa groundnut is grown mainly in Nigeria, Sudan, Senegal, Congo and

Cameroon. The total harvested area in Africa is 8.7 million ha with a total production

of 7.3 million mt. Average productivity in this region is 841 kg ha-l, which is poor

when compared to the USA where it is close to 2800 kg ha-l. Average productivity in

of Nigeria is 1124 kg ha"I, in Sudan it is 760 kg ha", while in Malawi, Senegal and

Congo it is close to 650 kg ha".

1.3.3 North America

The total groundnut harvested area in the USA is 0.57 million ha, with a total

production of 1.6 million mt. The average productivity of this region (2800 kg ha") is

1500 kg ha-l above the world average. Production is mainly concentrated in three

major geographic areas: the Southeast, which includes Alabama, Florida, Georgia and

South Carolina; the Southwest, which includes New Mexico, Oklahoma and Texas;

and Virginia-Carolina, which includes North Carolina and Virginia. The largest area

is found in Georgia, followed by North Carolina and Alabama.

 10

1.3.4 South America

The major countries growing groundnut in South America are Argentina, Brazil,

Paraguay and Bolivia. It is grown in an area of0.45 million ha, with a production of

0.62 million mt. The average productivity of this region is 1368 kg ha". Argentina is

the major groundnut growing country in this region, contributing 66% of area and

production. The SAT growing region in this continent is in Brazil, between 19° and

25°S.

1.3.5 Europe

The total production of groundnut in Europe is 12,286 mt, which comes from an

harvest area of 11,106 ha with an average productivity of 1229 kg ha". The major

groundnut growing country is Bulgaria followed by Greece, Spain, Yugoslavia, Italy

and Portugal. There has been a significant increase in the total groundnut area

harvested in this region in the last decade from 11,718 ha in 1980 to 13,301 ha in

1998 (i.e. by 17 %).

1.4 UTILIZA nON

Groundnuts are an important foodstuff in the SAT and are a subsistence food crop

throughout the tropics (Bunting et al., 1985). They are mainly grown for the kernels,

and the edible oil and meal derived from them and the vegetative residue (haulms).

Groundnut kernels typically contain 47-53% oil and 25-36% protein. Groundnuts are

used in various forms in SAT countries, which include groundnut oil, roasted and

salted groundnut, boiled or raw groundnut or as paste popularly known as groundnut

 11

(or peanut) butter (Singh, 1992). The tender leaves are used in certain parts of West

Africa as the vegetable in soups.

Groundnut oil is the most important product of the crop, which is used for

domestic and industrial purposes. Groundnut oil is the cheapest and most extensively

used vegetable oil in India. It is use~ mainly for cooking, for margarine and vegetable

ghee, salads, for deep-frying, for shortening in pastries and bread, for pharmaceutical

and cosmetic products, as a lubricant and emulsion for insecticides and as a fuel for

diesel engines (Duke, 1981). The press cake containing 40-50% protein is used

mainly as a high-protein livestock feed and as a fertiliser. The dry pericarp of the

mature pods (known as shells or husks) are used for fuel, as a soil conditioner, filler

in fertilisers and feeds, or are processed as substitute for cork or hardboard or

composting with the aid of lignin decomposing bacteria (Adams and Hartzog, 1980).

The foliage of the crop also serves as silage and forage.

1.S CONSTRAINTS TO PRODUCTIVITY IN THE SAT

The productivity of groundnut in the SAT regions of Asia and Africa is very Iow

« 700 kg ha") when compared to world's average of 1274 kg ha". This is due to

various abiotic and biotic constraints.

Abiotic factors of prime importance include temperature extremes, water

stress, soil factors such as alkalinity, poor soil fertility and nutrient deficiencies.

However, the major yield limiting factors in the SAT are water stress and heat stress.

Groundnuts grown in this region are often exposed to damaging hot soil and hot air

temperatures of 40°C during parts of their growing season. Water deficits, when they

occur at critical stages can reduce pod yield by >70% (Nageswara Rao et al., 1989;
 12

Wright and Nageswara Rao, 1994). The influence of abiotic stresses is complex given

that they are often confounded with each other e.g. hot temperatures are often

associated with water deficits and high light intensities. Heat stress and water stress

significantly affect physiological processes, e.g. inhibition of photosynthesis,

disruption of respiration, change~ in membrane permeability, and interfere with

nutrient mobility (Kvien, 1995).

Edaphic factors such as alkalinity and salt stresses are also known to reduce

pod yields. Alkaline, lime rich soil causes iron chlorosis, which leads to yield losses of

the order of 30% (potdar and Anders, 1995). In many groundnut growing regions

evaporation often exceeds precipitation and both irrigation water and soil are

moderately saline, resulting in the accumulation of salt in the pod zone of the soil

profile. This can decrease dinitrogen fixation and vegetative growth and increase the

percentage loss of immature pods thus leading to smaller yields (Kvien, 1995).

Groundnuts when effectively nodulated seldom respond to applications of nitrogen.

Symptoms ofP and K deficiency are rare. However, the crop has a large demand for

Ca, and any deficiency can reduce productivity by as much as 30% (Cox et al.,

1982). Constraints to podding also include Fe and S deficiency and aluminium and

zinc toxicity (Cox and Sholar, 1995).

Biotic constraints to yield include insect pests, diseases and weeds (Stalker,

1997). The impact of pests and diseases in the SAT reflects the use of locally grown

cultivars, which apart from having poor yield potential also lack resistance to diseases

and insects. The most serious pests in India include Spodoptera litura, Aproaerema

modicella, and Helicoverpa armigera. In certain areas the two-spotted spider mite

(Tetranychus sp.) is widespread and can cause severe yield losses, particularly when

groundnuts are grown in light, sandy soils that become drought stressed. The
 13

populations of spider mite can build up rapidly, particularly if insecticide sprays kill

natural predators. Intercellular feeders such as aphids (Aphis craccivora), thrips

(Frankliniella sp.) and leafhoppers (Empoasca sp.), and soil pests like termites

(Microtermes sp. and Odontotermes sp.) and white grubs (Eulepida marshona and

Lachnostema sp.) are also important pests in some of the SAT areas (Gibbons,

1986).

Groundnuts are also often attacked by fungi, bacteria and viruses. The most

common fungal pathogens known to drastically reduce yield and/or quality of the

crop include leaf spots (Cercospora sp.), rusts (Puccinia arachidis) and toxin

producing Aspergillus flavus. Then again, the soil-borne diseases stem rot

(Sclerotium rolfsii) and pod rot (Phythium sp.) also pose serious problems in some

areas. Root-knot nematodes (Meloidogyne sp.) are also an important yield limiting

factor. Bacterial disease caused by Pseudomonas solanacearum is economically

important and widespread in China and Indonesia (Gibbons, 1986). The important

viral disease is seed-borne peanut mottle virus (PMV), which is found in most of

groundnut growing countries. Finally, weeds are important biotic factors causing

significant yield losses; they compete with the crop for space, nutrients, light and

water, and also act as alternative hosts for various pests and diseases. Nutsedge

(Cyperus sp.), morning glory (Ipomea sp.), pigweed (Amaranthus sp.) and crabgrass

(Digitaria sp.) all cause significant yield losses in groundnuts (Brecke, 1995).

Among the various production constraints cited above environmental

constraints like water stress and hot air and soil temperatures are the major targets of

research in the SAT. These environmental constraints are of more concern because of

the likely effects of global warming and climate change on agriculture (Houghton et

al., 1990; Parry et al., 1990).

 14

CHAPTER 2 GROWTH, DEVELOPMENT, NITROGEN

NUTRITION AND SEED YIELD OF GROUNDNUT

2.1 GRO\VTH AND DEVELOPMENT

2.1.1 Growth stages

The growth stages of groundnut plants based on visual observations of vegetative

and reproductive growth have been described and defined by Boote (1982). This

widely adopted system describes a series of vegetative (V) and reproductive stages

(R), and all stages are discrete population-based events which are mostly determined

by field observations (Table 2.1). The reproductive stages are also shown in Fig 2.1.

Table 2.1 Growth stage descriptors for groundnut (Boote, 1982).

Stage Stage title Description

Vegetative stages
YE Emergence Cotyledons near the soil surface with the seedling partly visible
VO Cotyledons are flat and open at or below the soil surface
VI First tetrafoliolate One to N developed nodes on the main axis, a node is counted when its
tetrafoliolate is unfolded and its leaflets are flat
Reproductive stages
RI Beginning bloom One open flower at any node
R2 Beginning peg One elongated peg (gynophore)
R3 Beginning pod One peg in soil with turned swollen ovary at least twice the width of the
peg
R4 Full pod One pod fully expanded to dimensions characteristic for the cultivar
R5 Beginning seed One fully expanded pod in which seed cotyledon growth is visible when
the fruit is cut in cross-section
R6 Full seed One pod with cavity apparently filled by the seeds when fresh
R7 Beginning maturity One pod showing visible natural colouration or blotching of inner
pericarp or testa
R8 Harvest maturity 66-75% of all developed pods have testa or pericarp colouration.
R9 Over-mature pod One undamaged pod showing orange-tan colouration of the testa and/or
natural peg deterioration
15
 16

2.1.2 Seedling and vegetative growth

The groundnut seed consists of two cotyledons, a hypocotyl, epicotyl, and radicle.

All primordial leaves, which the seedling will develop within the first few days after

germination, are present in the seed (De Beer, 1963). Maeda (1972; 1973) found that
.
4-5 leaf primordia are present in the embryo of seed; five are well developed in large

seeds and four in small ones. Germination is epigeal, the cotyledons become green

soon after emergence. The seedling consists of cotyledons, vegetative axes, and the

main axis. The hypocotyl is white and is easily distinguished during the early stages of

growth, but becomes indistinguishable from the root as the plant matures.

Groundnuts take about 3-5 d for germination and emergence from the soil at

30D
e (Mohamed et al., 1988 a, b). The radicle emerges within 24 h or earlier for

vigorous Spanish types and within 36 to 48 h in Virginia types. The primary root

system is tap rooted but many lateral roots also appear about 3 dafter germination

(Gregory et al., 1973). Roots are concentrated in the 5 to 35 cm zone below the soil

surface, but penetrate the profile to the depth of 135 em (Intorzato and Tella, 1960).

Groundnut roots do not have typical root hairs, but rather tufts of hair, which are

produced in the root axils (Moss and Ramanatha Rao, 1995).

During the first few days the developing seedlings are dependent on

assimilates stored in cotyledons. After 5-10 d, depending upon cultivar and

envirorunental conditions, the seedling grows autotrophieally and is capable of

absorbing minerals via the roots whilst the epicotyl is exposed to light and capable of

photosynthesis. Stems are angular, green or pigmented and are initially solid, but as

the plants grow they tend to become somewhat hollow. The main stem develops from

a terminal bud of the epieotyl and two opposite cotyledonary laterals grow at soil
 17

level. The main stem can be upright or prostrate and from 12 to 35 cm long or may

exceed 1 m in runner types (Ramanatha Rao, 1988).

The growth and branching patterns differ between subspecies and botanical

types. Subspecies hypogaea has alternating pairs of vegetative and reproductive

nodes, while subsp. fastigiata has. a sequential pattern of reproductive nodes. The

early vegetative growth stage is mainly concerned with mainstem elongation and leaf

production, whereas the formation of lateral branches dominates later growth. Maeda

(1970) recorded that mainstem leaves account for >50% ofleaf area of plants for the

first 35 d, but at 90 d they account only for 10%. After flowering, dry matter

accumulation is mainly in reproductive structures.

2.1.3 Reproductive growth and maturity

Groundnut cultivars typically flower (RI) about 25 to 30 d after sowing depending

on cultivar and climatic conditions. The flowering pattern varies within and between

botanical types. The Spanish types flower relatively early and have a broader first

flowering peak whereas the Virginia types are later flowering and have multiple

flowering peaks. Cultivars within the subspecies also vary in their flowering patterns

(Seshadri 1962). Flowers are borne in the axils of leaves, usually with three flowers

per inflorescence. Generally one bud per inflorescence reaches anthesis on a given

day (Moss and Ramanatha Rao, 1995), but occasionally two or more buds may open

on the same day. Flower colour varies from yellow to orange to dark orange and

rarely white in colour. The style is contained within a calyx tube (hypanthium). The

bud is 6-10 mm long 24 h before anthesis and, during the day, the hypanthium

elongates slowly and the bud attains a length of 10-20 mm. During the night,
 18

elongation of the hypanthium is more rapid. The flower contains 10 anthers, five of

which are smaIl and globular and five are oblong. One of the anthers is usually sterile

and difficult to observe. The anther attains a maximum length of 5-7 mm at the time

of anthesis (Smith, 1950). Flowers open early in the morning as soon as they receive

light. The dehiscence of anther takes just before or when the flower opens or

sometimes much earlier (Bolhuis et al., 1965). The stigma is receptive from 24 h

before to 12 h after flower opening (Hassan and Srivastava, 1966). Groundnuts are

usually self-pollinated and pollination occurs just before the flowers open.

It has been observed that sporogenesis and gametogenesis occurs 3 to 6 d

prior to anthesis when buds are about 5 mm long (Xi, 1991; Martin et al., 1974).

Pollination occurs just before anthesis and after pollination the pollen tube grows at a

rate of 1 em h-t resulting in fertilization about 5-6 h after pollination (Lim and

Gumpil, 1984). This may vary with cultivar and environment.

After fertilization the flower withers and in doing so it activates the growth

and elongation of the intercalary meristem, which is located at the base of the ovary.

A stalk like structure, called a peg, becomes visible within 4-6 d after fertilization

under optimum environmental conditions. Peg extension is slow at first and takes

about 5-6 d to penetrate the bracts (Smith, 1956 a, b). Once pegs are 3-4 mm long

they become positively geotropic and start to grow towards the soil (R2, pegging

stage). The rate of elongation then increases rapidly between 5-10 d after fertilization

(periasamy and Sampoornam, 1984) and pegs can be as long as 15 cm. The peg bears

the ovary with the fertilized ovule at its tips. The peg typically reaches and penetrates

the soil surface in about 8-14 d after fertilization (Schenk, 1961). Once the peg enters

the soil and penetrates to a depth of 4-5 cm, the tip of the ovary begins to swell (R3,

podding stage) and turns horizontally away from the base of the plant and develops
 19

into a pod. The time from RI to R3 is usually 15 to 20 d (Boote, 1982), after which

the pod begins to expand rapidly until it reaches dimensions characteristic of the

cultivar. The R4 stage (first full pod stage) is defined as the time when 50% of the

plants have achieved fully expanded pods.

The number of flowers produced per day increases gradually; maximum

numbers typically appear at 14-28 d after flower initiation and numbers then decline

to zero during the pod filling stage (Smith, 1954; Williams et al., 1975a; Young et

al., 1979). However, cultivars within and between botanical types may vary in their

flowering behaviour. Pods become countable and pod weights become measurable at

about 60-70 d after planting. Pod number per plant rises rapidly to a maximum at 80-

120 d (Daughtry et al., 1975; Williams et al., 1975b) depending upon cultivar and

botanical type. Schenk (1961) reported that the fresh weight of the whole pod

increased very rapidly during the first 14 d of subterranean growth and that pods

attained their maximum size after 21 d. During the seed growth phase (R5 growth

stage), when seed cotyledon growth is visible in at least one pod on 50% of the

plants, the endocarp recedes as the ovule grows and has disappeared completely by

the time seeds are mature. During this period the inner phase of the shell is darkened

by tannin deposition and turns dark brown on maturation (Gregory et al., 1973). Pod

growth rates differ among cultivars (Duncan et al., 1978), and are affected by

temperature of the fruiting zone (Dreyer et al., 1981). Pod growth rates slow down

as pods mature (Boote and Ketring, 1990).

 20

2.2 NODULE FORMATION AND NITROGEN FIXA nON

Nodule initiation and subsequent growth and function is an interactive process

involving the eukaryotic host legume and the prokaryotic Rhizobium. The process is

complex, resulting in biochemical and morphological changes in both symbionts and

leading to the capacity of reducing atmospheric nitrogen (Elkan, 1995). Nodule

formation occurs generally in three sequential steps i.e., root colonisation and

infection, multiplication of bacteria and development of nodules.

2.2.1 Root colonisation and infection

Groundnuts are nodulated by a wide range of species of rhizobia. Proliferation of

compatible rhizobia in the rhizosphere of legumes is the first step towards nodule

formation (Broughton, 1978). Toomsan (1990) discovered that there is no clear

specificity in groundnut nodulation and root colonisation.

In most legumes rhizobia enter through root hairs via an "infection thread",

but in groundnut roots the invasion process is rather different. Normal root hairs are

absent (Gregory et al., 1973); instead tufted rosettes of hairs are found in the

junctions of root axils. It is at these junctions where nodulation occur (Allen and

Allen 1940; Inouye and Maeda, 1954).

2.2.2 Nodule development

Once the rhizobia have entered the root and occupied the space between the root hair

wall and the adjoining epidermal and cortical cells, the cells adjacent to the point of

Rhizobium penetration separate at their middle lamellas and the resultant spaces
 21

become filled with bacteria, forming intercellular zones of infection (Nambiar, 1988).

The bacteria penetrate into progressively deeper cell layers and intracellular infection

then occurs. Soon after intracellular infection, the bacteria multiply rapidly. Further

development of the nodule occurs by repeated division of the infected host cells, and

the bacteria become transformed into different morphological forms known as

bacteriods (Chandler, 1978). Staphorst and Strijdom (1972) have observed that

because the size of bacteriods is relatively large, the number of bacteriods per unit

nodule weight in groundnut is small when compared to other tropical legumes such

as cowpea (Vigna unguiculata L. Walp.). The initial spherical shape of nodules of

legumes can develop into various morphological variants, but in groundnut they

remain more or less spherical throughout.

2.2.3 Nodule function

The bacteriods contain the nitrogenase enzyme which reduces gaseous nitrogen. The

effectiveness of the nodule to fix nitrogen depends on the presence of

legheamoglobin, which gives a pink colouration to the nodule tissue. Schiffmann and

Lobel (1973) suggested that quantitative determination of legheamoglobin during

certain stages of plant growth (e.g. flowering) could be used as an indicator of

dinitrogen fixation in groundnuts. The host plant provides energy in the form of

carbohydrates for dinitrogen fixation and the carbon skeleton for the assimilation of

reduced nitrogen (NH4). The fixed nitrogen is transferred to shoots mainly in the

form of a-methylene glutamine (Fowden, 1954). Not all the rhizobia that produce

nodules fix nitrogen. It is important therefore to select good inoculant strains for
 22

survival in rhizosphere, competition with indigenous strains, infectivity and fixation

potential. Such information on groundnuts is very much lacking.

 23

CHAPTER 3 EFFECT OF TEMPERATURE ON GROWTH,

DEVELOPMENT, NITROGEN NUTRITION AND SEED YIELD

Temperature plays an important role in all aspects of plant growth and development.
.
It is a major environmental factor that determines the rate of groundnut crop

development (Ketring and Reid, 1995). Heat stress and water stress are the major

environmental factors limiting pod yields in groundnut (ICRISAT, 1994).

Groundnuts grown in the SAT are often exposed to temperatures short episodes or

continuous periods of air and soil temperature >35°C. Groundnuts are susceptible to

both hot air and hot soil temperatures because of aerial flowering and subterranean

fruiting habit of the crop. Accordingly, the research reported in this thesis

investigates the effects of heat stress on groundnut. Therefore, the effects of hot

temperatures on growth and development, dry matter production, partitioning of dry

matter and pod yields on groundnuts are reviewed here. The optimal temperature for

different aspects of growth and development are given in Table 3.1.

3.1 EFFECT OF TEMPERATURE ON GROWTH, DEVELOPMENT AND

SEED YIELD

3.1.1 Seed germination and seedling emergence

The optimal temperature for germination of 15 cultivars was in the range of 28° to

36°C (Mohamed et al., 1988a), which is similar to the values reported by Mills

(1964). Mohamed et al. (1988a) found that for seed germination of groundnuts base

temperature (Tb) ranges from 8° to 11.5 "C, optimum temperature (To) ranges
 24

Table 3.1 Optimum temperature for vegetative and reproductive growth and
development of groundnuts.

Optimum

Trait temperature Reference

Seed germination 28° to 30°C Mohamed et al., 1988

Seedling growth 28°C Leong and Ong, 1983

Leaf appearance and 28° to 30°C Fortanier, 1957; Suzuki, 1966; Cox,
leaf area development 1979; Leong and Ong, 1983; Ketring,
1984a; Ahring et al., 1987

Branching and Leong and Ong, 1983

stem growth

Flower production Fortanier, 1957; Wood, 1968; Cox,

1979;

Pod formation 23° to 26°C Williams et al., 1975; Cox, 1979;

pod growth and Dreyer et al., 1981; Ketring, 1984a

seed yields

Root growth 25°C Ahring et al., 1987

Nitrogen fixation 25°C Nambiar and Dart, 1983

 25

from 29° to 36.5°C, and maximum temperature (Tm) ranges from 41° to 47°C.

Temperature hotter than 36°C led to poorer rates of germination. The germination of

seeds generally commenced from 27 h after sowing within the range of temperature

from 25° to 41°C.

The optimum temperature for the germination and seedling development is

close to 30°C (Leong and Ong, 1983; Ketring and Reid, 1995). In another study

groundnut cultivars emerged between 4 to 6 d after sowing at 23°C, while at 27°C it

was about one day sooner (Mohamed et al., 1988b). In terms of thermal time all

cultivars emerged between 74 and 101°Cd (T, = lOOC) at 27°C and between 68 to

92° Cd at 23°C.

3.1.2 Vegetative growth

Temperature significantly influences the vegetative growth of the groundnuts. The

optimum temperature range for leaf and stem growth is 28° to 30°C (Fortanier, 1957;

Bolhuis and De Groot, 1959; Cox, 1979). Studies on stands of cv. Robut 33-1 have

shown that the rate of foliage development (i.e. the appearance and expansion of

leaves and the appearance and extension of branches) increases to a maximum at

28° to 30°C (Leong and Ong, 1983).

Under field conditions in Zimbabwe groundnut crop growth rate, leaf area

and total dry matter production were maximal at a site with mean daily maximum and

minimum temperature of 30° and 17°C, respectively (Williams et al., 1975 a, b). The

rate of crop photosynthesis was remarkably conservative over a range of mean air

temperature from 19° to 30°C and groundnuts attain their maximum leaf apparent
 26

photosynthesis at 30°C and this was reduced by 25% at 40°C (Bhagsari et al., 1976;

Ketring et al., 1982a).

When groundnut cv. Florigiant was grown at day/night temperatures ranging

from 34°/30°C to 18°/14°C, early growth, as determined by the accumulation of

shoot dry matter was greatest at 3Qo/26°C and virtually nil at 18°/14°C (Cox, 1979).

Further increase in temperature to 34°/30°C significantly reduced dry matter

production, which suggests that the optimum temperature regime for vegetative

growth is 300/26°C (Cox, 1979). Ketring (1984a) found that long duration of

exposure to hot temperatures (up to 35°C) reduced the number ofleaves, leaf growth

rate, specific leaf area and total leaf area per plant when compared to values obtained

at 30° or 32°C. It was concluded from these experiments that a day temperature of

35°C is supra-optimal for growth and development even under well watered

conditions.

Research has also shown that groundnuts accumulate less total biomass

whenever the night temperatures are consistently at or below 16°C. Several workers

have indicated that dry matter accumulation of groundnuts ceases at temperatures

cooler than 14°C (Mills, 1964; Cox and Martin, 1974; Ono et al., 1974; and Angus et

al., 1981). Similarly, controlled-environment experiments on groundnut cv. Florunner

have shown that the night temperature was negatively correlated with rates of growth

during germination and the vegetative growth stage. As the day/night temperature

decreased from 32°/26°C to l7°/HoC the duration of the germination period and the

vegetative stage increased from 31 to 75 d (Campbell, 1980).

During the early stages of growth, the optimum shoot growth of Spanish

cultivar Comet occurred at soil temperature between 31° and 37°C, whereas
 27

optimum root growth occurred between 25° and 31°C (Ahring et al., 1987). Suzuki

(1966) reported that the stern length and root dry matter increased with increasing

soil temperatures from 20° to 30°C.

3.1.3 Reproductive growth

The reproductive phases i.e., bud development, flowering, pegging, pod development

and pod growth in groundnuts are strongly influenced by temperature and highly

sensitive to heat stress.

3.1.3.1 Bud development

Hot temperatures are known to damage the flower buds of many crops including

groundnut (Tal war, 1997), cowpea (Mutters et al., 1989 a, b), and tomato

(Lycopersicum esculentum L.) (Kuo et al., 1986) resulting in bud abortion and male

sterility.

Groundnut buds of three sizes i.e., 12-15 mm [1 d before anthesis (DBA)],

9-10 mm (3 DBA) and below 7 mm (5 DBA) were tagged and exposed to hot

temperatures (35°/30°C) and the effects on fruit-set was recorded (Talwar, 1997). It

was observed that there was no effect of hot temperature on flower opening but fruit-

set was reduced when buds at the earlier stage of development (3 to 5 DBA)

experienced heat stress. Furthermore, cultivar differences were exposed in that, at hot

temperature fruit-set was significantly reduced cvs ICGS 44 and Chico, whereas

there was no effect on ICG 1236. Talwar (1997) found that the flower bud stage

most sensitive to hot temperature was from 3 to 5 d before flower opening. This
 28

stage in groundnut is known to coincide with microsporogenesis (Martin et al., 1974;

Xi, 1991). However, detailed histological studies are needed to identify the precise

processes that are affected during these stages, and the exact stage that corresponds

to the most sensitive phenological floral bud stage to hot temperatures.

3.1.3.2 Flowering

De Beer (1963) reported that flowering is more sensitive to heat stress than the other

phases of reproductive development. Studies in controlled environments have shown

that heat stress and water-deficits at flowering result in the largest reduction in pod

yields in both Spanish and Virginia types, suggesting that this stage of development is

more sensitive to hot temperature than the other post-flowering stages of

reproductive development (Craufurd et al., 1996).

Temperature optima for flowering in groundnut range from 20°C (Wood,

1968) to 30°C (Ono et al., 1974). Bagnall and King {1991 a, b} reported similar

findings, where the base temperatures identified were 13.6°, 12.5° and 11.4°C for

Spanish, Valencia and Virginia types, respectively. The thermal times for flowering

were significantly smaller for Spanish than for Virginia cultivars, reflecting the earlier

flowering of Spanish types. Bell et al. (1991) suggested that the base temperature

values for seedling emergence and flowering did not differ significantly for Spanish

cultivars (12.4° and 12.7°C, respectively), but in Virginia cultivars the base

temperature for flowering (S.2°C) was cooler than that for emergence (13.4°C).

Ketring (1984a) conducted experiments which involved three day

temperatures (30°, 32° and 35°C) with a constant night temperature of 22°C and

found that there were no significant effects of temperature on time to first flowering
 29

or on the total number of flowers produced. Craufurd et al. (1996) found that there

were significant increases in the rate of flower production following the episode of

heat stress, and that the cumulative number of flowers produced at hot temperature

(>35°C) were greater than those produced at the optimum temperature (28°C). In

contrast, data reported by Wood Cl 968) showed a negative relation between number

of flowers and mean diurnal temperature, and flower numbers decreased by 2.6

plant" °C-l over the range of 23° and 34°C. Similar observations on flower

production were also made by Fortanier (1957), who found that a higher day

temperature of 35°C relative to 29°C reduced flower numbers by about 50%. De

Beer (1963) found genotypic variation in flowering response to heat stress:

continuous exposure to mean temperature of 33°C compared to 24° or 28°C reduced

flower production in cv. Mallarca, while in cvs Schwarz 21 and Ukrain flower

production was significantlyincreased by about 300 and 100 % respectively.

Studies on the effects of heat stress on flower morphology have revealed that

the length of the hypanthium increases in plants grown at 35°/30°C as compared to

those grown at 25°/25°C in cultivars (Talwar, 1997). It was also observed that hot

temperature results in style elongation in certain cultivars e.g. lCGS 44 and Chico,

thus resulting in lower pollination leading to fewer peg and pod numbers. Talwar

(1997) also observed that when pollen grains obtained from plants grown at optimum

temperature (25°125°C) were heat stressed (40° or 45°C) for 30 min on the artificial

medium, pollen germination was reduced and delayed, and pollen tube growth was

significantlyreduced.
 30

3.1.3.3 Pegging

The optimum day/night temperature for number of developing pegs ranges from

25°/25°C (Wood, 1968) to 32°/23°C (Fortanier, 1957; Cox, 1979). Ketring (l984a)

observed that the number of subterranean pegs (i.e., those reaching the soil but less

than 5 mm in diameter) was significantly reduced as temperature increased from

30° to 35°C. In several other studies in controlled environments, the reproductive

growth of groundnuts was adversely affected when either the day or the night

temperature approached 35°C (Fortanier, 1957; Bolhuis and De Groot, 1959; Wood,

1968; Cox, 1979). It was observed that a constant day and night temperature of 33°C

during the flowering period resulted in pollen death (De Beer, 1963).

The onset of pegging and the development of pegs are especially sensitive to

hot temperature (Stirling and Black, 1990). Williams et al. (1975a) working at

several locations found the largest number of pegs on plants grown at an altitude of

900 m, with a mean daily temperature of 23.2°C, and the smallest number of pegs

were at 1620 m with a mean temperature of 17.9°C. They also found that the

optimum temperature for cv. Makulu Red was close to 20°C, which was about 4°C

cooler than the value obtained by Ong (1984) for cv. Robut 33-1, which was selected

for the hot and arid climate in India. The number of developing pegs was greater at

20°C than at 30°C and the percentage of the pegs converting into mature pods was

inversely related to temperature, which indicates that profuse peg production does

not by itself ensure large pod numbers and kernel yields (Wood, 1968).
 31

3.1.3.4 Podding

The optimum air temperature for pod initiation and development ranges between

24° to 28°C (Cox, 1979; Ketring, 1984a). Ketring (1984a) reported that increasing

the daytime temperature from 30° to 35°C significantly reduced the number of

subterranean pegs. It was also observed that at hot temperature (35°C) the numbers

of immature pods were significantlysmaller when compared to numbers produced at

30°C. Similarly, Ong (1984) reported that the mean diurnal temperature regime of

36°/25°C imposed from sowing until maturity reduced the number of pods by 50%

relative to those produced by plants grown at 27°117°C.

Ono et al. (1974) reported that the optimum soil temperature range for

podding in groundnut was from 31° to 33°C, and that soil temperatures above that

range significantly reduced the number of pegs forming the pods. A 6° to 9°C

increase in canopy temperature above 28°C (Sivakumar and Sarma, 1986) and a

3° to 4°C increase in podding zone temperature above 23°C (Sanders et al., 1985;

1986) during the reproductive period had adverse effects on pegging and pod

formation. Pod production may also be reduced in circumstances where vigorous

vegetative growth competes with reproductive organs for assimilates (Fortanier,

1957; Williams et al., 1978; Cox, 1979). Increased vegetative growth also often

involves greater stem elongation, which prevents the pegs from reaching the soil

surface (Leong and Ong, 1983).

3.1.3.5 Pod filling and yield

The optimum temperature for pod growth is about 23°C (Cox, 1979), but cultivars

may differ in sensitivity to temperature during reproductive stages (Bell et al.,

1993 a, b; Nigam et al., 1994). Cox (1979) reported that the rate of development of
 32

individual pods was smaller (0.026 g d-I) at 34°/30°C than at 26°/22°C (0.047 g d").

The pods grown and matured at 34°130°C weighed only 1.5 g pod", whereas those

which matured at 26°/22°C were 2.7 g pod", Ketring (1984a) observed that

increasing temperature from 30° to 35°C during the daytime significantly reduced

seed size and seed yields by about 10%.

The total biomass and seed yields of groundnut cvs ICGV -SM 86021 and 55-

437 were significantly reduced when they were exposed to 6 d of hot temperature

(45°C) at the start of pegging: the relative reduction (%) was greater in hot

temperature-sensitive cv. ICGV-SM 86021 than in hot temperature-tolerant cv. 55-

437 (Wheeler et al., 1997). The start of seed filling was delayed by hot temperature

episodes and was 11.7 d later in the susceptible cvs than in the tolerant cultivar.

However, the rate of increase of seed harvest index during seed filling was not

affected by hot temperature.

Wheeler et al. (1997) suggested that the longer the duration of flowering

which culminates with the start of seed filling, the greater is the probability that this

stage will coincide with stress temperatures in fluctuating field environments. They

suggested that a long period between the onset of flowering and the start of seed

filling might be a characteristic of temperature-sensitive cultivars. The conclusion was

that cultivar differences in the response of groundnut yield to episodes of hot

temperature were due to differences in the timing of seed filling rather than to cultivar

differences in the rate of dry matter partitioning to fruits. Similarly, Stirling and Black

(1990) concluded from their studies, that the major cause of variability in pod yields

and harvest indices between cultivars was due mainly to the differences in duration

between peg initiation and onset of rapid pod growth, because once pods were
 33

initiated the proportion of dry matter allocated to reproductive sinks was relatively

conservative.

Cultivars of the Virginia botanical type were relatively more sensitive to hot

temperatures and water deficits than the Spanish botanical types (Craufurd et al.,

1996). This was attributed to the. fact that Virginia types have a longer flowering

period and have a different reproductive strategy when compared to Spanish types

(Craufurd et al., 1996).

Dreyer et al. (1981) reported that increasing the fruiting zone temperature by

3 0, 7° or 1DoC above a mean of 23°C was sufficient to reduce the number of fruits,

fruit size, filling period and pod yields. Similarly, hot soil temperatures (>33°C)

significantly reduced pod dry weights and yields (Ono et al., 1974). Ono (1979) also

conducted experiments to examine the time when pod development is most seriously

affected by the soil environments in the podding zone. Hot soil temperatures (37° to

39°C) were applied to the podding zone at successive 10 d intervals after peg

penetration. It was concluded that pod development was suppressed by the

treatments given during the first 30 d, with greatest reductions occurring between 20

and 30 d after peg penetration.

The optimum soil temperature for pod yields was in the range of 30 to 33°C

(Ono, 1979). Golombek and Johansen (1997) reported that soil temperature had a

marked effect on reproductive growth and development of groundnuts, and day/night

soil temperatures of 38°/32°C compared with 26°/20°C or 32°/26°C, significantly

reduce pod yields. This reduction was primarily due to soil temperature effects on the

processes of pod initiation rate, pod growth rate, and 1~O-mature seed weight.
 34

3.2 EFFECf OF TEMPERATURE ON NODULATION AND NITROGEN

FIXATION

Grain legume production in most agricultural systems is heavily dependent on

symbiotic nitrogen fixation. The principal environmental stresses which occur in the

SAT are heat stress and moisture stress, and these stresses interfere with any or all of

the processes of root infection, nodule development or nitrogen fixation per se (Giller

and Wilson, 1991). Hot soil temperatures adversely affect the growth and survival of

rhizobia in soils and their symbiotic association with legumes and prevent nodulation

(pate, 1961; Dart, 1977; Nutman, 1975; Day et al., 1978).

Soil and therefore root temperatures in tropical and subtropical regions are

often in the range of 35° to 40°C (Chatel and Parker, 1973; Dugas, 1984) and are

detrimental to nodule formation and dinitrogen fixation (Arayangkoon et al., 1990).

At these hot temperatures, the groundnut root biomass is reduced and the roots are

thin, unbranched and with very few root hairs and so produce fewer nodules.

Kishinevsky et al. (1992) showed that the groundnut-Bradyrhizobium symbiosis was

completely inhibited by a soil temperature of 40°C. The effects were due not only to

the failure of nodulation but also to the inability of nodules to function even if they

were formed. Hot temperatures adversely affected the process of infection more than

the process of nodule growth. Continuous exposure to a root temperature of 37°C

reduced the total nitrogen by 49% due to impaired nodule function, but it did not

reduce nodulation (Kishinevsky et al., 1992).

In the SAT surface soil temperatures can occasionally reach 65°-70°C and

temperatures above SO°C are common at S cm depth (Dudeja and Khurana, 1989),

which is sufficient to inhibit germination of seeds and to kill many bacteria. Most of
 35

the heterotrophic free living nitrogen fixing bacteria and rhizobia are not resistant to

desiccation, and excessive soil temperature can therefore kill most of the bacteria in

the surface layers of dry soil (Marshall, 1964).

Many aspects of the Rhizobium-legume symbiosis are affected by hot root

temperatures, including: growth an~ survival of rhizobia (Day et al., 1978; Parker et

al., 1977); formation of root hairs (Frings, 1976; Lie, 1974); binding ofrhizobial cells

to the surface of root hairs (Frings, 1976); formation of infection threads (Ranga

Rao, 1977); structure and development of root nodules (Lie, 1974); legheamoglobin

content of nodules (Frings, 1976); activity of the nitrogenase enzyme (Dart and Day,

1971; Munns et al., 1977); and the nitrogen concentration and dry matter production

of nodulated plants (Dart et al., 1975; Day et al., 1978; Gibson, 1966; Herridge and

Roughley, 1976).

Nambiar and Dart (1983) reported that temperatures of 300e and 35°e

significantly reduced the nitrogenase activity of groundnut root nodules as compared

to those at a temperature of 25°C. The optimum temperature for the symbiotic

system in groundnut is comparable with other tropical and subtropical grain legumes.

The inhibition of nitrogen fixation due to heat stress varies between crops; for

soyabean it was 30°C, for pea (Pisum sativum L.) and lupin (Lupinus polyphy/Jus) it

was 25°e, and for alfalfa (Medieago sativa) and faba bean (Vieia/aba) it was 20°C.

Symbiotic nitrogen fixation is further influenced by Bradyrhizobium strain,

host cultivar, and their interaction. Kishinevsky et al. (1992) observed that sensitivity

to heat stress was strongly influenced by the strain of Bradyrhizobium fanning the

nodules. Three strains of Bradyrhizobium (280A, 2209A and 32Hl) were evaluated

for their ability to grow and survive at temperatures up to 42°C. Strain 32Hl was
 36

unable to grow at 37°C. At 30°C strain 280A formed 15xl09 and 22xl09 bacteriods

g nodule" on roots of Virginia and Spanish cultivars, respectively, whereas at 37°C

the corresponding numbers were l1xl09 and 9xl09 bacteriods g nodule". In contrast,

for strain 32Hl, the number ofbacteriods g nodule" were 30xl09 and 14xl09 at 30°C

and 4xl09 and 6xl09 at 37°C in Virginia and Spanish cultivars, respectively,

indicating the more extreme thermo sensitivity of this strain.

The tolerance and survival of 18 strains of groundnut rhizobia

(Bradyrhizobium sp.) at elevated temperature in peat was tested based on growth

response to increasing temperature from 27° to 48°C (Elsaeid et al., 1990). All the

strains tested survived well at 29°C; however, considerable variation in survival

occurred at either a constant temperature of 41°C or at day/night temperatures of 41 °

/29°C. Only two strains (NC 92 and TAL 25) survived well enough for 70 d in the 41

0/29°C regime and were suggested to be considered for widespread use as inoculants.

The strain NC 71 was able to survive well at constant 41°C for 70 d, suggesting that

it is suitable where diurnal variation in hot days and nights was less.

The sensitivity of groundnuts to heat stress could be correlated with the

sensitivity of the nitrogen fixing ability and growth of strains of Bradyrhizobium to

hot soil temperature. However there has been very little work done to identify

groundnut rhizobia able to effectively fix nitrogen at temperature extremes and their

suitability under field conditions (Elkan, 1995).

 37

CHAPTER 4 OBJECTIVES OF CURRENT RESEARCH

Seed yield in groundnuts depends primarily on the number of pegs produced, the

proportion of these pegs that produce mature pods (Enyi, 1977) and mean kernel

weight (Labana et al., 1980). The formation of pegs and the proportion which form

pods depend largely on the number of flowers which open and on the proportion of

those flowers that form pegs. These events occur during three distinct stages of

reproductive development; the production of flowers (Le. budding and flowering),

the development of pegs that carry the ovary below ground (Le. pegging and fruit-

set), and the subsequent formation and filling of pods (Le. pod filling). If heat stress

occurs during these stages seed yields and dry matter partitioning to fruits are

reduced. These effects vary with cultivar, the intensity of stress, the duration of

stress, and the timing of stress. Furthermore these stages may have different relative

sensitivity to heat stress.

Given these considerations, experiments were designed to study the effects of

hot air and/or hot soil temperatures imposed at different stages of crop development

(flowering, pegging or podding), for different durations (ranging from 2 d to 60 d),

and at different times of the day (e.g. day, night, mornings and evenings). The impact

of heat stress on nodulation, flower production, pollen production, pollen viability,

fruit set, fruit number, dry matter production and partitioning, and pod yield was

quantified.

The specific objectives of the research were:

 38

1. To determine the effect of hot air temperature from flower bud initiation

until maturity on total dry matter production, reproductive potential (i.e. the

respective numbers of flowers, pegs and pods) and realisation of that

potential (i.e. dry matter accumulation and partitioning to fruits);

2. To investigate whether the Spanish and Virginia botanical types have

different inherent sensitivities to heat stress and to assess the differences in

the mechanism of responsiveness to heat stress in putatively susceptible and

tolerant cultivars;

3. To determine how hot air and/or hot soil temperatures imposed from

podding until reproductive maturity affects dry matter production,

partitioning and pod yield and whether their effects were additive or

interactive;

4. To identify time(s) between floral bud initiation and pod initiation when

daytime hot air temperature stress causes the largest reduction in

reproductive yield of Spanish and Virginia cultivars;

S. To determine the effects of short episodes (2, 4, or 6 d) of exposure to hot

temperatures (34°, 42°, and 48°C) on flower production, fruit-set and the

number of pegs and pods;

6. To determine whether fruit-set in groundnut was relatively more or less

sensitive to hot temperature during different periods of daylight i.e. during

the first 6 h (AM) or the second 6 h of the daylight period (PM) in each

diurnal cycle;

7. To determine whether responses of flower production, fruit-set and number

of pegs and pods to temperature were quantitative or qualitative;

 39

8. To determine and quantify the effects of short episodes (6 d) of hot days

(34°, 42°, or 48°C) and warm night (28°C) temperature on pollen

production, pollen viability, fruit-set and fruit production;

9. To quantify the effects of hot air and/or hot soil temperature imposed from

flowering until maturity and their interaction on dry matter production and

pod yield of groundnut plants relying on inorganic N or on symbiotically

fixed N2 or combination of both; and

10. To determine whether plants dependent on symbiotic N2 are more sensitive

to hot temperatures than those dependent on inorganic N.

 40

CHAPTER 5 GENERAL MATERIALS AND METHODS

This Chapter describes programme of experiments, controlled environment facilities

and also gives the structure and presentation of results. All the experiments were

conducted at the Plant Environmental Laboratory of the Department of Agriculture,

The University of Reading between June 1996 to November 1997.

5.1 PROGRAMME OF EXPERIMENTS

First, two experiments were conducted to help understand the effects of continuous

hot air temperature on dry matter production, partitioning of dry matter to pods and

pod yields of Spanish and Virginia botanical types (Chapter 6). One experiment

examined the effects of hot air temperature imposed from the start of flower bud

initiation until reproductive maturity. Another experiment examined the effects of hot

air and/or hot soil temperature imposed from start of podding until reproductive

maturity. These experiments were followed by a reciprocal transfer experiment aimed

at identifying the time(s) during reproductive development when hot air temperature

has the maximum effects on reproductive potential and yield in Spanish and Virginia

types (Chapter 7). After identifying the sensitive phase, three detailed experiments

were conducted to quantify the effects of temperature on flower production, pollen

production, pollen viability, fruit-set and number of pegs and pods in Spanish cv.

ICGV 86015 (Chapter 8 and 9). Finally an experiment was conducted to determine

whether plants dependent on symbiotic N2 fixation were more sensitive than plants

dependent on inorganic N to hot air andlor hot soil temperature (Chapter 10).
 41

5.2 CULTIV ARS RESEARCHED

Two cultivars each of the Spanish and Virginia botanical type with similar times to

first flower, based on the findings of previous experiments conducted at the Plant

Environment Laboratory were used in the current research (Table 5.1). Seeds were

obtained from the ICRISAT Asia Centre in India and the ICRISAT Sahelian Centre

in Niger.

Table 5.1 Origin, botanical type and relative heat tolerance of the four cultivars

used in various controlled-environment experiments.

Cultivar Origin Botanical type Heat tolerance Days to first flowering

ICGV 86015 India Spanish Susceptible 28

ICGV 87282 India Virginia Susceptible 28

796 Senegal Spanish Tolerant 25

47-16 India Virginia Tolerant 30

5.3 CONTROLLED ENYmONMENT FACILITIES

Controlled environment facilities Le. polyethylene tunnels (poly-tunnels) and modified

Saxcil growth cabinets at the Plant Environment Laboratory, The University of

Reading, were used for the research. These facilities were linked to modern data

logging and control equipment, and were supported by an oil-fuelled generator,

which operates automatically in case of power failure from the national grid.
 42

5.3.1 Polyethylene tunnels

Plants were grown in pots kept in two poly-tunnels each 25 m long,' 8 m wide and

3 m high at apex (Fig. 5.1). Day and night temperature in the poly-tunnels was

controlled within ±IoC of the target values. On rare days, when the ambient

temperature was exceptionally hat, day temperature was sometimes 2° to 4°C

warmer than the set target temperature. Air temperatures were measured in each

poly-tunnel with screened and aspirated copper constantan thermocouples positioned

at the top of the plant leaf canopy. Readings were taken at lOs intervals and means

were stored for successive 10 min periods using a data logger (DL2, Delta-T Devices

Ltd, Cambridge, UK). Carbon dioxide was at ambient concentrations throughout.

The photoperiod in the poly-tunnels was controlled by a blackout facility operated

manually each morning and evening. The relative humidity was not precisely

controlled but was maintained close to 70% (±5) through automatic sprinklers and

ventilation. Relative humidity was measured every 10 s and the means were stored

for every 10 min period during the experimental period.

5.3.2 Blackout facility

To control the photoperiod in the poly-tunnels a blackout facility was created by

erecting a 500 J.1 gauge light proof black-polythene sheet on a 2 m high galvanised

steel pipe frame fixed at both the ends (Fig. 5.2). Opening and closing the blackout

was done manually each morning and evening such that the two leading edges

overlapped each other by at least 3 m in order to ensure the cover was light tight.

Bricks were placed along both sides of the blackout to trap the polythene sheet to the

ground. The night temperature inside the blackout was regulated by drawing in air
 43

Fig. 5.1 Photographs of (a) the polyethylene covered tunnel structure

and (b) groundnut plants growing in the pots inside the poly-tunnel.
 44

er>
c
'lfj
o
_.J

u
+-'
::J
o
.::£.
U
o
--'
CD

c
~
a..
o
.....,
:J
o
~
u
o
-..J

OJ
 45

maintained at an appropriate temperature from within the poly-tunnel using two fans

fitted at one end of the enclosure. The air was distributed uniformly to the whole area

through perforated polythene ducts connected to those fans. Temperature in the

blackout facility was continuously logged as described previously. The photo-

thermoperiod was coincident and ~aintained at 12 h d" in all the experiments.

5.3.3 Soil temperature control

Soil temperature control benches were specially constructed by fixing a frame on the

cast iron table structure 2.75 m long, 1.5 m wide and 0.5 m high (Fig. 5.3) and

installed in each of the poly-tunnels. Five tubular heaters (2.5 m long) with a total

wattage of 2.4 kW, were fixed longitudinally just below the surface of the bench.

These heaters were fitted to a switchboard that was automatically controlled by a

data logger (CRI0, Campbell Scientific Ltd, Shepshed, UK). On top of these benches

at a height of 45 cm, a frame like structure was made by wires running longitudinally

and horizontally at every 30 cm to create small compartments. The complete

structure was sealed and made airtight by two layers of heat resistant polyethylene

sheet. The upper surface was cut and pots were inserted into each of the

compartments, ensuring a tight fit so that there was no heat loss. The target daytime

soil temperature was set at 10°C above the ambient soil temperature and controlled

by a data logger. Soil temperature at night was not controlled and pots were allowed

to return to ambient temperature. Soil temperature during day and night was

measured by placing thermocouples at a depth of 5 em in four pots chosen at random

across the bench at every lOs and means were recorded for successive 10 min

intervals on a data logger.

 46

Fig. 5.3 Photographs showing the soil temperature control facility

constructed inside the polyethylene covered tunnel structure.
(a) cast iron table structure; (b) tubular heaters; (c) heat resistant
plastic; and (d) thermocouples.
 47

5.3.4 Growth cabinets

Plants were grown in modified Saxcil growth cabinets in which the day and night

temperatures were controlled precisely within close tolerances. The photo-

thermoperiod was equal and coincident at 12 h dol and was maintained in all the

cabinets by automatic switching on and off the main light source at pre-set times.

Relative humidity in the cabinets was maintained to give a vapour pressure deficit

(VPD) of 1.2 kPa. Illumination was provided by a combination of "cool white"

fluorescent tubes supplemented by incandescent lamps, giving a photosynthetic

photon flux density (PPFD) of 650 umol m2 fl at canopy level. Atmospheric carbon

dioxide concentration was maintained at 360 umol mol" air in each cabinet.

5.4 STRUCTURE AND PRESENTATION OF RESULTS AND DISCUSSION

The results of the experiments mentioned in Section 5.1 are presented as Chapters 6

through 10. Each of these Chapters is a complete formal scientific paper with,

Abstract, Introduction, Materials and methods, Results and Discussion, with an

exception of Chapter 6 which does not have an Abstract. Four of the Chapters have

been submitted to peer reviewed journals and Chapters 7 and 9 have been accepted

for publication as indicated below. The final Chapter discusses and evaluates the

overall objectives, methods, results, and implications of the current research

programme and future work (Chapter 11). All the references cited in the thesis are

given at the end.

Chapters 7 to 10 are submitted as formal scientific papers to the following

journals:
 48

Vara Prasad, P.V., Craufurd, P.Q. and Summerfield, RI. (1999) Effect of timing of

heat stress on groundnuts during reproductive development. Crop Science

39: (5) 000-000 (In press) (Chapter 7).

Vara Prasad, P.V., Craufurd, P.Q., Summerfield, RI. and Wheeler, T.R Effect of

short episodes of heat stre~s on flower production and fruit-set. Journal of

Experimental Botany. (Chapter 8).

Vara Prasad, P.V., Craufurd, P.Q. and Summerfield, R.J. Fruit production, pollen

number and viability of groundnut in response to short episodes of heat stress.

Annals of Botany (In press) (Chapter 9).

Vara Prasad, P.V., Craufurd, P.Q. and Summerfield, R.I. Effect of hot air and soil

temperature on pod yields of groundnuts dependent on symbiotic or inorganic

nitrogen. Plant and Soil. (Chapter 10).

 49

CHAPTER 6 EFFECTS OF CONTINUOUS HOT AIR AND HOT

SOIL TEMPERATURE ON DRY MATTER PRODUCTION,

PARTITIONING AND POD YIELDS

6.1 INTRODUCTION

Groundnut is an important source of oil and protein-rich food and feed for people

and livestock in the developing world. It is an important component of the cropping

systems of the semi-arid tropics, which are characterised by hot temperatures and low

and erratic rainfall. Hot temperature stress is one of the least well understood of all

the abiotic adversities that affect crops (paulsen, 1994) and is one of the major

uncontrollable factors affecting plant growth, development and productivity

(Marshall, 1982). Groundnut crops grown in the semi-arid tropics are often exposed

to air and soil temperatures warmer than 35°C during the reproductive phase,

circumstances which significantly reduce seed yield (ICRISAT, 1994). Groundnut

plants are susceptible to both hot air and hot soil temperatures, due to their aerial

flowering and subterranean fruiting habit. That said, most research in groundnut has

been done either on hot air (Ketring, 1984a; Ong, 1984) or on hot soil temperature

(Ono et al., 1974; Golombek and Johansen, 1997); studies on the combined effects of

both factors have received far less attention. This is unfortunate given that this

information is essential to identify heat-tolerant cultivars which could be used as

parents in breeding programmes.

The mean optimum air temperature range for vegetative growth in groundnut

is between 25° and 28°C (Wood, 1968; Cox, 1979) which is slightlywarmer than the
 50

optimum range reported for reproductive growth, i.e. between 22° and 24°C (Wood,

1968; Cox, 1979). Hot days >35°C during the reproductive phase reduce fruit-set,

and consequently the number of pods and ultimate seed yields (Ketring, 1984a).

The optimum soil temperature range for germination is between 29° and 36°C

(Mohamed et al., 1988a) and for root growth it is close to 30°C (Suzuki, 1966).

Similarly, the optimum soil temperature range for pod formation and development is

between 31 ° and 33°C and soil temperature >33°C significantly reduce the number of

mature pods and seed yields (Ono et al., 1974; Ono,1979; and Dreyer et al., 1981).

Golombek and Johansen (1997) found that most pods were produced at mean soil

temperatures between 23° and 29°C; temperatures of 17° and 35°C were sub- and

supra-optimal, respectively.

In the research reported here the effects of hot air in combination with hot

soil temperature on dry matter production and partitioning, and reproductive

development were investigated in Spanish and Virginia botanical type cultivars. The

objectives were: (a) to understand the responses of Spanish and Virginia cultivar to

hot air or hot soil temperatures; and (b) to determine how hot air and hot soil

temperature affected reproductive development and yield, and whether their effects

were additive or multiplicative.

6.2 :MATERIALS AND METHODS

Two experiments were conducted between June and September in each of 1996 and

1997 in the controlled environment facilities of the Plant Environment Laboratory,

Department of Agriculture, The University of Reading (51°27' lat. and 00°56' long.).

They were undertaken in two polyethylene-covered tunnels (poly-tunnels) each 25 m

 51

long by 8 m wide by 3 m high at the apex, maintained either at close-to-optimum

day/night temperatures of 28°122°C or at hot days combined with an optimum night

temperature, i.e. 38°122°C.

The diurnal photo- and thermo-period in both poly-tunnels were coincident

and equal at 12 h d-I (0750 to !950 h). The photoperiod was controlled by a

manually operated blackout facility. Air temperatures were measured in each poly-

tunnel with screened and aspirated copper constantan thermocouples positioned at

the top of the plant canopy. Readings were taken at 10 s intervals and means of

successive 10 min periods were stored using a data logger (Delta-T Devices Ltd,

Cambridge, UK). Carbon dioxide fluctuated at ambient concentrations, and relative

humidity during the day was controlled by automatic water sprinklers and ventilation

to give a VPD close to 2 kPa in both poly-tunnels. The poly-tunnels transmitted

about 75% of incoming photosynthetically active radiation such that the

photosynthetic photon flux density at canopy level averaged 504 and 597 umol m-2s-l

during the experiment undertaken in 1996 and 1997, respectively.

6.2.1 Cultivars and temperature treatments

6.2.1.1 Experiment J (1996)

Two cvs each of Spanish botanical type (ICGV 86015 and 796) and Virginia

botanical type (ICGV 87282 and 47-16) were used. All plants were grown at

optimum day/night temperature of 28°122°C from sowing until first flower bud

initiation at 21 DAS. Thereafter one-half of the plants of each cv. were transferred to
 52

the adjacent poly-tunnel maintained at the hot day/optimum night temperature of

38°/22°C, where they remained until final harvest at 90 DAS.

6.2.1.2 Experiment 2 (1997)

Two cvs, ICGV 86015 (Spanish) and ICGV 87282 (Virginia), were grown at

optimum day/night temperatures of 28°/22°C from sowing until first flower

appearance at 28 DAS. Thereafter, one-half of the plants of each cv. were transferred

to the adjacent poly-tunnel maintained at hot day/optimum night temperature of

38°/22°C, where they remained until final harvest at 90 DAS.

Within each air temperature regime, two soil temperature environments

(ambient and hot) were imposed from first pod initiation at 15 d after first flowering

(DAF) until final harvest. The hot soil temperature regime was imposed by placing

one-half of the plants on customised constructed benches 2.75 m long by 1.5 m wide

and 0.5 m high, fitted with 5 tubular heaters, 2.4 m long, with a total wattage of

2.4 kW. The target hot soil temperature was set at lOOC above ambient soil

temperature and was controlled by copper constantan thermocouples placed at a

depth of 10 cm using a data logger (CR 10, Campbell Scientific Ltd, Shepshed, UK).

Soil temperature during the night was not controlled and pots were allowed to return

to ambient temperature, typically achieved within four hours. Soil temperature during

the day and night in both the ambient and hot soil temperature regimes were recorded

by thermocouples placed at a depth of 5 cm. Readings were measured at lOs

intervals and means of successive 10 min periods were stored using a data logger.
 53

6.2.2 Plant husbandry

Uniform seeds of each cv. were selected and treated with Apron Combi 453 FS (Ciba

Agriculture, Cambridge, UK) as a precautionary measure against seed-borne

diseases. Seeds were pre-germinated at 25°C on moist filter paper in Petri dishes kept

in the dark for 2 d until radicles emerged. The germinated seeds were then sown one

per 15 L pot at a depth of 2.5 cm. The sides of pots were covered with aluminium

foil to reduce radiative soil heating. The rooting medium comprised sand, gravel,

vermiculite and loamless peat compost mixed in proportions of 4:2:2: 1, by volume,

respectively. A commercial controlled-release fertiliser (0.15 kg kg" N, 0.10 kg kg"

P, 0.12 kg kg" K, 0.02 kg kg" MgO plus trace elements; Osmocote Plus, Scotts UK

Ltd, UK) was incorporated into the mixture at the manufacturer's recommended rate

of 5 g L-l. Seeds were not inoculated with rhizobia and so plants were dependent on

inorganic nitrogen. All pots were soaked with tap water and allowed to drain for 24 h

before sowing; thereafter they were irrigated as necessary through an automatic drip

irrigation system (Hadley et al., 1982).

All plants were healthy and there were no serious pest or disease problems.

Releases of predators (Phytoseiu/us persimilis Athias-Henriot) and foliar sprays of

Torque (a.i. Fenbutatin Oxide) controlled a mild incidence of red spider mite

(Tetranychus urticae Koch). Thrips (Thrips tabaci Lindeman) were controlled by

release of the predator Amb/yseius cucumeris Oudemans. Plants were also sprayed

with Repulse (a.i. Chlorothalonil; 2,4,5,6-tetrachloro-l,3-benzenedicarbonitrile)

during 1996 to controlleafspot (Cercospora sp.).

 54

6.2.3 Observations and data analysis

Durations (d) from sowing to the appearance of the first fully opened flower (RI;

Boote, 1982) and then the first peg (R2) were recorded on all plants. Thereafter, the

numbers of flowers opening each day were counted until final harvest. At the final

harvest, all plants were removed from each pot without damaging the root systems

and were separated into roots, leaves, stems (including petioles), pegs and pods. The

numbers of pegs and pods per plant were recorded and roots were washed with water

to remove the potting medium. The respective weights of roots, leaves, stems, pegs

and pods per plant were recorded after oven-drying these components at 80°C to a

constant weight. Total dry weights (inclusive of senesced leaves and roots), pod

harvest indices (ratio of pod to total dry weight), and root-to-shoot (i.e. root dry

weight to leaf and stem dry weight) ratios were calculated from the weights of

individual components. Values for pod dry weight were adjusted by multiplying

recorded data by x 1.65 to allow for the oil content of the seeds (Duncan et al.,

1978).

At final harvest the total number of pegs and pods (reproductive number) per

plant were counted. The proportion of flowers setting pegs (fruit-set) was calculated

as the ratio of total cumulative flower number to reproductive number. Similarly,the

proportion of pegs forming pods was calculated as a ratio of reproductive number to

number of pods. The data on fruit-set and proportion of pegs forming pods were

subject to angular transformation before analysis to ensure homogeneity of variances.

Experiment 1 was analysed as a split-plot design with air temperature

treatments as main plots and cultivars as sub plots, replicated five times. Experiment

2 was analysed as a split-split plot design with air temperature as main plots, soil

temperature as sub-plots and cultivars as sub-sub plots, replicated five times. Analysis
 55

of variance for all the variables was performed using Genstat 5 (Genstat 5

Committee, 1987).

6.3RESULTS
.
Target temperatures in both experiments were within close tolerances (SD<1.3°C).

The mean (day/night) air temperature in the optimum and hot air temperature poly-

tunnels were 25° (27.9°/22.1°C) and 3D.3°C (38.3°/22.3°C), respectively, and were

similar in both experiments. Mean (day/night) ambient and hot soil temperatures were

6.3.1 Experiment 1 (1996)

There were significant differences (P<D.D01) between cultivars and their response to

air temperature for all traits given in Tables 6.1 and 6.2, unless specific mention

indicates otherwise.

6.3.1.1 Cultivar responses

Overall, the Spanish cvs ICGV 86015 and 796 produced larger (mean 54%)

cumulative numbers offlowers plant" (p<0.001) than the Virginia cvs ICGV 87282

and 47-16 (Table 6.1) due to greater rate of flower production (Fig. 6.1). However,

the Spanish cvs set 20% fewer pegs (P<0.001) than the Virginia cvs (Table 6.1), and

therefore had similar reproductive numbers (mean 123 plant"), Although Spanish cvs
 S6

"""e
"0
QJ
--
\C
I
***V'I *** * *** ***00
--
t"-

**V'I
~ ~ 0
~ "0 r.Il V'I
M

='" =
'"=
tI) N N M N
....'~"= M
OQ
M

-=~
'"=
r--
OQ

-
\0
-e- ~

- -
t"- 0 0

= UG~
CAl 0
t"-
I 0
N
d
V'I
M
V'I r-: \Ci
N
"¢

..--=
'"= ~
4'
QJ ~.
0
'"=
.•
....'" .•
I ·c
Cd
~
.•
--
CAl
cS ~ 0
.0
> Cd
·c
N

-
00
"0'" 0
.§> N
0
C. ! t"-
00 OO \0 M
\0
;>
-
V'I
N \Ci M M
"0 "0 > N "¢ "¢
N
= o M

=
'"=
'"
CAl
QJ
C.
-
\C
0\
e-
'"=
-
U

"0

.• ='"= .
-
CAl
=
- -- ""'
Cd

.... an '"
QJ
.2:
QJ
.:t:
'"
r!
0
\C
OQ
.~
'"= 8
C.
QJ
~ >
Co!)
QJ
~
0
= QJ

'-

=
0

.•
--
U
~

.~
.::
~
"0
;>

= ~
0
\0
0\
t"- - - -
0
M
~
"¢
N
\Ci
M
00
0
"¢
\0
V'I
V'I

0
1::
=
'c.
"=
'"=
....'" ca0
C
C.
tI) =
QJ ·C
c
~
0 e
.... ~Cd
0
!
...!w..
c. 'QJ"= .0
.c

=
'-
C
"0
....~
QJ
ell
·c -
V'I

-'"= ....~ ~
....=
Cd
c.. \0
00 00 N N
0\
00 N
c. \Ci
- 0\
;>. tI) \0 V'I
~ f > M M N "¢ V'I

-
e,:,
"0
QJ 0 QJ
.c c. c. u
e "0 e
==
~
= ....
'"=
~
QJ

0
QJ QJ
!

-- -
~ "0
'-
0 .5 0
~d
= '"= Cd
'"=
....
0
'"
QJ

t:
'"=
'"=
QJ

e
-a
""'0 .~
.:::-- -
0
.::
-e - -
0

--
QJ QJ
.:: ~ d
;
;> "0 .... e c.. v
-=e ~
0
'"= C. QJ
~ ~ ~
~
a: '"= ""'0 ~ '7-= td
........
QJ
....'"= ~ .:d
.-...=
"0 Cd

--= 0 bI)
c.. ~
'"= '"= ~ 'Q)
U -.
e
--...
0
~
..:; t:- r! .-....
0
>
Cd
0

t)
~
~
ell
QJ

t=Cd
~
-
"0
0
t+=l
._·C
bI)

-........ ....-.... 'e~... .-2

"0 '"= ell "0
;> .c ':;" tI)

...
QJ I
.c ~ ~
'"= "0 "0
'"=
= 0
***
:;) 0 0
~ c Col U ~ E-4 ~ ~
 57

c_
= ~
~
QO
~-=
s ....
.nco
......_ u
oCCJ
\0:> ~ c
QCo=
N
e-- > ......
c 041 ~
Ue~
-:~=
0
~ -= ~
CJCJ;
~ .c=~
.~ ~ 0

==C.c ~ ...
c..c ~
Vl~
ClIO
~ ~ '"
.c~~

~
M
-~~
--=
-c;;- r!. =
._
= :.•
fl
-. =CJ
~ ~ ·2

~
....=
~
C .6Al tl
=.!::~
'C..>-r;;
N 0 ...~.c.c
~ ~
~ c.~p..
~. 'l:'l: •
~ ~ ~-'"
CJ'l:e
QO ¢:: = c :s
'l:=_
~ <= ......= .
~
C.N ~ >.
~ ...QC c.Ql
>, t ~ S.~
N ~
r- ~QC~u
<=>~._ .... ~c.
~ -=... = ""
ou_~
0
~-O '"
~ ~ ~ '" ~
.,QCJ~=
S=o-
:s._ M e
=.5 -~
QO
~ ._
.c f.Ai~.c c.
S
~ ~>e ~ 0
... ~
.c - 5
~-.:s
=-u._ "'5
CJ-u'
~ ~."'~
a
M
~r----
cr~
COl

=CJ:S~ ~
-:.
•
NO
5 =
~ ~.c
c.;,);
5 ~ WJ
N
e·2 c.~
~ = 0 =
0 Vl.c~ =
Eo-< Co-
- . u·_CJ) c..
0
~
\C.c=
10 N QO ~ 10 ol:;:'~~
~ ~ ._ .c = ~
IiIoo _~ Co

( ARP lURid) uOllJnp0.ld .I3M.og JO 3lRlI

1- 1- •
 58

had significantly (p<0.001) smaller total dry matter than Virginia cvs (133 and 149 g

plant", respectively), the Spanish cvs partitioned more dry matter to pods and so had

90% more pod yields than the Virginia cvs (Table 6.1).

Among Spanish types, cv. 796 had fewer flowers than ICGV 86015 (310 and

368 plant", respectively) but the pr?portion of flowers setting fruits was significantly

(P<O.OOI) greater in cv. 796 (Table 6.1). The Spanish cv. 796 produced more total

dry matter than ICGV 86015 (136 and 129 g plant", respectively), whereas, ICGV

86015 partitioned significantly (p<0.001) more dry matter to pods (Table 6.1).

Similarly, among the Virginia types, cv. 47-16 produced significantly (p<O.OOI) more

total dry matter than ICGV 87282 (153 and 145 g plant", respectively), but

partitioned less dry matter to pods and so had significantly (p<O.OOI) smaller pod

yields than ICGV 87282 (Table 6.1). There were no significant differences between

cvs ICGV 87282 and 47-16 in the number of flowers produced, or in fruit-set or the

proportion of pegs forming pods (Table 6.1).

6.3.1.2 Response to air temperature

There were no significant effects of hot air temperature on the number of flowers

produced in cvs ICGV 86015, ICGV 87282 and 47-16, but hot air temperature

significantly (p<O.OI) increased flower production (by 56%) in cv. 796 compared to

optimum temperature (Fig. 6.2a). In contrast, fiuit-set was significantly (p<O.OOI)

reduced at hot air temperature by about 33% in all cvs (Fig. 6.2b). There was no

major detrimental effect of hot air temperature on the proportion of pegs forming
<

pods in any cultivar. Overall the imposition of hot temperature compared to optimum

temperatures significantly (p<O.OOI) reduced reproductive numbers by 44 and 28%

 59

(a) DOT
400 • HT
~
] 300
e
==
~
200
~ 100
~
e
~ 0
(b)
60
45
30
15
o
(c)
80
60

40
20
o
ICGV 86015 796 ICGV 87282 47-16

Spanish types Virginia types

1
Fig. 6.2 Numbers of flowers planf (a), the proportion of flowirs setting pegs and
pods (fruit-set, angular transformed) (b), and pod yield plant" (c) in Spanish and
Virginia botanical type evs grown in mean day/night (25°C, open bars) or in hot
(30°C, solid bars) air temperatures. Vertical bars are the SED for comparing the
two temperatures.
 60

0
0
.c
- .-= a
a
-- --
Wl
* * *
0 **00 **00 **\0 **0\
-
0 c.. ~
0
0 tI) r--: ~ 0
r-
'0
0
.-
{
.c
Cl)

U 11"1

=~~= '0=
~

- -
0 ~ 00 11"1
0\
0 00 ~ 0\

-
M

== \0

.- a
'0
=
- -==~ u • r--.
I

'<t'
~ U 0

- - - -
Wl 0\ \0
~ ~(.) 0
'<t'
\0 r--: r--: M
t os 11"1
~ r--.
.c
'0
= ~
0
r-
Cl)
..
cd
0
.c
0
e
- - --
... ..
c.. os
cd 00

- -
0
U ~
= =~
.- ~
Wl

o§ ~ 0
0
~
r--. ~ 0
"<:t'
0\
== "3 .-
I
00
0-;> ~ M

-
Col
r--.

--= --
c.. c.. Col
00
~
r- (;
-
r- \Q
~ ~ u
- - - -
00
I ~ U ~ \0 00
e- t.:
0
r--. 11"1
~
a "<:t'
'0
....
0
II')
~ r-- M

to == Wl
I-<
cd

--=
'0
N
00 ==~
a -
o~
8
~
-..
...-
0
N
r--.
00

>
(.!) -=
~
.c
~
r-
U
0
0
M
-- - -
0
~
~
~
r--.
00
M
II')

=
- .-=
COlS
c..
.-=
r-
u
t:::-
~
-= COlS
u
..
~
\0
0\
t"

U
-- «"1 N

-
11"1
~ 0
.c
-
'<t' 00
0

Cd 11"1 N 0\
a ff 00 ~ '<t' '<t'
-< (.)
°acd
=~= ~
Q
~
= ...
.-- !
0

-
~ Q'\ .c ~

- - - -
U r--.
.c
...
o~
-
Q 0\ '<t'
Col '0 <n 0 0 00

==
\Ci
= °8cd 0\

- a
11"1 0 c..>
N M "<:t'
'0 0
Go)
0 0.. \0 0..
r- 1.0 til fIl

- -=
c.. Q'\ 00 Go)
0
~
r-
r-- ~ (; -'"..'
-
Wl
'0
0
c..
'0
'0

In
Q
=
co:s

1.0
Wl

e~
-
u U
0
II')
~ -
N
\0
11"1
0\
-
M
N
0\
'<t'
~
II')
N
0
0
0
~
v
= QC

a~
c.. ""='
c::

-
co:s
>
-
cd
."
Cl) (.!) ~t= ........
~
- -
0
c.. U r-
.; ~c= 0
....
-
cd

0
r-
~
,Q
t:::-
.c
0-=
."

=
U
0

-..
Q
0..
I-<
u
-e ~
cd
0..
........
:::R
e_.,
~
-
........
~
o_0
v
~
tiS
dcd
a ...t 0-= -
E
f")
c.. ~
Go)
""=' c..>
Z=
00
0 0 t.I:l
0 .c ~ .~
oB ~ 0

-
~
....!
N 0
r-
Q
.c
<n
=' ~ ~
1.0
...
-= .--
~
.c
~ =
Q
Q

r-
u

-
0
In
N
....
°ca
I-<
~
""='

~
8
0..
""='
...
Cd
~
0
~
.c
""='
~
0
-
~
0

0
*
**
**
 61

in Spanish cvs ICGV 86015 and 796, respectively, and by 58 and 50% in Virginia cvs

ICGV 87282 and 47-16, respectively. (Table 6.2). Hot air temperature compared to

optimum temperature also reduced pod yields by 24, 27, 36 and 78% in cvs 796,

ICGV 86015, 47-16 and ICGV 87282, respectively (Fig. 6.2c). Hot air temperature

also significantly (P<0.001 to 0.~1) reduced total dry matter production, and

partitioning of dry matter to pods and roots in both cvs of Spanish and Virginia types

(Table 6.2) and the response of different cvs to hot temperature was similar to those

on pod yields. Of the four cultivars ICGV 87282 was clearly the most sensitive to hot

air temperature.

6.3.2 Experiment 2 (1997)

There were significant adverse effects of air temperature and the soil temperature, but

not of their interaction, on flower production, fruit-set, the proportion of pegs

forming pods, reproductive numbers, total dry matter production, pod harvest index,

and pod yields. Botanical type differences between cv. ICGV 86015 (Spanish type)

and cv. ICGV 87282 (Virginia type) were strikingly similar to those described for

Exp. 1 and so are not presented again here.

6.3.2.1 Response of cultivars to hot air temperature or hot soil temperature

Overall, hot air and hot soil temperature reduced pod yield in cv. ICGV 86015 by 21

and 33%, respectively, compared with a reduction of 49 and 64% in cv. ICGV

87282, respectively (Fig. 6.3c).

 62

Table 6.3 Proportion of flowers setting pegs and pods (fruit-set, angular

transformed), number of pegs and pods planf1(reproductive numbers), total

dry matter planf1, pod harvest index, root to shoot ratio and 100 seed weight of

groundnuts grown at mean day/night optimum air (2S0C) or hot air (30°C) or

at mean day/night ambient soil (is°C) or hot soil temperature (34°C). Data are

the mean of cultivars and five replicates.

Air temperature Soil temperature

Trait

Fruit-set e) 46.8 27.9 1.9*** 39.5 35.1 2.0*

Reproductive number plant" 126 76 5.0*** 120 83 4.0***

Total dry matter (g plant") 135.8 122.4 4.2* 151.6 106.5 3.6***

Pod harvest index (%) 38.1 28.0 0.8*** 37.1 28.5 1.3***

Root to shoot ratio (%) 17.l 15.4 0.4* 15.4 17.2 0.5**

100 seed weight (g) 33.9 31.7 1.1 36.3 29.3 2.3*

*, **, ***, Significant at P<0.05, P<O.OI and P<O.OOI, respectively.

Hot air and hot soil temperature had similar absolute effects and reduced

reproductive numbers from about 120 to 80 plant" (Table 6.3). This reduction

however, was achieved in different ways. Hot air temperature significantly (p<0.01)

increased cumulative flower number by 39%, but this response was offset by a

significant (p<0.001) reduction of 40% in fruit-set (Table 6.3, Fig. 6.3). There was
 63

...-t
I
...,_

.s=
(a) DOT
c. 400 .HT

300
200
100
o
(J'
"'-'" (b)
.-
...,_
Q 50

-e
..
~ 40
30
Q
e,
...,_
Q 20
ell
Co) 10
~
o
(c)
80
60
40
20
o
ICGV 86015 ICGV 87282 ICGV 86015 ICGV 87282

Air temperature Soil temperature

Fig. 6.3 Numbers of flowers plant" (a), the proportion of pegs forming pods (peg to pod
1
ratio, angular transformed) (b), and pod yield planf (c) in Spanish cv. ICGV 86015 and
the Virginia cv. ICGV 87282 grown in mean day/night optimum air (25°C, open bars) or
in hot air (30°C, solid bars) and at mean day/night ambient soil (25°C, open bars) or hot
soil (34°C, solid bars) temperature. Vertical bars are the SED for comparing the two
temperatures.
 64

no effect of hot air temperature on the proportion of pegs forming pods. In contrast,

hot soil temperature significantly reduced flower number by 21% (p<0.05), and the

proportion of pegs forming pods by 18% (p<O.OOI), but reduced fruit-set only by

11% (Fig. 6.3, Table 6.3).

Total dry matter production, partitioning of dry matter to pods and pod yield

were all significantly (p<0.05 to <0.001) reduced by hot air and hot soil temperatures

(Table 6.3 and Fig. 6.3c). Root-to-shoot ratio was significantly (p<0.01) affected by

hot air (decreased) and hot soil (increased), but these differences were small. Hot soil

temperature, but not hot air temperature also significantly (p<0.05) reduced 100 seed

weight by 20% (Table 6.3) and this was the principal cause of differences in pod yield

between treatments.

6.4 DISCUSSION

It is clear that hot air (38°/22°C; mean diurnal 30°C) temperature significantly

reduced total dry matter production and the partitioning of dry matter to pods and

therefore pod yields, as described elsewhere (Wood, 1968; Cox, 1979; Ong, 1984;

Ketring, 1984a). For example, Ketring (1984a) reported that the continuous

exposure of plants to hot days (35°C) reduced leaf area, number of pods and pod

yields, relative to values at 30° or 32°C.

Groundnuts attain their maximum leaf net photosynthesis at 30°C and this

was reduced by 25% at 40°C (Ketring et al., 1982a). The principal detrimental affect

of hot air temperature was on fruit-set and not on flower production or on the

proportion of pegs forming pods. Air temperatures of 33°C during flowering reduces
 65

fruit-set as a result of pollen death (De Beer, 1963), and associated reduction in pod

numbers and yields (Ong, 1984; Ketring, 1984a).

Virginia types appear to be more sensitive to heat stress than Spanish types.

This difference in sensitivity was due largely to different patterns of partitioning of

dry matter between botanical typesz rather than to sensitivity to hot temperature per

se. Spanish and Virginia types differ inherently in their growth habit, branching

patterns and rate of flower production (Bunting and Elston, 1980; Summerfield and

Roberts, 1985). Spanish types have an erect and upright growth habit and are

sequentially branched; the appearance of the first flower marks the end of the

vegetative phase, and all subsequent nodes, including the main axis are reproductive

(Bunting and Elston, 1980). Therefore, the rate of flower production in Spanish types

is rapid and reaches an exponential phase immediately after start of flowering (Figs

6.1a and b). This sequence results in the development and growth of a large number

of fairly synchronous fruits and seeds, i.e. to a large sink size, and consequently a

large partitioning coefficient.

Virginia types on the other hand have a running or spreading growth habit

and are alternatively branched; pairs of vegetative nodes alternate with pairs of

reproductive nodes and the main axis is always vegetative (Bunting and Elston,

1980). The rate of flower production is therefore relatively slower during the initial

stages than it is in Spanish types, since the number of reproductive nodes is fewer

(Fig. 6.1c and d). As a consequence of the branching pattern, the rate of

development,judged from the durations to R2, R3 and RS, is also slower in Virginia

types. Both vegetative and reproductive growth continues simultaneously, and so

accumulation of sinks is slower, and the partitioning coefficient is smaller.

 66

Within Spanish and Virginia botanical types, pod yields of cvs ICGV 86015

and ICGV 87282 were more adversely affected by hot air temperature than in 796

and 47-16, respectively. These differences were mainly related to the ability of

cultivars to produce flowers and set fruits, and so to maintain a greater pod harvest

index at hot temperatures (Table 6',2 and Fig. 6.2). The fact that pod harvest indices

of cvs ICGV 86015, 796 and 47-16 were reduced by hot temperatures to a similar

extent, suggests that pod yield also varies with total dry matter production. The

smaller yield of cv. 47-16 at optimum temperatures reflects the late maturity of this

cv. compared to other cultivars.

It is clear that cultivars that have greater yields at optimum temperature (i.e.

are inherently more productive when T=T0) are relatively more susceptible to hot

temperatures. Similar observations have also been reported in many drought studies,

where those cvs with high yields at nonstressed conditions are the most sensitive to

drought (Williams et al., 1986; Nageswara Rao et al., 1989). This association is

consistently strong, apparently because large pod yield potential is dependent on a

high partitioning to the pods, which leaves little flexibility for buffering of the fruit

growth by reduced vegetative growth (Williams and Boote, 1995). However, only

cv. ICGV 87282 was obviously more susceptible to hot air temperature; this cv. had

a similar yield potential to ICGV 86015 and 796 under optimum conditions, but

yields were reduced by hot temperature to the extent of 80% in ICGV 87282

compared with a 30% reduction in other cvs.

Hot air (38°/22°C; mean 30°C) andlor hot soil (38°/30°C; mean 34°C)

temperature imposed from the start of podding until reproductive maturity reduced

pod yields to a similar extent: the effects were additive and without interaction (Fig.

6.4). For example, pod yield was reduced by about 28% by hot air, 42% by hot soil
 67

• 30°C (Hot air)

D 34°C (Hot soil)

~ 30° + 34°C (Hot air + Hot soil)

Total dry matter Pod yield

Temperature treatments
Fig. 6.4 Relative difference (%) in total dry matter and pod yield plant"
at mean day/night hot air (300q or hot soil (34°q or hot air and hot
soil combined temperature treatments in relation to optimum air and
ambient soil temperature control treatment.
 68

and by 67% by a combination of hot air and soil temperature. This additive response

reflected the fact that the effects of hot air and hot soil temperature were different.

Smaller pod yields at hot air temperature were a consequence of reduced fruit-set,

fewer pegs and pods, and hence reduced partitioning of dry matter to pods. In

contrast, smaller pod yields at hot ,soil temperature were due to smaller numbers of

flowers and a reduced proportion of pegs forming pods, and smaller 100 seed

weights (Table 6.3 and Fig. 6.3). These differences in responses to hot air and hot

soil temperature are particularly important in groundnut, as flowers are pollinated and

pegs are formed above the soil surface while fruit growth and seed development

occurs below the soil surface. Other work has shown that hot soil temperature of

>33°C reduce the number of fruits, fruit growth rates, 100 seed weight and so pod

yields (Ono, 1979; Dreyer et al., 1981; Golombek and Johansen, 1997).

Hot air temperature reduced root-to-shoot ratio and this was due mainly to

reduced root growth rather than shoot growth, as found by the work of Wood

(1968), in which hot air temperature of 35°C relative to 20°C reduced root dry

matter by 65%. In contrast to hot air temperature, hot soil temperature increased

root-to-shoot ratios in the present study, as the effects of hot soils were relatively

small i.e. hot soils (34°C) relative to ambient soil (25°C) reduced root dry weight

only by 12%, while shoot dry weight (leaves and stems) was by 22%. Golombek and

Johansen (1997) reported an increase in the specific root length and root weight of

groundnut at hot soil temperature (38°/32°C) relative to cool soil temperature

(200/14°C).

Differences in root growth and root-to-shoot ratios in groundnuts cultivars

have been reported elsewhere (Ketring et al., 1982b; Ketring, 1984b) and it has been

observed that declines in partitioning of dry matter to roots at hot air temperature are
 69

more rapid in heat-susceptible cultivars compared to heat-tolerant ones (Wheeler et

al., 1997). Furthermore, the ability of cultivars to produce a greater root mass at hot

temperatures may be important for a more assured supply of water and nutrients, and

therefore increased pod yields (Gregory and Brown, 1989; Bassirirad et al., 1991).

Investigations on the combined effects of hot soil and hot air temperature on root

growth in groundnuts are needed to identify those cultivars, which have greater

partitioning to roots (and pods) at hot temperatures.

In the experiments described here the plants were not inoculated with rhizobia

and so were dependent solely on inorganic N. The response of plants dependent on

dinitrogen fixation to hot temperature may well be different, as shown for other grain

legumes such as cowpea (Summerfield et al., 1978) and chickpea (Rawsthome et al.,

1985 a, b). The effects of hot air and hot soil temperature on nodulation, dry matter

production, partitioning and yields of groundnuts plants dependent on symbiotic and

inorganic N are presented and discussed in Chapter 10 (pp 128 to 146).

In summary, these two experiments have shown that cultivars within and

between Spanish and Virginia types vary in their response to hot temperatures. These

differences were mainly due to botanical structural differences and the ability of

cultivars to produce flowers and set fruits at hot temperatures. Increasing hot air

andlor hot soil temperature lOoC above the ambient day temperature of 28° and

26°C, respectively, reduced pod yields to a similar extent and the effects were

additive without interaction. The principal impact of hot air temperature was on fruit-

set and the number of fruits to reach maturity, while those of hot soil temperature

were on flower production, the proportion of pegs forming pods and 100 seed

weight.
 70

CHAPTER 7 SENSITIVITY OF GROUNDNUT TO TIMING OF

HEAT STRESS DURING REPRODUCTIVE DEVELOPMENT

7.1 ABSTRACT

Groundnut crops grown in the semi-arid tropics are commonly exposed to damaging

hot temperatures of above 35°C. The objectives of this research were to identify the

time(s) during reproductive development when hot days reduce yield, and to examine

relations between flower production and sensitivity to heat stress. At the start of

flower bud initiation (21 d after planting, OAP) plants of the cvs ICGV 86015 and

ICGV 87282 were grown either at 28°122°C (optimum temperature, OT) or at 38°

122°C (hot temperature, HT) or were reciprocally transferred at 3 d intervals between

the OT to HT regimes and vice versa, until 46 OAP. Transferred plants remained in

the new temperature regime for 6 d before being returned to their original regime. All

plants were harvested at 67 DAP. In cv. ICGV 86015 transfers between 6 d before

and 15 d after flowering (DAF) significantly (p<0.001) affected total number of pegs

(i.e. pegs and pods) and reproductive (peg and pod) dry weight, with the greatest

effect occurring at 9 OAF. In cv. ICGV 87282, number of pegs and reproductive dry

weights were also significantly reduced by transfers at 9 and 12 OAF. Heat stress had

no effect on flower production or the proportion of pegs forming pods, but did

significantly reduce the proportion of flowers producing pegs. Data presented

suggests that it is heat stress during floral bud development that determines peg

number.
 71

7.2 INTRODUCTION

Groundnut is an important oilseed and forage crop grown in many countries in the

semi-arid tropics of Asia and Africa. Pod yields in the semi-arid tropics are poor,

averaging only 800 to 900 kg ha"I, compared with yields of> 2500 kg ha" in the

USA (FAO, 1998). Groundnuts ,grown in semi-arid tropical regions are often

exposed to air temperatures above 35°C (ICRISAT, 1994). Heat stress and moisture

stress have been identified as major constraints to productivity (ICRISAT, 1992).

Furthermore, given the present trend of global warming, temperatures are likely to

become even hotter (Houghton et al., 1990). It is important to identify heat-tolerant

groundnut cultivars for use in breeding programmes, but the effect of heat stress is

not uniform during all stages of the reproductive phase. To do so, it is necessary to

understand the effects of the timing of heat stress on reproductive development and

yield.

Temperature significantlyinfluences the rate of development and growth of

groundnut (Leong and Ong, 1983). The optimum temperature range for vegetative

and reproductive growth and development is between 25° and 30°C (Wood, 1968;

Cox, 1979), although the precise optimum has not been determined for most

processes (Williamsand Boote, 1995). Day temperatures ;;85°C decrease individual

leaf area, reduce the numbers of pegs and pods, and result in lower pod yields

(Ketring, 1984a). Some of these adverse effects may be associated with pollen

mortality, which occurs when temperature is ~33°C (De Beer, 1963). Cultivars differ

in their sensitivity to heat stress during both the vegetative and the reproductive

phases (Nigam et al., 1994). Heat-tolerant cultivars have been identified in West
 72

Africa based on their superior partitioning of dry matter to pods (Greenberg et a/.,

1992).

Although the reproductive phase is relatively more sensitive than the

vegetative phase to heat stress in many crop species (Hall, 1992), no information is

available on the effect of the timing of heat stress during reproductive development

on groundnut pod yield. Pod number is the end product of the number of floral buds

that produce flowers, the proportion of those flowers that are fertilized and produce

pegs, and the proportion of those pegs which develop into pods. Heat stress at any or

all of these stages may reduce pod yield. For example, in common bean (Phaseo/us

vulgaris L.) and cowpea plants are particularly sensitive to hot night temperature

during macro- and micro-sporogenesis, 6 to 8 and 6 to 12 d before anthesis,

respectively (Ahmed et a/., 1992; Gross and Kigel, 1994), and heat stress at these

stages of development causes male sterility. Plants are also sensitive to heat stress at

anthesis and in common bean the negative effects of heat stress at this stage are

associated with the function of the gynoecium (Gross and Kigel, 1994). However, the

post-fertilization and early seed development stages in common bean are more

tolerant to heat stress than the pre-fertilization stages.

The objectives of this research were to identify the time(s) between floral bud

initiation and pod initiation when daytime hot temperature stress causes the largest

reduction in reproductive yield of cultivars representative of Spanish and Virginia

botanical types of groundnuts; and to examine the relation between the sensitivity to

heat stress and flower production.

 73

7.3 MATERIALS AND METHODS

A reciprocal transfer experiment was conducted during the summer months of 1996

in the controlled environment facilities at the Plant Environment Laboratory of the

Department of Agriculture, The University of Reading, UK (51 °27'N lat. and 00°56'

W long.). The experiment involveclmoving plants reciprocally between optimum and

hot temperature regimes at different times and for a fixed period before returning

them to their original temperature regime.

7.3.1 Environmental conditions

The experiment was carried-out in two adjacent polyethylene covered tunnels (poly-

tunnels, 25 m long by 8 m wide by 3 m high at apex) aligned East-West, one

maintained at an optimum day/night temperature of 28°/22°C (OT) and the other at

hot day/optimum night temperature of 38°/22°C (HT). The photo- and thermo-

period in the poly-tunnels was 12 h d-I and the photoperiod was controlled by a

manually operated blackout facility. Air temperatures were measured in each poly-

tunnel with screened and aspirated copper constantan thermocouples positioned at

the top of the plant canopy. Readings were taken at lOs intervals and means for

successive 10 min periods were stored using a data logger (Delta- T Devices Ltd,

Cambridge, UK). Carbon dioxide was at ambient concentration, 360 umol mol", and

relative humidity was maintained at 70 (±5)% through automatic sprinklers during

the day. The poly-tunnels transmitted about 75% of the incoming photosynthetically

active radiation (PAR) and PPFD in both the poly-tunnels averaged about 500 umol

m-2 S-I during the experimental period.

 74

7.3.2 Cultivars and plant husbandry

One Spanish cv., ICGV 8601S, and one Virginia cv., ICGV 87282, were used, the

seeds of which were obtained from the ICRISAT Asia Centre located at Hyderabad

in India.

Uniform seeds of each, cultivar were selected and treated with

Apron Combi 453 FS (metalaxyIx [methyln-(2-methoxyacetyle)-n-(2,6-xylyI)-DL-

(alanite)]+[2-(thiazol-4-yl) benzimidazole: 2- (1,3- thiazol - 4 -yl) benzimidazole] +

[tetramethylthiuram disulfide: bis (dimethylthiocarbamoyl)disulfide]) (Ciba

Agriculture, Cambridge, UK) as a precautionary measure against seed-borne

diseases. They were pre-germinated on the moist filter paper in Petri dishes kept in

the dark for two days at 2SoC, until radicles became visible. The germinated seeds

were planted on 6 June 1996 (DAP=O), one per 15 L pot at a depth of2.5 cm. The

pots were covered with aluminium foil to reduce radiative soil heating. The rooting

medium comprised sand, gravel, vermiculite and loamless peat compost mixed in

proportions of 4:2:2:1, by volume, respectively. A commercial controlled-release

fertiliser (0.15 kg kg" N, 0.10 kg kg" P, 0.12 kg kg" K, 0.02 kg kg" MgO plus

trace elements; Osmocote Plus, Scotts UK Ltd, UK) was incorporated into the

mixture at the manufacturer's recommended rate of 5 g L-t• Seeds were not

inoculated with Rhizobium and plants were therefore wholly dependent on inorganic

nitrogen. All pots were soaked with tap water and allowed to drain for 24 h before

planting; thereafter they were irrigated as necessary through an automatic drip

irrigation system.

All plants were healthy and there were no serious pest or disease problems.

Release of predators (Phytosetulus persimilis Athias-Henriot) and foliar sprays of

Torque (a.i. Fenbutatin Oxide) controlled a mild incidence of red spider mite
 75

(Tetranychus urticae Koch). Thrips (Thrips tabaci Lindeman) were controlled by

release of the predator Amblyseius cucumeris Oudemans. Plants were also sprayed

with Repulse (a.i. Chlorothalonil; 2,4,5,6-tetrachloro-l, 3-benzenedicarbonitrile) to

prevent the occurrence ofleafspot (Cercospora sp.).

7.3.3 Reciprocal transfer treatments

From planting to 21 DAP, when the first flower buds were initiated, all plants were

grown at OT. At 22 DAP, one-half of the plants were transferred to HT. Thereafter,

plants were transferred at 3 d intervals from OT to HT, and from HT to OT, up to 46

DAP, giving a total of nine transfer treatments. Plants remained in the new

temperature regime for 6 d (Wheeler et al., 1997) before being returned to their

original regime, where they remained until harvest at 67 DAP (between R5 and R6;

Boote, 1982). Plants were harvested at these stages rather than at maturity, to ensure

that the effects of the transfer treatments could be clearly defined (Wheeler et al.,

1997). Plants remaining in the 28°/22°C and 38°/22°C environments from 22 to 67

DAP served as OT and HT controls, respectively. There were five replicate plants

for the transfer treatments and 10 replicate plants for each control treatment.

7.3.4 Observations and data analysis

Duration (d) from planting to the appearance of first fully opened flowers (RI), and

pegs (R2) were noted on all plants. The number of flowers that opened each day was

determined until final harvest. The times of pod (R3) and seed (R5) initiation were

determined from a destructive harvest in an adjacent experiment.

 76

At the final harvest (67 DAP), plants were removed from each pot and

separated into roots, leaves, stems, pegs and pods. The numbers of pegs and pods per

plant were recorded and roots were washed with water to remove the potting

medium. The respective weights of roots, leaves, stems, pegs and pods per plant

were recorded after oven-drying these components to a constant weight for 3 d at

80°C. Total dry weight and pod harvest index, the ratio of pod to total dry weight

(inclusive of senesced leaves and roots), were calculated from the weights of

individual components. Total dry weight values were adjusted to allow for the oil

content of the seeds (Duncan et al., 1978). Fruit-set was defined as the ratio of total

number of pegs to total number of flowers, i.e. proportion offertilized flowers. Pegs

forming pods was defined as the ratio of total number of pods to total number of

pegs, i.e. the proportion of fertilized flowers forming pods.

The 3 d moving averages of number of flowers produced per day during the

period from first flowering until 42 DAF were calculated based on the number of

flowers produced per day. The respective number of flowers produced during the 6 d

stress period and subsequently in the 6 d period following stress were calculated for

all transfer treatments. Analysis of variance was used to compare these values with

the number of flowers produced during the same period in the OT and HT controls.

All data are expressed on a per plant basis unless otherwise stated.

The experiment was designed as a randomised block with added control.

Each reciprocal transfer treatment was replicated five times and the controls were

replicated 10 times in order to increase the precision of comparisons with the transfer

treatments. The analysis of variance for all the variables was performed using Genstat

5 (Genstat 5 Committee, 1987).

 77

7.4 RESULTS

The mean day and night temperatures (± SD) in the OT poly-tunnel were 27.9 (±1.4)

°C and 22.1{±1.2)OC, respectively; corresponding values in the HT poly-tunnel were

38.4{±1.1)OC and 22.3{±O.9)OC, respectively. There were significant effects of

cultivar, temperature and transfers- on all of the traits given in Tables 7.1 and 7.2

unless specific mention indicates otherwise.

7.4.1 Cultivar responses

Durations from planting to the respective appearance of the first flower (RI), first

peg (R2), first pod (R3) and first seed (R5) were 28, 37, 45 and 52 d in the Spanish

cv. ICGV 86015. The first flower and peg in the Virginia cv. ICGV 87282 also

appeared after 28 and 37 DAP, respectively, but pods and seeds first appeared 4 and

7 d later, respectively than in ICGV 86015. There was no difference between cvs in

total dry weight, but ICGV 86015 had a significantly (P<O.OOI) greater pod harvest

index than ICGV 87282 (Table 7.1). The pattern of flower production over time was

similar in both cvs (Fig. 7.1). The number of flowers opening each day increased until

24 or 27 DAF, when seeds started to grow (R5), and declined thereafter. The

maximum number of flowers opening each day was greater in ICGV 86015 than in

ICGV 87282, and so significantly (P<O.OOI) more flowers were produced in ICGV

86015 (Table 7.1). Flower production had effectively stopped in ICGV 86015 by 40

DAF, while in ICGV 87282 more than 5 flowers dol were still being produced at 42

DAF. There was no significant (P>0.44) difference between cvs in the proportion of

flowers setting pegs (fruit-set), and so ICGV 86015 produced significantly «0.001)

more pods than ICGV 87282.

 78

Table 7.1 Total dry weight, pod harvest index, total flower number, proportion

of flowers setting pegs (fruit-set), and the number of pods in the Spanish cv,

ICGV 86015 and Virginia cv. ICGV 87282. Data are means of temperature and

transfer treatments.

Trait ICGV 86015 ICGV 87282 SEDt

Total dry weight (g plant") 103 100 1.58

Pod harvest index (%) 41 23 0.72***

Flower number plant" 339 262 7.60***

Fruit-set (%) 31 32 0.98

Pod number plant" 51 41 1.32***

*** Significant at P~O.OOI probability level.

t SED = Standard Error of Difference of means (1,172 df).

7.4.2 Effect of temperature

The effects of temperature on both cvs were similar, and there was no temperature x

cultivar interaction (P>O.75). There was little or no effect of HT on the times to

appearance of the first peg, pod or seed. Hot temperature significantly (P<O.OOI)

reduced total dry weight and pod harvest index (Table 7.2). Hot temperature also

increased the total number of flowers per plant (P<O.OS), an effect associated with an

increased rate, rather than duration, of flower production (Fig. 7.1).

 79

.._ ~_
en
QJ

••= ....
en
UU
0 0
t-i M
~ f1lQJ
•• .c
C'IC'I >~
QJ •
!::!!::!

--_
-= en
o 0
~ \0 .... ~
00 00
C'lM
('f') -=
.t:I....
"0 ...
•0 _
= Q.
QJ
~
t-i
=
('f')
.......e
'f)QJ
Cl C

~
~!
t-i
M
>~
C·S
t-i u;
~Q.
M QC) .e
~ C
QC) ~ .cCol
M III ...
r-
QC)
H ·2~
M -Q
H ~ ~~
G
U
H .=_
ell

a....
.cQJ
_
QJVJ
.........
_ en
c U•
0
~ ~ _"'M
-
~ \0
..c ~ =-M
.... .c: .........
0
c: ~-
.5 ...s_ ..., &0
M

= .....
VJ
J....
~ t: _
c~~
M
~
t=:
J....
c .~
.....
e 0 't
~.-
QJ I,::-QJ
I---i
\0
c::C': s uo
C M
c.
ell
~
~ ('f')
..::M
0--"0
..
VJ
t-i ~ :.000 ~
-I"') ell
C':
=
QJ "0 "0
e c =
('f')
~
'= "'-"0
CIS CC
QJ. C
>- C.
Cu ..
ltl ~ ... 0 ~
~ M I M QJ
"0 M C.
~.... c;-
00
....
-;M~........
III

QC)
~ c._C
l/')
~ "0 >. QJ
QJ
Col-III Col
Q ==_ =
\0
QC) H M
~
"oc
C =_ ...
-.5 ~
G
U
H ...
c. ....c.
C _
eIl=c.
QJ Col QJ
~=-=
~ c~ ....
~ Cc~
~ .......C
'-" C ~-=
"M III
0 QJOO~
.t:IM
l/') Q l/') <= <= §~"O
M M ~ ~ Z>;
-C
r-:UC2
(l-P IJURId ·OU)uO!JJnpold l~M.OU JO alRlI ..bD"~ ~..
~CoI
 80

Table 7.2 Total dry weight, pod harvest index, total flower number, proportion

of flowers setting pegs (fruit-set), proportion of pegs forming pods, and the

number of pods in the optimum (28°l2rC) and hot temperature (38°l2rC)

control treatments. Data are means of cultivars and transfer treatments.

Temperature

Trait SEDt

Total dry weight (g plant") 123 81 3.71***

Pod harvest index (%) 39 26 1.69***

Flower number plant" 279 336 17.9*

Fruit-set (%) 45 18 2.31 ***

Pegs forming pods (%) 50 57 1.79*

Pod number plant" 60 33 3.10***

.,**. Significant at P~0.05 and P~O.OOI probability levels, respectively.

t SED = Standard Error of Difference of means of transfer x temperature interaction

(9,172 dt).

7.4.3 Effect of reciprocal transfers

Transferring plants for 6 d from OT to HT, or from HT to OT, had significant

(p<0.001 and P<0.05, respectively) effects on the number of pegs and reproductive

dry weights (Fig. 7.2). When plants of ICGV 86015 were transferred from OT to

HT, and vice-versa, between 6 d before flowering and 15 OAF, the number of pegs
 81

bb;
eo
~ ~ .
.;~.
,.?
~
•• ••
• ••
.
~

r
+ ~ 1:
~T
•• ••
• ••
~
o ~?
.: p
M
I

\0
I
• • •• • ••
•
··e··
• •
'·0
••.•.••• •'<).•
• =
o
U

~ f?
s r?
~~:
u
~.. 1\ · · ••
- •• ••
\0
I

~
..!,
:·e·: :'0 e~ :'0 =
e
U

(IJURld ~) lq~!aA\
A.lP aA!JJnpo.ldalI
 82

was either significantly reduced (OT to HT) or significantly increased (ID to OT)

relative to the respective controls (Fig. 7.2a). Transfers at 18 DAF had no significant

effect on number of pegs. Irrespective of the direction of transfer to and from OT or

HT, the effect of temperature increased with time of transfer until 9 DAF, which

coincided with first peg appearanc~ (R2). The effect of transfers on reproductive dry

weights were similar to those on number of pegs; reproductive dry weights were

significantly (P<O.OI) reduced or increased relative to their respective controls when

the plants were transferred between start of flowering and 15 DAF.

In ICGV 87282, the effects of the transfers to or from OT and HT were less

marked than in ICGV 86015 (Fig. 7.2b). The only significant reduction in number of

pegs occurred when plants were transferred from OT to HT at 9 and 12 DAF, while

in transfers from HT to OT, number of pegs were significantly increased only at

6 and 9 DAF. Similarly, transfers from OT to HT at 12 and IS DAF only significantly

affected reproductive dry weights.

Hot temperature had a significant effect (P<O.OOI) on the total flower number

in the controls (Tables 7.2 and 7.3; Fig. 7.1) and in the reciprocal transfer treatments

(Table 7.3). Hot temperature increased flower production in the HT control and OT

to HT transfer treatments and vice versa in the HT to OT transfer treatments.

However, these changes in flower production only occurred in 6 d following transfers

to HT or OT (P<O.OI) (Table 7.3); during the 6 d OT or HT stress period,

temperature had no significant (P>O.35) effect on flower production (Table 7.3).

 83

Table 7.3 Number of flowers produced during the 6 d 'stress' period and in the

6 d following the 'stress' period in optimum (28°/22°C) and hot (38°122°C)

temperature control and reciprocal transfer treatments. Data are means of

cuItivars and transfer treatments.

Treatment During 6 d stress period In 6 d following stress period

Flowers plant"

OT Control 31 45

Transferred to HT 32 53

HT Control 40 58

Transferred to OT 35 48

SEDt 3.4 3.7**

** Significant at P$O.OI probability level.

OT= Optimum temperature and HT = Hot temperature treatment

t SED = Standard Error of Difference of means (1,468 dt).

7.S DISCUSSION

Temperature plays an important role in all aspects of crop growth and development

(Ong, 1986) and groundnut crops in the semi-arid tropics are rarely grown under

optimal conditions. It is clear from the present research that a day temperature of

38°C (mean diurnal temperature 30°C) imposed during the reproductive phase is

supra-optimal and reduces early reproductive yield in both Spanish and Virginia
 84

cultivars. Ketring (1984a) also reported that 35°/22°C reduced number of pegs by

33% relative to 30o/22°C. This is consistent with a 50% reduction at 38°/22°C

relative to 28°/22°C found in the present study.

The reciprocal transfers clearly demonstrated that hot days imposed from 6 d

before flowering until 15 DAP in.ICGV 86015, and from 6 to 15 DAF in ICGV

87282, significantly reduced number of pegs and reproductive dry weights. The

pattern of flower production and start of peg initiation was similar in both cvs. The

largest reduction in reproductive dry weight occurred at 9 DAF. This coincided with

first peg appearance (R2), and the period when flower production rates in the initial

20 DAP in OT were near maximal. Given that flower production started to decline

about the time seeds began to grow (R5), the end of the heat-sensitive period

probably reflects the changing pattern of resource allocation between developing and

growing fruits and vegetative growth (in ICGV 87282) and reproductive node

initiation (Bunting and Elston, 1980).

The duration of the sensitive period was much shorter in ICGV 87282 than in

ICGV 86015. This apparent difference in sensitivity was associated with slower rates

of development and flower production in ICGV 87282. However, Virginia cvs such

as ICGV 87282 generally produce fewer, but larger fruits and the difference in the

duration of the sensitive period probably reflects the different reproductive strategies

of Virginia and Spanish types, rather than sensitivity per se. The remainder of the

discussion will therefore focus on the more responsive Spanish cv. ICGV 86015.

The main effect of HT was on fruit-set, i.e. the proportion of flowers

producing pegs. Hot temperature had no deleterious effect on flower production or

on the proportion of pegs producing pods. These data therefore point towards

fertilization as the processes most affected by HT in groundnut. Fruit-set and

 85

fertilization have also been found to be particularly sensitive to HT in a number of

other legumes, including cowpea (Hall, 1992; Ahmed et al., 1992) and common bean

(Monterroso and Wien, 1990; Gross and Kigel, 1994). Detailed studies of individual

flower buds and flowers have shown that the most sensitive stage of development to

HT occurs at microsporogenesis, 6,to 8 and 10 to 12 d before anthesis in cowpea and

common bean, respectively. Hot temperature at microsporogenesis causes low pollen

viability, poor anther dehiscence and hence male sterility. The reciprocal transfer

treatments that started 6 d before flowering significantly reduced peg number (Fig.

7.1). These transfers ended at flowering and can therefore only have affected

processes that occurred during floral bud development and before anthesis. In

groundnut microsporogenesis occurs 3 to 6 d before anthesis (Martin et al., 1974; Xi,

1991). Therefore, it seems highly likely that microsporogenesis in groundnut is also

sensitive to hot temperature.

If floral buds at a stage 3 to 6 d before anthesis are sensitive to temperature,

then exposure to HT at this stage (Le. transfer from OT to HT) will result in male

sterility and lower peg numbers. Conversely, floral buds exposed to OT at this stage

(Le. transfers from HT to OT) should set fruits normally and have greater peg

numbers. To test this hypothesis, the total number of flowers opened in 6 d periods,

at 3 d intervals, starting from 3 d before to 12 d after the start of each HT or OT

reciprocal transfer treatment, were regressed against peg number at harvest expressed

as a percentage of the control value. Transfers at 15 and 18 DAF were excluded,

since after 12 DAF the direction of the response was reversed. The variances of these

regressions were compared (Fig. 7.3). The lowest variance for transfers from OT to

HT and HT to OT was found for flower production starting 6 d after the start of the
 86

6 d period relative to start of

temperature treatment

Fig. 7.3 Variance of regressions of number of pegs relative to control values

against the cumulative number of flowers opening in each 6 d period relative
to start oftemperature treatment in plants transferred (a) from 28°/22°e (OT)
to 38°/22°e (HT) and (b) from 38°/22°e to 28°/22°e.
 87

.....0 95
~ (a) OT to HT
>~~
.....
.....c

-- --
~...._"

85
.-=
~
• 0

~=
~
e 0
c.J 75
=E-4
=0 y = 99.5(:1:1.4) -O.5(:l:O.03)x

~
CJ)
~
65
2
r =0.98 •
15 30 45 60 75
Flowers plant" in 6 d following HT

.....0 180
.-
~
>~
~
.....c
(b) HT to OT
A

-- --
~...._"

160
.-=
~
•

~=
0

~
e 0
c.J
140
=E-4
==
CJ)
~
y = 115(±6.5) +0.7(±0.14)x
2
r =0.84
~ 120

15 30 45 60 75
Flowers plant" in 6 d following OT

Fig. 7.4 Relations between (a) total number of pegs relative to 28°/22°e (OT) control
values and the number of flowers produced in the 6 d period following an episode of
38°/22°e (HT) and (b) total number of pegs relative to 38°/22°e control values and
the number of flowers produced in 6 d period following an episode of 28°/22°C.
 88

HT or OT treatment. These data show that variation in peg number was most closely

associated with variation in flower production in 6 d period after exposure to HT or

~T. Although the fate of individual flowers was not followed in the current

experiment, the number of flowers opening in the 6 d period after the HT or OT

transfer treatment (Table 7.3) should reflect the number offloral buds exposed to HT

or OT at 0 to 6 d before anthesis. These relations therefore confirm indirectly that

floral bud development is most sensitiveto temperature.

There were strong linear relations between peg number relative to control

values and the number of flowers produced during the 6 d period following HT or

OT (Fig. 7.4). As expected from the hypothesis outlined in the previous paragraph,

transfers from OT to HT reduced peg numbers whereas transfers from HT to OT

increased peg numbers. Clearly, only a short exposure to favourable (OT) or

unfavourable (HT) temperature at the sensitive stages of development, 0 to 6 d

before anthesis, can have significant effects on reproductive potential. Furthermore,

the longer the period between first flowering and exposure to OT or HT, the greater

the effect on peg number, since floral bud and flower production increases

ontogenetically until pod and seed growth start (Figs 7.1 and 7.2). In other words,

plants are more sensitive to HT or OT at 9 DAF than 0 DAF because flower bud

production is greater.

In summary, this research has revealed that sensitivity to heat stress in

groundnut extends from 6 d before anthesis until 15 d after flowering, and that the

magnitude of sensitivity depends on the number of floral buds which are exposed to

heat stress in the 6 d period before anthesis.

 89

CHAPTER 8 EFFECTS OF SHORT EPISODES OF HEAT

STRESS ON FLOWER PRODUCTION AND FRUIT-SET OF

GROUNDNUT

8.1 ABSTRACT

Heat stress is one of major environmental constraint limiting seed yields of groundnut

crops grown in the semi-arid tropics. The objectives of this research were to

determine: (i) the effects of short episodes of exposure to hot temperature on flower

production (FN), the proportion of flowers forming fruits (fruit-set) and the number

of pegs and pods per plant (RNt); (ii) whether fruit-set is affected by hot temperature

during different periods of daylight in each diurnal cycle; and (iii) whether the

responses of FN, fruit-set and RNt to temperature were qualitative or quantitative.

Plants of cv. ICGV 86015 were grown in controlled environments at a day/night

temperatures of 28°/22°C from sowing until 9 d after flowering (DAF). Then,

cohorts of plants were: (a) exposed to day temperature of 28°, 34°,42° or 48°C for

2,4 or 6 d; or were (b) exposed to 34°, 42° or 48°C for 6 d either throughout a 12 h

day (0800 to 2000 h, WD), or only during the first 6 h (AM) or second 6 h (PM) of

the day. Values of RNt were significantly reduced by hot temperature, by duration of

exposure and timing of exposure. Hot temperature (>34°C) during AM and WD

significantly reduced fruit-set and hence RNt. In contrast, hot PM temperature had no

effect on fruit-set. Variation in FN was quantitatively related to day temperature over

the range 28° to 48°C. In contrast, variation in fruit-set was quantitatively related to

AM temperature above a critical value of 37.3 °C. These findings indicate that the
 90

response ofRNt to day temperature is quantitative and can be modelled by combining

the responses ofFN and fruit-set to temperature.

8.2 INTRODUCTION
.
Hot temperatures are a major constraint to crop adaptation and productivity,

especially when these temperature extremes coincide with drought and with critical

stages of plant development (McWilliam, 1980). Groundnuts are an important

subsistence and cash crop of the semi-arid tropics where they are often exposed to

maximum temperatures of >40°C for short periods during the growing season

(ICRISAT, 1994). Furthermore, with present trends of global warming, temperatures

are likely to become hotter, and an increase in mean temperature of 2° to 3°C is

predicted to reduce groundnut yields in India by 23 to 36% (Hundal and Kaur, 1996).

It is therefore important to investigate and quantify the effects of episodes of hot

temperature on reproductive yield of groundnut, both to improve our ability to

simulate and predict responses to environment, and to help design screening methods

for heat tolerance.

The optimum temperature for growth and development of groundnut is

between 25° and 30°C (Williams and Boote, 1995). For example, Ketring (1984a)

showed that the numbers of pegs and pods were reduced by 33% by exposure to a

day temperature of 35°C compared with 30°C. The reproductive phase of groundnut

is more sensitive to heat stress than the vegetative phase (Cox, 1979; Ketring,

1984a). The greatest sensitivity to hot days (38°C) occurs from 6 d before to 15 d

after flowering (Vara Prasad et al., 1998). In the latter experiment, the reduction in

the number of fruits (i.e. pegs and pods) following exposure to 38°C was primarily
 91

due to reduced fruit-set (proportion of flowers producing pegs or pods), rather than

to reduced flower production or to the proportion of pegs producing pods. Studies

on cowpea (Hall, 1992) and common bean (Gross and Kigel, 1994) have also shown

significant deleterious effects of heat stress on fruit-set.

Although episodes of hot t~mperatures (~ 35°C) are known to significantly

reduce the number of fiuits, and therefore seed yields in groundnut, the quantitative

sensitivity of flower production and fiuit-set to temperature has not been determined.

The objectives of the present research were to determine: (i) the effects of short

episodes (2, 4 or 6 d) of exposure to hot temperatures, i.e. 34°, 42° or 48°C, on

flower production, fruit-set and the number of pegs and pods per plant; (ii) whether

fruit-set in groundnut is relatively more or less sensitive to hot temperatures during

different periods of daylight in each diurnal cycle; and (iii) whether the responses of

flower production, fruit-set and the numbers of pegs and pods per plant to

temperature were quantitative or qualitative.

8.3 MATERIALS AND METHODS

Two experiments were conducted between June and August 1997 in the controlled

environment facilities of the Plant Environment Laboratory, Department of

Agriculture, The University of Reading (51°27'N lat. and OOo56'W long.). They were

undertaken in a polyethylene covered tunnel (poly-tunnel) maintained at a near

optimum day/night temperature of 28°/22°C and in four modified Saxcil growth

cabinets maintained at different day temperatures (Table 8.1).

 92

Table 8.1 Mean day and night air temperatures ee) in the poly-tunnel from

sowing to 37 d after sowing (DAS) and from 44 to 52 DAS, and in the four 6 d

temperature treatments imposed in the growth cabinets between 38 to 43 DAS.

Time (DAS) Location Target day temperature eC) Mean air temperature eC)

Day Night

o to 37 Poly-tunnel 28 28.2 22.3

38 to 43 Cabinets 28 (control) 27.8 22.3

34 33.8 22.7

42 41.8 22.8

48 47.8 23.1

44 to 52 Poly-tunnel 28 28.4 22.4

The photo- and thermo-period in both the poly-tunnel and each cabinets were

coincident and equal at 12 h dol. The photoperiod was controlled by a manually

operated blackout facility in the poly-tunnel, and by automatic time switches in the

cabinets. Air temperatures were measured in the poly-tunnel and the cabinets using

screened and aspirated copper constantan thermocouples positioned at the top of the

plant canopy. Readings were taken at 10 s intervals and means were stored for

successive 30 min periods using a data logger (Delta-T Devices Ltd, Cambridge,

UK). Carbon dioxide concentration in the cabinets was maintained at ambient

concentration, 360 umol mol" of air. Relative humidity during the day was
 93

maintained close to 70(±5)% in the poly-tunnel using water sprinklers and ventilation,

while in the cabinets VPD was maintained at 1.2 kPa in all temperature regimes. The

poly-tunnel transmitted 75% of incoming photosynthetically active radiation and

photosynthetic photon flux density (PPFD) averaged 590 umol m-2s-1 during the

experiment. The corresponding valve of PPFD in each growth cabinet was 650 urnol

m-2s-1 obtained from a combination of cool white fluorescent tubes and incandescent

lamps.

8.3.1 Temperature treatments

During the period from sowing until 9 d after flowering (OAF), all plants were grown

at a day/night temperature of28°/22°C in a poly-tunnel. Thereafter, cohorts of plants

were transferred to different growth cabinets and exposed to various temperatures

for a maximum of 6 d before being returned to the poly-tunnel maintained at

28°/22°C, where they remained until final harvest at 24 DAF. Plants maintained in the

28°/22°C cabinet for the entire 6 d treatment period served as controls for both

experiments. Night temperature in all the treatments was maintained at 22°C.

8.3.1.1 Experiment 1. Duration of exposure to hot temperature

At 9 DAF, replicate plants were transferred to four different growth cabinets and

exposed to hot days of 34°, 42° or 48°C for 2, 4 or 6 d. In the 2 and 4 d duration

treatments, plants were subsequently transferred to a cabinet maintained at 28°C for

the remainder of the 6 d period before being returned to the poly-tunnel.

 94

8.3.1.2 Experiment 2. Timing of exposure to hot temperature

At 9 DAF, replicate plants were exposed to hot days of 34°, 42° or 48°C during

either the first 6 h (0800 to 1400 h, AM) or for the second 6 h (1400 to 2000 h, PM)

of the 12 h light period, or for the whole day (0800 to 2000 h, WD) for 6 d (i.e., 9 to

15 DAF). In the AM: treatments plants were exposed to hot temperatures of 34°, 42°

or 48°C in different cabinets during the first 6 h of the light period and then

transferred to an optimum day temperature (28°C) cabinet for the second 6 h of the

light period. In contrast, in the PM treatment plants were exposed to optimum

temperature (28°C) during the first 6 h of the light period and then transferred to hot

temperatures of 34°, 42° or 48°C in different cabinets for the second 6 h of the light

period.

8.3.2 Cultivar and plant husbandry

Uniform seeds of the Spanish botanical type (A. hypogaea subsp. fastigiata) cv.

ICGV 86015 were selected and treated with Apron Comhi 453 FS (Ciha,

Agriculture, Cambridge, UK) as a precautionary measure against seed-borne

diseases. Seeds were pre-germinated on moist filter paper in Petri dishes kept in the

dark for 2 d at 25°C until radicles emerged. The germinated seeds were then sown on

3 June 1997, one per 2.5 L pot at a depth of2.S cm. The sides of pots were covered

with aluminium foil to reduce radiative heating. The rooting medium comprised sand,

gravel, vermiculite and loamless peat compost mixed in proportions of 4:2:2:1, by

volume, respectively. A commercial controlled-released fertiliser (0.15 kg kg" N,

0.10 kg kg" P, 0.12 kg kg" K, 0.02 kg kg" MgO plus trace elements; Osmocote

Plus, Scotts UK Ltd, UK) was incorporated into the mixture at the manufacturer's
 95

recommended rate of 5 g L-l. Seeds were not inoculated with rhizobia and plants

were dependent on inorganic nitrogen. All pots were soaked with tap water and

allowed to drain for 24 h before sowing; thereafter they were irrigated as necessary

through an automatic drip irrigation system in the poly-tunnels or were hand-watered

.
during the 6 d period in the cabinets. There were no disease problems and sporadic

pest infestations were controlled by releases of the predators Phytoseiu/us persimi/is

Athias-Henriot against red spider mite (Tetranychus urticae Koch) and Amblyseius

cucumeris Oudemans against thrips (Thrips tabaci Lindeman).

Both experiments were sown with six replicates of each temperature

treatment and with 12 replicates for the controls. Only uniform plants which flowered

28 d after sowing (DAS) were selected for the experiments to ensure no effects of

'escape' and the rest were discarded, which gave four replicates of each temperature

treatment and eight replicates of the controls. A subset of four plants from the

control treatment were harvested at 18 DAF, to estimate the numbers of pegs and

pods produced from those flowers which had opened before the imposition of the

target treatments.

8.3.3 Observations

Durations (d) from sowing to the appearance of the first open flower and to the first

peg ~ 3 mm long were recorded on all plants. Thereafter, the numbers of flowers

opening each day were counted until final harvest. Plants were harvested on two

occasions; first at 9 d after the start of the temperature treatments (18 DAF) and then

at final harvest taken 9 d after the end of the temperature treatments (24 DAF). At

both harvests the numbers of pegs and pods per plant were counted and the dry
 96

weights of roots, stems (including petioles), leaves, pegs, and pods were determined

after oven-drying at 80°C for 3d.

8.3.4 Data analysis

.
The fate of individual flowers (i.e., whether they produced a peg or a pod) was not

monitored and so the total number of pegs and pods at the final harvest, hereafter

referred as the final reproductive number (RNf), were produced from the flowers

opened at 28°C (i.e. in the poly-tunnel) and in the 6 d temperature treatments in the

cabinets. In order to determine the fate of those flowers which opened during the 6 d

temperature treatment, it was assumed that for cv. ICGV 86015 the time from flower

opening to peg appearance was 9 d (authors, unpublished data). Therefore, all

flowers that opened and were fertilized between the onset of flowering and 9 OAF

should have produced a peg or a pod by the time the first harvest was taken at 18

OAF. The number of pegs and pods at this harvest was referred as the initial

reproductive number (RNi). Similarly, any flower that opened between the end of the

temperature treatments (15 OAF) and the final harvest 9 d later (24 OAF), should not

have formed a peg. Therefore, the numbers of pegs and pods arising from those

flowers that opened during the 6 d temperature treatment (the treatment reproductive

number. RNt) were estimated as the difference between RNcand RNi.

Fruit-set during the temperature treatments was calculated as the ratio of RNt

to the cumulative flower number opened during the 6 d temperature treatment (FN).

The values of fiuit-set were subject to angular transformation before analysis to

ensure homogeneity of variance. Data on FN, fruit-set and RNt were analysed as a

split-plot design with four replicates using Genstat 5 (Genstat 5 Committee, 1987).
 97

8.4 RESULTS AND DISCUSSION

There was no significant difference among plants (p>O.50) in the number of flowers

produced before the start of the heat stress treatments at 9 DAF. The number of pegs

produced from flowers opening before 9 DAF (RNi), was 11(SD ±1.8).

8.4.1 Experiment 1. Duration of exposure to hot temperature

The values ofFN, fruit-set and RNt were all significantly affected by day temperature,

duration of exposure to temperature (p<O.OOl for each) and their interaction (p<O.05

to P<O.Ol). Values of RNt were reduced by temperatures ~42°C and by 4 or 6 d

exposure to hot temperature, with the relative decrease in RNt being greater when

plants were exposed to hotter temperature for a longer duration (Table 8.2).

Day temperature and duration of exposure to hot temperature had similar

effects on fruit-set (Table 8.3), although it was only in the most extreme hot

temperature regimes, where a 6 d exposure to 42°C and 4 and 6 d exposures to

48°C, significantly (p<0.001) reduced fruit-set to below the mean value of 50.2%.

Flower number was also affected in a similar negative manner by the temperature

treatments and there was a strong and positive linear relation (~=0.94, n=10;

P<O.OOl) between RNt and FN (Fig. 8.1). Previous studies by Ong (1984) have

shown for groundnut cv. Robut 33-1, a mean diurnal temperature regime of

36°/25°C imposed from sowing to maturity reduced the number of pods by 50%

relative to those produced by plants grown at 27°/17°C. Similarly, Ketring (1984a)

recorded that a temperature regime of 35°/22°C (day/night) from flowering to

maturity reduced fruit numbers by about 33% compared with values at 30o/22°C.
 98

However, in the present study, only a 4 or 6 d exposure to 42° and 34° C,

respectively, were required to reduce RNt by a similar extent (Table 8.3).

Table 8.2 Effect of 2, 4, or 6 d duration of exposure to hot days of 34°, 42° or

48°C on the numbers of pegs and pods produced during the 6 d treatment

period (RNt plant"), Control plants were maintained at 28°C for 6 d.

Temperature eC) Duration of exposure (d)

2 4 6 Mean

28 (control) 16.3 16.3

34 19.8 18.3 10.5 16.2

42 13.3 11.8 4.0 9.7

48 13.8 1.3 0.0 5.0

Mean 15.6 10.5 4.8

SED for duration of exposure (df 2, 18)=1.44*** and for temperature x duration

interaction (df4, 18) =1.79*

*, *** Significant at P<0.05 and P<O.OOI probability levels, respectively.

 99

25

,,-.....
20
~
I....

15
-=c.
~

'-'
.... 10
~ 5
0

o 10 20 30 40
-1
FN (plant)

Fig 8.1 Relation between the number of pegs and pods (RN. plant") and the
number of flowers (FN plant") in plants exposed to day temperture of 28° to
48°C for 2, 4, or 6 d. Fitted line: RN,= -6.41 (±1.57)+O.80(±O.07) FN, r2=O.94,
n=10, P<O.OOI.
 100

Table 8.3 Effect of 2, 4, or 6 d duration of exposure to hot days of 34°, 42° or

48°C on the proportion of flowers producing fruits (fruit-set, angular

transformed) during the 6 d treatment period. Control plants were maintained

at 28°C for 6 d.

Temperature eC) Duration of exposure (d)

2 4 6 Mean

28 (control) 51.2 51.2

34 52.5 49.3 46.9 49.5

42 50.5 49.3 25.9 41.9

48 51.8 15.8 0.0 22.5

Mean 51.6 38.1 24.3

SED for duration of exposure (df 2, 18)=5.43*** and for temperature x duration

interaction (df 4, 18)=5.16*

*, *** Significant at P<0.05 and P<O.OOI probability levels, respectively.

8.4.2 Experiment 2. Timing of exposure to hot temperature

Hot temperature imposed for only the first 6 h (AM) or the second 6 h (PM) of the

12 h light period in each diurnal cycle had no significant (P>0.10) effect on FN.

However, exposure to 34°, 42° or 48°C for the WO significantly reduced FN when

compared to plants maintained at 28°C (Fig. 8.2a). There was a strong and negative
 101

(a) I
.-...
~
..... 30

-=
C':
c-
.._., 20
Z
~
10

(b)
.-... 60 I
e....,
....
<lJ

.-=rIl
..!.
"-
40

~ 20

(c) I
~
.-... 20
• ....
-=-
e,
C':

.._.,
10
~

28 34 42 48
Maximum day temperature (CC)

Fig. S.2 Effect of timing of exposure to hot temperature on (a) number of flowers
(FN plant!); (b) proportion of flowers settin, pegs (fruit-set, angular transformed);
and (c) number of pegs and pods (RN, plant" ) produced during a 6 d treatment
period when plants were exposed to day temperature of 28°,34°,42° or 4SoC for
the whole 12 h day (0800 to 2000 h; WD, 0) or for only during the first 6 h
(OSOOto 1400 b; AM, A), or during the second 6 h (1400 to 2000 h; PM, C) of
the 12 h day. Vertical bars denote the SED for treatments.
 102

linear relation between FN and mean day temperature (?=0.98, n=4; P<O.OOI;Fig.

8.2a) and FN decreased by 1.1 plant" °Cl.

In contrast to FN, fruit-set was significantly affected by the timing of

exposure to hot temperature (Fig. 8.2b). When plants were exposed to hot

temperatures during PM, temperatu.re had no significant effect on fruit-set. However,

when plants were exposed to hot temperature during AM or for the WD, fruit-set

was reduced from 51% at 28°C to 0 % at 48°C. The effects of hot temperature

during AM or WD were similar (Fig. 8.2b) and fruit-set was decreased by about

2.5% planrloC-l between 28° and 48°C. Values of RNt were affected by the timing

of exposure to hot temperature (P<O.OOI)in a similar manner to that of fruit-set (Fig.

8.2c).

This experiment has revealed that fruit-set is strongly influenced by timing of

hot temperature, and that it is the temperature during AM, rather than the WD, that

determines the response. Groundnut flowers typically open early in the morning, self-

pollination occurs just before opening, and fertilization is completed within 5 to 6 h

later (Lim and Gumpil, 1984). Therefore, it is perhaps not surprising that exposure to

hot temperatures during AM, but not PM, reduces fruit-set and hence RNt• In

contrast, the similar effects of AM and PM temperatures on FN, suggest that it is the

mean day temperature rather than AM or PM temperature which most affects flower .

production.

8.4.3 Relations between FN, fruit-set, RNt and temperature

To examine the relation between FN and day temperature, and fruit-set and AM

temperature, data from both experiments have been combined in Fig. 8.3. There was
 103

Fig. 8.3 Relations between (a) number of flowers (FN plant") and mean day
(0800 to 2000 h) temperature; (b) percentage fruit-set (angular transformed)
and mean AM (0800 to 1400 h) temperature; and (c) number of pegs and pods
(RNt plant") and mean day (OSOOto 2000 h) temperature during a 6 d
treatment period when plants were exposed to temperatures of 2So, 34°, 42° or
4SoC for 2, 4 or 6 d periods (0), or exposed only during the first 6 h (OSOOto
1400 h; AM, L\) or the second 6 h (1400 to 2000 h; Pl\f,D) of a 12 h light period.
The solid symbols in (c) represent values where Al\f temperature was >37.3°C,
while the open symbols represent the values where AM temperature was ~
37.3°C. Fitted line in (a): y=6S.S(±5.35)-1.3(±0.15)x; n=16, ~= 0.S4; P<0.001
and in (b): y=2.29 (±0.024)-0.04S(±0.0055)x; n=16, ~=0.95; P<0.001.
 104

(a)
-....=
....
'
30

-to:
.._,
Z
Q; 20

~ 10

28 34 42 48
Mean day (0800 to 2000 b)
temperature ee)

(b)

-....
.._,
0

~
60

40
........
rI.l
I

~
...
== 20

28 34 42 48
Mean AM (0800 to 1400 b)
temperature ee)

(c)

-=
....
' ....
20 I-
octo
-
.._,
Q;
to:
10 l-
oro 'b
LlO

...
~

0 ~
I
A
•• •
I
I I

0 28 34 42 48
Mean day (0800 to 2000 b)
temperature ee)
 105

a strong and negative linear relation (~=0.84, n=I6; P<O.OOI) between FN and mean

day (0800 to 2000 h) temperature (Two) across all temperature treatments, whereby:

FN= 68.8{±5.35)-1.3{±O.I5)*T wo [Eq.I]

where Two is the mean day temperature, and FN decreased by 1.3 plant" °C-l rise in

.
temperature between 28°C and 48°C (Fig. 8.3a). The greater values of FN for the

AM and PM treatments in Fig. 8.2a for a given maximum day temperature were

therefore clearly due to the cooler mean day temperature, since plants were exposed

to 6 h of 34°, 42° or 48°C combined with 6 h of 28°e. Data presented by Wood

(I968) also shows a similar negative relation between FN and mean diurnal

temperature, and FN decreased by 2.6 plant" 0c-l

over the range of 23° to 34°e. The

night temperatures used by Wood (1968) were 25° and 30oe, compared with 22°e in

the present study, which may have contributed to the greater sensitivity. Similar

observations on FN were also made by Fortanier (1957), who found that a hot day

temperature of 35°C relative to 29°e reduced FN by about 50%.

Although FN was quantitatively related to mean day temperature, variation in

fruit-set was found to be most closely associated with variation in mean AM

temperature (Fig. 8.3b). When AM temperature (TAM)was ~7.3°e, temperature had

no effect on fruit-set, which averaged 50.2 (±O.77). However, when TAMexceeded

the critical value of37.3°e, fruit-set was reduced from the maximum value of 50% to

0% at 48°e. The proportion of flowers producing pegs and pods (fruit-set) is

therefore given by:

Fruit-set = 0.50 [Eq.2]

 106

when TAM>37.3°C,

Fruit-set = 2.29 (±O.024)-0.048(±O.0055)*TAM [Eq.3]

Research in common bean and cowpea has also shown that heat stress

.
reduces fiuit-set and that the response may be due to damage to the pollen mother

cells (e.g. Warrag and Hall, 1984), poor pollen viability (Halterlein et al., 1980; De

Beer, 1963), impaired style and ovule function (Gross and Kigel, 1994) and failure of

fertilization of pollen and ovule (Ormrod et al., 1967). In groundnut, Talwar (1997)

has. shown that exposure to day/night temperature of 35°125°C reduced the rate of

pollen tube growth when compared to 25°120°C, while De Beer (1963) reported that

no viable pollen was produced when plants are grown at a constant day/night

temperature of 33°C. The threshold value of 37.3°C for fiuit-set during the day

identified here, combined with a night temperature of 22°C, is not inconsistent with

these findings. The effects of day and night temperature on pollen production, pollen

viability and fiuit-set are reported and discussed in Chapter 9 (pp lOS to 127).

Across all temperature treatments there was a negative linear relation

(~=0.71, n=16; P<O.OI) between RNt and mean day temperature (Fig. S.3c). The

sensitivity of RNt to temperature over the range of 2So to 48°C was -1.1 plant" 0c-l

and RNt reached zero at 42.3°C. However, as RNt is the product of differential

temperature effects on FN (Fig. 8.2a) and fruit-set (Fig. 8.2b), the response ofRNt to

temperature is better described by combing Equations 1 to 3. Therefore, when TAM~

37.3°C, then RNt is determined by FN, whereby:

RN, = [68.8(±5.35)-1.3(±O.15)*T wo]* 0.50 [Eq.4]

Where Two is the whole day (0800 to 2000 h) temperature.

 107

However, when TAM>37.3°C, then RNt is determined by variation in FN and fruit-set,

whereby:

RNt= [68.8(±5.35)-1.3(±O.15)*TWD]*2.29(±O.024)-0.048(±O.0055)*TMf [Eq.5]

Where TAMis the AM (OSOOto 1400 h) temperature. The cohorts of data in Fig.

S.3b, where RNt is modelled by Equation 4 and 5, are shown by open and closed

symbols, respectively.

In summary, this research has shown that only 6 d exposure to day

temperature ~34°C at 9 DAF significantly reduced flower production, fruit-set and

fruit number. Variation in flower number was quantitatively related to mean day

temperature over the range of 28° to 48°C. In contrast, variation in fruit-set was

related to temperature during the first 6 h of the daylight period above a critical

temperature of37.3°C. The response offruit number to temperature can be modelled

by combining the quantitative responses of flower number and fruit-set to

temperature.
 108

CHAPTER 9 FRUIT NUMBER IN RELATION TO POLLEN

PRODUCTION AND VIABILITY IN GROUNDNUT EXPOSED

TO SHORT EPISODES OF HEAT STRESS

9.1 ABSTRACT

Hot days and warm nights are important environmental factors limiting fruit yields of

groundnuts in the semi-arid tropics. The objective of the present research was to

quantify the effects of short episodes of heat stress on pollen production and viability,

and fruit yield. Plants of cv. ICGV 86015 were grown at a day/night temperature of

28°/22°C from sowing until 9 d after flowering. Then, cohorts of plants were

exposed to a factorial combination of four day (28°, 34°, 42° and 48°C) and two

night (22° and 28°C) temperatures for 6 d. Thereafter, all plants were maintained at

28°/22°C until final harvest 9 d later. Number offlowers plant" (FN), the proportion

of flowers setting pegs (fruit-set), the number of pegs and pods plant" (reproductive

number, RNt), pollen production flower" and pollen viability were determined during

the 6 d stress period. There were strong negative linear relations between day

temperature over the range of 28° to 48°C and FN (slope, -1.1 OC-I), fruit-set

(-2.8% 0c-\ RN t (-0.90 OC-I),and pollen production (_390°C-I) and viability (-1.9%

OC-I). Warmer night temperature (22° cI28°C) had no effect on FN, but reduced

fruit-set (31 to 19%), RNt (8 to 5), and pollen production (4389 to 2800) and

viability (49 to 40%). There were no significant interactions between day and night

temperature. Reduced fruit-set was a consequence offewer pollen grains and reduced

pollen viability. The threshold temperature for pollen production and viability was
 109

34°C and there were strong negative linear relations between both pollen production

and pollen viability and accumulated temperature >34°C.

9.2 INTRODUCTION
.
Groundnut is an important oilseed and forage crop grown in the semi-arid tropical

countries of Africa and Asia. Heat andlor drought induced stresses are the major

environmental factors limiting pod yields in the semi-arid tropics (ICRISAT, 1994).

The optimum day/night temperature for vegetative and reproductive growth and

development in groundnut ranges from 25°125°C (Wood, 1968) to 30o/26°C (Cox,

1979) and from 25°120°C (Wood, 1968) to 26°/22°C (Cox, 1979), respectively.

Short episodes of hot days (>35°C) and warm nights (>25°C) are common in the

semi-arid tropics (e.g. in sub-Saharan Africa: Sivakumar et al., 1993), and where

these episodes coincide with critical stages of plant development they can be

detrimental to seed production.

Studies in controlled environments have shown that both continuous hot days

(35°C), and short episodes of hot days (~38°C for 6 d), reduce the number of fruits

in groundnut (Ketring, 1984a; Wheeler et al., 1997; Vara Prasad et al., 1998).

Groundnut plants are particularly sensitive to hot days from 6 d before until 15 d

after coming into flowering, with maximum effects occurring at 9 d after flowering

(Vara Prasad et al., 1998). However, the mechanisms by which hot temperature

reduce fruit numbers has not been identified. In grain legumes such as cowpea and

common bean reduced fruit numbers at hot temperatures are associated with a

reduction in the number of flowers produced and the proportion of flowers which set

fruits (Hall, 1992; Konsens et al., 1991). In cowpea, reduced fruit-set was associated
 110

with poor pollen viability and reduced anther dehiscence (Mutters et al., 1989 a, b;

Ahmed et al., 1992), particularly when hot temperatures were experienced at

microsporogenesis (Warrag and Hall, 1984).

The objective of the research reported here was to determine and quantify the

effects of short episodes of hot ~ays and warm nights on pollen production and

viability, and fruit number in groundnuts.

9.3 MATERIALS AND METHODS

The experiment was conducted between June and August 1997 in the controlled

environment facilities of the Plant Environment Laboratory, Department of

Agriculture, The University of Reading (51°27'N lat. and 00056'W long.). It was

undertaken in a polyethylene covered tunnel (poly-tunnel) maintained at near

optimum day/night temperature of 28°/22°C and in five modified Saxcil growth

cabinets each maintained at a different temperature regime. The actual temperatures

experienced in the different environments and treatments throughout the experiment

are summarised in Table 9.1.

The diurnal photo- and thermo-period in both the poly-tunnel and cabinets

were coincident and equal at 12 h dol (0800 to 2000 h). The photoperiod was

controlled by a manually operated blackout facility in the poly-tunnel, and by

automatic time switches in the growth cabinets. Air temperatures were measured in

the poly-tunnel and the cabinets using screened and aspirated copper constantan

thermocouples positioned at the top of the plant canopy. Readings were taken at 10 s
 111

Table 9.1 Mean day and night air temperatures (Oe) in the poly-tunnel from

planting to 36 d after sowing (DAS) and from 43 to 53 DAS, and in the eight

6 d temperature treatments imposed in the growth cabinets between 37 to 42

DAS.

Target temperature Actual mean air

Time (DAS) Location (day/night °C) temperature (day/night °C)

o to 36 Poly-tunnel 28/22 28.2/22.3

37 to 42 Cabinets 28/22 (control) 27.8/22.3

34/22 34.4/22.7

42/22 41.8/22.8

48/22 47.8/23.1

28/28 28.2/28.1

34/28' 34.4/28.1

42/28' 41.8/28.1

48/28' 47.8/28.1

43 to 53 Poly-tunnel 28/22 28.4/22.4

, Obtained by transfers between cabinets.

 112

intervals and means were stored for successive 30 min periods using a data logger

(Delta- T Devices Ltd, Cambridge, UK). Carbon dioxide concentration in the cabinets

was maintained at 360 umol marl of air. Relative humidity during the day was

maintained close to 70(±5)% in the poly-tunnel using water sprinklers and ventilation,

while in cabinets vapour pressure deficit was maintained at 1.2 kPa in all temperature

regimes. The poly-tunnel transmitted 75% of incoming photosynthetically active

radiation and photosynthetic photon flux density (PPFD) averaged 590 umol m2s-l

during the experiment. The corresponding value of PPFD in each growth cabinet was

650 umol m-2s-l obtained from a combination of cool white fluorescent and

incandescent lamps (95 and 5% by wattage, respectively).

9.3.1 Temperature treatments

During the period from sowing until 9 d after first flower appearance (OAF), all

plants were grown at a near optimum day/night temperature of28°/22°C in the poly-

tunnel. Thereafter, a factorial combination of four day (28°, 34°, 42° and 48°C) and

two night (22° and 28°C) temperature treatments were imposed for 6 d by

transferring plants to growth cabinets. The growth cabinets were maintained at

day/night temperatures of 28°/22°, 28°/28°, 34°/22°, 42°/22° and 48°/22°C (Table

9.1). The 34°/28°, 42°/28° and 48°/28°C temperature treatments were imposed by

transferring plants at 2000 h from 34°/22°, 42°/22° and 48°/22°C to 28°/28°C, and

back again at 0800 h each day. After the 6 d stress period in the cabinets, all plants

were returned to the poly-tunnel maintained at 28°/22°C, where they remained until

final harvest at 24 DAF. Plants maintained in the 28°/22°C cabinet for the 6 d

treatment period served as controls.

 113

9.3.2 Cultivar and plant husbandry

Uniform seeds of the Spanish botanical type (A. hypogaea subsp. jastigiata) cv.

ICGV 86015 were selected and treated with Apron Combi 453 FS (Ciba,

Agriculture, Cambridge, UK) as a precautionary measure against seed-borne

diseases. Seeds were pre-germinate? at 25°C on moist filter paper in Petri dishes kept

in the dark for 2 d until the radicles emerged. The germinated seeds were then sown

on 3 June 1997, one per 2.5 L pot at a depth of 2.5 cm. The sides of pots were

covered with aluminium foil to reduce radiative heating. The rooting medium

comprised sand, gravel, vermiculite and loamless peat compost mixed in proportions

of 4:2:2: 1, by volume, respectively. A commercial controlled-release fertiliser (0.15

kg kg" N, 0.10 kg kg" P, 0.12 kg kg" K, 0.02 kg kg" MgO plus trace elements;

Osmocote Plus, Scotts UK Ltd, UK) was incorporated into the mixture at the

manufacturer's recommended rate ofS g L-l. Seeds were not inoculated with rhizobia

and plants were dependent on inorganic nitrogen. All pots were soaked with tap

water and allowed to drain for 24 h before sowing; thereafter they were irrigated as

necessary through an automatic drip irrigation system in the poly-tunnel or were

hand-watered during the 6 d period in the cabinets. There were no disease problems

and sporadic pest infestations were controlled by releases of the predators

Phytoseiu/us persimilis Athias-Henriot against red spider mite (Tetranychus urticae

Koch) and Amblyseius cucumeris Oudemans against thrips (Thrips tabaci

Lindeman).

The experiment was sown with six replicates of each temperature treatment

and with 12 replicates for the controls. Onlyuniform plants that flowered on the same

day (28 d after sowing) were selected and transferred to cabinets to remove any

confounding effects of time of flowering, which gave four replicates of each

 114

temperature treatment and eight replicates of the controls. A subset of four plants

from the control treatment were harvested at 18 DAF to estimate the numbers of

pegs and pods produced from those flowers which had opened before the imposition

of the target heat stress treatments.

9.3.3 Observations

Durations (d) from sowing to the appearance of the first open flower and the first peg

~ 3 mm long were recorded on all plants. Thereafter, the number of flowers opening

each day were counted until final harvest. Plants were harvested on two occasions:

first at 9 d after the start of the temperature treatments (18 DAF) and then at final

harvest taken 9 d after the end of the temperature treatments (24 DAF). At both

harvests the numbers of pegs and pods per plant were counted and the dry weights of

roots, stems (including petioles), leaves, pegs, and pods were determined after oven-

drying at 80°C for 3d.

Individual flowers were collected between 0800 and 0815 h, each day during

the 6 d stress period in all the temperature regimes to count the number of pollen

grains and to test pollen viability. The number of pollen grains flower" was counted

using a heamocytometer (Fisher Scientific, Pittsburgh, USA) as described by Kearns

and Inouye (1993) and Freshney (1994). The viability of the pollen was determined

by the reaction of grains to triphenyl tetrazolium chloride as described by Keams and

Inouye (1993). Anthers were collected before dehiscence and were split open on a

glass slide and stained. Pollen grains that stained red were classified as viable,

whereas those that remained transparent were classified as dead. The numbers of

viable and non-viable grains were counted.

 115

9.3.4 Data analysis

The fate of individual flowers (Le. whether they produced a peg or a pod) was not

monitored and so the total number of pegs and pods at final harvest, hereafter

referred as the final reproductive number (RNr), were produced from flowers opened

at 28°C (i.e. in the poly-tunnel) and in the 6 d temperature treatments in the cabinets.

In order to determine the fate of those flowers which opened during the 6 d

temperature treatment, it was assumed that for cv. ICGV 86015 the time from flower

opening to peg appearance was 9 d (authors, unpublished data). Therefore, all

flowers that opened and were fertilized between the onset of flowering and 9 nAP

should have produced a peg or a pod by the time of first harvest was taken at 18

nAP. The number of pegs and pods at this time was referred as the initial

reproductive number (RNj). Similarly, any flower that opened between the end of the

temperature treatments (15 nAF) and the final harvest 9 d later (24 nAP), should

have not formed a peg. Therefore, the numbers of pegs and pods arising from those

flowers that opened during the 6 d temperature treatment (the treatment reproductive

number, RNt) were estimated as the difference between RNrand RNj• Fruit-set during

the temperature treatments was calculated as the ratio of RNt to the cumulative

number of flowers produced during the 6 d temperature treatment (FN). The values

of pollen viability and fruit-set were subject to angular transformation before analysis

to ensure homogeneity of variances.

The effects of day and night temperature on FN, fruit-set, RNt, and pollen

production and viability were examined by comparing linear regressions (Mead et al.,

1993) using routines in Genstat 5 (Genstat 5 Committee, 1987).

 116

9.4 RESULTS

There was no significant difference among plants in the number of flowers produced

(19±2.S) before the start of the heat stress treatments at 9 DAF and the number of

pegs and pods produced from those flowers was II (SD±I.8).

9.4.1 Number of flowers and fruit-set

There was a significant (P<O.OI) effect of day temperature but no significant

(P>O.IO) effect of night temperature (Table 9.2) or their interaction on FN, and

o accordingly a common line could be fitted to describe the response of FN to

temperature (Fig. 9.la). The FN was reduced by 1.1 plant" eCrl between a day

temperature of 28° and 48°C and the ceiling temperature (where y=O)was 54.SoC.

In contrast to FN, there were significant effects of day (P<O.OI) and night

temperature (P<O.OS)on fruit-set, but again no interaction (P>O.IO). Accordingly the

response of fruit-set to temperature was described by two parallel lines (Fig. 9.lb).

Thus, at both 22° and 28°C night temperature fiuit-set was reduced by 2.8% eCrl

between day temperature of 28° and 48°C, and the ceiling temperature was 49° and

44°C at 22° and 28°C night temperature, respectively. Overall, warm nights reduced

fruit-set from 31 to 19.5% (Table 9.2).

9.4.2 Number of pegs and pods

Number of pegs and pods plant" (RNt), as for fruit-set, was also affected by day

(P<O.OOI)and night (P<O.OI)temperature, but not by their interaction (P>O.IO), and

 117

Fig. 9.1 Relation between mean day temperature eC) and (a) number of
flowers opened during the 6 d stress period (FN); (b) proportion of those
flowers which set pegs or pods (fruit-set, angular transformed); and (c) number
of pegs and pods (RNt) of groundnut cv. ICGV 86015 grown at night
temperatures of 22° (.) or 28°C (0). Fitted lines (a) y=58.29 (±5.16)-1.07
(±O.133x,~=0.93, P<O.Ol; (b) at 22°C: e, y=137.8(±14.2)-2.81(±O.36)x, .-2=0.93,
P<O.Ol and at 28°C: 0, y=123.5(±14.3) -2.81(±O.36)x, .-2=0.92,P<O.OI; and
(c) at 22°C: ., y=40.21(±1.81)-0.856(±0.046)x, ~=0.98, P<O.Ol and at 28°C: 0,
y=36.42 (±0.66)-0.856(±0.046)x, ~=0.98, P<O.Ol. Vertical bars denote standard
errors and are shown where they exceed the size of the symbol.
 118

40
.-. (a)
~
I...... 30
- =~c.- 20
._,
Z 10
~
0

.-. 75 (b)
._, 60
°.......
CIJ
VJ
.......I
45
•• 30
~
I.= 15
0

.-.20
~
I.......
(c)
= 15
--
~
c.- 10
._,
5
~
0

28 35 42 49

Mean day temperature (OC)

 119

so the response of RNt to temperature was also described by two parallel lines (Fig.

9.1c). Thus, as day temperature increased from 28° to 48°C, RNt was reduced by 0.9

plant" coCrl at both 22° and 28°C night temperature. Overall, warm nights reduced

RNt from 7.7 to 5.0 plant" (Table 9.2). The ceiling temperature was 47° and 42.5°C

at 22° and 28°C night temperature, respectively.

Table 9.2 Effect of night temperature on the cumulative number of flowers

opened during the 6 d stress period (FN), the proportion of those flowers which

set pegs or pods (fruit-set, angular transformed), and the number of pegs and

pods (RN,), and pollen production flower" and proportion of viable pollen

(angular transformed) on day 6 of the stress period in groundnut. Data are the

means of four day temperature treatments and four replicates. Standard errors

are given in parentheses.

Night temperature caC)

Trait 22 28

FN (planrl) 17.8(±2.21) 17.7(±2.55)

Fruit-set CO) 31.0(±3.39) 19.5(±3.00)

RN, (plant") 7.7(±O.80) 5.0(±1.02)

Pollen production flower" 4389(±605) 2800(±267)

Pollen viability CO) 45.5(±3.03) 40.1(±2.02)

 120

9.4.3 Pollen production and viability

There were strong positive linear relations between fruit-set and both pollen

production (r2=0.95; Fig. 9.2a) and pollen viability (r=o.94; Fig. 9.2b) indicating that

day and night temperature had similar effects on pollen production and pollen

viability (both measured on day 6 ~f the stress period) as on fruit-set and RNt (Fig.

9.2; Table 9.2). Pollen production and viability were reduced by 390 flower" (OCr1

and 1.9% (OCr1 at both 22° and 28°C night temperature, respectively, as day

temperature increased from 28° to 48°C. Warmer nights reduced mean pollen

numbers from 4389 to 2800 flower" and mean pollen viabilityfrom 49 to 40% (Table

9.2). The ceiling day temperature for pollen production and viability was 49° and

67°C, respectively.

The viability of pollen grains is determined during the early stages of floral

bud development (De Beer, 1963) and the effects of temperature on pollen viability

vary with stage of pollen development (Sugiyama et al., 1966). In the present work,

the effects of day and night temperature on pollen production and viability varied

over time, but without interaction, and therefore the mean effects of night

temperature are presented in Fig. 9.3 to illustrate these trends. The respective values

for pollen production and viabilityat 28° and 34°C day temperatures were similar and

constant during the 6 d stress period. At 42°C, pollen production and viability were

significantly reduced after 3 and 4 d of exposure, respectively, and after 6 d only

1524 pollen grains flower" were produced compared with 7346 at 28°C (Fig. 9.3a).

However, the viability of the pollen grains produced at 42°C remained moderate at

43% (Fig. 9.3b). At 48°C, pollen production and viability were lower after 3 d of

exposure, and no flowers were produced after 4 d. The effects of 22° and 28°C night
 121

60
(a)
45
~
°.....
'-'
Q,) 30
._
~
v..l

~ -
= 15

o 2000 4000 6000 8000

1
Pollen production flo,ver-

60
(b)
45
~
°.....
'-'
Q,) 30
......_
v..l
I

~ -
= 15

35 45 55 65 75
Pollen viability (0)

Fig. 9.2 Relation between fruit-set (angular transformed) and (a) pollen production
1
flower- and (b) pollen viability (angular transformed) at a night temperature of
2tC (.) or 28°C (0). Fitted lines: pollen production y= -25.1(±5.2) + O.OlO(±O.OOl)x,
2
r =0.95, n=8, P<O.Ol; pollen viability y= -69.1(±9.73)+1.73(±O.17)x, r =O.94, n=8,
P<O.Ol.
 122

,.-4
I

"- (a)
~ 9000
Q
..:
=......
.- Q
6000
CJ

"C =
Q
"-
c.. 3000 ~
=
-
~
CIJ

Q
~

90
(b)
.-..
0
'-"
P
-
.-
.-
.c
60
~
~ KO!
.-> Cl:
~
~
~A
=
~
- CIJ

Q
30

0
1 2 3 4 5 6

Duration of temperature treatment (d)

Fig. 9.3 Effect of day temperature ofl8° (.),34° (0),42° (A), 48° (~) on (a) pollen
production flower-I; and (b) pollen viability (angular transformed) over time during
the 6 d stress period. The value of pollen viability of 4 d after the start of the temperature
treatment at 48°C day temperature is from one replicate only. Vertical bars denote the
standard errors and are shown where they exceed the size of the symbol.
 123

temperatures on pollen production and viability followed a similar trend, with values

at 28°C being consistently smaller than those at 22°C (Table 9.2). Clearly, pollen

production is relatively more sensitive to hot days and warm nights than pollen

viability.

9.5 DISCUSSION

Warmer days and nights, but not their interaction, significantly reduced FN, fruit-set,

RNt and pollen production and viability. The fact that these responses were all well

described by simple linear regressions over the range of 28° and 48°C should be

useful in the modelling of temperature responses in groundnut. The effects of

temperature on RNt reported here agree with the findings of other studies in

controlled environments (Wood, 1968; Cox, 1979; Ketring, 1984a). For example,

Ketring (1984a) reported that a day/night temperature regime of 35°/22°C relative to

300/22°C reduced fruit numbers by 33%, similar to the reduction of35% at 34°/22°C

relative to 28°/22°C in the present study. The reduction in RNt at hot days and warm

nights in the present study was mainly due to the production of fewer flowers and a

reduction in the proportion of those flowers that set fruits. This finding confirms that

a day temperature of 35°C is supra optimal for fruit production in groundnuts, even

under well watered conditions.

De Beer (1963) reported that the number of pollen grains of groundnut cv.

Schwarz 21 was reduced by 71%, from 3388 to 987 grains flower" when constant

day and night temperature was increased from 24°C to 33°C, and no viable pollen

was produced at 33°C. This is consistent with a 69% reduction, from 8156 to 2500

grains flower", when mean temperature was increased from 25°C (28°/22°C) to
 124

32°C (42°/22°C) and a 93% reduction, when mean temperature was increased to

35°C (42°/28°C). The variation in pollen production flower" at optimal environments

may have been due to cultivar differences (palmer et al., 1978), diurnal variation in

the temperatures or the length of the stamens (Trivedi and Verma, 1975). Other

studies on cowpea (Ahmed et al., !992) and common bean (Gross and Kigel, 1994)

have also shown that heat stress reduces fruit-set. In cowpea, poor fruit-set in hot

nights (30°C) was associated with impaired anther dehiscence and reduced pollen

viability (Warrag and Hall, 1984). This is also so in groundnuts, as shown by the

strong and significant positive relations between fruit-set and each of pollen

production and pollen viability(Fig. 9.2).

Hot days >34°C reduced pollen production and viability only after 3 d of

exposure (Fig. 9.3). This suggests that groundnut flowers are either sensitive to hot

days and warm nights at a specific stage 3 to 4 d before flower opening, or that floral

buds need to be exposed to day temperatures >34°C for more than 2 d to elicit

effects on pollen production and pollen viability. To test these hypotheses, the pollen

number data presented in Fig. 9.3a were re-plotted against accumulated day

temperature >34°C (Fig. 9.4). There was a strong and significant (r=0.88, n=18;

P<O.OOl)negative relation between pollen production and day temperature, such that

pollen number was reduced by 164 grains flower" eCr1 of accumulated temperature

>34°C. A comparison of regressions at 22° or 28°C night temperature revealed no

significant difference in the values of slope (p>0.10) or intercept (p>0.10) and so a

cornmon line described the data. No pollen grains were produced when flower buds

accumulated ~55°Cd above 34°C. Pollen viability was also linearly related to

cumulative temperature >34°C [y=72.4(±3.09)-O.S4(±O.lO)x;~=O.6S; P<O.Ol; n=16]

 125

....
I

-
~
~
0
c 6000
8000

.-= 4000
0
~
~
=
-=
"C
0
c. 2000
-
~
0
~

0 20 40 60

Cumulative temperature >34°C

1
Fig. 9.4 Relation between pollen production flower- and cumulative day temperature
(>34°C) after the start of the 6 d temperature treatment at a night temperature of
22° (.) and 28°C (0). Fitted line: y= 9055(±490) - 164(±14.2) x, l=O.88, n=18, P<O.OOI.
 126

although as noted previously the effects of hot temperature on pollen viability were

not as marked as those on pollen number. We conclude that pollen production and

viability are determined by the cumulative effects of temperature above a critical

value, not by temperature effects at a specific stage of development before flower

opening.

The precise mechanisms that reduce pollen production and viability in

groundnut are not known. In cowpea, reductions in pollen viability and poor anther

dehiscence reflect the premature degeneration of the tapetal layer (Ahmed et al.,

1992), which plays an important role in micro sporogenesis (Echlin, 1971). Recent

studies on groundnut have shown that flower buds are particularly sensitive to heat

stress at a stage 3 to 5 d before anthesis (Talwar, 1997), which is known to coincide

with micro sporogenesis (Xi, 1991). Studies on common bean (Gross and Kigel,

1994) have also shown that microsporogenesis is sensitive to heat stress.

In the present study plants were not given opportunity to acclimate to hot

temperatures and the transition from optimum to hot temperature was rapid. The

seasonal effects may be gradual and some plants may have the ability to acclimate to

hot temperature. Further research in groundnut is clearly called for to understand the

effects of acclimation and mechanisms responsible for reduced pollen production and

viability.

In summary, this research has shown that short episodes (6 d) of hot days

(>34°C) and/or warm nights (28°C) reduced flower production and fruit-set, and

hence the number of fruits in groundnut. Reduced fruit-set was a consequence of

fewer pollen grains and poor pollen viability. The threshold or critical day

temperature for pollen production and viability was 34°C and there were strong and
 127

linear negative quantitative relations between both pollen production and pollen

viability and accumulated temperature >34°C.

 128

CHAPTER 10 EFFECT OF HOT AIR AND SOIL

TEMPERATURE ON POD YIELD OF GROUNDNUTS

DEPENDENT ON SYMBIOTIC OR INORGANIC NITROGEN

10.1 ABSTRACT

Crops of groundnut are frequently exposed to hot air and/or hot soil temperatures in

the semi-arid tropics. The objectives of the present research were: (i) to determine

the response of groundnuts to different nitrogen sources; (ii) to quantify the effects of

hot air and soil temperatures and their interaction on nodulation, dry matter

production, partitioning and pod yields; and (iii) to discover whether plants

dependent on symbiotic dinitrogen fixation are more sensitive to heat stress than

those dependent on inorganic N. All plants were grown at optimum air and ambient

soil temperatures from sowing until the first flowering. Thereafter, plants were

exposed to a factorial combination of two air temperatures [optimum: 28°/22°C

(day/night) and hot: 38°/22°C], two soil temperatures (ambient: 26°/24°C and hot:

37°/30°C) and three N-sources [inoculated with Bradyrhizobium strain NC 92

(symbiotic N2); inoculated and supplied with 20 ppm inorganic N (symbiotic N2 plus

20 N); or not inoculated and supplied with 100 ppm inorganic N (inorganic N)]. At

optimum air and ambient soil temperature dry matter and pod yields were greatest in

plants dependent on inorganic N, intermediate in symbiotic N2 plus 20 N and least in

symbiotic N2• Hot air and/or hot soil temperature significantly (P<O.OOI) reduced pod

yield to a similar extent and their effects were additive and without interaction. Hot

soil, but not hot air temperature, significantly (P<0.001) reduced nodule numbers and
 129

nodule dry weight per plant and 100 seed weight. Hot air and/or hot soil temperature

had no effect on pod yield in plants dependent on symbiotic N2 or symbiotic N2 plus

20 N, but did significantly (P<0.05) reduce pod yield in plants dependent on

inorganic N. These results suggest that effectively nodulated plants are potentially

more adaptable to hot temperatures. than those relying on large quantities of inorganic

N.

10.2 INTRODUCTION

Groundnut is an important oilseed and cash crop grown throughout the semi-arid

tropics where heat and water stress are major environmental factors limiting pod yield

(ICRISAT, 1994). In the semi-arid tropics groundnut crops are often cultivated on

marginal lands and seldom receive either irrigation or inorganic fertiliser (Gibbons,

1986). Groundnuts are susceptible to both hot air (Ketring, 1984a; Wheeler et al.,

1997) and hot soil temperature (Ono, 1979; Dreyer et al., 1981; Golombek and

Johansen, 1997), reflecting the aerial flowering and subterranean fruiting habit of the

crop. Most research on heat stress in groundnut has been done either on hot air or on

hot soil temperature effects; studies on the combination of both factors have received

little if any attention.

The optimum air temperature during the day in groundnut ranges from 25° to

30°C for vegetative growth, and from 24° to 26°C for reproductive growth and

development (Wood, 1968; Cox, 1979) Similarly, the optimum soil temperature

range for reproductive growth and development is from 31 ° to 33°C (Ono, 1979).

Soil temperatures >33°C significantly reduces the number of pods and pod dry

weight (Ono et al., 1974; Williams and Boote, 1995) and soil temperatures >25°C
 130

can also significantly reduce. nodulation and dinitrogen fixation (Nambiar and Dart,

1983).

Groundnut can assimilate nitrogen either as inorganic N or through symbiotic

Nz fixation. Attempts to supplement Nz fixation using inorganic fertiliser have not

been successful (Reddy et al., Ip81; Selamat and Gardner, 1985) because the

addition of fertiliser-N tends to substitute for, rather than supplement, Nz fixation.

Nevertheless, it is generally accepted that symbiotically fixed Nz is inadequate for

realising maximum seed yields, and that an application of small amounts of "starter"

fertiliser N is needed to establish seedlings and promote early dinitrogen fixation

(Selamat and Gardner, 1985). However, applications of large quantities of inorganic

N inhibit the growth of rhizobia, nodulation and dinitrogen fixation (Lie, 1974).

Studies on other grain legumes have shown that in plants dependent on

symbiotic Nz, application of small quantities of inorganic N can stimulate nodulation,

increase Nz fixation (Eaglesham et al., 1983) and lead to greater seed yields even at

hot air temperatures (Summerfield et al., 1978; Rawsthome et al., 1985a). It is

clearly important to quantify the response of groundnuts to hot air and/or hot soil

temperatures when supplied with either inorganic N, symbiotic Nz or symbiotic Nz

plus small quantities of inorganic N.

The research described here had the following objectives: (i) to determine the

response of Spanish cv. ICGV 86015 to different N-sources; (ii) to quantify the

effects of hot air andlor hot soil temperatures and their interaction on nodulation,

total dry matter and pod yields; and (iii) to discover whether plants dependent on

symbiotic Nz are more sensitive to hot temperatures than those dependent on

inorganic N.
 131

10.3 MATERIALS AND METHODS

The experiment was conducted during the summer months of 1997 in the controlled

environment facilities at the Plant Environment Laboratory, Department of

Agriculture, The University of Reading (51°27'N lat. and 00056'W long.). Seeds of

the Spanish botanical type (A. hypogaea subsp. jastigiata) cv. ICGV 86015 were

obtained from ICRISAT Asia Centre located at Hyderabad in India.

The experiment was carried-out in two adjacent polyethylene-covered tunnels

(poly-tunnel) aligned East-West, one maintained at a near optimum day/night air

temperature of 28°/22°C and the other at hot day/optimum night air temperature of

38°/22°C. Air temperatures were measured in each poly-tunnel with screened and

aspirated copper constantan thermocouples positioned at the top of the plant canopy.

Readings were taken at 10 s intervals and means were stored for successive 30 min

periods using a data logger (Delta- T Devices Ltd, Cambridge, UK)

The photo- and thermo-period in both the poly-tunnels was coincident and

equal at 12 h doland the photoperiod was controlled by a manually operated blackout

facility. Carbon dioxide concentration was not controlled and remained at ambient

concentration. Relative humidity during the day in both poly-tunnels was controlled

using automatic sprinklers and ventilation to maintain vapour pressure deficit close to

2 kPa. The poly-tunnels transmitted about 75% of the incoming photosynthetically

active radiation such that photosynthetic photon flux density averaged 594 umol m-2

S-l during the experiment.

Within each poly-tunnel soil temperature treatments were imposed by placing

pots on specially constructed benches fitted (hot) with or without (ambient) tubular

heaters. The target hot daytime soil temperatures were set at 10°C above the
 132

respective ambient soil temperature in both poly-tunnels and were controlled by

automatically switching the heaters on and offusing a data logger (eRIO, Campbell

Scientific Ltd, Shepshed, UK). Soil temperature during the night was not controlled

and pots were allowed to return to ambient temperature. Soil temperatures during the

day and night in both ambient and hot soil treatments were recorded by copper

constantan thermocouples placed at a depth of 5 cm in the soil. Readings were

measured at IDs intervals and means were stored for successive 10 min periods using

a data logger.

10.3.1 Temperature and nitrogen treatments

All plants were grown in the poly-tunnels from sowing until appearance of the first

flower at 28 d after sowing (DAS) at a day/night temperature of 28°/22°C. From 29

DAS until final harvest they were exposed to a factorial combination of two air

temperatures, two soil temperatures and three N-sources. At 29 DAS, air

temperature in one poly-tunnel was increased to 38° 122°C, whereas, the other one

remained at 28°/22°e. One-half of the plants in each air temperature regime were

also exposed to hot soil temperature (looe above ambient soil temperature). Within

each air and soil temperature regime, plants were supplied with one of the three

sources of N: they were either inoculated with Bradyrhizobium strain NC 92

(Biocare Technology Pty, Somersby, NSW, Australia) and grown without inorganic

N (symbiotic N2); or inoculated and supplied with 20 ppm inorganic N (symbiotic N2

plus 20 N); or not inoculated and supplied with 100 ppm inorganic N (inorganic N).
 133

10.3.2 Cultivar and plant husbandry

Uniform seeds of cv. ICGV 86015 were selected and treated with Apron Combi 453

FS (Ciba Agriculture, Cambridge, UK) as a precautionary measure against seed-

borne diseases. The seeds were pre-germinated at 25°C on moist filter paper in Petri

dishes kept in the dark for 2 d until.radicle emerged. The germinated seeds were then

sown singlyper 15 L pot at a depth of2.5 cm. The pots were wrapped on sides with

aluminium foil to reduce radiative heating. The rooting medium comprised sand,

gravel, vermiculite and loamless peat compost mixed in proportions of 4:2:2.1 by

volume, respectively. Two-thirds of the seeds were inoculated with Bradyrhizobium

and sown as per the N-source treatment. All pots were soaked with tap water and

allowed to drain for 24 h before sowing. Thereafter, all plants were hand-watered

two to four times each day until final harvest with equal quantities of nutrient

solution containing either 0,20 or 100 ppm inorganic N as determined by treatment.

There were no disease problems and sporadic pest infestations were

controlled by releases of predators Phytoseiulus persimilis Athias-Henriot against red

spider mite (Tetranychus urticae, Koch) and Amblyseius cucumeris Oudemans

against thrips (Thrips tabaci, Lindeman).

10.3.3 Observations and data analysis

Duration (d) from sowing to the appearance of the first fully opened flower (corolla

colour visible)was recorded on all plants. Thereafter, the number of flowers opening

each day was recorded on each plant until final harvest taken at reproductive maturity

(90 DAS). At harvest, plants were carefully removed from each pot without

damaging the root systems and were separated into roots, leaves, stems (including
 134

petioles), pegs and pods. The number of pegs, pods, and seeds per plant were

counted. The roots were washed to remove the potting medium and nodules were

separated and counted. The respective dry weights of nodules, roots, leaves, stems,

pegs, pods, and seeds were recorded after oven-drying each component at 60°C for

7 d. Sub-samples of leaves were collected and leaf area was determined using a leaf

area meter (Delta-T Devices Ltd, Cambridge, UK). The dried leaf samples were

finely ground and analysed for total N concentration (Dumas, 1831) using a leco

nitrogen determinator (Leco Corporation, Michigan, USA).

Total dry matter yield per plant, pod harvest index (i.e. the ratio of pod dry

weight to total dry matter yield) and root-to-shoot ratio (i.e. the ratio of root dry

weight to total above-ground dry weight) were calculated from the weights of

individual components. Specific leaf weight (g mo2) was estimated as the ratio of leaf

dry weight per unit leaf area. Specific leaf nitrogen (SLN) was calculated by

multiplying values of specific leaf weight and leaf N concentration. Values of pod

yields were adjusted by multiplying with a factor of 1.65 to allow for the oil content

of seeds (Duncan et al., 1978). All data are expressed on a per plant basis unless

otherwise stated.

The experiment was laid out as a split-split plot design and replicated five

times, with air temperature as main plots, soil temperature as subplots and N-source

as sub-subplots. The analysis of variance for all the variables was performed using

Genstat 5 (Genstat 5 Committee, 1987).

 135

10.4 RESULTS

Target temperatures were maintained throughout the experiment and were within

close tolerances (SD<1.4°C). The mean (day/night) air temperatures in the optimum

and hot poly-tunnels were 25.1° (28.2°/22.0°C) and 30.4°C (38.4°/22.4°C),

respectively. Mean (day/night) ambient and hot soil temperatures were close to 25.3°

(26.2°/24.4°C) and 33.7°C (37.5°/29.9°C), respectively.

There were significant (P<0.05 to 0.001) effects of different N-sources and

hot air and/or hot soil temperature on all the traits given in Table 10.1 unless

indicated otherwise. The interaction between N-sources and temperatures were

significant (P<0.05) only for total dry matter production, pod yield and SLN.

10.4.1 Effects of nitrogen sources

There were significant (P<0.001) effects of the different N-sources on total dry

matter production, pod yield, partitioning of dry matter to pods and roots, SLN,

number of nodules and nodule dry weight per plant (Table 10.1). Total dry matter

production and pod yields were significantly (P<0.001) greater in plants dependent

on inorganic N than in those dependent on symbiotic N2 plus 20 N or on symbiotic

N2 alone. There was no significant difference in pod harvest index between plants

relying on symbiotic N2 plus 20 N or on inorganic N alone, but both were

significantly (P<O.OOI)greater than those dependent on symbiotic N2• In contrast,

plants relying on symbiotic N2 had significantly (P<O.OOI)greater root-to-shoot ratio

than those dependent on inorganic N or symbiotic N2 plus 20 N.

 136

Table 10.1 Total per plant values of dry matter production, pod yield, pod

harvest index, root-to-shoot .

ratio, specific leaf nitrogen (SLN), nodule number

and nodule dry weight at reproductive maturity (90 DAS) in different N-source

treatments. Data are the means of air and soil temperatures and replicates.

Nitrogen source

Symbiotic N2 SEDt

Trait Symbiotic N2 plus 20 N Inorganic N (df=2,32)

Total dry matter (g) 51 115 138 4.7***

Pod yield (g) 19 64 77 3.4***

Pod harvest index (%) 37 56 55 1.3***

Root-to-shoot ratio (%) 30 15 16 1.0***

SLN(gNm"2) 0.95 1.34 1.33 0.05***

Nodule number 263 441 0 23***

Nodule dry weight (mg) 226 374 0 19***

***, Significant at the P<O.OOI probability level.

tSED, Standard error of difference of means.

 137

Plants dependent on symbiotic N2 plus 20 N or on inorganic N had similar

values for SLN and were both significantly (P<0.001) greater than those dependent

on symbiotic N2. The number of nodules and nodule dry weight per plant were

significantly (P<0.001) greater in plants dependent on symbiotic N2 plus 20 N

compared to those dependent on symbiotic N2 alone (Table 10.1).

10.4.2 Effects of air and soil temperature

Hot air temperature (P<0.05) and hot soil temperature (P<O.OOI) significantly

reduced total dry matter production and pod yield (Table 10.2). There was no

interaction (P>O.50) between air and soil temperature and the effects were additive,

i.e. pod yield was reduced by about 18, 24 and 37%, by hot air, hot soil, and hot air

and hot soil combined, respectively.

Hot air temperature significantly (P<O.OI) reduced partitioning of dry matter

to pods but had no effect on root-to-shoot ratio (Table 10.2). In contrast, hot soil

temperature had no effect on partitioning to pods, but significantly (P<0.01)

increased root-to-shoot ratio. There were no significant effects of hot air (P>0.18) or

hot soil (P>0.14) temperature on SLN. The values of 100 seed weight were

significantly reduced by hot soil temperature (P<0.05), but not by hot air temperature

(Table 10.2). Similarly, the number of nodules and nodule dry weight per plant were

significantly (P<0.001) reduced by hot soil temperature, but not by hot air

temperature (Table 10.3).

 138

Table 10.2 Total per plant values of dry matter production, pod yield, pod

harvest index, root-to-shoot ratio, specific leaf nitrogen (SLN) and 100 seed

weight at reproductive maturity (90 DAS) in different mean (day/night) air and

soil temperature treatments. Data are the means of N-sources and replicates

within each temperature regime.

Mean air temperature eC) Mean soil temperature eC)

Trait 25 30 SEDt 25 34 SEDt

(df=I,4) (df=I,8)

Total dry matter (g) 107 97 2.8* 113 90 3.7"*

Pod yield (g) 59 48 2.5* 61 46 2.6·"

Pod harvest index (%) 55 49 0.8** 54 42 1.2

Root-to-shoot ratio (%) 21 20 0.8 19 22 0.8**

SLN(gNm·2) 1.23 1.18 0.04 1.24 1.17 0.04

100 seed weight (g) 49 47 2.1 52 43 3.0*

*, **, *•• , Significant at the P<0.05, P<O.OI and P<O.OOI probability levels,

respectively.

tSED, Standard error of difference of means.

 139

Table 10.3 Number of nodules and nodule dry weight per plant at reproductive

maturity (90 DAS) in different mean (day/night) air and soil temperature

treatments. Data are the mea~s of N-sources and replicates within each

temperature regime.

Treatments Nodule number Nodule dry weight

Plant" (mg plant")

Mean air temperature ee)

25 380 323

30 323 277

SEDt (df=l, 4) 29.5 36.4

Mean soil temperature ee)

25 407 366

34 297 234

SEDt(df=1,8) 20.2*** 20.8***

***, Significant at the P<O.OOI probability level.

tSED, Standard error of difference of means.

 140

10.4.3 Interaction between temperature regimes and nitrogen sources

The response of plants dependent on the different nitrogen sources to hot air and hot

soil temperature was significantly different for pod yield and SLN (Fig. 10.1) and

total dry matter production. There were no significant interaction effects on number
.
of nodules, nodule dry weight, partitioning of dry matter to pods and roots or on 100

seed weight.

At mean optimum air (25°C) and ambient soil temperature (25°C), plants

dependent on inorganic N had greatest pod yields (109 g plant") whereas plants

dependent on symbiotic N2 plus 20 N and symbiotic N2 had pod yields of 68 and 23 g

plant", respectively (Fig. 10.1). In plants dependent on symbiotic N2 or symbiotic N2

plus 20 N, hot air and/or hot soil temperatures had no significant effect on pod yield.

In contrast, in plants dependent on inorganic N, hot air (P<0.05), hot soil (P<O.OOl)

and hot air and hot soil combined reduced pod yields by 26, 38 and 52%,

respectively. As a consequence, at hot air and hot soil temperatures pod yields in

plants dependent on symbiotic N2 plus 20 N and those dependent on inorganic N

were similar, 59 and 52 g plant", respectively.

The effects of hot air and hot soil temperature on total dry matter production

and SLN in plants dependent on different N-sources were similar to those on pod

yields, and there was a strong and significant positive linear (r=0.90, n=12, P<O.OOl)

relation between pod yield and SLN (Fig. 10.1). There were no effects of hot

temperatures on SLN in plants dependent on symbiotic N2 or symbiotic N2 plus 20 N.

However, in plants dependent on inorganic N hot air, hot soil, and hot air and hot soil

combined treatments reduced SLN significantly, from 1.6 g N m2 to 1.4, 1.3 and

1.1 g N m2, respectively.

 141

120
~ ~
~
I~ ~ 125.0(±13.29) - 97.3(±16.28)x
= 2

-
r =0.90
eo: 90
~ £
eJ)
'-"

.--
'"0
eo)
60
~
'"0
0
~ 30

o
0.80 1.00 1.20 1.40 1.60
2
Specific leaf nitrogen (g N m- )

l
Fig.IO.1 Relation between pod yield (g planf ) and specific leaf nitrogen (SLN, g N m-2)
when plants dependent on symbiotic N2 fixation (squares), symbiotic N2 plus 20 N (circles)
and inorganic N (triangles) were grown at optimum air and ambient soil ( 0 ,0 , D. ),
hot air ( IJ ,() ,J:::.. ), hot soil ([J ,et , A. ) or hot air and hot soil ( • , • , A )
temperatures from first flower appearance (28 DAS) to reproductive maturity (90 DAS).
 142

10.5 DISCUSSION

At close-to-optimum temperatures, plants supplied with 100 ppm inorganic N were

superior in terms of total dry matter production and pod yields compared to those

dependent on symbiotic N2 plus 20 N. This is because biological N2 fixation is

energetically very expensive due to greater respiratory burden of nodulated roots

(Minchin et al., 1981). In addition, the easy and ready availability of nitrate in plants

dependent on inorganic N, which is reflected in their greater SLN (Fig. 10.1) may

have contributed to greater yields at optimum temperature. Nitrate dependent plants

also had greater vegetative growth, more branches, more reproductive nodes and

hence a higher number of pods and greater yields, as observed for cowpea (Minchin

et al., 1980) and chickpea (Cicer arietinum L.; Rawsthorne et al., 1985a). However,

in contrast to cowpea, increasing the inorganic N supply from 20 to 100 ppm had no

effect on pod harvest index or root-to-shoot ratio.

Effectively nodulated plants receiving no inorganic N had smaller pod yields

and total dry matter production, compared with those receiving 20 ppm inorganic N

throughout the growth period. These differences are mainly attributed to poor

establishmentand growth of plants and large demands for nitrogen especially during

the reproductive period (Minchin et al., 1980; 1981). This suggests that small

quantities of inorganic N are necessary for better early vegetative growth and seed

yields in groundnut. Studies in cowpeas (Summerfield et al., 1978) and soyabean

(Eaglesham et al., 1983) have indicated that nodule-dependent plants receiving

20 ppm inorganic N, a level at which a synergistic effect between applied Nand

symbiosisoccurs, results in improved seed yields.

 143

It is clear that hot air (day/night, 38°/22°C) andlor hot soil (38°/30°C)

temperature imposed from flowering until reproductive maturity significantly reduces

total dry matter production and pod yield of groundnuts as reported elsewhere

(Ketring, 1984a; Ono et al., 1974; Ono, 1979). Air temperatures above 35°C during

the day significantly reduce photosynthesis (Ketring et al., 1982a), vegetative and

reproductive growth (Cox, 1979) and pod yields (Ong, 1984). Studies have also

shown that an increase in podding zone temperature to 10°C above an ambient soil

temperature of 25° or 27°C from pegging or podding until maturity significantly

reduce number of mature pods per plant, mature single seed weight and therefore

seed yield (Dreyer et al., 1981; Golombek and Johansen, 1997).

The effects of air and soil temperature detected in the present research were

additive and without interaction, suggesting that effects and response of plants to hot

air and hot soils are different. Hot air temperature significantly reduced the

partitioning of dry matter to pods, as others have reported (Ong, 1984; Nigam et al.,

1994). In contrast, hot soil temperature significantly reduced 100 seed weight and

increased the root-to-shoot ratio. Golombek and Johansen (1997) also reported that

hot day/night soil temperature of 38°/32°C significantly reduced 100 seed weight.

Furthermore, it was observed in the current study that although both hot air and hot

soil temperature significantly reduced the number of pegs and pods per plant, hot soil

temperature but not air temperature significantly (P<O.05) reduced flower production

and the proportion of pegs forming pods. In contrast hot air temperature significantly

(P<O.05) increased flower production and proportion of pegs forming pods. This

suggests that hot air temperature effects the proportion of flowers that set pegs or
 144

pods (Vara Prasad et al., 1998), whereas hot soil temperature effects flower

production and the proportion of pegs which form pods.

Comparison of effects of hot air and hot soil temperatures on nodulation and

nitrogen fixation in pea has shown that the soil temperatures are more critical than air

temperatures (Frings, 1976). This i~ consistent with a significant reduction in number

of nodules and nodule dry weight at hot soil, but not at hot air temperature (Table

10.3). Various aspects of the Rhizobium-legume symbiosis are known to be affected

by hot soil temperatures including growth and survival of rhizobia, formation of root

hairs, formation of infection threads, structure and development of root nodules, and

activity of the nitrogenase enzyme, resulting in reduced nodule numbers and nodule

mass per plant and total dry matter production and seed yields (Lie, 1974; Giller and

Wilson, 1991).

Although overall hot temperatures significantly reduced total dry matter

production and pod yield (Table 10.2), there were no significant effects of hot

temperatures on pod yields or SLN in plants dependent on symbiotic N2 and

symbioticN2 plus 20 N (Fig. 10.1). In contrast, hot temperature significantlyreduced

pod yields and SLN in plants dependent on inorganic N. The response of symbiotic

N2 plus 20 N to hot temperatures could be associated with more efficient nitrogen

assimilation by Rhizobium (Minchin et al., 1980). In cowpea and chickpea, plants

grown at warm temperatures with inorganic N were far more dependent on

remobilized N from leaves during reproductive growth than plants dependent on

symbiotic N2 (Minchin et al., 1980; Rawsthome et al., 1985 a, b). Indeed, plants

dependent on symbiotic N2 fixed significant amounts of N2 during reproductive

growth even at warm temperature, and this may well account for greater SLN at hot

air and hot soil temperature observed in the present study. There is a strong linear
 145

relation between SLN and radiation use efficiency (RUE) in groundnuts (Wright et

al., 1993) and RUE declined from 1.35 to 0.66 g Wl as SLN is reduced from 1.75
o2
to 0.97 g N m • There is also a strong positive association between SLN and carbon

exchange rates in groundnuts (Sinclair et al., 1993). Furthermore, both RUE and

carbon exchange rates are strongly, associated with pod yields (Williams and Boote,

1995). Therefore, the effects of temperature on SLN in the inorganic N treatment

would have reduced RUE and hence yield by about 50%, as observed in the current

research (Fig. 10.1).

It is evident that nodule-dependent plants were potentially more adaptable to

hot temperature than plants dependent on large quantities of inorganic N, provided

the strain is effective and tolerant, as reported for cowpea (Summerfield et al., 1978)

and chickpea (Rawsthome et al., 1985 a, b). The Bradyrhizobium strain NC 92 used

in the 'present research has been reported to survive well for 70 d at day/night

temperature of 41 °/29°C (Elsaeid et al., 1990). Differential response of rhizobial

strains to heat tolerance and their association with cultivars has been reported in

soyabean (Munevar and Wollum, 1981). Differences in optimum growth temperature

and tolerances has been observed for peanut rhizobia (Elsaeid et al., 1990;

Kishinevsky et al., 1992). It is therefore desirable to screen and select the proper

Rhizobium strain for temperature tolerances and asses their suitability under field

conditions.

In summary, this research has shown that increasing air and/or soil

temperature to 1Doe above typical ambient temperatures of 28° and 26°e,

respectively, reduces total dry matter production and pod yields of groundnut to a

similar relative extent, and that these effects were additive and without interaction.

Although plants relying on inorganic N produced significantly greater pod yields

 146

under optimum conditions, but were relatively more sensitive to hot temperatures

than those dependent on symbiotic N2• At hot temperatures plants dependent on

symbiotic N2 plus 20 N had significantlygreater values of SLN than those dependent

upon inorganic N and this relation may have contributed to their greater tolerance to

hot temperature.
 147

CHAPTER 11 GENERAL DISCUSSION

Heat stress is one of the least understood of all the abiotic constraints that limit the

growth, development and yield of annual crops (paulsen, 1994). Heat stress is also

one of the least controllable factors which affect grain legume productivity in the

semi-arid topics (Marshall, 1988). Heat stress is usually associated with drought

stress and/or mineral nutrient disorders but the effects of heat as opposed to water or

mineral nutrition have rarely been investigated. Furthermore, groundnuts, in contrast

to other legumes, have aerial flowers but subterranean fruits and so extremes of both

air and soil temperature are likely to be especially important environmental

constraints.

Groundnut crops grown in the semi-arid tropics often encounter brief or

sometimes prolonged periods of hot temperatures during the growing season. In

these circumstances seed yields are reduced. For example, for groundnut crops in

India heat stress is a serious problem mainly during the vegetative phase (North India,

summer season), or during the late vegetative phase (South India, post-rainy season)

or not until the flowering phase (rainy season) (Srinivasan et al., 1996). About 85%

of the total groundnut area in India is grown during the rainy season (Sankara Reddi,

1988) and so hot temperatures at flowering are a major constraint to seed yield.

Similarly, in the Sahelian regions of West Africa, groundnut crops also experience

hot temperatures during the flowering period (lCRISAT, 1994). Accordingly, the

research presented in this thesis has concentrated on the effects of heat stress during

the period from onset of flowering until setting of fruits.

 148

To sustain and stabilise groundnut production in the semi-arid tropics it is

important to provide farmers with: (a) a choice of cuItivars which are tolerant to heat

stress; andlor (b) to provide them with suitable management packages to minimise

the effects of heat stress. Selection and breeding for heat andlor drought tolerance

offer a long-term solution for increasing productivity of groundnuts in the semi-arid

tropics. The availability of useful genetic diversity for these complex traits is of

course an essential element for success. Plants have evolved various mechanisms to

cope with stress which have been broadly categorised as stress avoidance (escape)

and stress tolerance (Levitt, 1972). An ability to recover after stress is also an

important component of stress tolerance (Levitt, 1980). Certain cultivars or species

can adopt these mechanisms to avoid or tolerate stress, while others cannot.

Therefore, it is important to identify those cultivars which can tolerate heat stress and

then use them in future breeding programmes.

Most research in groundnuts has been targeted on drought tolerance, rather

than on heat tolerance, and significant progress has undoubtedly been made (Wright

and Nageswara Rao, 1994). However, research at the ICRISAT Sahelian Centre in

Niger has shown that temperature tolerance is the dominant attribute for adaptation

of groundnut cultivars to drought-prone regions of West Africa (Greenberg et al.,

1992). Earlier, Ketring (1986) concluded that traits for heat and drought tolerance

are genetically transferable. Studies in cowpea and common bean have shown heat

tolerance is commonly associated with only one or two genes which are strongly

heritable (Hall, 1992). Screening for heat tolerance is easier than screening for

drought tolerance, since water-use and water-use efficiency do not have to be

controlled or measured. Therefore, selecting for greater heat tolerance may offer
 149

breeders the opportunity for improving adaptation of groundnuts to drought-prone

regions.

To design a breeding programme based on heat tolerance, a better

understanding of responses of plants to heat stress is essential. That said, to examine

.
heat stress under field conditions is difficult because of the lack of control over the

timing, duration and intensity of stress. In addition, there are confounding effects of

drought, nutrition, photoperiod responses, edaphic and environmental conditions, and

pests and diseases. Heat stress studies, therefore, need to be conducted in controlled

environments in order to understand the physiological mechanism(s) of tolerance.

However, it is then essential to test and validate these mechanisms under field

conditions. Studies in controlled environments, for example, have been widely used

to help understand the physiological basis of phenological response to temperature

and photoperiod under field conditions (Summerfield et al., 1989; Hall, 1992). In the

programme of research reported here, several experiments were carried-out in

controlled environments to study the effects of continuous and short episodes of heat

stress on reproductive development and yield of different cultivars.

The research reported in this thesis has clearly shown that both continuous

exposure and short episodes of hot air temperature (38°/22°C) during the

reproductive period significantly reduce seed yields (Chapters 6 through 10).

Although both Spanish and Virginia types were affected by heat stress (Chapters 6

and 7), the Virginia types were the more sensitive. This difference in sensitivity was

due mainly to botanical type differences in reproductive strategy rather than to

responses to hot temperature per se. Spanish types are 'sequentially branched', i.e.

the appearance of the first flower marks the end of vegetative phase, and all

subsequent nodes, including those on main axis, are reproductive (Bunting and
 150

Elston, 1980). Therefore, once the first flower has appeared subsequent flower

production enters an exponential phase and flower number and reproductive sinks

accumulate rapidly. However, once seed growth starts, the seeds become the

dominant sinks for assimilates and so internal competition for assimilates effectively

terminates flower production. In contrast, Virginia types are 'alternate branching',

.
i.e. pairs of reproductive nodes alternate with pairs of vegetative nodes and the main

axis is always vegetative (Bunting and Elston, 1980). Therefore, flowers initially

appear much more slowly than in the Spanish types, and so sinks also accumulate

more slowly and therefore vegetative and reproductive growth continue

simultaneously.However, even after the start of seed growth in Virginia types flower

production continues, albeit at reduced rates.

In both botanical types, hot temperatures cause flower bud abortion and so

reduce the number of reproductive sinks. In consequence, flower production is not

limited by internal competition and so over time more flowers are produced in hot

temperature regimes than at ambient temperatures. The differences in sensitivity of

cvs within botanical types were related to an ability to produce more flowers and to

set fruits and so to maintain greater partitioning of dry matter to pods and roots at

hot temperatures.

Insight into the stages of Spanish and Virginia types most sensitive to hot air

temperature during reproductive development was gained by designing the reciprocal

transfer experiment described in Chapter 7. This experiment showed that groundnuts

were especially sensitive to hot air temperatures (38°/22°C) from 6 d before until

15 d after flowering and that the magnitude of sensitivity was related the number of

floral buds exposed to heat stress before anthesis. Therefore, sensitivity will vary with

cultivar depending on the pattern of flower production.

 151

However, simply identifying the relatively most sensitive stage and those

reproductive traits especially affected by heat stress is not sufficient for designing an

effective screening programme. Information on the quantitative effects of

temperature on reproductive traits such as flower production and the number of fruits

set at the most sensitive stages is al~o needed in order to identify critical temperatures

and to asses the impact of different hot temperature regimes on seed yields. Cultivar

ICGV 86015 was used in the experiments described in Chapter 8 and 9 to identify the

critical temperatures and quantify the effects of those temperatures on flower

production, pollen production, pollen viability, fruit-set and the fruit yield. The results

of those experiments showed that exposure for only 6 d to a day temperature of ~

34°C at 9 DAF (start of pegging) significantly reduced flower production, fruit-set

and fruit yield. Fruit-set was particularly sensitive to temperatures of>37.3°C during

the first 6 h of the daylight period in each 12 h diurnal cycle (Chapter 8). Overall,

there were negative quantitative relations between day temperature in the range from

28° to 48°C and both flower production (-1.3 plant" °C-l) and number of fruits (-1.1

plant" OC-l).Data presented by Fortanier (1957), Wood (1968) and Ong (1984) also

show similar significant reductions in the number of flowers and fruits as day

temperature increase from 25° to 35°C.

The threshold critical day temperature for pollen production and viability was

34°C and there was a strong negative quantitative relation between both pollen grain

number and pollen viability with accumulated day temperature >34°C (Chapter 9).

De Beer (1963) found that pollen death occurs when groundnuts plants are grown at

a constant day/night temperature of 33°C and that the number of pollen grains per

flower were reduced by 71% when temperature was increased from 24° to 33°C. The
 152

quantitative relations described in this thesis for flower production, fiuit-set and

pollen production are unique and should prove extremely useful to crop physiologists

and crop modellers seeking to predict and quantify the effects of hot temperatures on

groundnut production in different regions. Furthermore, the information now

available on sensitive periods, cr!tical stages, and the consequences of different

durations and timing of heat stress will be an important basis on which to design

screening techniques for heat tolerance in controlled environments.

The data presented in this thesis clearly indicate that groundnuts are sensitive

to heat stress prior to anthesis (Chapter 7) due especially to reduced fruit-set. This

finding is supported by those discussed in Chapter 9 where reduced fruit-set was

associated with fewer pollen grains and reduced pollen viability. The reduction in

pollen numbers and pollen viability may be a consequence of effects of heat stress at

microsporogenesis. Heat stress during floral bud microsporogenesis reduced anther

dehiscence and pollen viability in cowpea (Warrag and Hall, 1984) and also in

common bean (Konsens et al., 1991). Then again, the fact that groundnuts were

sensitive to hot temperature during first 6 h of the daylight period, strongly suggests

that the processes like pollination and fertilization could also be affected. This is

because pollination in groundnuts occurs mainly early in the morning just before

anthesis and fertilization is completed 5 to 6 h later i.e. before mid day (Lim and

Gumpil, 1984). Recent studies have also revealed that pollen tube growth is reduced

when groundnut flowers are exposed to hot temperatures of 35°/30°C (Talwar,

1997). It is now clear, then, that both pre-anthesis and post-anthesis events are

sensitive to heat stress in groundnuts.

The time taken from microsporogenesis to fertilization in groundnuts varies

between 3 and 6 d (Martin et al., 1974; Xi, 1991). Therefore, in all of the
 153

experiments presented in this thesis, temperature effects on pre-anthesis (Le.

micro sporogenesis) and post-anthesis (fertilization) were confounded, given that heat

stress was imposed for 6 d and that fruit-set was then estimated from the flowers

which opened during the 6 d stress period. It remains essential to separate the effects

of heat stress during (a) the pre-~thesis stage (i.e. on micro sporogenesis, anther

indehiscence and pollen viability); and (b) the post-anthesis stage (i.e. on stigma

receptivity, pollination, pollen tube growth, fertilization and embryo development)

and to discover whether any of these stages is more critical than others. Methods of

screening for heat tolerance at these two stages are discussed later in this Chapter

(pp 159 to 161).

The effects of heat stress at different stages of reproductive development,

including: floral bud development, fruit-set, embryo development and fruit or seed

development, and the mechanisms affected at these stages have been studied in detail

in cowpea, common bean, tomato, cotton (Gossypium hirsutum L.) and rice (Oryza

sativa L.) (Hall, 1992). Studies in cowpea have revealed that floral buds are sensitive

to heat stress from 9 to 7 d before flower anthesis (Ahmed et al., 1992). This period

is just after meiosis of the pollen mother cells, and heat stress during this stage

resulted in reduced fruit-set. The poor fruit-set was a result of incomplete anther

dehiscence and reduced pollen viability, caused by the inhibition of proline

accumulation in pollen grains due to degeneration of the tapetal layer (Mutters et al.,

1989a; Ofir et al., 1993). The amino acid proline protects pollen grains from heat

stress during germination and contributes to protein synthesis during pollen tube

elongation (Zhang and Croes, 1983 a, b).

In common bean, plants were sensitive to heat stress at 10 d before anthesis

i.e. during micro sporogenesis. Anthers failed to dehisce at hot temperatures and also
 154

had poor pollen viability (Halterlein et al. 1980; Konsens et al., 1991; Gross and

Kigel, 1994). Similarly, in tomato, stamens are particularly sensitive to hot

temperature from 9 to 4 d before anthesis, Le. during and just after the stages of

meiosis when the microspores are released from tetrads (Iwahori, 1965). Tomato

flowers are also sensitive to heat == at 1 to 3 dafter anthesis leading to reduced

fruit-set (Iwahori, 1966). In summary, fruit-set is affected by a combination of the

lack of anther dehiscence due to the absence of endothecium formation in anthers

(Rudich et al., 1977); excessive style elongation and exertion of stigma which

prevents normal pollination (Rick and Dempsey, 1969); and retarded embryo

development and embryo abortion because of degeneration of the endosperm

(Iwahori, 1966).

Genetic variation in pollen shed and fruit-set has been reported in cowpea

(patel and Hall, 1990), common bean (Dickson and Petzoldt, 1989), tomato (Levy et

al., 1978; El-Ahmadi and Stevens, 1979), rice (Yoshida et al., 1981) and cotton

(Rodriguez-Garay and Barrow, 1988). Heat-tolerant cultivars have been identified

based on their ability to shed pollen and set fruits at hot temperatures (Hall, 1992).

Certain heat-tolerant cultivars of rice escape hot temperature stress by shedding

pollen during the cooler morning hours (Mackill et al., 1982). In cowpea (Marfo and

Hall, 1992) and in common bean (Shonnard and Gepts, 1994), heat tolerance during

the reproductive development phase is conferred by dominant genes that are

heritable. Similar studies have indicated that genetic differences in pollen shed and

ability to set seed under hot temperature were strongly heritable in rice (Mackill and

Coffinan, 1983) and cotton (Rodriguez-Garay and Barrow, 1988).

Most of the progress made in breeding heat-resistant cultivars has been by

incorporating heat tolerance during reproductive development through selection for

 155

characters such as fruit-set in hot field nurseries (Hall, 1992). Inheritance studies

indicate that different aspects of heat tolerance during reproductive development are

conferred by major genes, which are either recessive or dominant, and substantial

progress has been made using pedigree breeding methods (Hall, 1992). Breeding for

heat tolerance by selecting for repr~ductive traits such as fruit-set and pollen shed has

been successful in cowpea, cotton, and tomato. Therefore, it is important to identify

genetic variability for heat tolerance based on reproductive traits in the groundnut

germplasm and to use them in breeding programmes to develop heat- and drought-

tolerant cultivars for the hostile climates of the semi-arid tropics. Given the successes

achieved in other grain legumes, it is to be hoped that heat tolerance genes in

groundnut will also prove to be simply inherited.

Air and soil temperature >35°C in the semi-arid tropics commonly occur

during the reproductive period (ICRISAT, 1992). Although the effects of hot air

(Ketring, 1984a; Ong, 1984) and hot soil temperature (Ono et al., 1974; Golombek

and Johansen, 1997) have been studied separately, very little research has been done

on the combination of these two factors. Therefore experiments were conducted to

investigate the effects on pod yields of both hot air and hot soil temperature imposed

from the onset of podding (Chapter 6) and from onset of flowering to maturity

(Chapter 10). Increasing air and/or soil temperature by lOoC during the day above

ambient values of 28° and 26°C, respectively, reduced pod yields to a similar extent,

and the effects were additive and without interaction. The smaller pod yields at hot

air temperature were a consequence of reduced fruit-set, and hence fewer fruits and

so partitioning of less dry matter to pods. Hot soil temperature, in contrast, reduced

flower production, as well as the proportion of pegs forming pods and 100 seed

weight. In other words, hot air and hot soil temperatures were affecting different
 156

yield forming processes, resulting in additive rather than interactive effects. This is

understandable given that most but not all flowers are pollinated above ground (some

flowers may be covered by soil following weeding or earthing up) while pod and seed

development occurs below ground.

In the experiments describe~ in Chapters 6 to 9, groundnut plants were not

inoculated with Rhizobium and were so totally dependent on inorganic N supplied

from a slow-release fertiliser. This decision was taken to remove any confounding

effects of temperature on symbiotic N2 fixation. However, in the natural edaphic

environments of the semi-arid tropics, plants are mostly or even wholly dependent on

dinitrogen fixation (Giller et al., 1987). Hot temperatures are known to adversely

effect both nodulation and dinitrogen fixation in legumes (Arayangkoon et al., 1990).

The optimum temperature range for symbiotically-dependent legumes is narrower

than that of plants supplied with fertiliser nitrogen (Sprent and Sprent, 1990; Lie,

1971). However, there is very little or no information on this subject in groundnuts.

An experiment was therefore designed to investigate the effects of heat stress on

plants inoculated or not with strain of Bradyrhizobium Ne 92 or reliant on a

combination of both symbiotic N2 and inorganic N (Chapter 10).

At close-to-optimum temperatures plants dependent on 100 ppm inorganic N

produced significantly greater dry matter and pod yields than those dependent on

symbiotic N2 alone or on symbiotic N2 plus 20 ppm inorganic N. However, at hot

temperatures plants dependent on 100 ppm inorganic N and those dependent on

symbiotic N2 plus 20 ppm inorganic N produced similar pod yields. Hot temperatures

had no effects on pod yields of plants dependent on symbiotic N2 plus 20 ppm

inorganic N or on symbiotic N2 alone, but significantly reduced pod yields in plants

dependent on inorganic N. The superiority in terms of seed yield of nodule-dependent

 157

plants supplied with small quantities of inorganic N in warm temperatures has been

reported in cowpea (Summerfield et al., 1978) and chickpea (Rawsthome et al.,

1985a).

It is therefore possible that in hot environments the use of effective heat-

tolerant strains of rhizobia in combi.nation with a optimum dose of fertiliser nitrogen,

perhaps as a 'starter' dose, based on soil fertility status may be beneficial. Differential

responses of strains of rhizobia to heat have been reported in for common bean,

cowpea and soyabean (piha and Munns, 1987; Munevar and Wollum, 1981), but

such information is scarce for groundnuts. Further, the comparative efficiency of

native rhizhobia and selected strains of introduced Rhizobium to fix nitrogen at hot

temperatures needs to be investigated not only to identify tolerant strains but also to

assess their competitive ability and survival under field conditions.

The results obtained from different experiments presented in this thesis also

provide an opportunity to compare the effects of hot air and/or hot soil temperatures

imposed either from the start of flower bud initiation or from the start of podding

until maturity (as described in Chapter 6), or from the start of flowering until

maturity (as described in Chapter 10). These comparisons show that the effects of

continuous hot air and/or hot soil temperatures in the different experiments were

consistent (Fig. 11.1). The effects of hot air temperature imposed from flower bud

initiation until maturity (Exp. 1) or from first flower appearance until maturity

(Exp. 2) were relatively greater than those imposed from podding until maturity

(Exp. 3). This finding may reflect the fact that a greater number of those flowers

which appeared during the initial stages of flowering experienced heat stress or it may

simply reflect the longer duration of exposure to hot temperatures experienced by

plants in Exp. 1 and 2. The effects of hot soil temperature imposed from first flower
 158

• 30°C (Hot air)

tOO
D 34°C (Hot soil)
~ Hot air + Hot soil

75

50

25

o
Exp.t Exp.2 Exp.3

Temperature treatments by experiment

Fig. 11.1 Relative difference (%) in pod yield at mean day/night hot air
(30oq or hot soil (34oq or hot air and hot soil combined temperature
treatments (30o/34°q in relation to values at an optimum temperature
of 25°C, when temperature treatments were imposed from flower bud
initiation (Expo1), or from first flower appearance (Expo2) or from
pod initiation (Expo3) until reproductive maturity at 90 DASo
 159

appearance until maturity (Exp. 2), and from pod initiation until maturity (Exp. 3),

were similar. It is apparent, therefore, that hot air temperature is more critical from

flower bud initiation until the onset of podding, whereas hot soil temperature is

critical from pod initiation until maturity.

Information on the quantitative effects of hot soil temperatures on seed yields,

and on the mechanisms of tolerance, cuItivar variation in tolerance, and on the

inheritance of the traits affected by hot soil temperature on groundnuts are poorly

understood. These data are of critical importance because the pods are formed below

ground, and so soil temperature plays a major role in determining seed yields.

As indicated earlier in this Chapter there is a need to screen for tolerance to

heat stress both prior to and after anthesis under controlled conditions. Screening for

heat tolerance at the pre-anthesis stages can be done by exposing plants to 6 d of hot

air temperature of>37°C during first 6 h of the daylight period, between 6 d before

until15 d after flowering. All flowers which open between the start of flowering until

3 d after the start of the temperature treatment, and all flowers which open 3 dafter

the end of the temperature treatments need to be removed. The logic here is that the

buds which open between 3 d after the start and 3 d after end of the temperature

treatment would have experienced hot temperature at 3 to 6 d before anthesis, i.e.

during micro sporogenesis. The experiment should end when all the flower buds

exposed to heat stress have been given an opportunity to set pegs (i.e. about 9 dafter

opening of flower buds). By estimating the ability of those flower buds which

experienced hot temperature to set fruits, relative tolerance or susceptibility to hot

temperature at the pre-anthesis stages can be exposed and quantified.

Screening for tolerance during and after anthesis can also be carried-out using

a 3 d heat stress treatment of>37°C. However, in this case it is those flowers which
 160

open before the start of the temperature treatment and after the end of the

temperature treatments which are removed. Tolerance is then assessed based on the

ability of those flowers which open during the treatment period to set fruits. In this

approach the duration of exposure to hot temperature is only 3 d, because the

minimum period from microsporogenesis to fertilization is 3 d (Martin et al., 1974;

Xi, 1991). Therefore ifplants are exposed to hot temperature for longer than 3 d then

some of the floral buds may also be affected.

Another and more precise way to screen for heat tolerance at these two

stages of development would be to tag individual unopened flower buds or fully

opened flowers which have been exposed to hot temperature and then to monitor

their ability to set fruits. The relation between the size of groundnut floral buds and

stage of development during microsporogenesis has been documented (Wilcox et al.,

1990) and this information would enable the tagging of flower buds at specific stages.

However, tagging individual flower buds or flowers is more difficult than removing

them due to their small size and because all flowers wither within 24 h after opening.

Screening for adaptation to hot soil temperature can be done by exposing

different cultivars to those conditions (40o/30°C) from the onset of podding until

maturity. Tolerance can be measured by estimating the ability of cultivars to produce

flowers and to set pods, and/or to produce dry matter and maintain a relatively

greater partitioning to pods in the stress conditions. Hot soil temperatures could be

imposed by tubular heaters as used in the present research or by other comparable

methods such as soil warming cables (Dreyer et al., 1981; Sanders et al., 1985) or

hot water baths (Golombek and Johansen, 1997).

The performance of cultivars at hot air and/or hot soil temperature under field

conditions could also be evaluated by using appropriate sowing dates and analyses of
 161

corresponding weather data (Greenberg et al., 1992). The disadvantage of this

approach would be the real risk of confounding responses with seasonal changes in

photoperiod, drought, pests and diseases incidence and management practices. Air

temperatures during the critical stages of crop development in the field can be

increased by constructing plastic e~velopes over the crop. For example, Omanga et

al. (1996) found that simply covering pigeonpea (Cajanus eajan L.) plants with a

clear plastic polythene sheet could increase the air temperature by 7° to 10°C above

the ambient values in Kenya. This approach proved very successful in advancing

understanding of photothermal effects on flowering.

In the present research, plants were not allowed to acclimate to hot

temperatures and so the transition from optimum to stress conditions was rapid.

However, in screening for heat tolerance the effects of acclimation need to be

considered. This is because under natural conditions temperature increases relatively

slowly and so plants may be able to adapt or acclimate to hot temperatures. In

contrast, a sudden rise in temperature may induce a different response. Based on

experience where cultivar rankings of other grain legumes in the field were reliably

predicted by the relative responses of plants following abrupt changes in temperature

in controlled environments [e.g. for cowpea, chickpea and lentil see Summerfield et

al., 1983; 1984; 1989], it would be surprising if the effects observed in the current

research would become invalid for groundnut in the field.

Differences in heat acclimation response may be related to the production of

heat shock proteins (Necchi et al., 1987; Blumenthal et al., 1990). Heat acclimation

potential and production of heat shock proteins are reported to vary widely between

cultivars of the same species, e.g. in wheat (Triticum aestivum L.; Blumenthal et al.,

1991), sorghum (Sorghum bicolor L.; Ougham and Stoddart, 1986), maize (Zea
 162

mays L.; Ristic et al., 1991) and soyabean (Li et al., 1991). Therefore, the ultimate

selection of heat-tolerant cultivars to be used as parents in breeding programmes

should be based on the ranking of their heat tolerance after they have been given the

opportunity to acclimate, as well as on their performance at hot temperature under

field conditions.

Crop simulation models have the potential of integrating knowledge of crop

physiology and of being important tools for decision making and planning decision

support systems and contingencies. Various models have been incorporated into the

software package entitled Decision Support Service for Agrotechnology Transfer

(DSSAT) and a crop growth model for peanut (Le. PNUTGRO, Boote et al., 1986)

has been incorporated into DSSAT. Several other models are also available for

simulating growth, development and yield of groundnuts e.g. PEANUT (Young et

al., 1979); PEANUTZ (Duncan et al., 1978); and QNUT (Hammer et al., 1995). The

DSSAT model is probably the most widely used. Accordingly, a preliminary analysis

was done to test ifPNUTGRO was able to simulate the effects of short or continuous

episodes of hot temperatures and/or water deficits imposed during flowering.

Results of the simulations showed that the model predicts the effects of water

stress well and that these quantitative effects were similar to those reported in

controlled environments by Craufurd et al. (1996) and in field by Nageswara Rao et

al. (1989). But for hot temperatures, the simulated effects of short episodes of stress,

were much smaller than those obtained in the current research, i.e. model predictions

underestimated reality by 20 to 30% (Fig. 11.2). This suggests that the model is not

sufficiently robust to accurately predict the effects of hot temperatures. Therefore,

validation of the model for hot climates is required. Further detailed investigations on

the long term trends in climate data and groundnut yields for various locations in the
 163

,,-...
~u
..=1II
'-" 0
100
oN
~
"t ~
0"C
Q.c-
90
~
0 ••
'Q.c-
"" ~
~ 0 80
rn ~ '""
~
=
-~-=......
"'C
.~
0
c.J
~
70
"'C c.,.
~
o 0 60

0 -6 -3 0 3 6 9 12 15 18

Timing of temperature treatment

relative to onset of flowering (d)

Fig.ll.2 Relative difference (%) in pod yield when plants were exposed
to 6 d of mean day/night air temperature of 30°C at different times relative
to the onset of flowering in relation to mean day/night air temperature of
25°C, as recorded in the current controlled envrionment studies (.) and as
simulated by the PNUTGRO model (0).
 164

tropical and semi-arid tropical countries would also be useful to incorporate a

suitable routine to quantitatively predict the effects of heat stress.

Crop growth models of rice (e.g. ORYZAI and SIMRlW) have been used to

predict the effects of climate change (i.e. hot temperature and increase in CO2

concentration) on yields of rice in ,hot climates of Asia (Matthews et al., 1995). It

was clear that the response of spikelet sterility to hot temperature was a major factor

determining the reliability of predictions in this region. Due to the extreme sensitivity

of spikelet sterility to temperatures warmer than 33°C, small increases in mean and

maximum temperature gave large reductions in yield due to the filling of fewer grains.

This suggests that models can predict the effect of hot temperatures provided they

are validated and have routines which describe the sensitivity of specific reproductive

traits and the quantitative effects of temperature. Therefore, knowledge of the

quantitative effects of temperature at the most sensitive stages of development on

flower number, pollen production, pollen viability, fruit-set, fruit numbers and pod

yield gained by this research on groundnut should now be readily incorporated into

predictive models.

Once the models have been validated for hot environments, they can be

combined with Geographical Information Systems (GIS) techniques to provide

quantitative estimates of the effects of climate change and the severity and timing of

heat and drought stress on different groundnut growing regions on international,

national or regional scales. These predictions and estimates will provide a better

focus for breeding programmes and will be useful to identify improved combinations

of planting locations, appropriate cultivars and sowing dates. Integration of

information from general atmospheric circulation models (on climate predictions)

with crop simulation models (on the effects of climate change on potential yields)
 165

would be useful to: (a) predict the timing and severity of hot temperature ahead of

time based on the analyses of historical climate data; and (b) take appropriate crop

management decisions, such as the selection of appropriate cultivars and date of

sowing to best avoid heat stress during the most sensitive stages.

In conclusion, the research {lrogramme reported in this thesis has led to better

understanding of the effects of hot air and hot soil temperature on reproductive

growth and development of groundnut. In particular the sensitive stages have been

identified, and the quantitative effects of air temperature on reproductive yield at

these sensitive stages have been described. The information presented here will help

crop physiologists, plant breeders and agronomists to: (a) design effective screening

techniques to assess heat-tolerance; (b) incorporate quantitative relations into crop

growth models and so predict the consequences of global warming on groundnut

production; and (c) develop a suitable agronomic package for the harsh environments

of the semi-arid tropics.

11.1 FUTURE RESEARCH

Given that the research described earlier has provided information and a basis for

designing a technique of screening cultivars for responses to hot air and hot soil

temperatures under controlled environments, available germplasm should now be

assessed for heat tolerance to air temperature by exposing plants to day/night

temperatures of 38°/28°C between 6 d before and 15 OAF based on their ability to

set fruits. Similarly, the heat tolerance to hot soil temperature should also be assessed

based on the ability of cultivars to produce greater pod yields after exposure to
 166

day/night soil temperature of 38°/30°C, starting from pod initiation and continuing

until maturity.

The knowledge gained by this research on the quantitative effects of heat

stress on yield and yield components should be incorporated into crop simulation

.
models and use them to quantitatively predict the effects of seasonal variants and

global warming on yields in different groundnut growing regions of the world.

Further research on the effects of heat stress on reproductive processes

during the pre-anthesis stages (microsporogenesis, pollen shed and pollen viability)

and during the post-anthesis stages (stigma receptivity, pollination, pollen tube

growth, fertilization and embryo development) would be useful to better understand

the relative importance of these stages. This information will also help us to

understand the mechanism(s) of tolerances in different cultivars.

The research presented in this thesis has focused primarily on the effects of

heat stress on reproductive traits, and did not investigate the physiological basis for

general heat tolerance. Other studies on groundnuts have indicated that physiological

traits such as rate of photosynthesis, membrane thermo-stability and chlorophyll

fluorescence could be used to select heat-tolerant cultivars (Srinivasan et al., 1996;

Chauhan and Senbaku, 1997). Therefore, screening for heat-tolerance based on

physiological traits and their impact on reproductive efficiency at hot temperatures

needs to be investigated.

In the present research, plants were not allowed to acclimate to hot

temperatures and the transition from optimum to hot temperature was rapid. In the

field these changes may occur more slowly. Therefore, experiments are now needed

to identify the effects of acclimation to hot temperature and gradual increase to a

daily maximum temperature on the expression of tolerance and reproductive yield.

 167

Once heat-tolerant cultivars have been identified based on either reproductive traits

or physiological traits, detailed studies on the inheritance of heat tolerance based on

conventional breeding methods and molecular markers need to be undertaken.

Finally this research suggests that plants dependent on symbiotic N2 were

potentially more adaptable to hOot temperature than those dependent on large

quantities of inorganic N. Differences in optimum growth temperature and tolerance

have been reported for groundnut rhizobia (Elsaeid et al., 1990; Kishinevsky et al.,

1992). It is therefore desirable to screen different strains for heat-tolerance and to

assess their suitability and agronomic superiority (persistence and competitive ability)

under field conditions.

Nitrogen fixation in groundnut is strongly influenced by cultivar, strain of

Rhizobium, edaphic environment (such as heat or water-stress and by nutrient

regimes e.g. concentrations ofN03-N, P and Fe), and the interactions between these

factors (Elkan, 1995). Unfortunately, little is known about the impact of these

interactions on nodule-dependent plants and there is an urgent need to understand

these effects in order to develop suitable agronomic packages for warm climates.

Hot temperatures are often associated with water and/or nutrient deficits, dry

air and large radiation values and very little research has been done on these

interactions. This thesis has identified those stages of groundnut which are most

sensitive to stress, and the reasons which determine those responses. It provides

therefore a quantitative platform for future and integrated research on the effects of

abiotic and biotic factors alike.

 168

REFERENCES

ADAMS, F. and HARTZOG, D.L. (1980). The nature of yield of Florunner peanuts
to lime. Peanut Science 7: 120-123.

AHMED, F.E., HALL, A.E. and DEMAS ON, D.A. (1992). Heat injury during floral
development in cowpea (Vigna unguiculata, Fabaceae). American Journal of
Botany 79: 784-791.

AHRING, RM., BANKS, D.l and SPRINGER, T.L. (1987). Peanut seedling
responses to root temperature controlled by a thermogradient sand.
Proceedings of American Peanut Research and Education Society 19: 39.

ALEXANDROV, V.Y. (1964). Cytophysiological and cytoecological investigations

of resistance of plant cells toward the action of high and low temperature.
Quarterly Review of Biology 39: 35-77.

ALLEN, O.N. and ALLEN, E.K. (1940). Response of the peanut plant to inoculation
with rhizobia, with special reference to morphological development of the
nodules. Botanical Gazette 102: 121-142.

ANGUS, IF., CUNNINGHAM, RB., MONCUR, M.W. and MACKENZIE, D.H.

(1981). Phasic development in field crops. I. Thermal response in the seedling
phase. Field Crops Research 3: 365-378.

ARAYANGKOON, T., SCHOMBERG, H.H. and WEAVER, RW. (1990).

Nodulation and N2 fixation of gaur at high root temperature. Plant and Soil
126: 209-213.

BAGNALL, D.l and KING, RW. (1991a). Responses of peanut (Arachis

hypogaea) to temperature, photoperiod and irradiance. 1. Effect on
flowering. Field Crops Research 26: 263-277.

BAGNALL, D.J. and KING, RW. (1991b). Responses of peanut (Arachis hypogaea)
to temperature, photoperiod and irradiance. 2. Effect on peg and pod
development. Field Crops Research 26: 279-293.

BASSIRIRAD, H., RADIN, lW. and MATSUDA, K. (1991). Temperature

dependent water and ion transport properties of barley and sorghum roots.
Plant Physiology 97: 426-432.

BELL, M.l, SHORTER, R. and MAYER, R. (1991). Cultivar and environmental

effects on growth and development of peanuts (Arachis hypogaea L.) I.
Emergence and flowering. Field Crops Research 27: 17-33.
 169

BELL, M.I., WRIGHT, G.C. and HARCH, G. (1993a). Environmental and

agronomic effects on the growth of four peanut cultivars in a subtropical
environment. I. Dry matter accumulation and radiation use efficiency.
Experimental Agriculture 29: 473-490.

BELL, M.I., WRIGHT, G.C. and HARCH, G. (1993b). Environmental and

agronomic effects on the growth of four peanut cultivars in a subtropical
environment. II. Dry matter partitioning. Experimental Agriculture 29: 491-
501.

BERRY, I. and BIORKMAN, O. (1980). Photosynthetic response and adaptation to

temperature in higher plants. Annual Review of Plant Physiology 31: 491-
543.

BHAGSARI, AS., BROWN, R.H. and SCHEPERS, 1.S. (1976). Effect of moisture
stress on photosynthesis and some related physiological characteristics in
peanuts. Crop Science 16: 712-715.

BLUMENTHAL, C.S., BATEY, I.L., BEKES, F., WRIGLEY, C.W. and

BARLOW, E.W.R. (1991). Seasonal changes in wheat-grain quality
associated with high temperatures during grain filling. Australian Journal of
Agricultural Research 42: 21-30.

BLUMENTHAL, C.S., BEKES, F., WRIGLEY, C.W. and BARLOW, E.W.R.

(1990). The acquisition and maintenance of thermotolerance in Australian
wheats. Australian Journal of Plant Physiology 17: 37-47.

BOLHUIS, G.C. and DE GROOT, W. (1959). Observations on the effect of varying

temperatures on the flowering and fruit set in three varieties of groundnut.
Netherlands Journal of Agricultural Sciences 7: 317-326.

BOLHUIS, G.G., FRINKING, H.D., LEEWAUGH, 1., RENS, R.G. and

STARITSKY, G. (1965). Occurrence of flowers with short style in groundnut
(Arachis hypogaea). Oleagineux 20: 293-296.

BOOTE, K.I. (1982). Growth stages of peanut (Arachis hypogaea L.). Peanut
Science 9: 35-40.

BOOTE, K.I. and KETRING, D.L. (1990). Peanut. In Irrigation of Agricultural

Crops, pp. 625-717. (Eds B.A Stewart and D.R. Nielsen). American Society
of Agronomy, Madison, Wisconsin, USA

BOOTE, K.1., lONES, I.W., MISHOE, I.W. and WILKERSON, G.G. (1986).
Modelling growth and yield of groundnut. In Agrometeorology of Groundnut,
pp. 243-254. (Eds M.V.K. Sivakurnarand S.M. Virmani). International Crops
Research Institute for the Semi-Arid Tropics, Patancheru, India.
 170

BRECKE, B.I. (1995). Management of weeds. In Peanut Health Management, pp.

43-50. (Eds H.A Melouk and F.M. Shokes). The American
Phytopathological Society, Minnesota, USA.

BROUGHTON, W.I. (1978). A review - control of specificity in legume - Rhizobium

associations. Journal of Applied Bacteriology 45: 165-194.

BUNTING, A. H. (1955). A classification of cultivated groundnuts. Empire Journal

of Experimental Agriculture 23: 158~170.

BUNTING, AH. (1958). A further note on the classification of cultivated

groundnuts. Empire Journal of Experimental Agriculture 26: 254-258.

BUNTING, AH. and ELSTON, I (1980). Ecophysiology of growth and adaptation

in the groundnut: an essay in structure, partition and adaptation. In Advances
in Legume Science, pp. 495-500. (Eds RI Summerfield and AH. Bunting).
HMSO, London, UK.

BUNTING, AH., GIBBONS, RW. and WYNNE, IC. (1985). Groundnut (Arachis
hypogaea L.). In Grain Legume Crops, pp. 747-800. (Eds RI. Summerfield
and E.H. Roberts). Collins Publications, London, UK. .

CAMPBELL, I.S. (1980). Growth and development of Florunner peanuts as affected

by temperature. Dissertation Abstracts-International 41: 1596-1597.

CHANDLER, M.R. (1978). Some observations on infection of Arachis hypogaea L.

by rhizobium. Journal of Experimental Botany 29: 749-755.

CHATEL, D.L. and PARKER, C.A (1973). Survival of field-grown rhizobia over
the dry summer period in Western Australia. Soil Biology and Biochemistry
5: 415-423.

CHAUHAN, Y.S. and SENBAKU, T. (1997). Evaluation of groundnut genotypes

for heat tolerance. Annals of Applied Biology 131: 481-489.

COx, F.R (1979). Effect of temperature treatment on peanut vegetative and fruit
growth. Peanut Science 6: 14-17.

COx, F.R and MARTIN, C.K. (1974). Effect of temperature on time from planting
to flowering in Virginia type peanuts (Arachis hypogaea L.). Peanut Science
1: 86-90.

COx, F.R and SHOLAR, IR (1995). Site selection, land preparation, and
management of soil fertility. In Peanut Health Management, pp. 7-10. (Eds
H.A Melouk and F.M. Shokes). The American Phytopathological Society,
Minnesota, USA
 171

COx, F.R, ADAMS, F. and TUCKER, B.B. (1982). Liming, fertilisation and
mineral nutrition. In Peanut Science and Technology, pp. 139-163. (Eds H.E.
Pattee and C.T. Young). American Peanut Research and Education Society,
Stilwater, OK, USA.

CRAUFURD, P.Q., WHEELER, T.R, ELLIS, RH. and SUMMERFIELD, RI.

(1996). High temperature and water-deficit tolerance in groundnuts. Final
Technical Report, pp. 35. Department of Agriculture, The University of
Reading, Reading, UK.

DART, P.I (1977). Infection and development of leguminous nodules. InA Treatise
on Dinitrogen Fixation, Section III Biology, pp. 367-472. (Eds RW.F. Hardy
and W.S. Silver). John Wiley and Sons, New York, USA.

DART, P.I and DAY, IM. (1971). Effect of incubation temperature and oxygen
tension on nitrogenase activity of legume root nodules. Plant and Soil Special
Vol.: 167-184.

DART, P.I, ISLAM, Rand DOBEREINER, I. (1975). Symbiosis in tropical grain

legumes: some effects of temperature and composition of the rooting
medium. In Symbiotic Nitrogen Fixation in Plants Vol. 7, pp. 361-384. (Ed.
P.S. Nutman). Cambridge University Press, Cambridge, UK.

DAUGHTRY, C.S., BROWN, RH. and ETHREDGE, W.I (1975). Effect of time
of application of succinic acid 2,2-dimethylhydrazide on yields and associated
characteristics of peanuts. Peanut Science 2: 83-86.

DAY, IM., ROUGHLEY, RI., EAGLESHAM, ARl, DYE, M. and WillTE, S.P.
(1978). Effect of high soil temperatures on nodulation of cowpea, Vigna
unguiculata. Annals of Applied Biology 88: 476-481.

DE BEER, IF. (1963). Influences of temperature on Arachis hypogaea L. with

special references to its pollen viability. Thesis, State Agricultural University,
Wageningen, The Netherlands.

DICKSON, M.H. and PETZOLDT, R (1989). Heat tolerance and pod set in green
beans. Journal of American Society of Horticultural Research 114: 833-836.

DREYER, I, DUNCAN, W.G. and McCLOUD, D.E. (1981). Fruit temperature,

growth rates, and yield of peanuts. Crop Science 21: 686-688.

DUDEJA, S.S. and KHURANA, AL. (1989). Persistence of Bradyrhizobium sp.

(Cajanus) in a sandy loam. Soil Biology and Biochemistry 21: 709-713.

DUGAS, W.A. (1984). Agroclimatic atlas of Texas part 7: Soil temperature. Texas
Agricultural Experimental Station, Miscellaneous. Publication 1552: 132.

DUKE, lA (1981). Handbook of Legumes of World Economic Importance. Plenum

Press, New York, USA.
 172

DUMAS, 1.B.A (1831). Procedes del analyse organique. Ann. Chim. Phys. 247:
198-213.

DUNCAN, W.G., McCLOUD, D.E., McGRAW, RL. and BOOTE, K.!. (1978).
Physiological aspects of peanut yield improvement. Crop Science 18: 1015-
1020.

EAGLES HAM, ARl, HASSOUNA, S. and SEEGERS, R (1983). Fertilizer-N

effects on N2 fixation by co~ea and soyabean. Agronomy Journal 75: 61-66.

ECHLIN, P. (1971). The role of tapetum during micro sporogenesis of angiosperms.

In Pollen: Development and Physiology, pp. 41-61. (Ed. 1. Heslop-Harrison).
Butterworths, London, UK.

EL-AHMADI, AB. and STEVENS, M.A (1979). Reproductive responses of heat-

tolerant tomatoes to high temperatures. Journal of American Society of
Horticultural Science 104: 689-691.

ELKAN, G.H. (1995). Biological nitrogen fixation in peanuts. In Advances in Peanut

Science, pp. 286-300. (Eds H.E. Pattee and H.T. Stalker). American Peanut
Research and Education Society, Stilwater, OK, USA

ELSAEID, N.R, WOLLUM, AG. and MUSSELwmTE, M.S. (1990). Growth

temperature tolerance and survival of peanut rhizobia in plant inoculants.
Turrialba 40: 287-291.

ENYI, RA.C. (1977). Physiology of grain yield in groundnuts (Arachis hypogaea).

Experimental Agriculture 13: 101-110.

FAD. (1998). Production Year Book 1997. Food and Agricultural Organisation,
Rome, Italy.

FORTANIER, E.J. (1957). Control of flowering in Arachis hypogaea L.

Mededelingen van de Landbouwhogeschool te Wageningen 57: 1-116.

FOWDEN, L. (1954). The nitrogen metabolism of groundnut plants: The role of y-

methylene glutamine and y-methyleneglutamic acid. Annals of Botany 18:
417-440.

FRESHNEY, R.I. (1994). Culture of Animal Ce/ls-A Manual of Basic Techniques.

Wiley-Liss, New York, USA

FRINGS, 1.F.1. (1976). The rhizobium - pea symbiosis as affected by high

temperature. Mededelingen Landouwhogeschool, Wageningen, The
Netherland76: 1-76.

GENSTAT 5 COMMITTEE. (1987). Genstat Reference Manual. Clarendon Press,

Oxford, UK.
 173

GIBBONS, RW. (1986). Biological constraints to increased groundnut production

in the semi-arid tropics. In Agrometeorology of Groundnut, pp. 47-62. (Eds
M.V.K Sivakumar and S.M. Virmani). International Crops Research Institute
for the Semi-Arid Tropics, Patancheru, India.

GIBBONS, RW., BUNTING, A.H. and SMARTT, 1. (1972). The classification of

varieties of ground nut (Arachis hypogaea L.). Euphytica 21: 78-85.

GIBSON, A.H. (1966). Physical environment and symbiotic nitrogen fixation. m.

Root temperature effects 'on shoot and root development and nitrogen
distribution in Trifolium subterraneum. Australian Journal of Biological
Sciences 19: 219-232.

GILLER, KE. and WILSON, KJ. (1991). Nitrogen Fixation in Tropical Cropping
Systems. CAB International, Wallingford, Oxon, UK.

GILLER, KE., NAMBIAR, P.T.C., SRINIVASA RAO, B., DART, P.I and DAY,
I.M. (1987). Nitrogen fixation in genotypes of groundnut (Arachis hypogaea
L.) compared using l~-isotope dilution. Biology and Fertility of Soils 5: 23-
25.

GOLOMBEK, S.D. and IOHANSEN, C. (1997). Effect of soil temperature on

vegetative and reproductive growth and development in three Spanish
genotypes of peanut (Arachis hypogaea L.). Peanut Science 24: 67-72.

GREENBERG, D.C., WILLIAMS, I.H. and NDUNGURU, B.I. (1992). Differences

in yield determining processes of groundnut (Arachis hypogaea L.) genotypes
in varied drought environments. Annals of Applied Biology 120: 557-566.

GREGORY, P.I and BROWN, S.C. (1989). Root growth, water use and yield of
crops in dry environments: What characteristics are desirable? Aspects
Applied Biology 22: 235-243.

GREGORY, W. C., KRAPOVICKAS, A and GREGORY, M.P. (1980). Structure,

variation, and evolution and classification in Arachis. In Advances in Legume
Science, pp. 469-481. (Eds RI. Summerfield and AH. Bunting). Royal
Botanical Gardens, Kew, London, UK.

GREGORY, W.C. and GREGORY, M.P.(1976). Groundnut, Arachis hypogaea. In

Evolution of Crop Plants, pp. 151-154. (Ed. N.W. Simmonds). Longman
Group Limited, London, UK.

GREGORY, W.C., GREGORY, M.P., KRAPOVICKAS, A, SMITH, B.W. and

YARBROUGH, I.A. (1973). Structure and genetic resources of peanuts. In
Peanuts: Culture and Uses. American Peanut Research and Educational
Association, Stilwater, Florida, USA.
 174

GREGORY, W.C., SMITH, B.W. and YARBROUGH, I.A. (1951). Morphology,

genetics and breeding. In The Peanut- the Unpredictable Legume, pp. 28-88.
National Fertiliser Association, Washington DC, USA.

GROSS, Y. and KIGEL, J. (1994). Differential sensitivity to high temperature of

stages in the reproductive development of common bean (Phaseolus vulgaris
L.). Field Crops Research 36: 201-212.

HADLEY, P., RICHARDSON, A.C., DICKINSON, D., CIDVERS, K., GILL, S

and SUMMERFIELD, RI: (1982). Research in controlled environments: a
versatile, inexpensive system for automatic irrigation of plants grown in pots.
Laboratory Practice 31: 728-732.

HALL, AE. (1992). Breeding for heat tolerance. Plant Breeding Reviews 10: 129-
167.

HALTERLEIN, Al, CLAYBERG, C.D. and TEARE, lD. (1980). Influence of high
temperature on pollen grain viability and pollen tube growth in the styles of
Phaseolus vulgaris L. Journal of American Society of Horticultural Science
105: 12-14.

HAMMER G.L., SINCLAIR, T.R., BOOTE, K.l, WRIGHT, G.C. MEINKE, H.

and BELL, M.l (1995). A peanut simulation model. 1. Model testing and
development. AgronomyJournal87: 1085-1093

HAMMONS, RO. (1994). The origin and history of the groundnut. In The
Groundnut Crop: A SCientific Basis for Improvement, pp. 24-42. (Ed. J.
Smartt). Chapman and Hall, London, UK.

HASSAN, M.A and SRIVASTAVA, D.P. (1966). Floral biology and pod
development of peanut studied in India. Journal of Indian Botanical Society
45: 92-102.

HERRIDGE, D.F. and ROUGHLEY, RI. (1976). Influence of temperature and

Rhizobium strain on nodulation and growth of two tropical legumes. Tropical
Grassland 10: 21-23.

HOUGHTON, 1.T., JENKINS, G.J. and EPHRAUMS, 1.1. (1990). Climate Change:
The IPPC Scientific Assessment. Intergovernmental Panel on Climate
Change. Cambridge University Press, Cambridge, UK.

HUNDAL, S.S. and KAUR, P. (1996). Climate change and its impact on crop
productivity in Punjab, India. In Climate Variability and Agriculture, pp.
377-393. (Eds Y.P. Abrol, S. Gadgil and G.B. Pant). Narosa Publishing
House, New Delhi, India.
 175

ICRISAT. (1992). Workshop recommendations. In Groundnut - A Global

Perspective: Proceedings 0/ an International Workshop, 25-29 Nov 1991,
ICRlSAT, pp. 375-378. (Ed. S.N. Nigam). International Crops Research
Institute for the Semi-Arid Tropics, Patancheru, India.

ICRISAT. (1994).ICRlSAT West Africa Programs Annual Report, 1993, pp. 36-37.
International Crops Research Institute for the Semi-Arid Tropics, Sahelian
Centre, Niamey, Niger.

INOUYE, C. and MAEDA, K. (1954). Studies on root nodules of peanuts. Research

Report of the Kochi University, Japan 3: 1-4.

INTORZATO, R. and TELLA, R. (1960). Sistema radicular do amendoim.

Bragantia 19: 119-123.

IWAHORI, S. (1965). High temperature injuries in tomato. IV. Development of

normal flower buds and morphological abnormalities of flower buds treated
with high temperature. Journal of Japanese Society of Horticultural Science
34: 33-41.

IWAHORI, S. (1966). High temperature injuries in tomato. V. Fertilization and

development of embryo with special reference to the abnormalities caused by
high temperature. Journal of Japanese Society of Horticultural Science 35:
379-388.

KEARNS, C.A. and INOUYE, D.W. (1993). Techniques/or Pollination Biologists.

Colorado University Press, Colorado, USA.

KETRING, D.L. (1984a). Temperature effects on vegetative and reproductive

development of peanut. Crop Science 24: 877-882.

KETRING, D.L. (1984b). Root diversity among peanut genotypes. Crop Science 24:
229-232.

KETRING, D.L. (1986). Physiological response of groundnuts to temperature and

water deficits - breeding implications. In Agrometeorology of Groundnut, pp.
135-148. (Eds M.V.K. Sivakumar and S.M. Virmani). International Crops
Research Institute for the Semi-Arid Tropics, Patancheru, India.

KETRING, D.L. and REID, J.L. (1995). Peanut growth and development. In Peanut
Health Management, pp. 11-18. (Eds H.A. Melouk and F.M. Shokes). The
American Phytopathological Society Press, Minnesota, USA.

KETRING, D.L., BROWN, R.H., SULLIVAN, G.A. and JOHNSON, B.B. (1982a)
Growth physiology. In Peanut Science and Technology, pp. 411-457. (Eds
H.E. Pattee and C.T. Young). American Peanut Research and Education
Association, Stilwater, OK, USA.
 176

KETRING, D.L., JORDAN, W.R. SMITH, O.D. and SIMPSON, C.E. (1982b)
Genetic variability in root and shoot growth characteristics of peanut. Peanut
Science 9: 68-72.

KIGEL, J., KONSENS, I. and OFIR, M. (1991). Branching, flowering and pod-set
patterns in snap-bean (Phaseolus vulgaris L.) as affected by temperature.
Canadian Journal of Plant Science 71: 1233-1242.

KISmNEVSKY, B.D. SEN, D and WEAVER, R.W. (1992). Effect of high root
temperature on Bradyrhizobium - peanut symbiosis. Plant and Soil 143: 275-
282.

KONSENS, I., OFIR, M. and KIGEL, J. (1991). The effect of temperature on the
production and abscission of flowers and pods in snap bean (Phaseolus
Vulgaris L.). Annals of Botany 67: 391-399.

KRAPOVICKAS, A. (1969). The origin, variability and spread of the groundnut

(Arachis hypogaea L.) (translated by 1. Smartt). In The Domestication and
Exploitation of Plants and Animals, pp. 427-441. (Eds P.J. Ucko and I.S.
Fallk). Duckworth, London, UK.

KRAPOVICKAS, A. and GREGORY, W.C. (1994). Taxonomy of the genus

Arachis (Leguminosae). Bonplandia 8: 1-186.

KRAPOVICKAS, A. and RIGONI, V.A. (1960). La nomen datura de las

subespecies y variadades de Arachis hypogaea L Revista de Investigaciones
Agricolas 14: 197-228.

KUO, C.G., CHEN, H.M. and MA, L.H. (1986). Effect of high temperature on
proline content in tomato floral buds and leaves. Journal of American Society
of Horticultural Science 111: 746-750.

KVIEN, C. (1995). Physiological and environmental disorders of peanut. In Peanut

Health Management, pp. 33-42. (Eds H.A. Melouk and F.M. Shokes). The
American Phytopathological Society, Minnesota, USA.

LABANA, K.S., SINGH, M., SANGHA, A.S. and JASWAL, S.V. (1980).
Variability and inter-relations among characters in F2 progeny of groundnut.
Journal of Research, Punjab Agricultural University 17: 107-114.

LEONG, S.K. and ONG, C.K. (1983). The influence of temperature and soil water
deficit on the development and morphology of groundnuts (Arachis hypogaea
L.). Journal of Experimental Botany 34: 1551-1561.

LEVITT, J. (1972). Responses of Plants to Environmental Stresses. Academic Press,

New York, USA.
 177

LEVITT, I (1980). Stress terminology in adaptation of plants to water and high

temperature stress: summary and synthesis. In Adaptation of Plants to Water
and High Temperature Stress, pp. 437-439. (Eds N.C. Turner and RI.
Kramer). Wiley Interscience, New York, USA.

LEVY, A., RABINOWITCH, H.D. and KEDAR, N. (1978). Morphological and

physiological characters affecting flower drop and fruit-set of tomatoes at
high temperatures. Euphytica 27: 211-218.

LI, P.H., DAVIS, D.W. and SHEN, Z.Y. (1991). High-temperature-acclimation

potential of the common bean: can it is used as a selection criterion for
improving crop performance in high-temperature environments? Field Crops
Research 27: 241-256.

LIE, T.A. (1971). Symbiotic nitrogen fixation under stress conditions. Plant and Soil
Special Vol.: 117-127.

LIE, T.A. (1974). Environmental effects on nodulation and symbiotic nitrogen

fixation. In The Biology of Nitrogen Fixation, pp. 555-582. (Ed. A. Quispel).
North Holland Publishing Co., Amsterdam, Netherlands.

LIM, E.S. and GUMPIL, IS. (1984). The flowering, pollination and hybridization of
groundnuts (Arachis hypogaea L.). Pertanika 7: 61-66.

MACKILL, 0.1. and COFFMAN, W.R (1983). Inheritance of high temperature

tolerance and pollen shedding in a rice cross. ZeitschriJt Pf/anzensuchtg 91:
61-69.

MACKILL, 0.1, COFFMAN, W.R and RUTGER, IN. (1982). Pollen shedding and
combining ability for high temperature tolerance in rice. Crop Science 22:
730-733.

MAEDA, K. (1970). Growth analysis on the plant type in peanut varieties, Arachis
hypogaea L. I. Varietal difference and seasonal change of leaf number and
area of peanut individuals differing in plant type grown under various field
conditions. Proceedings of the Crop Science Society of Japan 39: 177-183.

MAEDA, K. (1972). Growth analysis on the plant type in peanut varieties, Arachis
hypogaea L. IV. Relationship between the varietal difference of the progress
of leaf emergence on the main stem during pre-flowering period and the
degree of morphological differentiation of leaf primordia in the embryo.
Proceedings of the Crop Science Society of Japan 41: 179-186.

MAEDA, K. (1973). Growth analysis on the plant type in peanut varieties, Arachis
hypogaea L. V. Air temperature and some internal factors affecting the
seasonal change of the growth rate of main stem and its difference between
intraspecific two varietal groups. Proceedings of the Crop Science Society of
Japan 42: 46-53.
 178

MARFO, K.O. and HALL, AE. (1992). Inheritance of heat tolerance during pod set
in cowpea. Crop Science 32: 912-918.

MARSHALL, H.G. (1982). Breeding for tolerance to heat and cold. In Breeding
Plants for Less Favourable Environments, pp. 47-70 (Eds M.N. Christiansen
and C.F. Lewis). John Wiley and Sons, New York, USA

MARSHALL, K.C. (1964). Survival of root-nodule bacteria in dry soils exposed to

high temperatures. Austral~an Journal of Agricultural Research 15: 273-
281.

MARTIN, IP., CAS, S. and RABECHAULT, H. (1974). The culture in vitro of

peanut stamens (Arachis hypogaea L.). I. Stages of development of the
flower buds and microsporogenesis. Oleagineux 29: 145-149.

MATTHEWS, RB., HORIE, T., KROPFF, M.I. BACHELET, D., CENTENO,

H.G., SHIN, IC., MOHANDASS, S., SINGH, S., DEFENG, Z. and LEE,
M.H. (1995). A regional evaluation of the effects of future climate change on
rice production in Asia. In Modelling the Impact of Climate Change on Rice
Production in Asia, pp. 95-139. (Eds RB. Matthews, M.I Kropff, D.
Bachelet and H.H. van Laar). CAB International, Wallingford, Oxon, UK.

McWilLIAM, J.R. (1980). Summary and synthesis - adaptation to high temperature

stress. In Adaptation of Plants to High Temperature Stress, pp. 444-446.
(Eds N.C. Turner and P.J. Kramer). John Wiley and Sons, New York, USA

MEAD, R, CURNOW, RN. and HASTED, AM. (1993). Statistical Methods in

Agriculture and Experimental Biology (Second Edition). Chapman and Hall,
London, UK.

MllLS, W.T. (1964). Heat unit system for predicting optimum peanut-harvesting
time. Transactions of American Society of Agricultural Engineering 7: 307-
309

MINCHIN, F.R, SUMMERFIELD, R.I. and NEVES, M.C.P. (1980). Carbon

metabolism, nitrogen assimilation, and seed yield of cowpea (Vigna
unguiculata L. Walp.) grown in adverse temperature regimes. Journal of
Experimental Botany 31: 1327-1345.

MINCHIN, F.R., SUMMERFIELD, RJ., HADLEY, P., ROBERTS, E.H. and

RAWSTHORNE, S. (1981). Carbon and nitrogen nutrition of nodulated
roots of grain legumes. Plant Cell and Environment 4: 5-26.

MOHAMED, H.A, CLARK, lA and ONG, C.K. (1988a). Genotypic differences in

the temperature responses of tropical crops I. Germination characteristics of
groundnuts (Arachis hypogaea L.) and pearl millet (Pennisetum typhoides
S.& H.). Journal of Experimental Botany 39: 1121-1128.
 179

MOHAMED, H.A, CLARK, lA and ONG, C.K. (1988b). Genotypic differences in

the temperature responses of tropical crops TI. Seedling emergence and leaf
growth of groundnuts (Arachis hypogaea L.) and pearl millet (Pennisetum
typhoides S.& R.). Journal of Experimental Botany 39: 1129-1135.

MONTERROSO, V.A, and WIEN, H.C. (1990). Flower and pod abscission due to
heat stress in beans. Journal of American Society of Horticultural Science
115: 631-634.

MOSS, IP. and RAMANATHA'RAO, V. (1995). The peanut - reproductive

development to plant maturity. In Advances in Peanut Science, pp. 1-13.
(Eds H.E. Pattee and H.T. Stalker). American Peanut Research and
Education Society, Stilwater, OK, USA

MUNEVAR, F. and WOLLUM, AG. (1981). Effect of high root temperature and
Rhizobium strain on nodulation, nitrogen fixation, and growth of soyabeans.
Soil Science Society America Journal45: 1113-1120.

MUNNS, D.N. FOGLE, V.W. and HALLOCK, B.G. (1977). Alfalfa root nodule
distributionand inhibitionof nitrogen fixation by heat. Agronomy Journal69:
377-380.

MUTTERS, R.G., FERREIRA, L.G.R and HALL, AE. (1989a). Proline content of
the anthers and pollen of heat-tolerant and heat-sensitive cowpea subjected to
differenttemperatures. Crop Science 29: 1497-1500.

MUTTERS, R.G., HALL, AE. and PATEL, P.N. (1989b). Photoperiod and light
quality effects on cowpea floral development at high temperatures. Crop
Science 29: 1501-1505.

NAGESWARA RAO, RC., WILLIAMS, IH. and MURARI SINGH. (1989).

Genotypic sensitivity to drought and yield potential of peanut. Agronomy
JournaI81:887-893.

NAMBIAR, P.T.C. (1988). Nodulation and nitrogen fixation. In Groundnut, pp.

542-554. (Ed. P.S. Reddy). Indian Council of Agricultural Research, New
Delhi, India.

NAMBIAR, P.T.C. and DART, P.J. (1983). Factors influencing nitrogenase activity
(acetylene activity) by root nodules of groundnut, Arachis hypogaea L.
Peanut Science 10: 26-29.

NEccm, A., PONGA, N.E. and MAPELLI, S. (1987). Early and late heat shock
proteins in wheats and other cereal species. Plant Physiology 84: 1378-1384.

NIGAM, S.N., NAGESWARA RAO, RC., WYNNE, J.C., WILLIAMS, r.n.,

FITZNER, M. and NAGABHUSHANAM, G.V.S. (1994). Effect and
interaction of temperature and photoperiod on growth and partitioning in
 180

three groundnut (Arachis hypogaea L.) genotypes. Annals of Applied

Biology 125: 541-552.

NUTMAN, P.S. (1975). Rhizobium in the soil. In Soil Microbiology, pp. 111-131.
(Ed. N. Walker). John Wiley and Sons, New York, USA.

OFIR, M., GROSS, Y., BANGERTH, F. and KIGEL, I. (1993). High temperature
effects on pod and seed production as related to hormone levels and
abscission of reproductive structures in conunon bean (Phaseo/us vulgaris
L.). Scientia Horticulturae 55: 201-211.

OMANGA, P.A, SUMMERFIELD, RI and QI, A. (1996). Flowering in pigeonpea

(Cajanus cajan) in Kenya: responses of medium-and late-maturing genotypes
to location and date of sowing. Experimental Agriculture 32: 111-128.

ONG, C.K. (1984). The influence of temperature and water deficit on the partitioning
of dry matter in groundnut (Arachis hypogaea L.). Journal of Experimental
Botany 35: 746-755.

ONG, C.K. (1986). Agroclimatological factors affecting phenology of groundnut. In

Agrometeorology of Groundnut, pp. 115-125. (Eds M.V.K. Sivakumar and
S.M. Virmani). International Crops Research Institute for the Semi-Arid
Tropics, Patancheru, India.

ONO, Y. (1979). Flowering and fiuiting of peanut plants. Japanese Agriculture

Research Quarterly 13: 226-229.

ONO, Y., NAKAYAMA, K. and KUBOTA, M. (1974). Effect of soil temperature

and soil moisture in podding zone on pod development of peanut plants.
Proceedings of the Crop Science Society of Japan 43: 247-251.

ORMROD, D.P., WOOLLEY, C.J., EATON, G.W. and STOBBE, E.H. (1967).
Effect of temperature on embryo sac development in Phaseolus vulgaris L.
Canadian Journal of Botany 45: 948-950.

OUGHAM, H.J. and STODDART, J.L. (1986). Synthesis of heat-shock protein and
acquisition of thermo tolerance in high-temperature tolerant and high-
temperature susceptible lines of Sorghum. Plant Science 44: 163-167.

PALMER, RG., ALBERTSEN, M.C. and HEER, H. (1978). Pollen production in

soybeans with respect to genotype, environment, and stamen position.
Euphytica 27: 427-433.

PARKER, C.A, TRINICK, M.I and CHATEL, D.L. (1977). Rhizobia as soil and
rhizosphere inhabitants. In A Treatise on Dinitrogen Fixation. IV. Agronomy
and Ecology, pp. 311-352. (Eds RW.F. Hardy and AH. Gibson). John Wiley
and Sons, New York, USA.
 181

PARRY, M., PORTER, lR. and CARTER, T.R (1990). Climatic changes and its
implications for agriculture. Outlook on Agriculture 19: 9-15.

PATE, I.S. (1961). Temperature characteristics of bacterial variation in legume

symbiosis. Nature (London) 192: 637-639.

PATEL, P.N. and HALL, A.E. (1990). Genotypic variation and classification of
cowpea for reproductive responses to high temperature under long
photoperiods. Crop SCience.30: 614-621.

PAULSEN, G.M. (1994). High temperature responses of crop plants. In Physiology

and Determination of Crop Yield, pp. 365-389. (Eds K.I. Boote, I.M.
Bennett, T.R Sinclair, and G.M. Paulsen). American Society of Agronomy,
Wisconsin, USA.

PERIASAMY, K. and SAMPOORNAM, C. (1984). The morphology and anatomy

of ovule and fruit development in Arachis hypogaea L. Annals of Botany 53:
399-411.

PIHA, M.I. and MUNNS, D.N. (1987). Sensitivity of the common bean (Phaseolus
vulgaris L.) symbiosis to high soil temperature. Plant and Soil 98: 183-194.

POTDAR, M.V. and ANDERS, M.M. (1995). On-farm performance of groundnut

genotypes under different land configurations and foliar iron sprays for the
correction of iron chlorosis on calcareous soils in India. In Iron Nutrition in
Soils and Plants, pp. 111·118. (Ed. 1. Abadia). Kluwer Academic Publishers,
Netherlands.

RAMANATHA RAO, V. (1988). Botany. In Groundnut, pp. 24-64. (Ed. P.S.

Reddy). Indian Council of Agricultural Research, New Delhi, India.

RANGA RAO, V. (1977). Effect of root temperature on the infection process and
nodulation in Lotus and Stylosanthes. Journal of Experimental Botany 28:
241-259.

RAWS THORNE, S., HADLEY, P., SUMMERFIELD, RI. and ROBERTS, E.H.
(1985a). Effects of supplemental nitrate and thermal regime on the nitrogen
nutrition of chickpea (Cicer arietinum L.). I. Growth and development. Plant
and Soil 83: 265-277.

RAWSTHORNE, S., HADLEY, P., SUMMERFIELD, R.I. and ROBERTS, E.H.

(1985b). Effects of supplemental nitrate and thermal regime on the nitrogen
nutrition of chickpea (Cicer arietinum L.). II. Symbiotic development and
nitrogen assimilation. Plant and Soil83: 279-293.

REDDY, V.M., TANNER, I.W., ROY, R.C. and ELLIOT, I.M. (1981). The effects
of irrigation, inoculants and fertiliser nitrogen on peanuts (Arachis hypogaea
L.). ll. Yield. Peanut Science 8: 125-128.
 182

RICK, C.M. and DEMPSEY, W.H. (1969). Position of the stigma in relation to fruit
setting of the tomato. Botanical Gazette 130: 180-186.

RISTIC, Z., GIFFORD, D.l and CASS, D.D. (1991). Heat shock proteins in two
lines of Zea mays L. that differ in drought and heat resistance. Plant
Physiology 97: 1430-1434.

RODRIGUEZ-GARAY, B. and BARROW, lR (1988). Pollen selection for heat

tolerance in cotton. Crop SC_ience28: 857-859.

RUDD, V.E. (1981). Aescynomeneae. In Advances in Legume Systematics, Part l,

pp. 347-354. (Eds R.M. Polhill and P.H. Raven). Royal Botanical Gardens,
Kew, London, UK.

RUDICH, I, ZAMSKI, E. and REGEV, Y. (1977). Genotypic variation for

sensitivity to high temperature in the tomato: pollination and fruit set.
Botanical Gazette 138: 448-452.

SANDERS, T.H., BLANKENSHIP, P.D., COLE, RI. and lllLL, RA. (1985).
Temperature relationships of peanut leafcanopy, stem and fruit in soil varying
temperature and moisture. Peanut Science 12: 86-89.

SANDERS, T.H., BLANKENSHIP, P.D., COLE, RI. and SMITH, 1.S. (1986).
Role of agrometeorological factors in postharvest quality of groundnuts. In
Agrometeorology of Groundnut, pp. 185-192. (Eds M.V.K. Sivakumar and
S.M. Virmani). International Crops Research Institute for the Semi-Arid
Tropics, Patancheru, India.

SANKARAREDDI, G.H. (1988). Cultivation, storage and marketing. In Groundnut,

pp. 318-383. (Ed. P.S. Reddy). Indian Council of Agricultural Research, New
Delhi, India.

SCHENK, RU. (1961). Development of the peanut fruit. Georgia Agricultural

Experiment Station, Technical Bulletin N.S. 22, University of Georgia,
Athens, USA.

SCHIFFMANN, J. and LOBEL, R (1973). Seasonal changes in symbiotic nitrogen

fixation and haemoglobin content in nodules of peanuts. Plant and Soil 39:
329-340.

SELAMAT, A and GARDNER, F.P. (1985). Growth, nitrogen uptake, and

partitioning in nitrogen-fertilized nodulating and nonnodulating peanut.
Agronomy Journal 77: 862-867.

SESHADRI, C.R. (1962). Groundnut. Indian Journal of Plant Physiology 11: 158-
163.

SHONNARD, G.C. and GEPTS, P. (1994). Genetics of heat tolerance during

reproductive development in common bean. Crop Science 34: 1168-1175.
 183

SINCLAIR, T.R., BENNETT, 1.M. and BOOTE, K.I. (1993). Leafnitrogen content,
photosynthesis and radiation use efficiency in peanut. Peanut Science 20: 40-
43.

SINGH, B. (1992). Aspects of groundnut utilization and possible improvement of

indigenous foods in some countries of semi-arid tropical Africa. In Groundnut
- A Global Perspective, Proceedings of an International Workshop, 25-29
November 1991, ICRISAT, pp. 257-263. (Ed. S.N. Nigam). International
Crops Research Institute for the Semi-Arid Tropics, Patancheru, India.

SIVAKUMAR, M.V.K. and SARMA, P.S. (1986). Studies on water relations of

groundnuts. In Agrometeorology of Groundnuts, pp. 83-98. (Eds M.V.K.
Sivakumar and S.M. Vinnani). International Crops Research Institute for the
Semi-Arid Tropics, Patancheru, India.

SIVAKUMAR, M.V.K., MAIDOUKIA, A and STERN, R.D. (1993).

Agroclimatology of West Africa: Niger (2nd Edition). Information Bulletin
No.5. International Crops Research Institute for the Semi-Arid Tropics,
Patancheru, India

SMARTT, 1. (1961). Groundnut varieties of northern Rhodesia and their

classification. Empire Journal of Experimental Agriculture 29: 153-158.

SMARTT, J. (1990). Grain Legumes: Evolution and Genetic Resources. Cambridge

University Press, Cambridge, UK.

SMITH, B.W. (1950). Arachis hypogaea L. Aerial flower and subterranean fruit.
American Journal of Botany 37: 802-815.

SMITH, B.W. (1954). Arachis hypogaea. Reproductive efficiency. American

Journal of Botany 41: 607-616.

SMITH, B.W. (1956a). Arachis hypogaea. Normal megasporogenesis and syngamy

with occasional single fertilization. American Journal of Botany 43: 81-89.

SMITH, B.W. (1956b). Arachis hypogaea. Embryogeny and the effect of peg
elongation upon embryo and endosperm growth. American Journal of Botany
43: 233-240.

SPRENT, 1.1. and SPRENT, P. (1990). Nitrogen Fixing Organisms: Pure and
Applied Aspects. Chapman and Hall, London, UK.

SRINIVASAN, A., TAKEDA, H. and SENBOKU, T. (1996). Heat tolerance in food

legumes as evaluated by cell membrane thermo stability and chlorophyll
fluorescence techniques. Euphytica 88: 35-45.

STALKER, H.T. (1997). Peanut (Arachis hypogaea L.). Field Crops Research 53:
205-217.
 184

STALKER, H.T. and SIMPSON, C.E. (1995). Germplasm resources in Arachis. In

Advances in Peanut SCience, pp. 14-53. (Eds H.E. Pattee and H.T. Stalker).
American Peanut Research and Education Society, Stilwater, OK, USA

STAPHORST, I.L. and STRIJDOM, B.W. (1972). Some observations on the

bacteriods in nodules of Arachis sp. and isolation of Rhizobia from these
nodules. Phytophylactica 4: 87-92.

STIRLING, C.M. and BLACK, C.R (1990). Stages of reproductive development in

groundnut (Arachis hypogaea L.) most susceptible to environmental stress.
Tropical Agriculture 68: 296-300.

SUGIYAMA, T., IWAHORI, S. and TAKAHASm, T. (1966). Effect of high

temperature on fruit setting of tomato under cover. Acta Horticulturae 4: 63-
69.

SUMMERFIELD RI. and ROBERTS, E.H. (1985). Arachis hypogaea L. In

Handbook of Flowering. Vol. 1, pp. 74-82. (Ed. AH. Halevy). CRC Press,
Florida, USA

SUMMERFIELD, RI. and MUEHLBAUER, F.J. (1982). Controlled environments

as an adjunct to field research on lentils (Lens culinaris). II. Research
strategy. Experimental Agriculture 18: 3-15

SUMMERFIELD, RI., HADLEY, P., ROBERTS, E.H., MINCHIN, F.R and

RAWSTHORNE, S. (1984). Sensitivity of chickpeas (Cicer arietinum) to hot
temperatures during the reproductive period. Experimental Agriculture 20:
77-93.

SUMMERFIELD, RI., MINCIllN, F.R, ROBERTS, E.H. and HADLEY, P.

(1983). Cowpea. In Proceedings of Symposium on Potential Productivity of
Field Crops Under Different Environments, pp. 249-280. (Ed. S. Yoshida).
International Rice Research Institute, Manila, Philippines. .

SUMMERFIELD, RI., MINCHIN, F.R, STEWART, K.A and NDUNGURU, B.1.

(1978). Growth, reproductive development and yield of effectively nodulated
cowpea plants in contrasting aerial environments. Annals of Applied Biology
90: 277-291.

SUMMERFIELD, RI., MUEm..BAUER, F.J. and SHORT, RW. (1989).

Controlled environments as an adjunct to field research on lentils (Lens
culinaris). V. Cultivar responses to above- and below-average temperatures
during the reproductive period. Experimental Agriculture 25: 327-341.

SUZUKI, M. (1966). Studies on thermoperiodicity of crops. II. The effect of soil

temperature to fiuctification on peanuts. Chiba University Technical Bulletin
13: 95-101.
 185

TALWAR, H.S. (1997). Physiological basis for heat tolerance during flowering and
pod setting stages in groundnut (Arachis hypogaea L.). JIRCAS Visiting
Fellowship Report 1996-97. JIRCAS, Okinawa, Iapan.

TOOMSAN, B. (1990). Groundnut microbiology research at Khon Kaen University.

In Groundnut Improvement Project at Khon Kaen University. Report of
Workfor 1986-1990, pp. 89-111. (Ed. A Patanothai). Khon Kaen University,
Thailand.

TRIVEDI, B.S. and VERMA, H. (1975). A study of pollen production and

stainability in Cassia tora L. and Arachis hypogaea L. Journal of Palynology
11: 139-142.

VARA PRASAD, P.V., CRAUFURD, P.Q., SUMMERFIELD, RI. and

WHEELER, T.R. (1998). Sensitivity of fruit-set to heat stress in groundnut.
Journal of Experimental Botany 40: 30.

VIRMANI, S.M. and SINGH, P. (1986). Agroclimatological characteristics of the

groundnut-growing regions of the semi-arid tropics. In Agrometeorology of
Groundnut, pp. 21-26. (Eds M.V.K. Sivakumar and S.M. Virmani).
International Crops Research Institute for the Semi-Arid Tropics, Patancheru,
India.

WARRAG, M.O.A and HALL, AE. (1984). Reproductive responses of cowpea

[Vigna unguiculata L. (Walp.)] to heat stress. II. Response to night air
temperature. Field Crops Research 8: 17-33.

WARRICK, RA (1990). Carbon dioxide, climate change and agriculture.

GeographyJournallS4: 221-233.

WHEELER, T.R., CHATZIALIOGLOU, A, CRAUFURD, P.Q., ELLIS, RH. and

SUMMERFIELD, RI. (1997). Dry matter partitioning in groundnut exposed
to high temperature stress. Crop Science 37: 1507-1513.

WILLCOX, M.C., REED, S.A, BURNS, lA and WYNNE, IC. (1990).

Microsporogenesis in peanut (Arachis hypogaea L.). American Journal of
Botany 77: 1257-1259.

WILLIAMS, I.H. and BOOTE, K.I. (1995). Physiology and modelling - predicting
the "unpredictable legume". In Advances in Peanut Science, pp. 301-353.
(Eds H.E. Pattee and H.T. Stalker). American Peanut Research and
Education Society, Stilwater, OK, USA

WILLIAMS, I.H., WILSON, I.H. and BATE, G.C. (1975a). The growth of
groundnuts (Arachis hypogaea L. cv. Makulu Red) at three altitudes in
Rhodesia. Rhodesian Journal of Agricultural Research 13: 33-43.
 186

WILLIAMS, I.H. WILSON, lH. and BATE, G.C. (1975b). The growth and
development of four groundnut (Arachis hypogaea L.) cultivars in Rhodesia.
Rhodesian Journal of Agricultural Research 13: 131-144.

WILLIAMS, lH., Hll..DERBRAND, G.L. and TATTERSFIELD, I.R (1978). The

effect of weather and genotype x environment interaction on the yields of
groundnut (Arachis hypogaea L.). Rhodesian Journal of Agricultural
Research 16: 193-204.

WILLIAMS, lH., NAGESWARA'RAO, RC., MATTHEWS, R and HARRIS, D.

(1986). Responses of groundnut genotypes to drought. In Agrometeorology
of Groundnut, pp. 99-106. (Eds M. V.K. Sivakumar and S.M. Virmani).
International Crops Research Institute for the Semi-Arid Tropics, Patancheru,
India.

WOOD, lM.W. (1968). The effect of temperature at early flowering on the growth
and development of peanuts (Arachis hypogaea). Australian Journal of
Agricultural Research 19: 241-251.

WRIGHT, G.C. and NAGESWARA RAO, R.C. (1994). Groundnut water relations.
In The Groundnut Crop- A Scientific Basis for Improvement, pp. 281-335.
(Ed. I. Smartt). Chapman and Hall, London.

WRIGHT, G.C., BELL, MJ. and HAMMER, G.L. (1993). Leaf nitrogen content
and minimum temperature interactions affect radiation-use-efficiency in
peanut. Crop Science 33: 476-481.

XI, X.Y. (1991). Development and structure of pollen and embryo sac in peanut
(Arachis hypogaea L.). Botanical Gazette 152: 164-172.

YOSHIDA, S., SATAKE,1. and MACKILL, DJ. (1981). High-temperature stress in

rice. IRRI Research Paper Series 67. International Rice Research Institute,
Manila, Philippines.

YOUNG, I.H., COx, F.R and MARTIN, C.K. (1979). A peanut growth and
development model. Peanut Science 6: 27-36.

ZHANG, H.Q. and CROES, A.F. (1983a). Protection of pollen germination from
adverse temperatures: a possible role for proline. Plant Cell and Environment
6: 471-476.

ZHANG, H.Q. and CROES, A.F. (1983b). Proline metabolism in pollen:

degeneration of proline during germination and early tube growth. Planta
159: 46-69.

View publication stats