Taro

TARO (COLOCASIA ESCULENTA L.
SCHOTT) YIELD AND QUALITY RESPONSE

TO PLANTING DATE AND ORGANIC FERTILISATION
RORISANG MARE
MSc. Crop Science (University of KwaZulu-Natal)
Submitted in fulfillment of the requirements for the degree of

Doctor of Philosophy (Crop Science)
in the
School of Agricultural Sciences and Agribusiness
University of KwaZulu-Natal
Pietermaritzburg
DECEMBER 2009
DECLARATION
The experimental work presented in this thesis was carried out at the University of
KwaZulu-Natal, Pietermaritzburg, from February 2007 to December 2009 under the
supervision of Professor Albert T. Modi.
These studies are my original work except where acknowledged and have not been
submitted in any form for any degree at any other university.
………………………………
Rorisang Mare
December 2009
I declare that the above statement is correct.
………………………………..
Professor Albert T. Modi
December 2009
i
ACKNOWLEDGEMENTS
I would like to thank Almighty God for giving me strength and perseverance to complete
this study.
I am grateful to my family and friends for their love, encouragement and support.
I express my gratitude to my supervisor, Professor Albert T. Modi for his determined

supervision, guidance and encouragement.
I extend my sincere appreciation to the technical staff for their assistance on the
experimental farm.
I would also like to thank Mr. Shakes Mabhida for allowing me to use his land for the
Umbumbulu trial.
This study was funded by SANPAD (South Africa-Netherlands Research Programme on

Alternatives in Development): Project Number 05/32.
ii
DEDICATION
I would like to dedicate this thesis to my sons, Mpho and Tšepo for understanding
when I couldn`t be there when they needed me most.
iii
ABSTRACT
Despite the importance of taro (Colocasia esculenta L. Schott) as a food security crop,
scientific research on it is scanty in South Africa. Production site, planting date and
fertiliser regime affect crop performance and quality, particularly that of cultivars,
because they tend to be adapted to specific localities. Storage temperature and packaging
method on the other hand affect the shelf-life. To investigate performance and quality of
three taro cultivars in response to planting date and fertilisation, a study was carried out at
two sites in KwaZulu-Natal, South Africa (Ukulinga and Umbumbulu), during the
2007/2008 growing seasons. The effect of two storage temperatures (12oC and ambient
temperature) and three packaging methods (polyethylene bags, mesh bags and open
boxes) on cormel quality following storage was also investigated for three cultivars.
Delayed planting negatively affected the number of cormels plant-1 and fresh cormel
mass plant-1. Fertilisation and cultivar affected the number of cormels plant-1 and fresh
cormel mass plant-1 only when planting was done in October and November at both sites.
Fertilisation increased the number of cormels plant-1 for all cultivars except Dumbe-
dumbe. Dumbe-dumbe had the lowest number of cormels plant-1 but the highest number
of marketable cormels plant-1. Dumbe-dumbe showed the lowest fresh cormel mass
plant-1 in October and the highest in November at Ukulinga. Fertisation increased fresh
cormel mass plant-1 in October at Umbumbulu. Dry matter content was negatively
affected by fertilisation at Ukulinga. The response of dry matter content, specific gravity,
protein, minerals, reducing sugars and starch content was variable depending on cultivar.
Delayed planting negatively affected starch content for Dumbe-dumbe and Pitshi at
Ukulinga. Fertilisation decreased starch content of Pitshi, while delayed planting
increased sugar content for Dumbe-dumbe and decreased it for Mgingqeni and Pitshi at
Umbumbulu. Dumbe-dumbe had higher starch content and higher reducing sugars.
Considering all growth and quality parameters, it is recommended that Dumbe-dumbe is
the best taro cultivar for crisping and the best time to plant it is October with 160 kg N
ha-1 of organic fertiliser and November with 320 kg N ha-1 at Ukulinga whereas at
Umbumbulu the best time to plant Dumbe-dumbe is October with 320 kg N ha-1 of the
fertiliser. Starch granules degradation, alpha-amylase activity and sprouting increased
iv
with storage time and storage temperature. Cormels of Mgingqeni stored in polyethylene
bags showed highest alpha-amylase activity and sprouting. Reducing sugar content
increased and starch content decreased with time in storage and decline in storage
temperature. It is recommended that taro cormels be stored in mesh bags at 12oC.
The chapters of this thesis represent different studies presented as different papers.
Chapter 1 is a general introduction to explain the study background and hypothesis.
Chapter 2 is a general review of literature. Chapter 3 is on growth, development and yield
of taro in response to planting date and fertilisation. Chapter 4 is on the influence of
planting date and organic fertiliser on crisping quality of taro cormels. Chapter 5 is on
changes in the surface morphology of starch granules and alpha-amylase activity of taro
during storage. Chapter 6 is on the effects of pre- and post-harvest practices on starch and
reducing sugars of taro. The last chapter is a general discussion and conclusions.
v
TABLE OF CONTENTS
DECLARATION i
ACKNOWLEDGEMENTS ii
DEDICATION iii
ABSTRACT iv
TABLE OF CONTENTS vi
LIST OF FIGURES xii
LIST OF TABLES xviii
LIST OF APPENDICES xx
1. GENERAL INTRODUCTION 1
References 4
2. LITERATURE REVIEW 6
2.1 Quality of taro cormels for crisp making 6

2.1.1 Yield 6
2.1.2 Chemical composition of taro cormels 7
2.1.2.1 Specific gravity 7
2.1.2.2 Dry matter content 7
2.1.2.3 Starch content 8
2.1.2.4 Alpha-amylase activity 8
2.1.2.5 Reducing sugar content 9
2.1.2.6 Protein and mineral content 9
2.2 Taro growth stages 10
2.2.1 Establishment 10
2.2.2 Vegetative growth and corm initiation 10
2.2.3 Corm bulking and maturation 11
2.3 Factors influencing taro quality 12
2.3.1 Genetic factors 12
vi
2.3.2 Planting date 13
2.3.2.1 Temperature 13
2.3.2.2 Rainfall 15
2.3.3 Fertilisation 16
2.4 Shelf life 19
2.4.1 Factors affecting shelf life of taro 19
2.4.1.1 Storage temperature 20
2.4.1.2 Packaging 21
2.5 Conclusion 22
References 23
3 GROWTH, DEVELOPMENT AND YIELD OF TARO IN RESPONSE TO

PLANTING DATE AND FERTILISATION 35
Abstract 35
3.1 Introduction 36
3.2 Materials and methods 38
3.2.1 Planting material 38
3.2.2 Site description 38
3.2.3 Experimental design and agronomic practices 42
3.2.4 Data collection and analysis 42
3.3 Results 43
3.3.1 Emergence 43
3.3.2 Growth parameters 45
3.3.2.1 Number of leaves 45
3.3.2.2 Plant height 48
3.3.2.3 Leaf area 51
vii
3.3.3 Yield 53
3.3.3.1 Number of cormels 53
3.3.3.2 Grading of cormels 56
3.3.3.3 Fresh cormel mass 61
3.4 Discussion 62
3.5 Conclusions and recommendations 64
References 65
4 INFLUENCE OF PLANTING DATE AND ORGANIC FERTILISER ON
SPECIFIC GRAVITY, DRY MATTER CONTENT AND MINERAL CONTENT
OF TARO CORMELS 69
Abstract 69
4.1 Introduction 70
4.2.1 Planting material 72
4.2.2 Site description 72
4.2.3 Experimental design 73
4.2.4 Data collection and analysis 73
4.2.4.1 Dry matter content 74
4.2.4.2 Specific gravity 74
4.2.4.3 Protein content and total protein 74
4.2.4.4 Mineral content 74
4.3 Results 75
4.3.1 Dry matter content 75
viii
4.3.2 Specific gravity 76
4.3.3 Protein content and total protein 79
4.3.3.1 Protein content 79
4.3.3.2 Total protein 82
4.3.4 Mineral content of taro cormels 84
4.3.4.1 Nitrogen content 84
4.3.4.2 Phosphorus content 87
4.3.4.3 Potassium content 89
4.3.4.4 Calcium content 91
4.3.4.5 Magnesium content 92
4.3.4.6 Zinc content 93
4.3.4.7 Copper content 95
4.3.4.8 Manganese content 96
4.4 Discussion 99
References 101
5 CHANGES IN THE SURFACE MORPHOLOGY OF STARCH GRANULES,
ALPHA-AMYLASE ACTIVITY AND SPROUTING OF TARO DURING
STORAGE. 104
Abstract 104
5.1 Introduction 104
5.2.1 Plant material used 106
5.2.2 Scanning electron microscopy 106
ix
5.2.3 Alpha-amylase activity 106
5.2.4 Sprouting 107

5.2.5 Statistical analysis 107
5.3 Results 108
5.3.1 Scanning electron microscopy 108
5.3.2 Alpha-amylase activity 112
5.3.3 Sprouting 114
5.4 Discussion 115
References 118
6 EFFECTS OF PRE AND POST-HARVEST PRACTICES ON

CARBOHYDRATES OF TARO (COLOCASIA ESCULENTA L. SCHOTT) 121
Abstract 121
6.1 Introduction 121
6.2.1 Reducing sugar determination 123
6.2.2 Starch content determination 124
6.2.3 Statistical analysis 124
6.3 Results 124
6.3.1 Reducing sugar content 124
6.3.2 Starch content 127
6.4 Discussion 130
6.5 Conclusion and recommendations 132
x
References 133
7 GENERAL DISCUSSION AND CONCLUSIONS 136
References 146
xi
LIST OF FIGURES
Figure 3.1 Taro cultivars of KwaZulu-Natal 38

Figure 3.2 Mean temperature distribution at Ukulinga for a 10 year period to 2006 39
Figure 3.3 Mean temperature distribution at Umbumbulu for a 10 year period to 2006 39
Figure 3.4 Mean rainfall distribution at Ukulinga for a 10 year period to 2006 40
Figure 3.5 Mean rainfall distribution at Umbumbulu for a 10 year period to 2006 40
Figure 3.6 Effect of planting date and cultivar averaged across Gromor Accelerator
application rate on emergence at Umbumbulu 44
Figure 3.7 Effect of cultivar and Gromor Accelerator application rate averaged across
planting date on emergence at Umbumbulu 45
Figure 3.8 Effect of planting date averaged across cultivar and Gromor Accelerator
application rate on number of leaves plant-1 at Ukulinga 46
application rate on number of leaves plant-1 at Umbumbulu 47
Figure 3.10 Effect of cultivar averaged across planting date and Gromor Accelerator
application rate on number of leaves plant-1 at Umbumbulu 47
application rate on plant height at Umbumbulu 49
Figure 3.12 Effect of planting date and Gromor Accelerator application rate averaged
across cultivar on plant height at Umbumbulu 50
Figure 3.13 Effect of Gromor Accelerator application rate and cultivar averaged across
planting date on plant height at Umbumbulu 50
application rate on leaf area plant-1 at Ukulinga 51
application rate on leaf area plant-1 at Umbumbulu 52
Figure 3.16 Effect of planting date and Gromor Accelerator averaged across cultivar on
leaf area plant-1 at Umbumbulu 53
application rate on number of cormels plant-1 at Ukulinga 54
xii
application rate on number of cormels plant-1 at Umbumbulu 54
Figure 3.19 Effect of planting date and Gromor Accelerator averaged across cultivar on
number of cormels plant-1 at Umbumbulu 55
Figure 3.20 Effect of Gromor Accelerator and cultivar averaged across planting date on
number of cormels plant-1 at Umbumbulu 56
Figure 3.21 Number of cormels plant-1 for three taro cultivars (D = Dumbe-dumbe, M =
Mgingqeni and P = Pitshi) planted with three Gromor Accelerator
application rates (0 = 0 N kg ha-1, 1 = 160 N kg ha-1 and 2 = 320 N kg ha-1)
obtained for October planting at Ukulinga 57
obtained for October planting at Umbumbulu 57
obtained for November planting at Ukulinga 58
obtained for November planting at Umbumbulu 58
obtained for December planting at Ukulinga 59
obtained for December planting at Umbumbulu 59
xiii
obtained for January planting at Ukulinga 60
obtained for January planting at Umbumbulu 60
application rate on fresh cormel mass plant-1 at Ukulinga 61
across cultivar on fresh cormel mass plant-1 of taro cormels at
Umbumbulu 62
Figure 4.1 Effect of Gromor Accelerator application rate averaged across planting date
and cultivar on dry matter content of taro cormels at Ukulinga 75
application rate on dry matter content at Ukulinga 76
application rate on specific gravity at Ukulinga 77
across cultivar on specific gravity at Ukulinga 77
application rate on specific gravity at Umbumbulu 78
across cultivar on specific gravity at Umbumbulu 79
application rate on protein content of taro cormels at Ukulinga 80

Figure 4.8 Effect of Gromor Accelerator application rate averaged across planting date
and cultivar on protein content of taro cormels from Ukulinga 80
xiv
and cultivar on protein content of taro cormels from Umbumbulu 81
application rate on protein content of taro cormels from Umbumbulu 81
application rate on total protein of taro cormels from Ukulinga 82
across cultivar on total protein of taro cormels from Umbumbulu 83
Figure 4.13 Effect of cultivar and Gromor Accelerator application rate averaged across
planting date on total protein of taro cormels from Umbumbulu 84

Figure 4.14 Effect of planting date averaged across Gromor Accelerator application rate
and cultivar on nitrogen content of taro cormels at Ukulinga 85
and cultivar on nitrogen content of taro cormels at Ukulinga 85
and cultivar on nitrogen content of taro cormels at Umbumbulu 86
application rate on nitrogen content of taro cormels at Umbumbulu 87
application rate on phosphorus content of taro cormels at Ukulinga 88
application rate on potassium content of taro cormels at Ukulinga 90
application rate on potassium content of taro cormels at Umbumbulu 91
and cultivar on potassium content of taro cormels at Umbumbulu 91
application rate on calcium content of taro cormels at Umbumbulu 92
application rate on magnesium content of taro cormels at Ukulinga 93
xv
application rate on zinc content of taro cormels at Ukulinga 94
application rate on zinc content of taro cormels at Umbumbulu 94
and cultivar on zinc content of taro cormels at Umbumbulu 95
application rate on manganese content of taro cormels at Ukulinga 97
application rate on manganese content of taro cormels at Umbumbulu 98

Figure 4.29 Effect of planting date and Gromor Accelerator application rate on
manganese content of taro cormels at Umbumbulu 98
Figure 5.1 Scanning electron micrographs showing starch granules of Dumbe-dumbe at
harvest 109
Figure 5.2 Scanning electron micrographs showing starch granules of Mgingqeni at
harvest 109
Figure 5.3 Scanning electron micrographs showing starch granules of Pitshi at harvest
110
Figure 5.4 Scanning electron micrographs showing starch granules of taro cultivars
stored in polyethylene bags for four months at ambient temperature 111
Figure 5.5 Alpha-amylase activity in taro cormels at harvest up to four months after
storage in polyethylene bags at ambient temperature 112
Figure 5.6 Sprouting in taro cormels at harvest up to four months after storage in
polyethylene bags at ambient temperature 114
Figure 6.1 Reducing sugar content of taro cormels planted at different planting dates
averaged across cultivar at Ukulinga 125
averaged across cultivar and Gromor Accelerator application rate at
Umbumbulu 126
Figure 6.3 Starch content of taro cultivars planted at different planting dates averaged
across Gromor Accelerator application rates at Ukulinga 128
xvi
Figure 6.4 Starch content of taro cultivar planted at different planting dates at
Umbumbulu 129
Figure 6.5 Starch content of taro cultivars planted with different Gromor Accelerator
application rates averaged across planting dates at Umbumbulu 129
xvii
LIST OF TABLES
Table 3.1 Mean temperature and mean rainfall data for Ukulinga and Umbumbulu for the
duration of the experimental period 41
Table 3.2 Physical and chemical properties of soils from Ukulinga and Umbumbulu 41
Table 3.3 Nutritional composition of Gromor accelerator 42
Table 3.4 Effect of planting date, Gromor Accelerator application rate and cultivar on
emergence at Ukulinga. 44
Table 3.5 Effect of planting date, Gromor Accelerator application rate and cultivar on
plant height at Ukulinga 48
Table 4.1 Effect of planting date and Gromor Accelerator application rate on
phosphorus content of taro cormels of different cultivars at Umbumbulu 89
Table 4.2 Effect of planting date and Gromor Accelerator application rate on copper
content of taro cormels of different cultivars at Umbumbulu 96
Table 5.1 Alpha-amylase activity of taro cultivars stored at different temperatures in
different packaging materials 113
Table 5.2 Sprouting of taro cormels stored at different temperatures in different
packaging materials 115
Table 6.1 Reducing sugar content of taro cultivars stored at different temperatures in
different packaging materials 127
Table 6.2 Starch content of taro cultivars stored at different temperatures in different
packaging materials 130
Table 7.1 Arbitrary scoring of crisping quality of taro cultivars planted in October at
Ukulinga 140
Table 7.2 Arbitrary scoring of crisping quality of taro cultivars planted in November at
Ukulinga 140
Table 7.3 Arbitrary scoring of crisping quality of taro cultivars planted in December at
Ukulinga 141
Table 7.4 Arbitrary scoring of crisping quality of taro cultivars planted in January at
Ukulinga 141
Table 7.5 Arbitrary scoring of crisping quality of taro cultivars planted in October at
Umbumbulu 142
xviii
Table 7.6 Arbitrary scoring of crisping quality of taro cultivars planted in November at
Umbumbulu 142
Table 7.7 Arbitrary scoring of crisping quality of taro cultivars planted in December at
Umbumbulu 143
Table 7.8 Arbitrary scoring of crisping quality of taro cultivars planted in January at
Umbumbulu 143
xix
LIST OF APPENDICES
Appendix 3.1 Analysis of variance of the growth and yield parameters of taro obtained
at Ukulinga 148
Appendix 3.2 Analysis of variance of the growth and yield parameters of taro obtained
at Umbumbulu 151
Appendix 3.3 Regression analysis of growth and yield parameters 154
Appendix 3.4 Net returns hectare-1 of taro in KwaZulu-Natal 156
Appendix 4.1 Analysis of variance of dry matter content of taro cormels obtained at
Ukulinga 157
Appendix 4.2 Analysis of variance of dry matter content of taro cormels obtained at
Umbumbulu 157
Appendix 4.3 Analysis of variance of specific gravity of taro cormels obtained at
Ukulinga 158
Umbumbulu 158
Appendix 4.5 Analysis of variance of protein content of taro cormels obtained at
Ukulinga 159
Umbumbulu 159
Appendix 4.7 Analysis of variance of total protein of taro cormels obtained at
Ukulinga 160
Umbumbulu 160
Appendix 4.9 Analysis of variance of mineral content of taro cormels obtained at
Ukulinga 161
Umbumbulu 166
xx
Appendix 5.1 Analysis of variance of alpha-amylase activity of taro cormels of three
taro cultivars (Dumbe-dumbe, Mgingqeni and Pitshi) stored at 12oC and
ambient temperature in three packagings (Polyethylene bag, open box
and mesh bag) for four months 171
Appendix 5.2 Analysis of variance of sprouting of taro cormels of three taro cultivars
(Dumbe-dumbe, Mgingqeni and Pitshi) stored at 12oC and ambient
temperature in three packagings (Polyethylene bag, open box and mesh
bag) for four months 172
Appendix 6.1 Analysis of variance of reducing sugar content of taro cormels obtained
at Ukulinga 173
Appendix 6.2 Analysis of variance of reducing sugar content of taro cormels obtained
at Umbumbulu 173
Appendix 6.3 Analysis of variance of reducing sugar content of taro cormels of three
taro cultivars (Dumbe-dumbe, Mgingqeni and Pitshi) stored at 12oC
and ambient temperature in three packagings (Polyethylene bag, open
box and mesh bag) for four months 174
Appendix 6.4 Analysis of variance of starch content of taro cormels obtained at
Ukulinga 175
Umbumbulu 175
Appendix 6.6 Analysis of variance of starch content of taro cormels of three taro
landraces (Dumbe-dumbe, Mgingqeni and Pitshi) stored at 12oC and
ambient temperature in three packagings (Polyethylene bag, open box
and mesh bag) for four months 176
Appendix 7.1 Arbitrary scores for taro crisping quality parameters 177
xxi
CHAPTER 1
GENERAL INTRODUCTION
Taro, Colocasia esculenta (L.) Schott belongs to the monocotyledonous family Araceae
of the order Arales whose members are known as aroids (Henry, 2001; Van Wyk, 2005).
Araceae includes about 100 genera and 1500 widely distributed species (Merlin, 1982;
Vinning, 2003). Taro is one of the few edible species in the genus Colocasia within the
sub-family Colocasioideae (Ezumah, 1972) and the most widely cultivated species
(Vinning, 2003). The species is considered to be allogamous and polymorphic (Ivancic et
al., 2003). Colocasia species can be classified as follow: (1) Colocasia esculenta (L)
Schott var. esculenta (produces a large cylindrical central corm with very few cormels
and is known as dasheen); and (2) Colocasia esculenta (L) Schott var. antiquorum
(produces a small globular central corm surrounded by numerous side cormels and is also
known as eddoe) (Purseglove, 1972; Lebot and Aradhya, 1991).
Originating in Asia, this root crop is now found primarily in tropical and subtropical
regions of the world (Bradbury et al., 1988; Macleod, 1990). In South Africa, taro is
mainly a KwaZulu-Natal coast and hinterland traditional crop (Modi, 2004), hence the
Zulu name amadumbe. It is an important staple crop in the subtropical coastal area
starting at Bizana district in the Eastern Cape and the rest of coastal KwaZulu-Natal.
There is less cultivation of the crop in the Midlands and generally none in the northern
parts of the province where the climate is drier and cooler. The crop is also cultivated in
the subtropical and tropical parts of Mpumalanga and Limpopo provinces (Shange,
2004).
Taro is used as food, prepared the same way as potatoes. Its flour is considered good
baby food because its starch is easily digestible; and it helps with digestive problems and
supplements iron (Onwueme, 1999; Shange, 2004; Van Wyk, 2005). According to
Vinning (2003), taro starch digestibility is as high as 98%. It is also suitable as a specialty
food for allergic infants and persons with alimentary disorders and for those allergic to
cereal starch as well as those sensitive to animal milk ((Salunkhe and Kadam, 1998;
Onwueme, 1999; Vinning, 2003). Various parts of the plant are also used in traditional
1
medicine practice (Tsitsiringos, 2002). Taro is also used as an ornamental in Australia,
Japan, Italy and elsewhere. Although, it is irritating to the human skin, mouth and throat,
acridity protect taro against herbivores. The crop is thus well suited for organic farming
(Krech et al., 2004). Taro is regarded as potential and important because of its nutritional
status and the role it plays in food security, especially in the rural areas. In many rural
areas, taro industry provides meaningful employment and plays an important part in the
cultural practices and as a vehicle for rural development (Onwueme, 1999). Taro is an
excellent multipurpose food crop for subsistence agriculture and home gardens. Its ability
to tolerate salinity makes it suitable for localities where few other crops grow (Grubben
and Denton, 2004). As such, it merits more attention in research focusing on yield
(Grubben and Denton, 2004).
Most taro production in South Africa is consumed as subsistence food on the farms and a
small proportion finds its way to the market (Shange, 2004). According to Modi (2003),
only Umbumbulu farmers market it. Taro is of increasing importance as a subsistence
crop in the rural areas of KwaZulu-Natal. The rise in importance of the crop can be
attributed to the fact that 7 years ago the subsistence farmers from Umbumbulu started to
supply fresh certified organic taro to Woolworths Foods and Pick`n ‘Pay chain stores
which sell it in Durban, Johannesburg and Cape town. The subsistence farmers are only
able to supply taro after harvest. They do not meet a regular continuous demand of
consumers by providing availability of taro for longer periods of time. This is mainly
because storage life of taro is usually rather short owing to its high moisture content. This
short storage life of taro has emphasized the need to improve its storage potential. The
potential of local cultivars for food processing needs to be explored, to improve
marketing opportunities and household income for the farmers. Their potential for
processing into crisp chips has been identified in KwaZulu-Natal to add value to the
product.
Despite its contribution to food security over centuries, including the times before the
advent of commercial crops originating in Europe, which predominate traditional
agriculture today (e.g. potatoes), agronomic research into taro is very recent in South
2
Africa (Mare, 2006, Modi, 2003; 2007; Shange, 2004). Among the aspects that have not
been studied in detail regarding South African taro include the relationship between
agronomy and quality with respect to storability and possibly food processing, as is the
case in root crops such as potato.
A neglect of the crop is found in the extension services, so that the technical knowledge
base of the producers is very low. Most of the above constraints in the taro production
system can be effectively tackled and possibly solved through research. A lot of taro
production still relies on age-old traditional production methods. Research into various
agronomic and storage practices is needed to improve the productivity and storability of
taro. There is need to establish, through research, the best pre- and post-harvest practices
with respect to planting date, fertilisation and storage requirements of taro among others.
All the above research priorities need to be addressed in order to sustain the taro industry.
With the potential production of chips from taro, there is a need for an increased
understanding of how corm chipping quality is impacted by planting date, landrace,
fertilisation and storage. The specific objectives of the study were
1. To investigate the effect of planting date and fertilisation on growth and yield of taro
cultivars found in Umbumbulu, KwaZulu-Natal.
2. To determine the effects of planting date, fertilisation on chipping quality of taro
cormels.
3. To examine changes in the surface morphology of starch granules, alpha-amylase
activity and sprouting of taro cormels during storage.
4. To investigate effects of pre- and post-harvest practices on starch and reducing

sugars of taro.
This dissertation is presented in the form of manuscripts, written according to the style of
the African Crop Science Journal, for convenience. The last chapter is a generalized
discussion and conclusion.
3
References
Bradbury, J.H., Holloway, W.D., Bradshaw, K., Jealous, W. and Phimpisane, T. 1988.
Effect of cooking on nutrient content of tropical root crops from the South
Pacific. Journal of Science of Food and Agriculture 43: 333–342.
Ezumah, H.C. 1972. The growth and development of taro, Colocasia esculenta (L.)
Schott, in relation to selected cultural management practices. Ph.D. Thesis.
University of Hawaii.
Grubben, G.J.H. and Denton, O.A. 2004. Vegetables. Plant Resources of Tropical Africa.
PROTA.
Henry, R. J. 2001. Plant genotyping: The DNA fingerprinting of plants. CAB Publishing,
Southern Cross University, Australia.
Ivancic, A., Quero-Garcia, J. and Lebot, V. 2003. Development of visual tools for
selecting qualitative corm characteristics of taro (Colocasia esculenta (L.)
Schott). Australian Journal of Agriculture 54: 581-587.
Krech, S., McNeill, J.R. and Merchant, C. 2004. Encyclopedia of world environmental
history. Routledge
Lebot, V. and Aradhya, K. M. 1991. Isozyme variation in taro (Colocasia esculenta (L.)
Schott) from Asia and Oceania. Euphytica 56: 55-66.
Macleod, G. 1990. Combined gas chromatography-mass spectrometric analysis of the
aroma components of cooked taro (Colocasia esculenta L.). Food Chemistry
38: 89–96.
Mare, R.M. 2006. Phytotron and field performance of taro [Colocasia esculenta (L.)
Schott] landraces from Umbumbulu. MSc. Thesis. University of KwaZulu-
Natal, Pietermaritzburg, South Africa.
Merlin, M. 1982. The origins and dispersal of true taro. Native Planters: Ho`okupu Kalo.
Modi, A.T. 2003. What do subsistence farmers know about indigenous crops and organic
farming? Preliminary case in KwaZulu-Natal. Development Southern Africa
20: 673-682.
4
Modi, A.T. 2004 . Short-term preservation of maize landrace seed and taro propagules
using indigenous storage methods. South African Journal of Botany 70 (1): 16-
23.
Onwueme, I.C. 1999. Taro cultivation in Asia and the Pacific. FAO PAR publication
1996/16. Bangkok, Thailand
Purseglove, J.W. 1972. Tropical crops: Monocotyledons. Wiley, New York.
Salunkhe, D.K. and Kadam, S.S. 1998. Handbook of vegetable science and technology:
Production, composition, storage and processing. Marcel Dekker. USA.
Shange, L.P. 2004. Taro (Colocasia esculenta (L.) Schott) production by small-scale
farmers in KwaZulu-Natal: Farmer practices and performance of propagule
types under wetland and dryland conditions. MSc. Thesis. University of
KwaZulu-Natal, Pietermaritzburg.
Tsitsiringos, V.K. 2002. Financial engineering, E-commerce and supply chain. Kluwer
Academic Publishers, The Netherlands.
Van Wyk, B.E. 2005. Food plants of the world: Identification, culinary uses and
nutritional value. Briza Publications, Pretoria, South Africa.
Vinning, G. 2003. Select markets for taro, sweet potato and yam. RIRDC Project No.
UCQ-13A.
5
CHAPTER 2
LITERATURE REVIEW
2.1 Quality of taro cormels for crisp making
Quality refers to the suitability of cormels for a particular manufacturing process. It is

important with regard to food processing and it is a combination of characteristics that
give a commodity value. The relative importance given to a specific quality characteristic
varies in accordance with the commodity concerned and with the individual or market
concerned with quality assessment (Kader and Rolle, 2004). According to Hollyer et al.
(2000), making taro chips is similar to making potato chips so the quality parameters
required for chip making are similar. Potato processors require precise tuber
characteristics to optimize product quality and quantity (Williams and Cobb, 1993). In
potatoes, high yield is an important quality attribute to producers while tuber size is
important to processors (Kader and Rolle, 2004; Carputo et al., 2005) Potato industry
requires large grade tubers with better crisping quality (Kader and Rolle, 2004). Chemical
composition is of utmost importance in tubers destined for crisp making, and in potatoes,
these includes specific gravity, dry matter content, starch content, starch quality, sugar
content (sucrose, glucose and fructose), protein content, minerals and shelf life (Eskin,
1989; Kader and Rolle, 2004; Carputo et al., 2005; Gebhardt et al., 2005; O`Keefe et al.,
2005).
2.1.1 Yield
Yield traits include total weight of cormels plant-1, number of cormels plant-1 and mass of
individual corms. Smith (1987) stated that the size of the tuber is the key criterion for
determining suitability of a potato cultivar for crisping. Potato processors prefer medium
sized tubers, 170-284g in weight to produce uniform crisps (Smith, 1987). The cultivar
suitable for crisping must, therefore, have high yields with a high proportion of usable
tubers having a good size. In potatoes, a 45 kg sample with about 250 tubers will have
many tubers in this size range and an average weight close to 184 g per tuber (Smith,
6
1987). According to Panhwar (2005), tubers of about 50-350 g in weight are preferred
because they produce the crisp of the right size and pass through the processing line with
less need for hand trimming.
2.1.2 Chemical composition of taro cormels
2.1.2.1 Specific gravity
Specific gravity is a major characteristic of tubers for processing. It influences the

processing quality of potatoes and is used by processors to assess the acceptability
(Laboski and Kelling, 2007). High specific gravity indicates maturity and corresponds to
high quality (Bowers et al., 1964), but specific gravity should also not be excessively
high since tubers with excessively high specific gravity are susceptible to bruising
(Mosley and Chase, 1993). Tubers with high specific gravity produce high chip yield,
absorb less oil and need less energy to process and hence less processing costs (Smith,
1987; Xiong and Tai, 2003). According to Hollyer et al. (2000), specific gravity of raw
taro corms varies in a narrow range of 0.94 – 0.98, with more mature corms having the
greater value. Taro corms selected for crisping must have moderately high specific
gravity. Tubers with very high specific gravity bruise easily. In potatoes, tubers with
specific gravity of 1.070 – 1.109 can be used for crisping but excellent crisps can be
obtained with specific gravity of 1.090 – 1.099 (Smith, 1987). Panhwar (2005) suggested
the minimum acceptable levels of 1.07.
2.1.2.2 Dry matter content
Dry matter needed in the raw product depends on its processing destination. Quality
characteristics of potatoes processed into crisps are largely dependent on the dry matter
content (Talburt and Smith, 1987). In potatoes, tubers high in dry matter are suitable for
the manufacture of crisps (Lisińska and Laszczyński, 1989) and this attribute need to be
present at harvest and remain through storage (Smith, 1987). Higher dry matter content in
raw products improves recovery rate during processing and directly influences texture
7
and appearance of and indirectly colour of potato crisps (Smith, 1987). Hollyer et al.
(2000) also reported that chips made from the bottom of the corm are rated better in
appearance than those made from the top part because the bottom part is dryer with
around 5 percent greater dry matter content than the top part. According to Smith (1987),
dry matter content for crisping should be in the range of 20-23% whereas Kita (2002), on
the other hand, suggested that percentage of dry matter in potatoes for crisp production
should be 20 – 25%. This was also confirmed by Panhwar (2005) who suggested that the
minimum dry matter content acceptable for crisping should be 20%.
2.1.2.3 Starch content
Starch content was reported to influence corm quality for making crisp chips in taro
(O`Keefe et al, 2005). Kita (2002) observed the texture of crisps to be dependent on the
content of starch in potato tubers. The ideal crisping potato must have high starch and dry
matter content, it should be high at harvest and through storage (Smith, 1987; O`Keefe et
al., 2005). According to Kita (2002), starch content for crisping should be more than
15%.
2.1.2.4 Alpha-amylase activity
Alpha amylase is a starch degrading enzyme that takes part in the breakdown of starch in
potatoes (Witt and Sauter, 1996; Ramachandran et al., 2004). In cereal grains it is of
prime importance in initial stages of starch degradation (Macgregor, 1983). It is one of
the three enzymes that can convert starch back to glucose (Kays, 1991). Alpha amylase is
inhibited by calcium (Duffus and Duffus, 1984; Witt and Sauter, 1996; Ramachandran et
al., 2004). Kumar and Ramesh (undated) stated that the enzyme is heterogeneous and that
some of its components might be more efficient than others in hydrolyzing starch
granules. The activity of the enzyme is therefore of utmost importance to monitor its part
in starch degradation. The low thermal stability and the high affinity to granules at low
temperature might suggest a role for the enzyme especially in cold-induced sweetening.
8
2.1.2.5 Reducing sugar content
Sugar content of potato tubers is another important factor accounting for their usability as
processing raw material (Talburt and Smith 1987). Reducing sugars are responsible for
enzymatic browning during frying which is a negative aspect for the potato processing
industry (Allison et al., 1999). O’Keefe et al. (2005) reported that chips produced from
the top of the taro corm were a darker colour than chips produced from the middle of the
corm due to a higher reducing sugar content in the top part than the middle of the corm
(an average of 0.15% higher). According to Roe et al. (1990) 90% of the variation in
colour of potato chips could be accounted for by variation in sugar content. Potatoes
suitable for crisping must have low reducing sugars and sucrose contents to minimize the
browning during frying and like with other attributes they need to be low at harvest and
throughout storage (Smith, 1987). Smith (1987) reported that the best levels for potato
chip colour occur when sucrose levels are equal to or below 1.5mg/g tuber and glucose
levels are equal to or below 0.35 mg/g tuber. Kumar et al. (2007) stated that tubers with
high reducing and total sugar (>2%) were found to be unsuitable for crisp-making.
According to Lisińska and Leszczyński (1989), reducing sugar should not exceed 0.25 –
0.5% whereas according to Kita (2002), the content of total sugars should be less than
0.23% and reducing sugars less than 0.12%. Panhwar (2005), on the other hand,
suggested that reducing sugars should not exceed 0.2%.
2.1.2.6 Protein and mineral content
Protein nitrogen could influence the quality of crisps. Besides starch content of the potato
tubers, crisp texture also depends on the sum of other components including protein
nitrogen (Kita, 2002). Tuber proteins also include amino acids which together with sugars
are responsible for the production of the colour of fried chips. According to Roe et al.
(1990), 8% of the variation in the colour of crisps could be accounted for by amino acids.
Total proteins and levels of minerals in taro are important because they are components
of the human diet. And foodstuff is regarded as important based on its composition of
components of the human diet. Mineral content also affect the colour of chips by
9
affecting the reducing sugar content. Success in making taro chips varies with varieties
due to the acridity nature of the taro corms (Hollyer et al., 2000; Bradbury and Nixon,
1998). Ndimantang and Obiakor (2006) reported that frying in oil to doneness reduced
oxalates. Therefore frying is greatly recommended for the quality processing of
Cocoyams. High temperatures are known to cause the calcium oxalate containing cells
(raphides) to collapse leading to the break down of oxalate structure. The mechanism of
oxalate reduction by heat has not been fully clarified (Ndimantang and Obiakor, 2006).
2.2 Taro growth stages
Corm quality is determined at different growth stages in taro. According to Sivan (1982)
there are three growth stages in taro, namely establishment, vegetative growth and corm
initiation and bulking through maturation.
2.2.1 Establishment
The period of establishment is the root formation and leaf production during the first
month (Sivan, 1982). This stage is characterized by sprouting and root growth.
Successful establishment is a critical prerequisite for efficient crop production and is
primarily determined by propagule quality (Modi, 2007). In taro, propagule size is crucial
for successful establishment because at this stage plants are entirely supported by
available carbohydrates from the seed piece up to a plant leaf area of 400cm2 plant-1
(Singh et al., 1998). A study by Modi (2007) showed that large propagules in taro
improved stand establishment by increasing the number of plants reaching the third leaf
stage and leaf area plant-1 one month after planting.
2.2.2 Vegetative growth and corm initiation
This is a period of rapid root and shoot development with initiation of corm development
during two to four months (Sivan, 1982). The stage is marked by increase in plant height,
number of leaves and leaf area and slow corm growth (Tumuhimbise et al., 2007; Silva et
10
al, 2008). The leaf and stem are the dominant sinks for assimilates at this stage (Singh et
al., 1998). Maximal total leaf area indices were obtained at about 117 days after planting
and then declined sharply (Goenega, 1995). This was confirmed by Mare (2006) who
found leaf number, plant height and leaf area to reach maximum at 120 days after
planting. Corm formation commences at about three months after planting and cormel
formation follows soon afterwards in cultivars that produce appreciable cormels
(Onwueme, 1999).
2.2.3 Corm bulking and maturation
This is the period of a climax of root and shoot growth with a rapid increase in corm
formation during five to six months and a senescence period of decreasing root and shoot
growth with continued increase in corm size during six through nine months (Sivan,
1982). The leaf development decreases in intensity and the plant growth is reduced (Silva
et al., 2008). This was also confirmed by Onwueme (1999), who reported that the rapid
decline in shoot growth and total shoot dry weight was shown at about six months after
planting. And this was characterized by a reduction in the number of active leaves,
decrease in the mean petiole length, a decrease in the total leaf area per plant, and a
decrease in the mean plant height on the field (Onwueme, 1999). According to Goenega
(1995), corm bulking occurred after the attainment of maximal leaf area indices and the
partitioning of dry matter to the corms remained constant almost especially after 150 days
after planting. Tumuhimbise et al. (2007) also reported that it is a period of growth in
which corm diameter and length increased rapidly throughout the 150 days. Singh et al.
(1998) on the other hand reported assimilate partitioning to corms to be maximum at 120
days after planting and levelled until 160 days after planting under non limiting water and
nitrogen conditions. Corms and cormels are given the first priority and become the main
sink for available assimilates at this point and grow very rapidly (Singh et al., 1998;
Onwueme, 1999).
11
2.3 Factors influencing taro quality
The yield of taro is affected by planting date with temperature being the most important
factor affecting growth (Lu et al., 2001). Water availability and fertilisation also affect
yield and quality of taro (Chun-Feng and Kun, 2004; Scheffer et al., 2005). In potatoes,
yield and chemical composition varies with variety (Karam et al., 2009). It is also
influenced by climatic conditions, which are dependent upon the site and planting date.
Storage environment also bring changes to the chemical composition of potato tubers
(Lisińska and Laszczyński, 1989).
2.3.1 Genetic factors
Yield and chemical composition of root crops differs with different cultivars. Proietti et
al. (2005) suggested quality attributes in potatoes to be affected mainly by cultivar. Osiru
et al. (2009) recorded variation in root yield among sweet potato genotypes. Babaji et al.
(2009) also reported Nicola, RC 767-2 and WC 732-1 which are Irish potato cultivars to
have produced more tubers and unmarketable tuber yield than Greta, whereas Greta and
RC 767-2 produced larger tubers than Nicola and WC 732-1. Variation in water content
(55.8 to 74.4 g 100 g-1), starch content (20.0 to 35.1 g 100 g-1), protein content (0.5 to 2.1
g 100 g-1) and mineral content (Wills et al., 1983) and starch content (509.1 to 705.7 g kg-
1
dry weight) (Agbor-Egbe and Rickard, 1990) between taro cultivars was recorded.
According to Smith (1987), specific gravity is an inherited trait determined by the genetic
makeup of a cultivar. They reported that some cultivars of potatoes develop higher
specific gravity than others and that a cultivar with inherently low specific gravity cannot
produce tubers with a high specific gravity through cultural management (Smith 1987).
Long et al. (2004) also reported that genotype was the major factor that influenced
specific gravity, sucrose and glucose content of Michigan potato cultivars. Highly
significant differences in dry matter content between tested potato varieties were reported
by Musa et al. (2009). Wills et al. (1983) indicated starch properties in taro to be
influenced by genetic variation. Significant variations in amylase activity were reported
between cultivars in sweet potatoes (Nandutu et al., 2000).
12
Sugar content was highly varied between potato cultivars and the ratios of reducing
sugars to sucrose and fructose to glucose differed from variety to variety (Lisińska and
Laszczyński, 1989). Similarly, protein content and mineral levels in different cultivars of
taro were also found to be variable (Wills et al, 1983; Bártová et al., 2009). This
confirmed previous studies that found wide differences in the content of protein,
carbohydrate, iron, calcium and phosphorus of 9 cultivars of edible aroids (Rashid and
Daunicht, 1979). Islam et al. (2007) found that in sweet potatoes, different cultivars were
different in terms of storability.
2.3.2 Planting date
Planting date influences yield of tuber crops (Khan et al., 2003; Martin et al., 2005).
Planting date effects on the yield and chemical composition of a crop are dependent on
the environmental conditions prevailing during the crop life cycle. A differential response
of the genotypes to the varying climatic conditions at different locations was also
reported for cocoyam (Reyes - Castro et al., 2005). These environmental conditions or
weather variations among locations and the planting dates include temperature and
rainfall or water availability. The optimum planting date is related to soil and air
temperatures as well as soil moisture content (Lisińska and Laszczyński, 1989). The soil
moisture in turn depends on rainfall or water availability. The ideal planting date is the
one that does not allow the stages of crop growth to coincide with the periods when the
crop is very sensitive to temperature or moisture stress to avoid drastic effects on yield
and quality. Lu et al., (2001) found that in Taiwan final harvest index in taro was highest
for January and March plantings, whereas July and September plantings had the lowest
harvest index, and that was attributed to high temperature during the vigorous top-growth
stage for January and March crops and declining temperature during the vigorous top-
growth stage for July and September crops.
2.3.2.1 Temperature
Temperature was found to be the primary factor governing growth rate in taro (Chan et
al., 1998). Transpiration, translocation, photosynthesis and respiration which all affect
13
yield and chemical composition of root crops are temperature dependent (Smith, 1987).
According to Bazzaz and Sombroek (1996), all stages of crop development are sensitive
to temperature and the sensitivity differs with phenological stages and genotype. They
also stated that in general, development accelerates as temperature increases.
Wolf et al. (1990) found that translocation of assimilates to the vegetative organs was
greater at higher temperatures while translocation to the tubers was less in both potato
varieties studied. This was confirmed by Almekinders and Struik (1996) who reported
increased number of leaves per stem and stems per shoot in potatoes at warmer
temperatures which altered assimilate production and partitioning and hence yield.
Ghosh et al. (2000) observed maximum yield reduction at high temperature at vegetative
to tuber initiation which was most critical for tuber yield, and highest yield at low
temperature at tuber initiation to initial tuber bulking which was considered as more
advantageous step towards yield improvement. Many studies reported that high
temperature reduce yield (Rykaczewska, 2004). High temperature (30oC) compared to 15
o
C and 25oC was found to degrade potato tuber quality by reducing specific gravity
(Ghosh et al., 2000).
Dry matter content is reduced by temperature and water stresses (Smith, 1987; Bakayoko
et al., 2009). Mittra and George (2000) reported lower tuber dry matter content in sweet
potatoes planted in June, July and August in India compared to those planted in
September and October and this was attributed to high dry matter accumulation in leaves
and vines in plants grown during the period of high temperature (June, July and August).
The same observations were made by Colla et al. (2005) who found that dry matter of
potatoes grown in 2003 was lower than in 2004 due to the very high temperature,
especially night temperature, which may have reduced the rate of photosynthesis and
increased respiratory losses. These findings were confirmed by Hammer et al. (2007)
who found dry matter of cassava storage roots to be highest during the cooler months
when canopy vigour was lowest.
The influence of environmental factors on starch properties in chickpea are greater than
varietal differences (Debon and Tester, 2000). Excessively high day temperatures
14
decrease photosynthesis and increase respiration and these causes loss of starch. Starch in
the tubers particularly near the stem end is then converted into sucrose and the dry matter
of the tuber decreases whereas the sugar levels are raised. High air temperatures may be
accompanied by water stress, resulting in uneven tuber growth or bulking rate (Smith,
1987). Temperature influences the uptake and metabolism of mineral nutrients by plants
by increasing transpiration rates (Kader and Rolle, 2004).
2.3.2.2 Rainfall
Water availability influences yield and tuber size distribution in potatoes (Casa et al.,
2005). Al-Omran (1991) recorded maximum average yields at the highest irrigation level
in potatoes. It was also reported that in taro inadequate rainfall during the time of greatest
water need resulted in lower yield and percentage corm dry matter (Miyasaka et al.,
2001). Similar results were reported by Yuan et al. (2003) who found that total fresh
tuber yields and marketable tuber yields increased with increasing amounts of irrigation
water and that highest yields were obtained at highest irrigation level. They also found
that irrigation increased tuber number and mean weight of the tubers (Yuan et al., 2003).
Reduced potato yields were also reported with water stress (Rykaczewska, 2004;
Alsharari et al., 2007).
Adequate moisture produces potatoes with high specific gravities and starch content
whereas water deficiency leads to low specific gravity (Smith, 1987). Yuan et al. (2003)
reported irrigation to decrease specific gravity. Irrigation influences dry matter content
(Casa et al., 2005). According to Smith (1987), water stress reduces dry matter content.
Mittra and George (2000), on the contrary reported lower tuber dry matter content in
sweet potatoes planted during the period of high rainfall. This is in line with what Hollyer
et al. (2000) found that wet-grown taro varieties have too high a moisture content to
make a good chip. Irrigation influences sugar content. Water deficiency lead to high
sugar content (Smith, 1987).
15
2.3.3 Fertilisation
Nitrogen, phosphorus and potassium are the three major nutrients, which individually
and/or together maintain growth, yield and quality of plants (Mazid, 1993; Ivonyi et al.,
1997). Additional nutrients are applied to the soil to produce optimum yields and quality
since many soils do not supply enough phosphorus and potassium (Smith, 1987). Crops
respond differently to different fertiliser elements, and proper fertiliser management for a
plant species is important for increasing yield and quality.
Fertilisation influences the water and nutrient supply to the plant, which can in turn affect
the nutritional quality of the harvested plant part. The effects of mineral and elemental
uptake from fertilisers by plants are, however, significant and variable (Kader and Rolle,
2004). Proper nitrogen fertilisation is important to achieve optimum yields and quality of
potatoes for processing (Smith, 1987). Deficits or excesses in nitrogen availability are
both negative for optimization of yield and quality (Allison et al., 1999). Many tuber
quality attributes are promoted by an adequate nitrogen supply, but decreased by an
excess (Casa et al., 2005). However, many of the effects of nitrogen on tuber quality are
strongly influenced by other factors, such as variety and water status (Casa et al., 2005).
Potassium fertilisation also affects yield in potatoes (Kumar et al., 2004).
Nitrogen is responsible for 26-41% of crop yields (Maier et al., 1994; Mazid, 1993).
Nitrogen excesses is negative for tuber yield in the sense that it increases leaf area and
prolong leaf production leading to delay in tuber maturity. According to Allison et al.
(1999), nitrogen deficiencies decrease the fraction of larger size tubers. This was
confirmed by Casa et al. (2005) who reported that nitrogen deficiency in addition to
causing yield decreases, also affects tuber number and size distribution. Zelalem (2009)
also reported that nitrogen fertilisation increase tuber number and average tuber yield in
potatoes.
Phosphorus indirectly promotes plant growth and absorption of potassium as well as

other nutrients (Oya, 1972). Root crops have a large requirement for phosphorus and it
contributes to early tuberisation and increase the number of tubers per plant (Jenkins and
16
Ali, 2000). This confirmed earlier report by Ali (1998) that phosphorus increases the
number of tubers per plant. Ali et al. (2004) compared tuber number and yield of potato
cv. Desiree fertilised with six levels of phosphorus fertiliser and observed that tuber
number m-2 increased up to 200 kg P2O5 ha-1. Dubetz and Bole (1975), on the other hand,
previously reported number and tuber yield to increase up to 224 kg P2O5 ha-1 with mean
weight per tuber continuing to increase up to 448 kg P2O5 ha-1. According to Smith
(1987) deficiency of phosphorus also leads to very small tubers.
Potatoes use large amounts of potassium (Smith, 1987). Tuber size is one of the tuber
quality parameter most affected by potassium fertilisation (McDole et al., 1978; Sharma
and Arora, 1987; Chapman et al., 1992; Westermann et al., 1994). An increase in the
proportions of large tubers relative to small ones was reported in response to potassium
fertilisation (Martin-Prevel, 1989; Singh et al., 1996; Karam et al., 2005). Tawfik (2001)
reported that high rate of potassium increased yield of medium (28-60 mm) and over-
sized tubers (> 60 mm) by about 15 and 40% respectively. The increase in large size
tubers in response to potassium nutrition was attributed to an increase in water
accumulation in tubers (Perrenoud, 1993). Smith (1987) also reported the deficiency of
potassium to lead to many small sized tubers. Chapman et al., 1992 and Westermann et
al., 1994 reported total tuber yield increase with increase in potassium fertilisation. Plants
fertilised with high rate of potassium (120 kg K. Fed-1) showed 25-30% increase in fresh
weight of tubers compared to those of low potassium (60 kg K. Fed-1) (Tawfik, 2001). A
sufficient supply of potassium also promotes nitrogen uptake efficiency of plants due to
its stimulant effect on plant growth (Oya, 1972). Deficiencies in sulfur, calcium, boron
and zinc may also lower yield and lead to earlier vine death (Smith, 1987). Increasing soil
calcium may increase average tuber size and decrease tuber number (Ozgen et al., 2003).
Adequate soil fertility produces potatoes with high specific gravities whereas nutrient
excess and deficiency lead to low specific gravity (Smith, 1987). Specific gravity often
decreases as rate of applied nitrogen increases (Laurence et al., 1985; Porter and Sisson,
1991; Feibert et al., 1998; Sparrow and Chapman, 2003). Dahlenberg et al. (1990),
however, reported lower specific gravity in nitrogen deficient tubers as well as in those
17
from plants where nitrogen was in excess of that needed for maximum yield, in 3 out of 4
sites in South Australia. This is in line with findings that that nutrient deficiency leads to
low specific gravity (Smith, 1987) and that excessive nitrogen promotes excessive vine
growth late in the season resulting in potato tubers having much lower specific gravity
(Smith, 1987; Laboski and Kelling, 2007).
Excessive application rates of potassium may reduce tuber solids and hence lower
specific gravity (Laboski and Kelling, 2007). Specific gravity and dry matter content of
potatoes were not affected by phosphorus fertilisation when 0, 20, 40 and 60 kg ha-1 of
phosphorus were applied on vertisols of Debre Berhan in the highlands of central
Ethiopia (Zelalem, 2009). According to (Laboski and Kelling, 2007), phosphorus
applications may improve specific gravity when soil test phosphorus levels are low. A
high potassium level has been associated with slightly lower specific gravity (Smith,
1987). Potassium did not influence tuber specific gravity and dry matter content when 0-
166 kg K ha-1 was applied (Kumar et al., 2004).
Fertiliser application to nutrient deficient soils will usually improve dry matter content.
Tubers from nutrient-deficient plants are usually lower in dry matter content. Excessive
nitrogen also results in potato tubers having much lower dry matter content (Smith
(1987); Allison et al. (1999); Casa et al. (2005)) but this can be improved by phosphorus
application (Smith, 1987). Hollyer et al. (2000) found similar results in taro that the more
nitrogen, the lower the dry matter. Deficiency of phosphorus and potassium, and excess
potassium fertilisation also lowers dry matter content in potatoes (Smith, 1987;
Perrenoud, 1993). Starch content was found to decrease with increasing nitrogen fertiliser
rates (Shan et al., 2004).
Nutrient deficiency stresses raise sugar levels. Nutrient deficiency leads to high sugar
content (Smith, 1987). Nitrogen excesses have been shown to delay tuber maturity and
this lead to higher reducing sugars content in the harvested tubers (Smith, 1987; Allison
et al., 1999). Potassium application has been found to reduce reducing sugar content of
18
tubers (Kader and Rolle, 2004) but potassium did not influence reducing sugars when 0-
166 kg K ha-1 was applied (Kumar et al., 2004).
Colla et al. (2005) reported increased nitrogen concentration of tubers with nitrogen
increases in the fertiliser rate in agreement with the findings of Meyer and Marcum
(1998). Millard (1985) also reported increased tuber nitrogen concentrations from 0.68-
0.81 to 1.27-1.49% dry matter with increasing nitrogen fertiliser over the range 0-250 kg
ha-1. This also means increased protein content since protein content is calculated from
the nitrogen content.
2.4 Shelf life
Shelf life is crucial in the use of root parts as raw materials for processing. In potatoes,
crisping industries use stored tubers for most of the processing since potatoes are not
harvested throughout the year (Smith, 1987). Likewise, taro is harvested from August to
February in Umbumbulu and there is a need therefore to store it for the five months when
it is not harvested. Proper storage management helps retain the quality standard important
to processing. The major quality concerns for stored potatoes destined for processing are
specific gravity, dry matter content, starch content, alpha amylase activity, sprouting and
sugar accumulation (Smith, 1987). The ideal taro cultivars should be able to store and
maintain the quality of corms for crisping as long as possible. In potatoes, tubers must
also exhibit long dormancy and amenable to storage at low temperatures for up to 240
days (Williams and Cobb, 1993).
2.4.1 Factors affecting shelf life of taro
The shelf life of root crops is influenced among others by genotype, storage temperature
and packaging (Kay, 1987; Brown et al., 1990; Yosuke et al., 2000). The chemical
composition of potato tubers changes during storage (Lisińska and Laszczyński, 1989).
According to Bowers et al. (1964) corms with high specific gravity tend to have the
longer the shelf life. Smith (1987) stated that in Russet Burbank potatoes, less glucose
19
tend to accumulate in storage when tubers have a high specific gravity. Biemelt et al.
(2000) reported that alpha-amylase did not change significantly during storage and that it
was stimulated by 30% after the onset of sprouting. They therefore concluded that starch
degradation is not a prerequisite for the initiation of sprouting since no change in starch
degrading enzymes was detected prior to visible sprout growth. High levels of calcium in
tubers can improve its storability by reducing respiration rates (Ozgen et al., 2003; Kader
and Rolle, 2004).
2.4.1.1 Storage temperature
Temperature is the most important environmental factor that influences the deterioration
of harvested commodities (Kader and Rolle, 2004). It has great influence on many factors
that cause loss during storage (Diop, 1998). Low temperature storage causes tubers to
develop reducing sugars and decrease starch content hence become unsuitable for
processing into crisps (Wong Yen Cheong and Govinden, 1998; Pal et al., 2008). This
was confirmed by Yosuke et al. (2000) by reporting that sugar accumulation tends to
increase and starch content to decrease more at a lower temperature compared to a higher
temperature. Agbor-Egbe and Rickard (1990) reported a significant reduction to 39.8-
47.4 g kg-1 in starch content and increase to 8.0-11.6 g kg-1 in total sugars content to
occur during storage of cormels for two weeks under tropical ambient conditions (24-
29°C; 86-98% RH).
Higher temperatures, on the other hand, cause sprouting and weight loss to occur earlier
(Wong Yen Cheong and Govinden, 1998; Pal et al., 2008). Sprouting can result in
dramatic loss as the stored starch is transformed to sugars and utilized by the elongating
shoots with appreciable loss of both food value and moisture (FAO, 1981). Storage at
20oC, particularly under high relative humidity conditions resulted in excessive sprouting
(Midmore et al., 2006). This was in line with what was reported earlier by Wong Yen
Cheong and Govinden (1998) that tubers stored at 8-10 °C or at ambient temperature lost
much weight because of excessive sprouting as from about 12 weeks compared to those
stored at 2-4°C. High temperatures reduce specific gravity by increasing respiration rates
whereas low relative humidity increases specific gravity by allowing tubers to lose more
20
water. An increase in temperature causes an increase in the rate of natural breakdown of
all produce as food reserves and water content become depleted. High temperature can
greatly increase the loss of water from stored produce beyond that unavoidably lost from
natural causes (FAO, 1989).
Cocoyam has a considerably longer shelf life of several weeks and this can be extended
further with refrigerated storage (O'Hair, 1990). Agbo-Egbe and Rickard (1991) reported
taro to be successfully stored for 5-6 weeks at 15°C and 85% relative humidity. Midmore
et al. (2006), on the other hand reported storage of corms to be possible at 7–12°C in a
dark well ventilated room for up to eight weeks without quality compromise whereas
Hicks and Nguyen (2004) recommended storage of corms under the same conditions and
same time at 7–15°C. According to Opara (1998), if storage environment can be
maintained at 11-13°C and 85-90% relative humidity, the length of storage of taro can be
extended to about 150 days. This implies that storage life can be generally improved at
conditions of lower temperature and high humidity. The cooling of produce extends its
life by slowing the rate of breakdown. Losses have been decreased with improved storage
systems that control temperature and humidity. Maximum storage life can be achieved
by storing only undamaged produce at the lowest temperature and highest relative
humidity tolerable by the crop (FAO, 1989), however care should be taken not use very
low temperatures. For sweet potato, storage is particularly successful when temperature is
kept relatively cool, although if held below 12°C chilling damage can occur (FAO,
1981). A previous Japanese study by Rhee and Iwata (1982) had found taro showing
chilling at 4°C storage.
2.4.1.2 Packaging
Packaging methods affect storage of harvested crops by influencing air flow rates around
the commodities, thereby affecting temperature and relative humidity (Kader and Rolle,
2004). Kader and Rolle (2004) further stated that relative humidity can be controlled by
adding polyethylene liners in packing containers and using perforated polymeric films for
packaging. Aroids can be successfully stored in plastic bags. It was observed in Fiji that
21
packaging of taro in plastic bags enhances keeping quality (FAO, 1981). Compared to the
open box, the conditions created inside the bag reduce moisture loss. Quevedo et al.
(1991) reported that packing taro corms in plastic bags and closely tying the open end
with rubber bands reduced weight loss. Kay (1987) reported that the storage life of corms
in polyethylene bags was 26-40 days over those packed in cartons. Taros stored in
polyethylene bags showed a 6% loss in fresh weight and 50% decay (Passam, 1982).
2.5 Conclusion
It is clear from the preceding review of literature that the agronomy of taro is well
understood. However, the majority of agronomic knowledge about the crop was derived
from environments outside South Africa. Therefore, basic agronomic studies on taro
species occurring in South Africa are still relevant. With respect to quality attributes
associated with processing of taro, there is still a paucity of knowledge, hence the
tendency to use potato as a model crop in this review. The gap of knowledge about the
influence of agronomic practices, specifically fertilisation, on taro cormel quality is the
focus of the current study.
22
References
Agbor-Egbe, T. and Rickard, J.E. 1990. Evaluation of the chemical composition of fresh
and stored edible aroids. Journal of the Science of Food and Agriculture 53 (4):
487-495.
Agbor-Egbe, T. and Rickard, J.E. 1991. Study of the factors affecting storage of edible
aroids. Annals of Applied Biology 119: 121-130.
Ali, H. 1998. Studies on phosphorus nutrition of potatoes. Ph.D. Thesis. University of
Wales, Aberystwyth, UK.
Ali, H., Anjum, M.A. and Randhawa, S.A. 2004. Influence of phosphorus on yield
potential of potato (Solanum tuberosum L.) crops. International Journal of
Agriculture and Biology 6 (1): 165-167.
Allison, M.F., Allen, E.J. and Fowler, J.H. 1999. The nutrition of the potato crop. British
Potato Council Research Review 1991/1. BPC Oxford, UK.
Almekinders, C.J.M. and Struik, P.C. 1996. Shoot development and flowering in potato
(Solanum tuberosum L.). Potato Research 39 (4): 581-607.
Al-Omran, 1991. Effect of deficit irrigation on potatoes production. Journal of King Saud
University Agricultural science 3 (1): 139-147.
Alsharari, S.F., Alsadon, A.A. and Alharbi, A.R. 2007. Evaluation of drought tolerance
of potato cultivars under greenhouse conditions. Acta Horticulturae 747: 67-74.
Babaji, B.A., Amans, E.B., Chiezey, U.F., Falak, A.M., Tanimu, B. and Mukhtar, A.A.
2009. Unmarketable tuber yield and other agronomic parameters of four varieties
of Irish potato (Solanum tuberosum L.) as influenced by NPK fertilizer rate and
type of seed tuber at Samaru, Nigeria. Asian Journal of Crop Science 1(1): 26-33.
Bakayoko, S., Tschannen, A., Nindjin, C., Dao, D., Girardin, O. and Assa, A. 2009.
Impact of water stress on fresh tuber yield and dry matter content of cassava
(Manihot esculenta Crantz) in Côte d’Ivoire. African Journal of Agricultural
Research 4 (1): 21-27.
Bártová, V., Bárta, J., Diviš, J., Švajnar, J. and Peterka, J. 2009. Crude protein content in
tubers of starch processing potato cultivars in dependence on different agro-
ecological conditions. Journal of Central European Agriculture 10 (1): 57-66.
23
Bazzaz, F. and Sombroek, W. 1996. Global climate change and agricultural production.
Direct and indirect effects of changing hydrological, pedological and plant
physiological processes. John Wiley & Sons Ltd. Baffins Lane, Chichester.
Biemelt, S., Hajirezaei, M., Hentschel, E. and Sonnewald, U. 2000. Comparative analysis
of abscisic acid content and starch degradation during storage of tubers harvested
from different potato varieties. Potato Research 43 (4): 297-393.
Bowers, F.A., Plucknett, D.L. and Young, O.R. 1964. Specific gravity evaluation of corm
quality in dasheen. Circular 61. Hawaii Agricultural Experimental Station,
College of Tropical Agriculture and Human Resources, University of Hawaii.
Bradbury, J. H. and Nixon, R. W., 1998. The acridity of raphides from the edible aroids.
Journal of the Science of Food and Agriculture 76 (4): 608-616.
Brown, J., Mackay, G.R., Bain, H., Griffith, D.W. and Allison, MJ. 1990. The
Processing potential of tubers of the cultivated potato, Solanum tuberosum L.,
after storage at low temperature. 2. Sugar concentration. Potato Research 33:
219 - 227.
Carputo, D., Aversano, R. and Frusciante, L. 2005. Breeding potato for quality traits.
Acta Horticulturae 684: 55-64.
Casa, R., Pieruccetti, F., Sgueglia, G. and LoCascio, B. 2005. Potato tuber quality
improvement through nitrogen management optimization: Review of
methodologies. Acta Horticulturae 684: 65-71.
Chan, L.F., Lu, C.T., and Lu, H.Y. 1998. Growth analysis of wetland taro [Colocasia
esculenta (L.) Schott] under various crop seasons. Journal of Agricultural
Research of China 47: 220–241.
Chapman, K.S.R., Sparrow, L.A., Hardman, P.R., Wright, D.N. and Thorp, J.R.A. 1992.
Potassiun nutrition of Kennebec and Russet Bursank potatoes in Tasmania:
Effect of soil and fertilizer potassium on yield, petiole and tuber potassium
concentrations, and tuber quality. Australian Journal of Experimental Agriculture
32: 521-527.
Chun-Feng, S. and Kun, X. 2004. Effect of nitrogen and potassium on yield and quality
of taro. Plant nutrition and fertilizer science 10 (2): 167-170.
24
Colla, G., Cardarelli, M., Ruophael, Y., Fiorillo, A. and Cirica, B. 2005. Response of
potato to organic fertiliser rate: yield, tuber quality and nutrient uptake.
Dahlenburg, A.P., Maier, N.A. and Williams, C.M.J. 1990. Effect of nitrogen on the size,
specific gravity, crisp colour and reducing sugar concentration of potato tubers
(Solanum tuberosum L.) cv Kennebec. Australian Journal of Experimental
Agriculture 30: 123-130.
Debon, S.J.J. and Tester, R.F. 2000. “In vivo” and “In vitro” annealing of starches. In:
P.A. Williams and G.O. Phillips, Editors, Gums and stabilizers for the food
industry, The Royal Society of Chemistry, Cambridge (2000), pp. 270–276.
Diop, A. 1998. Storage and Processing of Roots and Tubers in the Tropics. FAO
Corporate Document Repository. Rome. FAO.
Dubetz, S and Bole, J.B. 1975. Effect of nitrogen, phosphorus and potassium fertilizer on
yield components and specific gravity of potatoes. American Potato Journal 52:
399-405.
Duffus, C.M. and Duffus, J.H. 1984. Carbohydrate metabolism in plants. Longman Inc.,
New York. USA.
Eskin, N.A.M. 1989. Quality and preservation of vegetables. CRS Press, Florida, USA.
FAO. 1981. Food loss prevention in perishable crops. 1981. FAO Agricultural Services
Bulletin No.43.
FAO. 1989. Prevention of post-harvest food losses fruits, vegetables and root crops: A
training manual. FAO Training Series No.17/2.
Feibert, E.B.G., Shock, C.G. and Saunders, L.D. 1998. Nitrogen fertilizer requirements of
potatoes using carefully scheduled sprinkler irrigation. Horticultural science 33:
262-265.
Gebhardt, C., Menendez, C., Chen, X., Li, L. Schäfer-Pregl, R. and Salamini, F. 2005.
Genomic Approaches for the Improvement of Tuber Quality Traits in Potato
25
Ghosh, S.C., Asanuma, K., Kusutani, A. and Toyota, M. 2000. Effects of temperature at
different growth stages on nonstructural carbohydrate, nitrate reductase activity
and yield of potato. Environment Control in Biology 38 (4): 197-206.
Goenega, R. 1995. Accumulation and partitioning of dry matter in taro [Colocasia
esculenta (L.) Schott]. Annals of Botany 76: 337-341.
Hammer, G.L., Hobman, F.R. and Shepherd, R.K. 2007. Effects of planting time and
harvest age of cassava (Manihot esculenta) in Northern Australia. I. Crop growth
and yield in moist environments. Environmental Agriculture 23: 401- 414.
Hicks, D. and Nguyen, V.Q. 2004. Japanese taro. In Asian vegetables: The new industries
handbook, RIRDC.
Hollyer, J., Robert Paull, R. and Huang, A. 2000. Processing Taro Chips.
Islam, S., Porter, O.A., Izekor, S.E. and Garner, J.O. 2007. Quality characteristics and
yield attributes of sweetpotato (Ipomoea batatas L.) genotypes grown under
sustainable cultural conditions. International Journal of Food, Agriculture and
Environment 5: 220-224.
Ivonyi, I., Izsoki, Z. and van der Werf, H.M.G. 1997. Influence of nitrogen supply and P
and K levels of the soil on dry matter and nutrient accumulation of fiber hemp
(Cannabis sativa, L.). Journal of the International Hemp Association 4: 84-89.
Jenkins, P.D. and Ali, H. 2000. Phosphorus supply and progeny tuber numbers in potato
crops. Annals of Applied Biology 136: 41-46.
Kader, A.A. and Rolle, R.S. 2004. The role of post-harvest management in assuring the
quality and safety of horticultural produce. FAO Agricultural Services Bulletin
152.
Karam, F., Lahoud, R., Masaad, R. and Stephan, C. 2005. Yield and tuber quality of
potassium treated potato under optimum irrigation conditions. Acta Horticulturae
684: 103-108.
Karam, F., Rouphael, Y., Lahoud, R., Breidi, J. and Colla, G. 2009. Influence of
genotypes and potassium application rates on yield and potassium use efficiency
of potato. Journal of agronomy 8 (1): 27-32.
26
Kay, D.E. 1987. Crop and product digest No. 2 Root crops. 2nd Edition. London:
Tropical Development and Research Institute. 380 pp.
Kays, S.J. 1991. Post harvest physiology of perishable plant products. Van Nostrand
Reinhold, New York.
Khan, I.A., Deadman, M.L., Al-Nabhani, H.S. and Al-Habsi, K.A. 2003. Interactions
between temperature and yield components in exotic potato cultivars grown in

Oman. Acta Horticulturae 619: 353-359.
Kita, A. 2002. The influence of potato chemical composition on crisp texture. Food
Chemistry 76 (2): 173-179.
Kumar, P., Pandey, S. K., Singh, S. V., Rawal, S. and Kumar D. 2004. Effect of
potassium fertilization on processing grade tuber yield and quality parameters in
potato (Solanum tuberosum). Indian Journal of Agricultural Science 74 (4): 177-
179.
Kumar, S., Khade, H.D., Dhokane, V.S., Behere, A.G. and Sharma, A. 2007. Irradiation
in combination with higher storage temperatures maintains chip-making quality
of potato. Journal of food science 72(6): 402-406.
Kumar, B. and Ramesh, B. (undated). Activity of alpha-amylase in induced mutants of
barley, Department of Genetics & Plant Breeding, Ch. Charan Singh University,
Meerut (U P) – 250004, India.
Laboski, C.A.M. and Kelling, K. A. 2007. Influence of fertilizer management and soil
fertility on tuber specific gravity: A review. American Journal of Potato
Research 84 (4): 283-290.
Laurence, R.C.N., Armour, J.D., Shepherd, R.K., Loader, L.R. and Dwyer, M.J. 1985.
Nitrogen fertiliser requirements of irrigated potatoes on the Atherton Tableland,
North Queensland. Australian Journal of Experimental Agriculture 25: 954-958.
Lisińska, G. and Laszczyński, W. 1989. Potato Science and technology. Elsevier Science
Publishers Ltd, England.
27
Long, C.M., Snapp, S.S., Douches, D.S. and Chase, R.W. 2004. Tuber yield, storability,
and quality of Michigan cultivars in response to nitrogen management and
seedpiece spacing. American Journal of Potato Research 81 (5): 347-357.
Lu, H.Y., Lu, C.T., Chan, L.F. and Wei, M.L. 2001. Seasonal Variation in linear increase
of taro harvest index explained by growing degree days. Agronomy Journal
93: 1136-1141.
MacGregor, A.W., 1983. Cereal alpha-amylases: Synthesis and action pattern. In: Seed
Proteins (J. Daussant; J. Mosse and J. Vaughan, eds.). Annual Proceedings of the
Phytochemical Society of Europe, 20, 1-34. Academic Press, London.
Maier, N.A., Dahlenburg, A.P. and Williams, C.M.J. 1994. Effects of nitrogen,
phosphorus and potassium on yield, specific gravity, crisp colour and tuber
chemical composition of potato (Solanum tuberosum L.) cv. Kennebec.
Australian Journal of Experimental Agriculture 34: 813-824.
Schott] landraces from Umbumbulu. MSc. Thesis. University of KwaZulu-Natal,
Pietermaritzburg, South Africa.
Martin, R.J., Scheffer, J.J.C., Deo, B. and Halloy, S.R.P. 2005. Effect of planting date on
Ulluco yield. Acta Horticulturae 670: 181-187.
Martin-Prevel, P.J. 1989. Physiological processes related to handling and storage quality
of crops. P. 219-248. in: Proceedings of the 21st IPI Colloquium on Methods of K
research in plants, Louvain-la-Neuve, Belgium, 19-21 June 1989. Institute of
Potash Institute, Bern, Switzerland.
Mazid, M.A. 1993. Sulfur and nitrogen for sustainable rainfed lowland rice. PhD Thesis,
University Philippines at Los Banos. 1-232.
McDole, R.E., Stallknecht, G.F., Dwelle, R.B. and Pevek, J.J. 1978. Response of four
potato varieties to potassium fertilization in a seed growing area of eastern Idaho.
American Potato Journal 55: 495-504.
Meyer, R.D. and Marcum, D.B. 1998. Potato yield, petiole nitrogen and soil nitrogen
response to water and nitrogen. Agronomy Journal 90: 420-429.
28
Midmore, D., White, D., Nguyen, V., Hicks, D., Coleman, E., Newman, S., Wilk, P.,
Reeve, D. and McLaughlin, P. 2006 Development of taro, yam, yam bean and
sweet potato exports to Japan and USA. RIRDC Publication No 06/101.
Millard, P. 1985. The nitrogen content of potato (Solanum tuberosum L.) tubers in
relation to nitrogen application - the effect on amino acid composition and yields.
Journal of the Science of Food and Agriculture 37 (2): 107 – 114.
Mittra, B. N. and George, J. 2000. Effect of time of planting on the growth and dry matter
partitioning in sweet potato. Journal of Root Crops 26 (2): 57-60.
Miyasaka, S.C., Hollyer, J.R. and Cox, L.J. 2001. Impacts of organic inputs on taro
production and returns. CTAHR SCM-3.
Modi, A.T. 2003. What do subsistence farmers know about indigenous crops and organic
farming? Preliminary case in KwaZulu-Natal. Development Southern Africa
20: 673-682.
Modi, A.T. 2007. Effect of indigenous storage method on performance of taro [Colocasia
esculenta (L.) Schott] under field conditions in a warm subtropical area. South
African Journal of Plant and Soil 24: 214-219.
Mosley, A. R. and Chase, R. W. 1993. Selecting cultivars and obtaining healthy seed lots.
Pages 20-21 in: Potato Health Management. R. C. Rowe, ed. The American
Phytopathological Society, St. Paul, MN.
Musa, F., Berisha, D., Kelmendi, B., Rusinovci, I., Zhitia, O., Bekqeli, R. and Lushi, I.
2009. Tuber yield and other relevant parameters of some Netherlands potato
varieties in agro-climatic conditions of Kosova. 44th Croatian & 4th International
Symposium on Agriculture. Faculty of Agriculture. University of J.J. Strossmayer
in Osijek.
Nandutu, A., Carasco, J. and Hagenimana, V. 2000. Using sweet potato amylase extracts
for the determination of starch in foodstuffs. The Journal of food technology in
Africa 5 (2): 66-69.
Ndimantang, B., Asinobi, C. O., Obiakor, N. 2006. The effect of different processing
methods on some anti-nutritional factors content of Ede uhie (Xanthosoma
sagittifolium) and Ede ocha (Colocasia esculenta). International Journal of
Agriculture and Rural Development. ISSN: 1595 – 9716.
29
O'Hair, S.K. 1990. Tropical Root and Tuber Crops. p. 424-428. In: J. Janick and J.E.
Simon (eds.), Advances in new crops. Timber Press, Portland, OR.
O`Keefe, O., Mason, G., Willis, A. and Bell, G. 2005. Commercial taro chip development
using agri-chain partnerships: A processing, marketing and financial analysis.
RIRDC Publication No 05/144. RIRDC Project No DAQ-296A.
Onwueme, I. 1999. Taro Cultivation in Asia and the Pacific. FAO RAP Publication:
1999/16.
Opara, L.U. 1998. Edible aroids: Post harvest operation. Rome. FAO.
Osiru, M.O, Olanya, O.M., Adipala, E., Kapinga, R. and Lemaga, B. 2009. Yield stability
analysis of Ipomoea batatus L. cultivars in diverse environments. Australian
Journal of Crop Science 3 (4): 213-220.
Oya, K. 1972. Evaluation of potassium availability of four Michigan soils. The science
bulletin of the College of Agriculture, University of Ryukyus 19: 123-257.
Ozgen, S., Palta, J.P and Kleinhenz, M.D. 2003. Influence of supplemental calcium
fertilization on potato tuber size and tuber number. Acta Horticulturae 619: 329-
336.
Pal, S., Bhattacharya, A., Konar, A., Mazumdar, A. and Das, A.K. 2008. Chemical
composition of potato at harvest and after cold storage. International Journal of
Vegetable Science 14 (2): 162 – 176.
Panhwar, F. 2005. Potato (Solanum tuberosum L.) – Cultural practice in Sindh Pakistan.
Digitalverlag GmbH, Germany.
Passam, H.C. 1982. Experiments on the storage of eddoes and tannia (Colocasia and
Xanthosoma spp.) under tropical ambient conditions. Tropical Science 24 (2) 39-
46.
Perrenoud, S. 1993. Fertilizing for high yield potato. IPI Bulletin 8, 2nd Edition.
International Potash Institute, Basel, Switzerland.
Porter, G.A. and Sisson, J.A. 1991. Response of Russet Burbank and Shepody potatoes to
nitrogen fertilizer in two cropping systems. American Potato Journal 68: 425-
443.
30
Proietti, S., Moscatello, S., Battistelli, A., Mauromicale, G. and Ierna, A. 2005. Quality of
early potato in Sicily as affected by cultivar, sowing time and irrigation. Acta
Horticulturae 684: 171.
Quevedo, M.A., R.T. Sanico, M.E. Baliao. 1991. The control of post-harvest diseases of
taro corms. Tropical Science 31: 359.
Ramachandran, S., Patel, A.K., Nampoothiri, K.M., Chandran, S., Szakacs, G.,
Soccol, C.R. and Ashok Pandey, A. 2004. Alpha-amylase from a fungal
culture grown on oil cakes and its properties. Brazilian Archives of
Biology and Technology 47 (2): 309-317.
Rashid, M.M. and Daunicht, J.J. 1979. Chemical composition of nine edible
aroid cultivars of Bangladesh. Scientia Horticulturae 10 (2): 127-134
Reyes - Castro. G., Nyman, M. and Rönnberg-Wästljung, A.C. 2005.
Agronomic performance of three cocoyam (Xanthosoma violaceum
Schott) genotypes grown in Nicaragua. Euphytica 142 (3): 265-272.
Rhee, J.K. and Iwata, M. 1982. Histological observations on the chilling injury of taro
tubers during cold storage. Journal of the Japanese Society for Horticultural
Sciience 51 (3): 362-368.
Roe, M.A., Faulks, R.M. and Belsten, J.L. 1990. Role of reducing sugars and
amino acids in fry colour of chips from potatoes grown under different
nitrogen regimes. Journal of the Science of Food and Agriculture 52:
207–214.
Rykaczewska K. 2004. Effect of high temperature during vegetation on potato
(Solanum tuberosum L.) yield, period of tuber dormancy and seed tuber
yielding ability. Part I. Plant development and yield. Zeszyty
Problemowe Postepow Nauk Rolniczych 496: 185-198.
Scheffer, J.J.C., Douglas, J.A. and C.M. Triggs, C.M. 2005. Factors Affecting
the Production and Quality of Japanese Taro Cormels. Acta
Horticulturae 670: 167-172.
31
Shan, L., Sattelmacher, B., Kutzmutz, E., Mühling, K. H. and Dittert, K. 2004.
Influence of nitrogen nutrition on tuber quality of potato with special
reference to the pathway of nitrate transport into tubers. Journal of Plant
Nutrition 27 (2): 341-350.
Shange, L.P. 2004.Taro (Colocasia esculenta (L.) Schott) production by small-
scale farmers in KwaZulu-Natal: Farmer practices and performance of
propagule types under wetland and dryland conditions. MSc. Thesis.
University of KwaZulu-Natal, Pietermaritzburg.
Sharma, U.C. and Arora, B.R. 1987. Effect of nitrogen, phosphorus and
potassium application on yield of potato tubers (Solanum tuberosum L.).
Journal of Agricultural Science 108: 321-329.
Singh, J.P., Marwaha, J.S. and Grewal, J.S. 1996. Effect of sources and levels
of potassium on potato yield, quality and storage behaviour. Journal of
Indian Potato Association 23: 153-156.
Singh, U., Matthews, R.B., Griffin, T.S., Ritchie, J.T., Hunt, L.A. and
Goenaga, R. 1998. Modeling growth and development of root and tuber
crops. In: Understanding options for agricultural production. Ed Tsuji,
G.Y., Hoogenboom, G. and Thornton, P.
Silva, E .E. da, Azevedo, P. H. S de, Almeida, M. M. T. B., De-Polli, H. and Guerra, J. G.
M. 2008. Influence of intercropping and irrigation frequencies in the leaf
development and taro (Cocoyam) productivity under organic management. 16th
IFOAM Organic World Congress, Modena, Italy, June 16-20, 2008.
Sivan, P. 1982. Growth and development of taro under dryland conditions in Fiji. p. 637–
646. In Proc. Int. Symp. Trop. Root and Tuber Crops, 5th, Los Banos, Philippines.
17–21 Sept. 1979. Philippine Council for Agriculture and Resource Research. Los
Banos, Philippines.
Smith, O. 1987. Potato chips. In: W. Talburt and O. Smith, Editors, Potato processing,
Van Nostrand Reinhold Company Inc, New York.
Sparrow, L.A. and Chapman, K.S.R. 2003. Effects of nitrogen fertilizer on potato
(Solanum tuberosum L., cv Russet Burbank) in Tasmaria. 1. Yield and quality.
Australian Journal of Experimental Agriculture 43: 631-641.
32
Talburt, W.F. and Smith, O. 1987. Potato Processing 4th edition, NY. 491-534.
Tawfik, A.A. 2001. Potassium and calcium nutrition improves potato production in drip-
irrigated sandy soil. African Crop Science Journal 9 (1): 147-155.
Tumuhimbise, R., Talwana, H.L., Osiru, D.S.O., Serem, A.K., Ndabikunze, B.K., Nandi,
H. and Palapala, V. 2007. Growth and development of wetland-grown taro under
different plant populations and seedbed types in Uganda. African Crop Science
Journal 17: 49 – 60.
Westermann, D.T., Tindall, T.A., James, D.W. and Hurst, R.L. 1994. Nitrogen and
potassium fertilization of potatoes: yield and specific gravity. American Potato
Journal 71: 417-432.
Williams, R.O. and Cobb, A.H. 1993. The relationship between storage temperature,
reconditioning and tuber storage. Potato Research 36: 373-392.
Wills, R.B.H., Lim, J.S.K., Greenfield, H. and Bayliss-Smith, T. 1983. Nutrient
composition of taro (Colocasia esculenta) cultivars from the Papua New Guinea
highlands. Journal of the Science of food and Agriculture 34 (10): 1137 – 1142.
Witt, W. and Sauter, J.J. 1996. Purification and properties of a starch granule-degrading
alpha-amylase from potato tubers. Journal of experimental botany 47 (304): 1789-
1795.
Wolf, S., Marani, A. and Rudich, J. 1990. Effects of temperature and photoperiod on
assimilate partitioning in potato plants. Annals of Botany 66: 513-520.
Wong Yen Cheong, J.K.C and Govinden, N. 1998. Quality of potato during storage at
three temperatures. AMAS 1998. Food and Agricultural Research Council, Réduit,
Mauritius.
Xiong, X. and Tai, G.C.C. 2003. Comparison of specific gravity profiles inside tubers of
potatoes. Acta Horticulturae 619:
Yosuke, M., Yaptenco, K.F., Tomohiro, N., Toshiro, S., Hiroaki, S., Shinji, M. and
Katsumi, T. 2000. Property changes in potato tubers (Solanum tuberosum L.)
during cold storage at 0 and 10oC. Food Preservation Science 26 (3): 153-160.
Yuan, B.Z., Nishiyama, S. and Kang, Y. 2003. Effects of different irrigation regimes on
the growth and yield of drip-irrigated potato. Agricultural Water Management 63:
153-167.
33
Zelalem, A., Tekalign, T. and Nigussie, D. 2009. Response of potato (Solanum
tuberosum L.) to different rates of nitrogen and phosphorus fertilization on
vertisols at Debre Berhan, in the central highlands of Ethiopia. African Journal of
Plant Science 3 (2): 16-24.
34
CHAPTER 3
GROWTH, DEVELOPMENT AND YIELD OF TARO IN RESPONSE TO

PLANTING DATE AND FERTILISATION
Abstract
Despite the importance of taro (Colocasia esculenta L. Schott) as a food security crop,
scientific research on it is scanty in South Africa. Production site, planting date and
fertiliser regime affect crop performance, particularly that of cultivars, because they tend
to be adapted to specific localities. To investigate performance of three taro cultivars, a
study was carried out at two sites in KwaZulu-Natal, South Africa (Ukulinga and
Umbumbulu), during the 2007/2008 growing season. A split-split-plot design was used
with four planting dates (October 2007, November 2007, December 2007 and January
2008) as main plots, three taro cultivars (Dumbe-dumbe, Mgingqeni and Pitshi) as sub-
plots and three organic fertiliser (Gromor Accelerator) application rates (0, 160 and 320
N kg ha-1) as sub-sub-plots. Treatments were replicated three times. Delaying planting
significantly decreased crop emergence (%), leaf number plant-1, plant height plant-1, leaf
area plant-1, number of cormels plant-1 and fresh cormel mass plant-1 at both sites.
Mgingqeni had the highest emergence for earlier planting dates. Dumbe-dumbe plants
were taller for November than for October planting when 160 N kg ha-1 was applied at
Ukulinga. Application of Gromor Accelerator enhanced leaf number plant-1, plant height
plant-1, leaf area plant-1, number of cormels plant-1 and fresh cormel mass plant-1 at
Umbumbulu. Taro cultivars responded differently at both sites. Dumbe-dumbe had
significantly lowest number of cormels and lowest proportion of cormels falling under
the small size class. In this study, environmental variability due to site and planting date;
and fertilisation had an effect on taro growth and yield. It is recommended that taro
producers should plant in October at both sites with no fertiliser at Ukulinga and 160 N
kg ha-1 of Gromor Accelerator at Umbumbulu.
Key Words: Colocasia esculenta, fertilisation, planting date, site, yield.
35
3.1 Introduction
Taro (Colocasia esculenta L. Schott) is an important crop grown for its starchy corms
and it is a traditional staple food throughout the rural subtropical and tropical regions of
the world, most notably in the Pacific and Caribbean islands and West Africa (Vinning,
2003; Hancock, 2004). In South Africa, it is grown in the subtropical coastal area,
starting at Bizana district in the Eastern Cape and the rest of the coastal KwaZulu-Natal.
The crop is also cultivated in the subtropical and tropical parts of Mpumalanga and
Limpopo provinces (Shange, 2004). Taro is grown organically under dryland conditions
from July to January by the EFO (Ezemvelo Farmers` Organisation) farmers who market
it on contract to two large South African grocery store chains; Pick ‘n’ Pay and
Woolworths (Modi, 2003; Agergaard and Birch-Thomsen, 2006; Mare, 2006).
According to Miyasaka et al. (2003) taro grown under rainfed conditions can take 6 to 12
months to reach maturity.
Root and modified stem crop yield is a positive function of the number of
roots/tubers/corms and weight of roots/tubers/corms per planting hill (Khan et al., 2003;
Kader and Rolle, 2004). The number and size of tubers vary with cultivar (Khan et al.,
2003). Planting date and fertiliser regime affect crop performance, regardless of whether
the crop is a landrace or an improved cultivar. Yield of root crops has been shown to be
affected by water availability and temperature at different sites and planting dates (Lu et
al., 2001; Khan et al., 2003; Scheffer et al., 2005; Kumar et al., 2007; Hagman et al.,
2009). Time of planting and fertilisation of taro are, therefore, likely to be critical factors
affecting crop performance in response to agronomic practices. Tuber total yields were
found to be significantly higher in planting date four compared to other planting dates
when potatoes were planted on 11 January 2005 and every two weeks until 22 March
2005 in Florida (Worthington and Hutchinson, 2006). Sangakkara (1993) reported
highest yields when sweet potatoes were planted with the onset of rains in October in
India and this was attributed to adequate rainfall and higher diurnal variation in
temperatures over the growth cycle. Miyasaka et al. 2003 confirmed this and reported
that corm fresh weight increased significantly when taro was planted in Spring and
36
Summer compared to Winter and fall plantings. Supplementary irrigation gave a total
yield of 62t ha-1 compared to 26 t ha-1 under rainfed conditions (Scheffer et al., 2005).
This confirmed the findings of Shock (1992), Bussell and Bonin (1998), Bèlanger et al.
(2002) and Yuan et al. (2003) that supplementary irrigation had positive effect on total
fresh tuber yields, and that it increased yields by increasing tuber number and mean
weight of the tubers. Increased corm weight in the high watering-level treatment was
attributed to the earlier increased leaf and root growth (Bussell and Bonin, 1998). Wolfe
et al. (1983), on the other hand, reported a slight yield depression in the wettest
treatment, possibly due to poor soil aeration and nitrogen deficiency near the end of the
season. Higher temperatures were found to speed up development between emergence
and tuber initiation whereas total and tuber dry mass and canopy leaf area were reported
to decrease with increasing temperatures in potatoes (Kooman et al., 1996, Fleisher et al.,
2006). Timlin et al. (2006) also found end-of-season tuber mass to decrease with
increasing temperature above 24oC in potatoes.
Nitrogen fertilisation was reported to increase the average fresh tuber weight with distinct
effects for one cultivar than for the other (Bèlanger et al., 2002). According to Kumar et
al. (2007) plant height, leaf number and tuber weight per plant responded positively to
nitrogen application. Leaf area increased with increase in nitrogen and then decreased at
4.0 mM nitrogen in taro (Osorio et al., 2003). O'Beirne and Cassidy (1990) also reported
yield to increase with the use of 100-150 kg ha-1 nitrogen after which it remained
constant. In contrast to this, it was found that nitrogen fertilisation had a negative effect
on time of emergence and little or no effect on yield whereas phosphorus and potassium
increased yields (Wadas et al., 2004; Hagman et al., 2009). Zelalem et al. (2009) on the
other hand reported that average number and tuber weight were not influenced by
phosphorus fertilisation.
Since tuber yield is influenced by temperature and moisture, fertilisation and cultivar,
planting taro at appropriate times and using suitable fertilisation and cultivars can
enhance taro growth and hence yields. The objective of this study was to determine the
37
growth and yield of three taro cultivars grown at different planting dates using different
organic fertiliser regimes in KwaZulu-Natal, South Africa.
3.2 Materials and methods
3.2.1 Planting material
Three taro cultivars of the eddoe type Dumbe-dumbe, Mgingqeni and Pitshi were used for
this study (Fig. 3.1). Cormel size for the taro cultivars vary between and within cultivars.
The cormels planted for this study were selected to be in the following size ranges
(cormel -1): 21 – 60 g for Dumbe-dumbe, 11 – 40g for Mgingqeni and 5 – 20g for Pitshi.
The cormels were 2007 harvest from Ukulinga except for Dumbe-dumbe which was
obtained from farmers of Ezigeni, Umbumbulu district, KwaZulu-Natal, South Africa.
Dumbe-dumbe Mgingqeni Pitshi
Figure 3.1 Taro cultivars of KwaZulu-Natal.
3.2.2 Site description
Field experiments were carried out at two sites, Ukulinga (29o 37′S 30o 16′E) and
Umbumbulu (29o 36′S 30o 25′E) with different environmental conditions (Table 3.1) and
soil physical and chemical properties (Table 3.2). Temperature and rainfall data for both
sites were obtained from Weather South Africa for the past 10 years (Figs. 3.2, 3.3, 3.4
and 3.5) and for the duration of the experimental period (Table 3.1). Soil samples were
collected from both sites and analysed according to Naramabuye et al. (2008). The pH of
the soils from both sites (Table 3.2) were below 6 to 6.5 which is the soil pH at with taro
38
is known to grow best. Addition of organic manure was expected to increase the soil pH
as it was reported by Pocknee and Summer (1997), Wong et al. (1998), Mokolobate and
Haynes (2002) and Naramabuye et al. (2008) that organic manure has a rapid liming
effect. However, no soil analysis was performed at harvest.
1997 1998
25 1999 2000
Mean temperature ( C)
2001 2002
2003 2004
o
23 2005 2006
21
19
17
15
13
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Figure 3.2 Mean temperature distribution at Ukulinga for a 10 year period to 2006.
27
1997 1998
1999 2000
Mean temperature ( C)
25 2001 2002
2003 2004
o
2005 2006
23
21
19
17
15
Month
Figure 3.3 Mean temperature distribution at Umbumbulu for a 10 year period to 2006.
39
400
1997 1998
1999 2000
2001 2002
2003 2004
Mean rainfall (mm)
300
2005 2006
200
100
0
Month
Figure 3.4 Mean rainfall distribution at Ukulinga for a 10 year period to 2006.
300
1997 1998
1999 2000
2001 2002
Mean rainfall (mm)
2003 2004
2005 2006
200
100
0
Month
Figure 3.5 Mean rainfall distribution at Umbumbulu for a 10 year period to 2006.
40
Table 3.1 Mean temperature and mean rainfall data for Ukulinga and Umbumbulu for the duration of the experimental period.
Site Total Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sept
Ukulinga Rainfall(mm) 883.4 171.2 159 58.2 178.2 78.6 77.4 86.2 0 17 0 4 53.6
Temperature (oC) 19.6 18.6 19.6 21.9 23.7 24.3 22.7 19.3 18.8 15.3 15.4 18 18.4
Umbumbulu Rainfall(mm) 1334.1 139.5 301.2 57.3 133.8 78.7 196.2 122.1 28.4 171.2 3.5 1.8 100.4
Temperature (oC) 20.8 20.4 21.4 22.8 24.3 25 23.8 20.7 20 17.3 17 18.3 18.5
Table 3.2 Physical and chemical properties of soils from Ukulinga and Umbumbulu.
Site pH Sample Exch. Total P K Ca Mg Zn Mn Cu Acid MIR MIR MIR
-1
(KCl) density Acidity cations (mg L ) saturation Organic C clay N
-1 -1
(g ml ) (cmol L ) (%)
Ukulinga 5.07 1.08 0.05 25.22 10 223 2780 1303 4.6 27 7.1 0 1.8 31 0.25
Umbumbulu 4.16 0.96 1.68 6.72 3 97 658 183 1.3 8 5.4 25 4.5 59 0.31
*MIR – Mid infrared
41
3.2.3 Experimental design and agronomic practices
The experiment was designed as a split-split-plot design with four planting dates:
October 2007, November 2007, December 2007 and January 2008 (Table 3.3) as
main plots, three taro cultivars: Dumbe-dumbe, Mgingqeni and Pitshi (Fig. 3.1) as
sub-plots and three organic fertiliser (Gromor Accelerator) application rates (0, 160
and 320 N kg ha-1) as sub-sub-plots. The nutritional composition of Gromor
Accelerator is shown in table 3.3. Each planting date was replicated three times. Plot
size was 4 m2 containing 16 plants spaced 0.5m between and within rows. Sowing
was done by hand on ploughed and harrowed fields. Planting holes were opened with
a hand-hoe and organic fertiliser was mixed with soil before one cormel was planted
per hole. Weeds were controlled by hand-hoeing at one, two, three and four months
after planting.
Table 3.3 Nutritional composition of Gromor accelerator.

N P K Mg Ca S Fe Cu Zn B Mn Mo
(g kg-1) (mg kg-1)
30 15 15 5 20 0.6 2000 40 250 40 400 4
3.2.4 Data collection and analysis
Emergence percentage was calculated from the final emergence recorded four months
after planting. Plant vigour parameters (number of leaves plant-1, plant height plant-1
and leaf area plant-1) were determined from sampling unit (four innermost plants of
each experimental plot) at monthly intervals from one month after planting to four
months thereafter when positive change in growth was no longer occurring (Mare,
2006). Leaf area plant-1 (A) was determined non-destructively by regressing the
product of leaf length (L) and leaf width (W) (Loomis & Connor, 1992; Modi, 2007).
A = b (LW), where b is the coefficient of regression.
Yield was determined by the number of cormels plant-1 and fresh cormel mass plant-1
from sampling unit nine months after planting and cormels were weighed individually
and graded into sizes <20, 21-40, 41-60, 61-80, 81-100, 101-120, 121-140, 141-160
42
and 161-180 g. Analysis of variance was performed using GenStat Version 11.1
(2008). Least significant difference (LSD) was used to separate means. A probability
level of 0.05 was considered to be statistically significant.
3.3 Results
3.3.1 Emergence
The interaction of planting date, cultivar and organic fertilisation were observed to
play a significant role in taro emergence at Ukulinga (P = 0.006). At Umbumbulu,
emergence was affected by the interactions of planting date and cultivar (P < 0.001)
and cultivar and Gromor Accelerator application rate (P = 0.047). Emergence for
earlier planting dates was significantly higher than that of the latter planting dates at
both sites (Table 3.5 and Fig. 3.6). Taro cultivars showed the same trend of
emergence for October plantings at both sites but the trend changed at Umbumbulu
for the latter plantings, whereas at Ukulinga the same trend was displayed for other
planting dates. The different trends were probably due to the different response of
different taro cultivars to different sites. Application of the organic matter did not
enhance emergence of all cultivars at both sites (Table 3.4 and Fig. 3.7). Pitshi
generally displayed lowest emergence whereas Mgingqeni showed highest
emergence.
43
Table 3.4 Effect of planting date, Gromor Accelerator application rate and cultivar
on emergence at Ukulinga. Means of the interaction effect within columns and rows
(LSD(0.05) = 18.01) followed by the same letter are not significantly different
Planting date Gromor Taro cultivar

Accelerator Dumbe-dumbe Mgingqeni Pitshi
( N kg ha-1)
October 0 95.83k 100.00k 89.67jk
160 91.67k 100.00k 75.00ij
320 89.67 jk 100.00k 68.67hi
November 0 58.67ghi 100.00k 69.00hi
160 56.67gh 100.00k 75.33ij
320 52.00efgh 91.67k 67.00hi
December 0 16.83ab 48.00efg 6.67a
160 12.33ab 54.00fgh 11.17ab
320 18.83abc 37.67def 19.42abc
January 0 16.83ab 18.67abc 16.50ab
160 12.50ab 34.93cde 22.87abcd
320 16.67ab 25.20bcd 8.34ab
120 LSD (0.05) = 14.15

Dumbe-dumbe
e e
100 Mgingqeni
de Pitshi
Emergence (%)
80 cd
c
c c
60
b b
40 b
b
20 a
0
October November December January
Planting date
application rate on emergence at Umbumbulu.
44
120
LSD(0.05) = 8.60 Dumbe-dumbe
100 Mgingqeni
Pitshi
Emergence (%)
80 c
bc bc b b
bc
60
a a
40 a
20
0
0 160 320
-1
Gromor Accelerator (N kg ha )
Figure 3.7 Effect of cultivar and Gromor Accelerator application rate averaged
across planting date on emergence at Umbumbulu.
3.3.2 Growth parameters
3.3.2.1 Number of leaves
Figure 3.8 shows the number of leaves plant-1 of taro cultivars planted at four
different dates at Ukulinga. Planting date was observed to play a significant role in
number of leaves plant-1 (P < 0.001). The first two planting dates showed significantly
higher number of leaves than the latter planting dates. This is an indication that
delaying planting by two months at Ukulinga decreases the number of leaves plant-1.
Cultivar and Gromor Accelerator application rate and their interactions with planting
date had no effect on the number of leaves plant-1 at Ukulinga.
45
9
b LSD(0.05) = 1.17
-1 8 b
Number of leaves plant
7
6
5 a
4 a
3
2
1
0
Planting date
application rate on number of leaves plant-1 at Ukulinga.
At Umbumbulu, planting date (P = 0.009) and cultivar (P = 0.001) had significant

main effects on number of leaves plant-1. Delaying planting by one month from
October led to reduction in the number of leaves plant-1 from 5.13 to 3.76, but a
further delay had no effect (LSD(0.05) = 1.20) (Fig.3.9). Application of Gromor
Accelerator had no effect on the number of leaves plant-1. Dumbe-dumbe displayed a
significantly higher number of leaves plant-1 (4.45 leaves) than Mgingqeni (3.42
leaves) and Pitshi (3.01 leaves), and the latter two were not significantly different
from each other (LSD(0.05) = 0.70) (Fig. 3.10).
46
9
-1
LSD(0.05) = 1.20
Number of leaves plant 8
7
6 b
5
a
4
a a
3
2
1
0
Planting date
application rate on number of leaves plant-1 at Umbumbulu.
9
LSD(0.05) = 0.70
-1
8
Number of leaves plant
7
6
5 b
4 a
a
3
2
1
0
Taro cultivar
application rate on number of leaves plant-1 at Umbumbulu.
47
3.3.2.2 Plant height
There was a significant interaction of planting date, Gromor Accelerator application

rate and cultivar on plant height at Ukulinga (P = 0.002). The tallest plants were
obtained with all treatments except for Pitshi planted with no fertiliser and Dumbe-
dumbe planted with 160 N kg ha-1 organic fertiliser for October planting and Dumbe-
dumbe planted in November (Table 3.5). Application of organic fertiliser enhanced
plant height for Dumbe-dumbe planted in January and Pitshi planted in December.
Table 3.5 Effect of planting date, Gromor Accelerator application rate and cultivar
on plant height at Ukulinga. Means of the interaction effect within columns and rows
(LSD(0.05) = 8.28) followed by the same letter are not significantly different.
Planting date Gromor Taro cultivar

Accelerator Dumbe-dumbe Mgingqeni Pitshi
(N kg ha-1)
October 0 41.92ghij 45.00ij 35.58fg
160 33.75f 48.75j 42.08ghij
320 44.75hij 43.92hij 47.92j
November 0 48.17j 38.00fghi 36.58fgh
160 42.17ghij 33.42f 31.25f
320 49.25j 35.08fg 32.50f
December 0 15.33bcde 19.08de 9.62abc
160 10.83abcd 17.25cde 13.56bcde
320 13.00bcde 17.83cde 19.33e
January 0 15.92bcde 7.72ab 18.17de
160 19.03de 10.89abcd 2.96a
320 10.25abc 14.90bcde 16.90cde
The interaction of planting date and cultivar (P < 0.001), planting date and Gromor
Accelerator application rate (P = 0.003) and Gromor Accelerator application rate
and cultivar (P = 0.050) significantly affected plant height of taro at Umbumbulu.
Plant height of all cultivars was negatively affected by delayed planting (Fig. 3.11).
Delaying planting by one month from October decreased plant height for Dumbe-
dumbe, whereas the decrease was observed for Mgingqeni and Pitshi with two and
48
three months delay in planting respectively (Fig. 3.11). Pitshi had shorter plants than
Dumbe-dumbe and Mgingqeni which were not significantly different from each other
for October and December plantings. For November planting, Mgingqeni had tallest
plants compared to both Dumbe-dumbe and Pitshi which were not significantly
different from each other. Dumbe-dumbe showed tallest plants followed by Pitshi and
Mgingqeni respectively when planting was done in January.
40 LSD(0.05) = 3.60 Dumbe-dumbe

35 e e Mgingqeni
e Pitshi
30
Plant height (cm)
d d
25 cd cd
c c
20
b b
15
10 a
5
0
Planting date
application rate on plant height at Umbumbulu.
Fertilisation enhanced plant height at all planting dates except for December where
there was no significant difference between all fertiliser application rates (Fig. 3.12).
Application of 160 N kg ha-1 of the fertiliser significantly increased plant height but
incorporating more fertiliser had no effect. When no fertiliser was applied, a decrease
in plant height was observed with delay in planting until January. Plant height
decreased when planting was delayed by one and two months with application of 160
N kg ha-1 of the fertiliser, and by one, two and three months when 320 N kg ha-1 of
the fertiliser was applied.
49
40
f f LSD(0.05) = 4.67 0 N kg ha-1
35 160 N kg ha-1
30 320 N kg ha-1
Plant height (cm)
e de
25 c cd c
20 bc bc bc
b
15
10 a
5
0
Planting date
across cultivar on plant height at Umbumbulu.
Fertiliser application enhanced plant height of all taro cultivars (Fig. 3.13).
Application of 160 N kg ha-1 increased plant height for all cultivars but adding more
fertiliser was not effective. There was no difference in plant height between cultivars
when no fertiliser was added. When 160 N kg ha-1 of the fertiliser was applied, Pitshi
had shorter plants than other cultivars which were not different whereas Mgingqeni
had tallest plants compared to the other two which were also not significantly
different when 320 kg ha-1 of the fertiliser was added.
40
35 LSD(0.05) = 4.05 0 N kg ha-1
f 160 N kg ha-1
30
Plant height (cm)
ef ef 320 N kg ha-1
25 de
cd bcd
20 abc ab
a
15
10
5
0
Taro cultivar
Figure 3.13 Effect of Gromor Accelerator application rate and cultivar averaged
across planting date on plant height at Umbumbulu.
50
3.3.2.3 Leaf area
Leaf area plant-1 was affected by planting date (P < 0.001). There was no significant
effect of cultivar on leaf area plant-1, although the effect of planting date was
dependent on cultivar (Planting date x cultivar interaction: P = 0.024) at Ukulinga. At
Umbumbulu, leaf area plant-1 was affected by planting date (P < 0.001), cultivar (P <
0.001), fertiliser (P < 0.001), interactions of planting date and cultivar (P < 0.001) and
planting date and fertiliser (P < 0.001). Leaf area plant-1 of taro cultivars was
negatively affected by delayed planting at both sites (Fig. 3.14 and 3.15). At
Ukulinga, significantly higher leaf area plant-1 was observed when cultivars were
planted in October followed by November, which was also significantly higher than
the last two planting dates (Fig. 3.14). The exception was Dumbe-dumbe which
showed significantly highest leaf area plant-1 for the November planting. Dumbe-
dumbe showed lowest leaf area plant-1 compared to Mgingqeni and Pitshi which were
not different from each other when planting was done in October. For November
planting, Dumbe-dumbe had significantly higher leaf area plant-1 than Mgingqeni but
not Pitshi which was also not significantly different from Mgingqeni. There was no
difference in leaf area plant-1 between the cultivars when they were planted in
December and January.
2500
LSD(0.05) = 400.70
d d
cd
Leaf area plant-1(cm2)
2000 Dumbe-dumbe
bc
b b Mgingqeni
1500 Pitshi
1000
a
500 a a a
a
a
0
Planting date
application rate on leaf area plant-1 at Ukulinga.
51
At Umbumbulu, leaf area plant-1 decreased for all cultivars with one month delay in
planting (Fig. 3.15). Further delay in planting by two and three months for Mgingqeni
and by one month for Pitshi was also significant in decreasing leaf area plant-1. Pitshi
had lowest leaf area plant-1, Mgingqeni had highest leaf area plant-1 and Pitshi had
lowest leaf area plant-1 for October, November and December plantings respectively
compared to the other cultivars which were not significantly different from each
other. For January planting, Dumbe-dumbe had higher leaf area plant-1 than
Mgimgqeni but not Pitshi which was also not significantly different from Mgingqeni.
Delaying planting by one and two months decreased leaf area plant-1 when fertiliser
was applied, whereas when no fertiliser was applied the decrease was only observed
with three months delay in planting (Fig. 3.16). Fertilisation enhanced leaf area plant-1
for October planting only.
1500
e
LSD(0.05) = 147.30
e
Leaf area plant-1 (cm2)
Dumbe-dumbe
1000 d
d Mgingqeni
Pitshi
c
500 c c bc bc
ab
a a
0
Planting date
application rate on leaf area plant-1 at Umbumbulu.
52
d
1500 d LSD(0.05) = 303.20
Leaf area plant-1 (cm2)
0 N kg ha-1
1000 160 N kg ha-1
320 N kg ha-1
c
bc bc
500 abc
ab ab ab ab
ab
a
0
Planting date
Figure 3.16 Effect of planting date and Gromor Accelerator averaged across cultivar
on leaf area plant-1 at Umbumbulu.
3.3.3 Yield
3.3.3.1 Number of cormels
The number of cormels plant-1 was significantly affected by the interaction of planting
date and cultivar at Ukulinga (P = 0.004) and Umbumbulu (P = 0.034). The
interactions of planting date and fertiliser (P < 0.001) and cultivar and fertiliser also
affected the number of cormels plant-1(P = 0.020) at Umbumbulu. The number of
cormels plant-1 significantly decreased when planting was delayed by two months,
two and three months, and one and two months for Dumbe-dumbe, Mgingqeni and
Pitshi respectively at Ukulinga (Fig. 3.17). Pitshi had highest number of cormels
plant-1 followed by Mgingqeni and Dumbe-dumbe respectively for October planting.
When cultivars were planted in November, Mgingqeni and Pitshi which were not
significantly different from each other had significantly higher number of cormels
plant-1 than Dumbe-dumbe.
53
25
h LSD(0.05) = 3.29
-1
Number of cormels plant
20 g
Dumbe-dumbe
fg Mgingqeni
15 ef
Pitshi
de
10 cd
bc bc
ab
5 ab ab
a
0
Planting date
application rate on number of cormels plant-1 at Ukulinga.
At Umbumbulu, number of cormels plant-1 decreased for all cultivars when planting
was delayed by one month from October (Fig. 3.18). A further decrease for
Mgingqeni and Pitshi was also observed with a further one month delay in planting.
Dumbe-dumbe had the fewest cormels plant-1 for both October and November
plantings.
15
d LSD(0.05) = 2.19
-1
c
Dumbe-dumbe
10 Mgingqeni
b Pitshi
b b
5 a
a a a a
a
a
0
Planting date
application rate on number of cormels plant-1 at Umbumbulu.
54
Number of cormels plant-1 was decreased by delaying planting by one and two
months when fertiliser was applied. When no fertiliser was applied, delaying planting
had no effect on the number of cormels plant-1. Application of organic fertiliser
enhanced the number of cormels plant-1 for October and November plantings only
(Fig. 3.19). Addition of 160 N kg ha-1 of the fertiliser increased the number of cormels
plant-1 but adding more had no additional effect. Application of organic fertiliser had
no effect when planting was delayed until December and January.
Number of cormels plant-1 was enhanced for both Mgingqeni and Pitshi by
application of 160 N kg ha-1, whereas for Dumbe-dumbe it was increased by
application of 320 N kg ha-1 (Fig. 3.20). There was no significant difference in
number of cormels plant-1 between cultivars when no fertiliser was applied, but when
fertiliser was applied Dumbe-dumbe showed significantly lower number of cormels
plant-1 than Mgingqeni and Pitshi which were not significantly different from each
other.
15 c
c LSD(0.05) = 2.06
-1
0 N kg ha-1
10 160 N kg ha-1
320 N kg ha-1
b b
5 a a a a
a a a a
0
Planting date
Figure 3.19 Effect of planting date and Gromor Accelerator averaged across cultivar
on number of cormels plant-1 at Umbumbulu.
55
15
-1
Number of cormels plant Mgingqeni
Pitshi
10
c c c
c
b
5 ab
ab a
a
0
0 160 320
-1
Figure 3.20. Effect of Gromor Accelerator and cultivar averaged across planting date
on number of cormels plant-1 at Umbumbulu.
3.3.3.2 Grading of cormels
Most of the cormels were very small when they were graded according to size. About
70%, 80% and 85% of Dumbe-dumbe, Mgingqeni and Pitshi cormels respectively,
from both sites, planting dates, organic fertiliser application rates and cultivars, fell
under the >20 – 40g size class (Fig. 3.21, 3.22, 3.23, 3.24, 3.25, 3.26, 3.27 and 3.28).
56
60 >20
21-40
50 41-60
Number of cormels
61-80
40 81-100
101-120
30 121-140
141-160
20
10
0
D0 D1 D2 M0 M1 M2 P0 P1 P2
Figure 3.21 Number of cormels plant-1 for three taro cultivars (D = Dumbe-dumbe, M
= Mgingqeni and P = Pitshi) planted with three Gromor Accelerator application rates
(0 = 0 N kg ha-1, 1 = 160 N kg ha-1 and 2 = 320 N kg ha-1) obtained for October
planting at Ukulinga.
50
>20
21-40
40 41-60
Number of cormels
61-80
81-100
30 101-120
121-140
20 141-160
10
0
D0 D1 D2 M0 M1 M2 P0 P1 P2
(0 = 0 N kg ha-1, 1 = 160 N kg ha-1 and 2 = 320 N kg ha-1) obtained for October
planting at Umbumbulu.
57
50.00 >20
21-40
41-60
40.00
Number of cormels
61-80
81-100
30.00 101-120
121-140
141-160
20.00 161-180
10.00
0.00
D0 D1 D2 M0 M1 M2 P0 P1 P2
(0 = 0 N kg ha-1, 1 = 160 N kg ha-1 and 2 = 320 N kg ha-1) obtained for November
25 >20
21-40
20 41-60
61-80
Number of cormels
81-100
15 101-120
10
0
D0 D1 D2 M0 M1 M2 P0 P1 P2
(0 = 0 N kg ha-1, 1 = 160 N kg ha-1 and 2 = 320 N kg ha-1) obtained for November
58
25 >20
21-40
20 41-60
Number of cormels
61-80
15
10
0
D0 D1 D2 M0 M1 M2 P0 P1 P2
(0 = 0 N kg ha-1, 1 = 160 N kg ha-1 and 2 = 320 N kg ha-1) obtained for December
10.00 >20
21-40
41-60
Number of cormels
61-80
81-100
101-120
5.00
0.00
D0 D1 D2 M0 M1 M2 P0 P1 P2
(0 = 0 N kg ha-1, 1 = 160 N kg ha-1 and 2 = 320 N kg ha-1) obtained for December
59
12
>20
21-40
10 41-60
61-80
Number of cormels
0
D0 D1 D2 M0 M1 M2 P0 P1 P2
(0 = 0 N kg ha-1, 1 = 160 N kg ha-1 and 2 = 320 N kg ha-1) obtained for January
6
>20
5 21-40
41-60
Number of cormels
4 61-80
81-100
3
0
D0 D1 D2 M0 M1 M2 P0 P1 P2
(0 = 0 N kg ha-1, 1 = 160 N kg ha-1 and 2 = 320 N kg ha-1) obtained for January
60
3.3.3.3 Fresh cormel mass
Fresh cormel mass plant-1 was negatively affected by delayed planting at Ukulinga (P
< 0.001). There was no significant effect of cultivar on fresh cormel mass plant-1,
although the effect of planting date depended on cultivar (Planting date x cultivar
interaction: P = 0.011). Fresh cormel mass plant-1 significantly decreased for Dumbe-
dumbe when planting was delayed by two months, whereas for both Mgingqeni and
Pitshi, the decrease was observed when planting was delayed by one month (Fig.
3.29). Dumbe-dumbe had the lowest fresh cormel mass plant-1 for October planting,
but the highest mass for November planting. There was no significant effect of
fertilisation for fresh cormel mass plant-1.
500 g
LSD(0.05) = 77.92
-1
fg
Fresh cormel mass plant
400 ef
de Dumbe-dumbe
300 cd c Mgingqeni
Pitshi
200
b
ab
100 ab ab ab
a
0
Planting date
application rate on fresh cormel mass plant-1 at Ukulinga.
At Umbumbulu, similar to the site at Ukulinga, fresh cormel mass plant-1 was
negatively affected by delayed planting (P < 0.001). There was also no significant
effect of cultivar on fresh cormel mass plant-1. In contrast to the site at Ukulinga, fresh
cormel mass plant-1 was significantly affected by fertilisation (P < 0.001). Fertilisation
increased fresh cormel mass plant-1, although its effect depended upon planting date
(Planting date x fertilisation interaction: P < 0.001. Fertilisation enhanced fresh
61
cormel mass plant-1 only for the October and November plantings (Fig. 3.30). For both
October and November plantings, the fresh cormel mass plant-1 was enhanced when
160 N kg ha-1 of the organic fertiliser was added, but increasing the amount of
fertiliser to 320 N kg ha-1 was not effective.
500
LSD(0.05) = 47.52
-1
Fresh cormel mass plant
400 f f
0 N kg ha-1
300 160 N kg ha-1
320 N kg ha-1
200 e
de
cd
100 bc abc ab ab
ab ab a
0
Planting date
Figure 3.30. Effect of planting date and Gromor Accelerator application rate
averaged across cultivar on fresh cormel mass plant-1 of taro cormels at Umbumbulu.
All variables (emergence, leaf number plant-1, plant height, leaf area plant-1 and
number of cormels plant-1) except for number of cormels plant-1 at Ukulinga and leaf
area plant-1 at Umbumbulu significantly affected fresh cormel mass plant-1 of taro
when eliminating the effect of other variables (Appendix 3.3).
3.4 Discussion
The weather during the year of the field trial was normal and it is not shifting over
time due to global climate change (Figs. 3.2, 3.3, 3.4 and 3.5). It is evident from Table
1 that higher rainfall and temperature were experienced during the first two planting
dates and emergence of taro was greatly affected by rainfall and temperature. This
confirms statements by Horton (1987) and Ali (1989) that a liberal amount of water
and temperature should be available in the soil at planting to enhance rapid
emergence. Almekinders and Struik (1996) observed an increase in the number of
leaves at warmer temperatures. Deblonde and Ledent (2001) found drought to reduce
62
the number of green leaves in two out of six potato cultivars. The different number of
leaves plant-1 displayed by the cultivars in the present study was attributed to the
difference between genotypes.
The taller plants observed with the early two planting dates when rainfall was high at
both sites confirms the findings of Deblonde and Ledent (2001) that plant height was
sensitive to moderate drought conditions. This was also in line with what was found
by Zrust (1995) that water stress adversely affects plant height. Alsharari et al. (2007)
also found plants exposed to low water stress to be the tallest. The plants at
Umbumbulu were taller than those at Ukulinga probably because of the higher rainfall
at Umbumbulu (Table 3.1). Organic fertiliser application on the other hand enhanced
plant height. This finding was in line with findings of Allison et al. (1999). It is
evident from Figure 6 that plant height of taro corms was enhanced by fertiliser
application at Umbumbulu, confirming the findings of Mondal and Sen (2005) that
plant height of eddoe taro increased with increase in fertiliser levels.
The increased leaf area in response to earlier planting dates may have been attributed
to higher rainfall and temperature experienced over the growth period (Bussell and
Bonin 1998; Rinaldi, 2003). Ojala et al. (1990) previously reported leaf area to be
reduced by drought. Figure 8 shows that Mgingqeni and Dumbe-dumbe may have had
larger leaf area ascribed to the larger seed corms that were used as planting material
compared with Pitshi, which has a naturally small corm size. The effect of corm size
on plant leaf area confirms previous findings that early plantings and large gloves
produced larger leaf area in garlic (Stahlschmidt et al., 1997). In another study,
Jefferies (1992) also showed that in all the potato genotypes that were studied, final
size of leaves was reduced by drought, but the magnitude of the effect differed
significantly with genotype.
The number of cormels plant-1 and total fresh cormel mass plant-1 decreased with
delay in planting, improved with application of Gromor Accelerator and vary with
cultivars. There are many reports that yield of root crops, which is a positive function
of the number of tubers and weight of tubers planted hill-1 is affected by water
availability and temperature due to planting date (Lu et al., 2001; Khan et al., 2003;
63
Scheffer et al., 2005). In this study, the number of cormels plant-1 and total fresh
cormel mass plant-1 were highest at earlier planting dates which were characterized by
higher rainfall and lower temperature. The increased cormel mass in earlier planting
dates may have been attributed to the earlier increased leaf growth as a result of
higher rainfall and temperatures (Bussell and Bonin, 1998). This observation supports
the earlier findings that highest yields obtained when sweet potatoes were planted
with the onset of rains, when taro was planted in Spring and Summer compared to
Winter and fall plantings and when taro was planted under supplementary irrigation
rather than rainfed conditions were attributed to adequate rainfall or water availability
over the growth period (Sangakkara, 1993; Miyasaka et al., 2003; Scheffer et al.,
2005).
It was also shown in this study that organic fertiliser increased yields of taro. As was
the case with growth parameters, the effect of organic fertiliser was more pronounced
for earlier planting dates. This observation is consistent with findings reported in
other studies (Kumar et al., 2007; Hagman et al., 2009)
3.5 Conclusions and recommendations

In this study, it has been shown that taro growth and yield are affected by planting
date and fertilisation rate, but the response varies with cultivar. Taro growth and yield
parameters decrease with delay in planting and are positively affected by increasing
fertiliser rates. Mgingqeni and Pitshi had significantly higher fresh cormel mass plant-
1
than Dumbe-dumbe but a higher proportion of the cormels were small in size so it
was concluded that Dumbe-dumbe was the best performer. It is then recommended
that in order to improve taro yield at Ukulinga, Dumbe-dumbe should be planted in
October and November without organic fertiliser, whereas at Umbumbulu, on the
other hand, Dumbe-dumbe should be planted in October with 160 N kg ha-1 Gromor
Accelerator. Appendix 3.3 shows net returns hectare-1 of taro in KwaZulu-Natal
when 160 N kg ha-1 Gromor Accelerator is applied and this shows that a smallholder
taro enterprise is profitable by returning R12.89 for every R1.00 spent. It is not clear
why there is a negative effect of fertiliser on yield the later the crop is planted. A
future study will investigate the physiological characteristics of taro cormels that
influence sprouting, emergence and storability.
64
References
Agergaard, J. and Birch-Thomsen, T. 2006. Transitional rural landscapes: The role of
smallscale commercial farming in former homelands of Post-Apartheid
KwaZulu-Natal. Danish Journal of Geography 106 (2): 87-102.
Ali, A.M. 1989. Effects of planting on different sides of the ridge on emergence, plant
growth, yield and quality of potato (Solanum tuberosum) in the Sudan.
Experimental Agriculture 25: 409-415.
Allison, M.F., Allen, E.J. and Fowler, J.H. 1999. The nutrition of the potato crop.
British Potato Council Research Review 1991/1. BPC Oxford, UK.
Almekinders, C.J.M and Struik, P.C. 1996. Shoot development and flowering in
potato (Solanum tuberosum L.). Potato Research 39 (4): 581-607.
Alsharari, S.F., Alsadon, A.A. and Alharbi, A.R. 2007. Evaluation of drought
tolerance of potato cultivars under greenhouse conditions. Acta Horticulturae
747: 67-74.
Bèlanger, G., Walsh, J.R., Richards, J.E., Milburn, P.H. and Ziadi, N. 2002. Nitrogen
fertilisation and irrigation affects tuber characteristics of two potato cultivars.
American Journal of Potato Research 79 (4): 269-279.
Bussell, W.T. and Bonin, M.J. 1998. Effects of high and low watering levels on
growth and development of taro. New Zealand Journal of Crop and
Horticultural Science 26: 313-317.
Deblonde, P.M.K. and Ledent, J.F. 2001. Effects of moderate drought conditions on
green leaf number, stem height, leaf length and tuber yield of potato cultivars.
European Journal of Agronomy 14: 31–41.
Fleisher, D.H., Timlin, D.J. and Reddy, V.R. 2006. Temperature influence on potato
leaf and branch distribution and on canopy photosynthetic rate. Agronomy
Journal 98:1442–1452.
Hagman, J.E., Mårtensson, A. and Grandin, U. 2009. Cultivation practices and potato
cultivars suitable for organic potato production. Potato Research DOI
10.1007/s11540-009-9128-3.
Hancock, J.F. 2004. Plant evolution and the origin of crop species. CABI Publishing.
Wellingford. UK, p. 222.
Horton, D.E. 1987. Potatoes: production, marketing and programs for developing
countries, Westview Press, London, p. 39.
65
Jefferies, R.A. 1992. Responses of potato genotypes to drought. I. Expansion of
individual leaves and osmotic adjustment. Annals of Applied Biology 122 (1):
93-104.
Kader, A.A. and Rolle, R.S. 2004. The role of post-harvest management in assuring
the quality and safety of horticultural produce. FAO Agricultural Services
Bulletin 152.
Khan, I.A., Deadman, M.L., Al-Nabhani, H.S. and Al-Habsi, K.A. 2003. Interactions
between temperature and yield components in exotic potato cultivars grown
in Oman. Acta Horticulturae 619: 353-359.
Kooman, P.L., Fahem, M., Tegera, P. and Haverkort, A.J. 1996. Effects of climate on
different potato genotypes 2. Dry matter allocation and duration of the growth
cycle. European Journal of Agronomy 5: 207-217.
Kumar, P., Pandey, S.K., Singh, B.P. Singh, S.V. and Kumar, D. 2007. Effect of
nitrogen rate on growth, yield, economics and crisps quality of Indian potato
processing cultivars. Potato Research 50 (2): 143-155.
Loomis, R.S. and Connor, D.J. 1992. Crop ecology: productivity and management in
agricultural systems. Cambridge University Press, New York.
Lu, H.Y., Lu, C.T. Chan, L.F and Wei, M.L. 2001. Seasonal variation in linear
Increase of taro harvest index explained by growing degree days. Agronomy
Journal 93: 1136-1141.
Naramabuye, F.X., R.J. Haynes, R.J and Modi, A.T. 2008. Cattle manure and grass
residues as liming materials in a semi-subsistence farming system.
Agriculture, Ecosystems and Environment 124: 136-141.
Schott] landraces from Umbumbulu. MSc. Thesis. University of KwaZulu-
Natal.
Miyasaka, S.C., Ogoshi, R.M., Tsuji, G.Y. and Kodani, L.S. 2003. Site and planting
date effects on taro growth: Comparison with aroid model predictions.
Agronomy Journal 95: 545–557.
Modi, A.T. 2003. What do subsistence farmers know about indigenous crops and
organic farming? Preliminary case in KwaZulu-Natal. Development Southern
Africa 20: 673-682.
66
Modi, A.T. 2007. Effect of indigenous storage method on performance of taro
[Colocasia esculenta (L.) Schott] under field conditions in a warm subtropical
area. South African Journal of Plant and Soil 24: 214-219.
Mokolobate, M.S. and Haynes, R.J. 2002. Comparative liming effect of four organic
residues applied to an acid soil. Biology and Fertility of Soils 35: 79–85.
Mondal, S. and Sen, H. 2005. Growth and productivity of eddoe taro (Colocasia
esculenta var. antiquorum) as influenced by fertiliser levels and spacings.
Journal of root crops 31(1). .
O'Beirne, D. and Cassidy, J. C. 1990. Effects of nitrogen fertiliser on yield, dry matter
content and flouriness of potatoes. Journal of the Science of Food and
Agriculture 52 (3): 351-363.
Ojala, J.C., Stark, J.C. and Kleinkopf, G.E. 1990 Influence of irrigation and nitrogen
management on potato field and quality. American Potato Journal 67: 29–43.
Osorio, N.W., Shuai, X., Miyasaki, S., Wang, B., Shirey, R. L. and Wigmore, W. J.
2003. Nitrogen level and form affect taro growth and nutrition. Horticultural
Science 38 (1): 36-40.
Pocknee, S. and Summer, M.E. 1997. Cation and nitrogen contents of organic matter
determine its liming potential. Soil Science Society of American Journal 61:
86-92.
Rinaldi, M. 2003. Variation of specific leaf area for sugar beet depending on sowing
date and irrigation. Italian Journal of Agronomy 7 (1): 23-32.
Sangakkara, U.R. 1993. Response of sweet potato (Ipomoea batatas L.) to different
planting times. Journal of Agronomy and Crop Science 172 (2): 113 – 118.
Scheffer, J.J.C., Douglas, J.A. and Triggs, C.M. 2005. Factors affecting the
production and quality of Japanese taro cormels. Acta Horticulturae 670: 167-
172.
Shange, L.P. 2004. Taro (Colocasia esculenta (L.) Schott) production by small-scale
KwaZulu-Natal.
Shock, C.C., Zalewski, J.C., Stieber, T.D. and Burnett, D.S. 1992. Impact of early-
season water deficits on Russet Burbank plant development, tuber yield and
quality. American Journal of Potato Research 69 (12): 793-803.
67
Stahlschmidt, O., Cavagnaro, J.B. and Borgo, R. 1997. Influence of planting date and
seed cloves size on leaf area and yield of two garlic cultivars (Allium sativum
L.). Acta Horticulturae 433: 519-522.
Timlin, D., Lutfor-Rahman, S. M., Baker, J., Reddy, V. R. Fleisher, D. and
Quebedeaux, B. 2006. Whole plant photosynthesis, development, and carbon
partitioning in potato as a function of temperature. Agronomy Journal 98:
1195–1203.
Vinning, G. 2003. Select markets for taro, sweet potato and yam. RIRDC Project No.
UCQ-13A.
Wadas, W., Jablońska-Ceqlarek, R. and Kosterna, E. 2004. Effect of plastic covering
and nitrogen fertilisation on yield and quality of early potatoes. Folia
Horticulturae 16 (2): 41-48.
Wolfe, D.W., Fereres, E. and Voss, R.E. 1983. Growth and yield response of two
potato cultivars to various levels of applied water. Irrigation Science 3 (4):
211-222.
Wong, M.T.F., Nortcliff, S. and Swift, R.S. 1998. Method for determining the acid
ameliorating capacity of plant residue compost, urban waste compost,
farmyard manure and peat applied to tropical soils. Communication in Soil
Science and Plant Analysis 29: 2927-2937.
Worthington, C.M. and Hutchinson, C.M 2006. Yield and quality of `Atlantis` and
`Harley Blackwell` potatoes as affected by multiple planting dates, nitrogen
rates and growing degree days. Proceedings of the Florida State Horticultural
Society 119: 275-278.
Yuan, B.Z., Nishiyama, S. and Kang, Y. 2003. Effects of different irrigation regimes
on the growth and yield of drip-irrigated potato. Agricultural Water
Management 63: 153–167.
tuberosum L.) to different rates of nitrogen and phosphorus fertilization on
vertisols at Debre Berhan, in the central highlands of Ethiopia. African
Journal of Plant Science 3 (2): 16-24.
Zrust, J. 1995. The effect of drought on the potato leaf area. Scientia Agriculturae
Bohemica. 26 (3):177–188.
68
CHAPTER 4
INFLUENCE OF PLANTING DATE AND ORGANIC FERTILISER ON

SPECIFIC GRAVITY, DRY MATTER CONTENT AND MINERAL
CONTENT OF TARO CORMELS
Abstract
Taro (Colocasia esculenta L. Schott) is an important traditional crop in Africa, but its
contribution to food security is limited by lack of research on its agronomy and
commercialisation. A study was carried out to investigate the performance of three
taro cultivars at two sites in KwaZulu-Natal, South Africa (Ukulinga and
Umbumbulu), during the 2007/2008 growing season, with respect to chemical quality
of cormels. A split-split- plot design was used with four planting dates (October 2007,
November 2007, December 2007 and January 2008) as main plots, three cultivars
(Dumbe-dumbe, Mgingqeni and Pitshi) as sub-plots and three organic fertiliser
(Gromor Accelerator) application rates (0, 160 and 320 N kg ha-1) as sub-sub-plots.
At Ukulinga, application of 320 N kg ha-1 of Gromor Accelerator significantly (P =
0.007) decreased dry matter content. Dry matter content was significantly highest for
Dumbe-dumbe with November planting, whereas for both Mgingqeni and Pitshi the
highest dry matter content was obtained with December planting and declined
thereafter at Ukulinga. At Ukulinga, specific gravity was significantly lowest for
Dumbe-dumbe with January planting, whereas for Mgingqeni it was significantly
highest with December planting. Application of 5330 kg ha-1 of Gromor Accelerator
significantly decreased specific gravity only when planting was done in January. At
Umbumbulu, delaying planting until December significantly decreased (P = 0.013)
specific gravity for Dumbe-dumbe and Pitshi. Specific gravity was significantly
highest for October planting with 0 N kg ha-1 of Gromor Accelerator and lowest for
December planting with 320 Nkg ha-1 of the organic fertiliser at Umbumbulu. Mineral
content of cormels was significantly (P < 0.05) affected by both planting date and
fertiliser application. It is recommended that to improve taro chemical composition,
the crop should be planted in October at the study sites.
Key Words: Colocasia esculenta, cormel quality, fertilisation, planting date, site.
69
4.1 Introduction
Taro (Colocasia esculenta L. Schott) is an important traditional staple crop in the

rural areas of South Africa and other African countries, but its contribution to food
security is limited by lack of research on its agronomy and commercialisation. It is
grown in the subtropical coastal area of South Africa, starting at Bizana district in the
Eastern Cape and the rest of coastal KwaZulu-Natal. The crop is also cultivated in the
subtropical and tropical parts of Mpumalanga and Limpopo provinces (Shange, 2004).
Taro has an emerging market in KwaZulu-Natal and there is an initiative to process it
into crisps. Corm quality is the main concern when taro is considered for processing.
Like in potatoes, the processing industry requires corms of higher dry matter content
and specific gravity in order to qualify for crisping. Different taro cultivars are grown
organically at different planting dates. The different environmental conditions due to
different planting dates and the different organic fertiliser rates interact to influence
the corm quality of these cultivars. An understanding of these factors is essential for
management so that corm quality can be altered or maintained to suit the crisping
industry without compromising the nutritional quality of the corms.
Studies have been reported on dry matter content, specific gravity and nutritional
composition of root crops under different management practices, environmental
conditions (Mohamed, 1985; Casa et al., 2005; Long et al., 2004; Haase et al., 2007;
Huang et al., 2007; Kumar et al., 2007; El-Sirafy et al., 2008). They found planting
date, fertilisation, cultivar and environmental conditions to be the most important
factors that affect corm or tuber quality.
Response to planting date is, however variable depending on cultivar. Dry matter was
found to be highest at early planting for two of the potato cultivars while it was
highest for the other cultivar when it was planted at a later date. Protein content of the
tubers, on the other hand was higher when planting was delayed after the first planting
date regardless of the cultivar (Mohamed, 1985). Tuber specific gravity and dry
matter content were highest for three of the seven examined potato cultivars for spring
planting whereas the same cultivars and an additional cultivar emerged superior for
fall planting (Tawfik, 2002). Mittra & George (2000) found tuber dry matter content
to be lower in plants planted during the period of high rainfall and temperature (June,
70
July and August) compared to those planted in September and October in sweet
potatoes in India.
El-Sirafy et al. (2008) found application of farm yard manure to increase dry matter
content, protein content, specific gravity as well as nitrogen, phosphorus and
potassium content of potato tubers whereas soil potassium increase protein content.
They also reported that the interactions of farm yard manure and soil potassium
decrease dry matter content but increase protein content, specific gravity and nitrogen,
phosphorus and potassium content of potato tubers. McPharlin & Lancaster (2005)
reported that potato varieties grown at different sites respond differently to nitrogen
fertilisation and further stated that higher specific gravity in one cultivar was
associated with a higher calcium content and lower nitrogen content in the tubers.
Mondy & Koch (1978) reported potassium content to decrease and dry matter to
increase with increasing levels of nitrogen fertilisation in potatoes whereas Manrique
(1994) found excessive nitrogen fertilization to reduce dry matter allocated to the
corms in taro. This was in line with what Bèlanger et al. (2002) and Zelalem (2009)
found that nitrogen fertilisation significantly reduced dry matter content and tuber
specific gravity of potatoes. Bèlanger et al. (2002) further reported that tuber specific
gravity was strongly related to tuber nitrogen concentration, which in turn depended
primarily on nitrogen fertilisation and that the risks of low specific gravity are greater
when fertilisation exceeds the nitrogen requirements to reach maximum tuber yield.
On the contrary, Wadas et al. (2004) reported that nitrogen fertilisation had no effect
on quality of potato tubers. Phosphorus was found to have no effect on dry matter
content and specific gravity of sweet potatoes and potatoes respectively (Rashid and
Waithaka, 1985; Zelalem, 2009).
Upland-cultivated taro corms were found to retain higher moisture content and tended
to have higher mineral content than paddy taro (Huang et al., 2007). Abundant rainfall
during the month of June resulted in lower dry matter content (Fontana et al., 1990).
Supplemental irrigation had a limited effect on specific gravity and tuber nitrogen
concentrations (Bèlanger et al., 2002). The potassium fertilisation effect on specific
gravity depended upon the K concentration in the harvested tuber (Westermann et al.,
1994).
71
Midseason stress also appeared to reduce specific gravity (Lynch et al., 1995).
Accumulation of dry matter in the tubers was increased by water deficit (Nadler and
Heuer, 1995). Significant increases in specific gravity, dry matter and significant
decreases in protein content were observed due to the frequent irrigation at planting-
stolon initiation and stolon initiation-tuber bulking stages. Frequent irrigations at
tuber bulking stage were found to have deleterious effects on specific gravity and dry
matter, especially when irrigation continued until maturity (Günel and Karadoĝan,
1998). Nitrogen fertilisation, irrigation, and cultivars affect tuber characteristics such
as specific gravity, and nitrogen concentration. Nitrogen fertilisation increased tuber
N and N03-N concentrations, and decreased specific gravity.
Suitable planting date, fertiliser application should improve the quality of taro corms.
The objective of this study was to determine the dry matter content, specific gravity,
protein content and mineral composition of corms of three taro cultivars grown at
different planting dates using different organic fertiliser regimes in KwaZulu-Natal,
South Africa.
4.2.1 Planting material
Three taro cultivars of eddoe type Dumbe-dumbe, Mgingqeni and Pitshi were used for
this study (Fig. 3.1). Cormel size for the taro cultivars vary between and within a
cultivar. The cormels planted for this study were selected to be in the following size
ranges (cormel -1): 21 – 60 g for Dumbe-dumbe, 11 – 40g for Mgingqeni and 5 – 20g
for Pitshi. The cormels were harvested in 2007 from Ukulinga except for Dumbe-
dumbe which was obtained from farmers of Ezigeni, Umbumbulu district, KwaZulu-
Natal, South Africa.
4.2.2 Site description
Field experiments were carried out at two sites, Ukulinga (29o 37′S 30o 16′E) and
Umbumbulu (29o 36′S 30o 25′E) with different environmental conditions (Table 3.1)
and soil physical and chemical properties (Table 3.2). Temperature and rainfall data
72
for both sites were obtained from Weather South Africa for the past 10 years (Figs.
3.2, 3.3, 3.4 and 3.5) and for the duration of the experimental period Table 3.1). Soil
samples were collected from both sites and analysed according to Naramabuye et al.
(2008). The pH of the soils from both sites (Table 3.2) were below 6 to 6.5 which is
the soil pH at with taro is known to grow best. Addition of organic manure was
expected to increase the soil pH as it was reported by Pocknee and Summer (1997),
Wong et al. (1998), Mokolobate and Haynes (2002) and Naramabuye et al. (2008)
that organic manure has a rapid liming effect. However, no soil analysis was
performed at harvest.
4.2.3 Experimental design
The experiment was designed as a split-split-plot design with four planting dates:
October 2007, November 2007, December 2007 and January 2008 (Table 3.3) as
main plots, three taro cultivars: Dumbe-dumbe, Mgingqeni and Pitshi (Fig 3.1) as sub-
plots and three organic fertiliser (Gromor Accelerator) application rates (0, 160 and
320 N kg ha-1) as sub-plots. The nutritional composition of Gromor Accelerator is
shown in table 3.3. Each planting date was replicated three times. Plot size was 4 m2
containing 16 plants spaced 0.5m between and within rows. Sowing was done by
hand on ploughed and harrowed fields. Planting holes were opened with a hand-hoe
and organic fertiliser was mixed with soil before one cormel was planted per hole.
Weeds were controlled by hand-hoeing at one, two, three and four months after
planting. Harvesting was done eight months after planting.
4.2.4 Data collection and analysis
Dry matter content and specific gravity were determined on freshly harvested taro
cormels at harvest. Other cormels were peeled, cut into slices, immediately frozen in
liquid nitrogen, freeze dried and ground into powder for mineral content analysis.
Analysis of variance was performed using GenStat Version 11.1 (2008). Least
significant difference (LSD) was used to separate means. A probability level of 0.05
was considered to be statistically significant.
73
4.2.4.1 Dry matter content
Cormels were weighed to find fresh weight. They were then sliced before they were
oven dried at 70oC for 96 hours to a constant weight and weighed again. Dry matter
was determined by dividing the dry weight by fresh weight then multiplied by 100.
(Dry weight / Fresh weight) x 100 = % Dry matter.
4.2.4.2 Specific gravity
Cormels were weighed and then submerged in water to measure their volumes by
displacement. Cormel weight was divided by volume to determine specific gravity
(Yuan et al., 2003; Robin, 2004).
4.2.4.3 Protein content and total protein
The protein content was calculated by multiplying the nitrogen value, which was
determined by running on a LECO CNS instrument calibrated with an imported
sample and checked against known standard samples, by the conversion factor 6.25.
Total protein is the amount of protein in the fresh cormel mass plant-1.
4.2.4.4 Mineral content
For mineral content determination, the ground samples were analysed for nitrogen,
phosphorus, potassium, calcium, magnesium, sodium, zinc, copper, manganese and
iron content. Nitrogen was run on a LECO CNS instrument calibrated with an
imported sample and checked against known standard samples. The analyses for
calcium, magnesium, potassium, sodium, zinc, copper, manganese, iron, phosphorus
and aluminium were done using the I.C.P instrument which was calibrated on four
different levels of imported standards for each of the elements. Internal controls were
run every tenth sample and the instrument was checked regularly using an imported
multi element standard (Naramabuye et al., 2008).
74
4.3 Results
4.3.1 Dry matter content
Dry matter content of taro cormels was significantly affected by Gromor Accelerator
application rate (P = 0.020) and the interaction of planting date and cultivar (P =
0.031) at Ukulinga. Application of 320 N kg ha-1 of Gromor Accelerator significantly
decreased dry matter content from 27.56% to 24.55% (Fig. 4.1).
28 b
Dry matter content (%)
LSD(0.05) = 2.07
27
ab
26
25 a
24
23
0 160 320
Gromor Accelerator (N kg ha-1)
Figure 4.1 Effect of Gromor Accelerator application rate averaged across planting
date and cultivar on dry matter content of taro cormels at Ukulinga.
Delaying planting by one month from October significantly increased dry matter
content of Dumbe-dumbe and significantly decreased it again with January planting
(Fig. 4.2). For Mgingqeni, dry matter content was significantly increased when
planting was delayed by two months. Planting date had no effect on dry matter
content of Pitshi. Dumbe-dumbe displayed significantly highest dry matter content
for November planting, whereas for January planting Mgingqeni showed significantly
higher dry matter content than Dumbe-dumbe but not Pitshi which was also not
significantly different from Dumbe-dumbe. There was no significant difference in dry
matter content of the cultivars when they were planted in both October and December.
Gromor Accelerator and/or planting date had no effect on dry matter content of all
taro cultivars at Umbumbulu.
75
50
Dry matter content (%) Mgingqeni
40
Pitshi
d
bcd cd cd
30 abc bcd abc
ab abc ab ab
a
20
10
0
Planting date
application rate on dry matter content at Ukulinga.
4.3.2 Specific gravity
At Ukulinga, specific gravity was significantly affected by the interactions of planting

date and cultivar (P = 0.001) as well as planting date and Gromor Accelerator
application rate (P = 0.013). Cormels of Dumbe-dumbe from Ukulinga displayed a
significant decline in specific gravity with delay in planting by three months from
October (Fig. 4.3). Dumbe-dumbe had significantly higher specific gravity than Pitshi
but not Mgingqeni for October planting, whereas Mgingqeni showed significantly
higher specific gravity than Dumbe-dumbe but not Pitshi for January planting. The
significant difference in specific gravity was only significant between cultivars when
planting was done in October and January.
76
1.17 Dumbe-dumbe LSD(0.05) = 0.05
Mgingqeni
1.12 Pitshi
d
Specific gravity
d
cd cd bcd
1.07 bcd bcd
abc
ab ab ab
1.02
a
0.97
0.92
Planting date
application rate on specific gravity at Ukulinga.
Delaying planting by three months from October significantly decreased specific

gravity when 160 N kg ha-1 of Gromor Accelerator was applied (Fig. 4.4). The
significantly lowest specific gravity was obtained with 160 N kg ha-1 of Gromor
Accelerator for January planting.
1.15 0 N kg ha-1 LSD(0.05) = 0.07

160 N kg ha-1
1.1 320 N kg ha-1 c
bc
bc bc bc bc
Specific gravity
1.05 bc bc bc bc
ab
1
a
0.95
0.9
0.85
Planting date
across cultivar on specific gravity at Ukulinga.
77
Specific gravity responded significantly to the interactions of planting time and
cultivar (P = 0.033) and planting date and Gromor Accelerator application rate (P =
0.002) at Umbumbulu. Delaying planting by two months significantly decreased
specific gravity of Dumbe-dumbe and Pitshi but it increased again for Pitshi with
January planting (Fig. 4.5). Specific gravity was significantly lowest for Pitshi with
December planting.
1.14 LSD(0.05) = 0.06 Dumbe-dumbe

Mgingqeni
1.09 d cd Pitshi
cd cd
bcd cd
bcd
Specific gravity
bcd
1.04 bc bcd
b
0.99
0.94 a
0.89
0.84
Planting date
application rate on specific gravity at Umbumbulu.
Delaying planting by one month from October significantly decreased specific gravity
when no organic fertiliser was applied (Fig. 4.6). When 320 N kg ha-1 of Gromor
Accelerator was applied specific gravity was decreased with delay in planting by two
months, and increased thereafter with January planting. The significantly highest
specific gravity was obtained with October planting when organic fertiliser was not
applied, whereas the lowest was obtained with December planting when 320 N kg ha-1
was applied.
78
1.15 LSD(0.05) = 0.04 0 N kg ha-1
e 160 N kg ha-1
1.1 320 N kg ha-1
cd d cd
bcd
Specific gravity
1.05 bcd bcd bcd bcd

b bc
1
a
0.95
0.9
0.85
Planting date
across cultivar on specific gravity at Umbumbulu.
4.3.3 Protein content and total protein
4.3.3.1 Protein content
Planting date (P < 0.001) and Gromor Accelerator application rate (P = 0.032)
significantly affected protein content of taro cormels at Ukulinga. Delaying planting
significantly increased protein content of taro cormels. January planting displayed
significantly highest protein content (16.70%) followed by December planting
(14.69%), and both November (12.76%) and October (11.99%) plantings which were
not significantly different from each other (Fig. 4.7). Adding 160 N kg ha-1 of Gromor
Accelerator significantly increased protein content from 13.55% - 14.54%, but
further incorporation of 320 N kg ha-1 of the organic fertiliser did not further increase
protein content (14.01%) (Fig. 4.8). Cultivar and the interactions of planting date,
cultivar and fertiliser had no effect on protein content of cormels at Ukulinga.
79
20
LSD(0.05) = 1.32
c
b
Protein content (%)
15
a
a
10
0
Planting date
application rate on protein content of taro cormels at Ukulinga.
20
LSD(0.05) = 0.73
Protein content (%)
b ab
15 a
10
0
0 160 320
date and cultivar on protein content of taro cormels from Ukulinga.
At Umbumbulu, protein content significantly (P < 0.001) decreased with application

of Gromor Accelerator. Addition of 160 N kg ha-1 of organic fertiliser significantly
decreased protein content of cormels from 11.50 % to 9.41 %, but further increase of
the organic fertiliser had no further significant effect (9.21 %) (Fig. 4.9). The
interaction of planting date and cultivar also significantly (P = 0.002) affected protein
content of taro cormels at Umbumbulu. Protein content of Dumbe-dumbe was
80
significantly increased by delaying planting until November (Fig. 4.10). For
Mgingqeni, protein content was significantly increased when planting was delayed
from October to December whereas for Pitshi the increase was observed with delay in
planting from November to December and decreased thereafter. The significantly
highest protein content was obtained with Pitshi for December planting.
20
LSD(0.05) = 0.56
Protein content (%)
15
b
a a
10
0
0 160 320
date and cultivar on protein content of taro cormels from Umbumbulu.
20 Dumbe-dumbe
LSD(0.05) = 1.43 Mgingqeni
Pitshi
Protein content (%)
15 e
bcd cd d d
bcd cd
10 abc bcd
a ab ab
0
Planting date
application rate on protein content of taro cormels from Umbumbulu.
81
4.3.3.2 Total protein
The interaction of planting date and cultivar significantly affected total protein plant-1
at Ukulinga (P = 0.004). Total protein plant-1 was increased and decreased for Dumbe-
dumbe and Pitshi respectively when planting was delayed by one month from October
(Fig. 4.11). Delaying planting by one month had no effect on total protein plant-1 of
Mgingqeni. Further delaying planting by two months significantly reduced total
protein plant-1 for all cultivars. Pitshi had higher total protein plant-1 than other
cultivars when they were planted in October, whereas Dumbe-dumbe had higher
protein content plant-1 when planting was done in November. There was no difference
in total protein plant-1 between cultivars when they were planted in December and
January.
16
d Dumbe-dumbe
14 cd Mgingqeni
Total protein plant (g)
12 Pitshi
-1
bc
10 b
b b LSD(0.05) = 2.83
8
6 a
a a
4 a a a
2
0
Planting date
application rate on total protein of taro cormels from Ukulinga.
At Umbumbulu, delaying planting by two months decreased total protein plant-1 when
0 or 320 N ka ha-1 of fertiliser was applied, whereas when 160 N kg ha-1 was applied
the decrease was observed with delay in planting by one and two months (Fig, 4.12)..
Total protein plant-1 was significantly affected by the interactions of planting date and
fertiliser (P < 0.001) and cultivar and fertiliser (P = 0.022). Application of 160 N kg
ha-1 increased total protein plant-1 of taro cormels when planting was done in October,
whereas when planting was done in November, it was increased by adding 320 N kg
82
ha-1. Fertiliser had no effect on total protein plant-1 when planting was done in
8 e
LSD(0.05) = 1.21 0 N kg ha-1
7 de 160 N kg ha-1
6 d 320 N kg ha-1
-1
5
4 c
c
3 bc
2 ab ab a
a a a
1
0
Planting date
Figure 4.12 Effect of planting date and Gromor Accelerator application rate
averaged across cultivar on total protein of taro cormels from Umbumbulu.
Application of fertiliser increased total protein plant-1 for all cultivars, but further
increase of the fertiliser decreased total protein plant-1 for Pitshi (Fig.4.13). Mgingqeni
had higher total protein plant-1 than other cultivars when 320 N kg ha-1 was applied
whereas there was no difference in total protein plant-1 between cultivars when 0 and
160 N kg ha-1 was applied.
83
5 LSD(0.05) = 1.05 d
Dumbe-dumbe
Mgingqeni
4 cd
Pitshi cd
bc
-1
bc
3
ab ab
2 a a
0
0 160 320
-1
Figure 4.13 Effect of cultivar and Gromor Accelerator application rate averaged
across planting date on total protein of taro cormels from Umbumbulu.
4.3.4 Mineral content of taro cormels
Mineral content of taro cormels was significantly (P < 0.05) affected by planting date,
Gromor Accelerator application rate and cultivar with the exception of calcium,
copper and iron content of cormels from Ukulinga, and magnesium, iron and sodium
content of cormels from Umbumbulu which were not affected.
4.3.4.1 Nitrogen content
Nitrogen content of taro corms was positively affected by delayed planting (P <
0.001) and fertilisation (P = 0.032) at Ukulinga. Cultivar and interactions of planting
date, cultivar and fertilisation were not significant for nitrogen content. January
planting displayed significantly highest nitrogen content (2.67%) followed by
December planting (2.35%), and both November (2.04%) and October (1.92%)
plantings which were not significantly different from each other Fig. 4.14). Adding
160 N kg ha-1 of Gromor Accelerator significantly increased nitrogen content but
further incorporation of 320 N kg ha-1 of the organic fertiliser had no further effect
(Fig. 4.15).
84
3
LSD(0.05) = 0.21 c
2.5 b
Nitrogen content (%)
a
a
2
1.5
0.5
0
Planting date
Figure 4.14 Effect of planting date averaged across Gromor Accelerator application
rate and cultivar on nitrogen content of taro cormels at Ukulinga.
2.35 b
LSD(0.05) = 0.12
2.3
2.25 ab
2.2
a
2.15
2.1
0 160 320
date and cultivar on nitrogen content of taro cormels at Ukulinga.
At Umbumbulu, nitrogen content was significantly affected by Gromor Accelerator

application rate (P < 0.001) and the interaction of planting date and cultivar (P =
0.002). Addition of 160 N kg ha-1 of Gromor Accelerator significantly decreased
85
nitrogen content of cormels from 1.84 % to 1.50 % but further increase of the organic
fertiliser was not significant (Fig. 4.16).
2 b LSD(0.05) = 0.09
a a
1.5
0.5
0
0 160 320
date and cultivar on nitrogen content of taro cormels at Umbumbulu.
Delaying planting by one month from October significantly increased nitrogen

content of Dumbe-dumbe (Fig.4.17). Nitrogen content was increased for Mgingqeni
when planting was delayed until December, whereas for Pitshi, it was significantly
increased with a delay in planting by two months, and it decreased again with an
additional one month delay for Pitshi only. Pitshi showed significantly higher
nitrogen content than Dumbe-dumbe but not Mgingqeni which was also not
significantly higher than Dumbe-dumbe for October planting. For December planting,
Pitshi had significantly highest nitrogen content followed by Mgingqeni. Dumbe-
dumbe displayed significantly lowest nitrogen content for January planting.
86
2.5 Dumbe-dumbe LSD(0.05) = 0.23
e
Mgingqeni
Pitshi
2 d
d
bcd bcd cd cd
abc bcd
1.5 a ab ab
0.5
0
Planting date
application rate on nitrogen content of taro cormels at Umbumbulu.
4.3.4.2 Phosphorus content
Phosphorus content of taro cormels was negatively affected by delayed planting (P =

0.001). The interaction of planting date and cultivar significantly (P = 0.027) affected
phosphorus content of taro cormels at Ukulinga. Delaying planting from October to
November significantly decreased phosphorus content of all taro cultivars (Fig. 4.18).
A further delay in planting to December significantly decreased phosphorus content
of Mgingqeni only, whereas a further delay to January was not significant for all
cultivars. Dumbe-dumbe displayed significantly highest phosphorus content at all
planting dates, with the exception of November, where it was not significantly higher
than Mgingqeni. Phosphorus content of Mgingqeni and Pitshi was not different for all
planting dates. There was no significant effect of fertilisation for phosphorus content.
87
Dumbe-dumbe
0.5
LSD(0.05) = 0.04 Mgingqeni
Phosphorus content (%) g Pitshi
0.4 fg
ef def def
de cd
0.3 bc
a ab ab ab
0.2
0.1
0
Planting date
application rate on phosphorus content of taro cormels at Ukulinga.
At Umbumbulu, phosphorus content of taro cormels was positively affected by

fertilisation (P < 0.001). There was no significant effect of planting date and cultivar
on phosphorus content, although the effect of fertilisation depended on both planting
date and cultivar (Planting date x cultivar x fertiliser interaction: P = 0.014). Applying
160 N kg ha-1 of fertiliser enhanced phosphorus content of taro cormels except for
Dumbe-dumbe planted in December and Mgingqeni and Pitshi planted in January
where phosphorus content was enhanced by application of 320 N kg ha-1 of the
organic fertiliser (Table 4.1). Pitshi had significantly lower phosphorus content than
Dumbe-dumbe when the cultivars were planted in December without fertiliser.
Phosphorus content was significantly decreased for Mgingqeni when planting with
160 N kg ha-1 of the organic fertiliser was delayed from October to January. When
320 N kg ha-1 of the organic fertiliser was applied, delaying planting by one and two
months increased phosphorus content for Dumbe-dumbe and Pitshi respectively and
decreased it thereafter for Dumbe-dumbe only.
88
Table 4.1 Effect of planting date and Gromor Accelerator application rate on
phosphorus content of taro cormels of different cultivars at Umbumbulu. Means of the
interaction effect within columns and rows (LSD (0.05) = 0.05) followed by the same
letter are not significantly different.
Planting time Cultivar Gromor Accelerator (N kg ha-1)
0 160 320
October Dumbe-dumbe 0.10bcd 0.18fgh 0.16efgh
Mgingqeni 0.09abc 0.19hi 0.17fgh
Pitshi 0.09ab 0.17fgh 0.15defgh
November Dumbe-dumbe 0.10bcd 0.19hi 0.27j
Mgingqeni 0.08ab 0.14cdefgh 0.17fgh
Pitshi 0.09abc 0.17fgh 0.16efgh
December Dumbe-dumbe 0.11bcde 0.16efgh 0.17fgh
Mgingqeni 0.09abc 0.15defgh 0.18hi
Pitshi 0.04a 0.16efgh 0.23ij
January Dumbe-dumbe 0.08ab 0.14cdefgh 0.19hi
Mgingqeni 0.09ab 0.13bcdefg 0.18gh
Pitshi 0.09abc 0.12bcdef 0.19hi
4.3.4.3 Potassium content
Potassium content was affected by delayed planting (P < 0.001) at Ukulinga. There
was no significant effect of cultivar on potassium content, although the effect of
planting date depended on cultivar (Planting date x cultivar interaction: P = 0.011).
Delaying planting from October to November decreased potassium content of all
cultivars (Fig. 4.19). Potassium content was decreased further for Mgingqeni and
Pitshi when planting was further delayed by one month and increased thereafter for
Dumbe-dumbe and Pitshi. Mgingqeni had higher potassium content than Pitshi for
October planting whereas t had lowest potassium content than both Dumbe-dumbe
and Pitshi for January planting. There was no difference in potassium content
between the cultivars when they were planted in November and December.
Fertilisation had no significant effect on potassium content.
89
2.5 f Dumbe-dumbe
LSD(0.05) = 0.31
Potassium content (%)
ef Mgingqeni
2 de Pitshi
cd cd cd
1.5 bc bc
ab ab ab
a
1
0.5
0
Planting date
application rate on potassium content of taro cormels at Ukulinga.
At Umbumbulu, cultivar (P < 0.001) and Gromor Accelerator application rate (P <
0.001) displayed significant effect on potassium content of taro cormels. There was no
significant effect of planting date and the interactions of planting date, cultivar and
fertiliser. Dumbe-dumbe (1.73%) and Mgingqeni (1.73%) had significantly higher
potassium content than Pitshi (1.40%) but they were not significantly different from
each other (Fig. 4.20). Application of Gromor Accelerator enhanced potassium
content. Adding 160 N kg ha-1 of the organic fertiliser significantly increased the
potassium content of corms but further increasing of Gromor Accelerator to 320 N
kg ha-1 did not further increase potassium content of cormels (Fig. 4.21).
90
2.5
LSD(0.05) = 0.16
2 b b
1.5 a
0.5
0
Taro cultivar
application rate on potassium content of taro cormels at Umbumbulu.
2.5
LSD(0.05) = 0.16
2 b
b
1.5 a
0.5
0
0 160 320
-1
date and cultivar on potassium content of taro cormels at Umbumbulu.
4.3.4.4 Calcium content
All treatments and their interactions had no significant effect on calcium content at
Ukulinga, whereas at Umbumbulu only cultivar had significant effect on calcium
content (P = 0.002). Pitshi (0.0926%) displayed significantly highest calcium content
91
followed by Mgingqeni (0.0731%) and Dumbe-dumbe (0.0549%) respectively at
Umbumbulu (Fig. 4.22).
0.1 c
LSD(0.05) = 0.02
0.09
Calcium content (%)
0.08 b
0.07
0.06 a
0.05
0.04
0.03
0.02
0.01
0
Taro cultivar
application rate on calcium content of taro cormels at Umbumbulu.
4.3.4.5 Magnesium content
All treatments and their interactions had no significant effect on calcium content at
Umbumbulu, whereas at Ukulinga only cultivar had significant effect on calcium
content (P < 0.001). Cormels of Dumbe-dumbe (0.2243%) showed a significantly
higher magnesium content than Mgingqeni (0.1839%) and Pitshi (0.1710%) which
were not significantly different from each other (Fig. 4.23).
92
0.25 LSD(0.05) = 0.02
Magnesium content (%) b
0.2 a
a
0.15
0.1
0.05
0
Taro cultivar
application rate on magnesium content of taro cormels at Ukulinga.
4.3.4.6 Zinc content
Planting date significantly affected zinc content of taro cormels at Ukulinga (P =

0.041) and Umbumbulu (P = 0.026). Zinc content was also affected by fertilifation at
Umbumbulu (P = 0.034). Delaying planting from October to December increased zinc
content from 57.6 to 86.2 mg kg-1 but further delay in planting had no effect at
Ukulinga (Fig. 4.24). At Umbumbulu, zinc content was significantly increased when
planting was delayed by one month from October but further delay in planting
decreased zinc content to level obtained with October planting (Fig. 4.25). 320 N kg
ha-1 of fertiliser yielded higher zinc content than 160 N kg ha-1 of fertiliser but both
were not significantly higher than zinc content obtained when no fertiliser was applied
at Umbumbulu (Fig. 4.26).
93
100
LSD(0.05) = 19.11 b
b
Zinc content (mg kg )
-1
80 ab
a
60
40
20
0
Planting date
application rate on zinc content of taro cormels at Ukulinga.
20 b LSD(0.05) = 4.46
Zinc content (mg kg )
-1
15 a
a
a
10
0
Planting date
application rate on zinc content of taro cormels at Umbumbulu.
94
20 LSD(0.05) = 3.48
Zinc content (mg kg ) b
-1
ab
15
a
10
0
0 160 320
date and cultivar on zinc content of taro cormels at Umbumbulu.
4.3.4.7 Copper content
There was significant (P = 0.012) interaction between planting date, cultivar and
Gromor Accelerator application rate in the copper content of the corms at
Umbumbulu. When no fertiliser was applied, delaying planting of Pitshi from
October to December decreased copper content of corms but further delay of planting
to January increased it again (Table 4.2). When 320 N kg ha-1 of Gromor Accelerator
was applied, delaying planting of Mgingqeni from November to December
significantly decreased copper content and increased it again when planting was
further delayed until January. For Pitshi, when 320 N kg ha-1 of Gromor Accelerator
was applied, delaying planting until December significantly increased copper content
of corms. Dumbe-dumbe had significantly higher copper content with 320 N kg ha-1
than with 160 N kg ha-1 of Gromor Accelerator for November planting. For
December planting Mgingqeni displayed significantly higher copper content with 0 kg
ha-1 than with 320 N kg ha-1 of Gromor Accelerator, whereas for Pitshi applying the
organic fertiliser enhanced copper content of corms. For January planting Dumbe-
dumbe showed significantly highest copper content with 320 N kg ha-1.
95
Table 4.2 Effect of planting date and Gromor Accelerator application rate on copper
content of taro cormels of different cultivars at Umbumbulu. Means of the interaction
effect within columns and rows (LSD (0.05) = 1.67) followed by the same letter are
not significantly different.
Planting time Cultivar Gromor Accelerator (N kg ha-1)
0 160 320
October Dumbe-dumbe 4.83ab 5.53aef 5.80ah
Mgingqeni 7.87cdj 8.57gj 7.60hij
Pitshi 7.97dg 7.30fg 6.83ghi
November Dumbe-dumbe 5.27ab 4.93be 7.00ahi
Mgingqeni 7.47cd 8.20cg 9.27cjk
Pitshi 7.70fcd 7.63fg 7.83fij
December Dumbe-dumbe 4.90abe 5.87ef 6.27ehi
Mgingqeni 8.83d 7.57dfg 6.93ghi
Pitshi 6.56bc 8.64gk 10.14k
January Dumbe-dumbe 4.33ae 4.33e 6.43hi
Mgingqeni 9.27dl 9.03gl 9.27jkl
Pitshi 8.27cdm 8.93gm 8.99jkm
4.3.4.8 Manganese content
Manganese content of taro cormels was significantly (P = 0.028) affected by the

interaction of planting date and cultivar at Ukulinga. Delaying planting from
November to December significantly decreased manganese content of Pitshi, whereas
for Mgingqeni manganese content was significantly decreased by delaying planting
from October to December and from December to January (Fig. 4.27). The lowest
manganese content was obtained with January planting.
96
20
Manganese content (mg kg )
-1
Mgingqeni
15 d Pitshi
d cd d
10 bc b bc
ab ab
ab
ab
a
5
0
Planting date
application rate on manganese content of taro cormels at Ukulinga.
Cultivar (P < 0.001) and the interaction of planting date and Gromor Accelerator
application rate (P = 0.028) significantly affected manganese content at Umbumbulu.
Dumbe-dumbe showed significantly lower (11.80 mg kg-1) manganese content than
Mgingqeni (17.81 mg kg-1) and Pitshi (17.78 mg kg-1) which were not significantly
different from each other (Fig. 4.28). Delaying planting by one month from October
significantly increased manganese content of taro cormels when no fertiliser was
added followed by a significant decline when planting was done in December (Fig.
4.29). When 320 N kg ha-1 of the organic fertiliser was applied, manganese content
was increased by delaying planting until November and it thereafter decreased. There
was no difference between planting dates when 160 N kg ha-1 of the fertiliser was
applied. There was also no significant difference in manganese content between
fertiliser application rates when planting was done in October, December and January.
When planting was done in November, applying 160 N kg ha-1 of the fertilizer yielded
lowest manganese content compared to the other application rates.
97
20
-1 b b
LSD(0.05) = 2.80
15
a
10
0
Taro cultivar
application rate on manganese content of taro cormels at Umbumbulu.
LSD(0.05) = 5.62 0 N kg ha-1

30
160 N kg ha-1
-1
c c
25 320 N kg ha-1
bc
20 ab
ab
15 a a ab ab
a a a
10
0
Planting date
Figure 4.29 Effect of planting date and Gromor Accelerator application rate on
manganese content of taro cormels at Umbumbulu.
98
4.4 Discussion
Mohamed (1985) found dry matter to be highest at early planting for two of the potato
cultivars while it was highest for the other cultivar when it was planted at a later date.
It has been indicated that excessive nitrogen fertilisation reduces the proportion of dry
matter allocated to the corms (Manrique, 1994). Allison et al. (1999) also found tuber
dry matter content to decrease at increased nitrogen rates. Bèlanger et al. (2002), Casa
et al. (2005) and Zelalem et al. (2009) obtained similar results in potatoes.
Feltran et al. (2004) found specific gravity to be influenced by cultivar. Similar results
were also found by Long et al. (2004) that the major factor influencing specific
gravity in potatoes was genotype. Contrasting results have been reported by different
studies. Sparrow and Chapman (2003) reported that specific gravity often decreases
as the rate of nitrogen increases, whereas El-Sirafy et al. (2008) found that application
of farm yard manure increased specific gravity of tubers. However, it was shown that
specific gravity was lower in nitrogen deficient tubers as well as in those from plants
where nitrogen was in excess of that needed for maximum yield, at three out of four
sites in South Australia (Dahlenberg et al., 1990).
The chemical composition of potatoes varies with variety, area where they are grown
and cultural practices. Previous studies have shown that fertilisation influences the
water and nutrient supply to the plant, which can in turn affect the nutritional quality
of the harvested plant part (Kader and Rolle, 2004). Huang et al. (2007) on the other
hand, indicated that upland-cultivated taro corms tended to have higher mineral
content than paddy taro which might explain the higher mineral content of corms
from Ukulinga due to lower rainfall. Kader and Rolle (2004) stated that temperature
influences the uptake and metabolism of mineral nutrients by plants and this possibly
also explain the high content of other minerals at Umbumbulu as a result of higher
temperatures. Increase in the fertiliser application was also found to increase nitrogen
concentration of tubers (Colla et al., 2005).
99
The chemical composition of taro is influenced by both planting date and organic
fertiliser amount, but the response differs with cultivar. Organic fertiliser level has no
linear positive effect on its improvement of taro corm quality, but dry matter content
and specific gravity are generally negatively affected by increasing fertiliser levels.
Dumbe-dumbe performed better than the other two cultivars, especially when planted
in October with no fertiliser. It is recommended that to improve taro specific gravity,
dry matter content and mineral content, the crop should be planted in October at both
study sites.
100
References
Allison, M.F., Allen, E.J. and Fowler, J.H. 1999. The nutrition of the potato crop.
British Potato Council Research Review 1991/1. BPC Oxford, UK.
Bèlanger, G., Walsh, J.R., Richards, J.E., Milburn, P.H. and Ziadi, N. 2002. Nitrogen
fertilisation and irrigation affects tuber characteristics of two potato cultivars.
American Journal of Potato Research 79 (4): 269-279.
improvement through nitrogen management optimisation: Review of
methodologies. Acta Horticulturae 684: 65–71.
Colla, G., Cardarelli, M., Rouphael, Y., Fiorillo, A. and Cirica, B. 2005. Response of
potato to organic fertiliser rate: yield, tuber quality and nutrient uptake. Acta
Dahlenberg, A.P., Maier, N.A. and Williams, C.M.J. 1990. Effect of nitrogen on the
size, specific gravity, crisp colour and reducing sugar concentration of potato
tubers (Solanum tuberosum L.) cv Kennebec. Australian Journal of
Experimental Agriculture 30: 123-130.
El-Sirafy, Z.M., Abbady, K.A., El-Ghamry, A.M and El-Dissoky, R.A. 2008. Potato
yield quality, quantity and profitability as affected by soil and foliar potassium
application. Research Journal of Agriculture and Biological Sciences 4 (6):
912-922.
Feltran, J.C., Lemos, L.B. and Vieites, R.L. 2004. Technological quality and
utilization of potato tubers. Scientia Agricola 61 (6): 598-603.
Fontana, F., D'Amato, A., Cremaschi, D. and Salmi, A. 1990. Water requirements and
nitrogen fertilization in potato growing in the Padana Plain. Acta
Günel, E. and Karadoĝan, T. 1998. Effect of irrigation applied at different growth
stages and length of irrigation period on quality characters of potato tubers.
Potato Research 41 (1): 9-19.
Haase, T., Schüler, C., Haase, N.U. and Heß, J. 2007. Suitability of organic potatoes
for industrial processing: Effect of agronomical measures on selected quality
parameters at harvest and after storage. Potato Research 50 (2): 115-141.
101
Huang, C.C., Che, W.C. and Wang, C.C.R. 2007. Comparison of Taiwan paddy and
upland-cultivated taro (Colocasia esculenta L.) cultivars for nutritive values.
Food Chemistry 102: 250-256.
Bulletin 152.
Kumar, P., Pandey, S.K., Singh, B.P., Singh, B.P. and Kumar, D. 2007. Effect of
nitrogen rate on growth, yield, economics and crisps quality of Indian potato
processing cultivars. Potato Research 50 (2): 143-155.
Long, C.M., Snapp, S.S., Douches, D.S. and Chase, R.W. 2004. Tuber yield,
storability and quality of Michigan cultivars in response to nitrogen
management and seed spacing. American Journal of Potato Research 81: 347-
357.
Manrique, L. A. 1994. Nitrogen requirements of taro. Journal of plant nutrition 17(8):
1429-1441.
McPharlin, I.R. and Lancaster, R.A. 2005. Applied nitrogen effects on yield and
crisping quality of ‘Atlantic’, ‘Bliss ‘ and ‘Dawmor’ potatoes. Acta
Mittra, B. N. and George, J. 2000. Effect of time of planting on the growth and dry
matter partitioning in sweet potato. Journal of Root Crops 26 (2): 57-60.
Mohamed, A.I. 1985. Dry matter, protein content and yield of potato as affected by
cultivar and planting date in Sudan. Acta Horticulturae 158: 139- 141.
Mokolobate, M.S. and Haynes, R.J. 2002. Comparative liming effect of four organic
residues applied to an acid soil. Biology and Fertility of Soils 35: 79–85.
Mondy, N.I. and Koch, R.L. 1978. Influence of nitrogen fertilisation on potato
discoloration in relation to chemical composition. 1. Lipid, potassium, and
dry matter content. Journal of Agricultural Food Chemistry 26 (3):666 – 669.
Nadler, A. and Heuer, B. 1995. Effect of saline irrigation and water deficit on tuber
quality. Potato Research. 38:119–123.
Naramabuye, F.X., R.J. Haynes, R.J and Modi, A.T. 2008. Cattle manure and grass
residues as liming materials in a semi-subsistence farming system.
Agriculture, Ecosystems and Environment 124: 136-141.
102
Pocknee, S. and Summer, M.E. 1997. Cation and nitrogen contents of organic matter
determine its liming potential. Soil Science Society of American Journal 61:
86-92.
Rashid, K.and Waithaka, K. 1985. The effect of phosphorus fertilization on growth
and tuberization of sweet potato , Ipomoea batatas L. Acta Horticulturae 153:
345-353.
Shange, L.P. 2004.Taro (Colocasia esculenta (L.) Schott) production by small-scale
KwaZulu-Natal, Pietermaritzburg.
Sparrow, L.A. and Chapman, K.S.R. 2003. Effects of nitrogen fertiliser on potato
(Solanum tuberosum L., cv. Russet Burbank) in Tasmania. 1. Yield and
quality. Australian Journal of Experimental Agriculture 43: 631- 641.
Tawfik, A.A., Mansour, S.A., Ramadan, H.M. and Fayad, A.N. 2002. Processing
quality of selected potato varieties for chip and French fry industries in Egypt.
African Crop Science Journal 10 (4): 325-333.
Wadas, W., Jablońska-Ceqlarek, R. and Kosterna, E. 2004. Effect of plastic covering
and nitrogen fertilisation on yield and quality of early potatoes. Folia
Horticulturae16 (2): 41-48.
Westermann, D.T., Tindall, T.A., James, D.W. and Hurst, R.L. 1994. Nitrogen and
potassium fertilization of potatoes: yield and specific gravity. American Potato
Journal 71: 417-432.
Wong, M.T.F., Nortcliff, S. and Swift, R.S. 1998. Method for determining the acid
ameliorating capacity of plant residue compost, urban waste compost,
farmyard manure and peat applied to tropical soils. Communication in Soil
Science and Plant Analysis 29: 2927-2937.
tuberosum L.) to different rates of nitrogen and phosphorus fertilisation on
vertisols at Debre Berhan, in the central highlands of Ethiopia. African
Journal of Plant Science 3 (2): 16-24.
103
CHAPTER 5
CHANGES IN THE SURFACE MORPHOLOGY OF STARCH GRANULES,

ALPHA AMYLASE ACTIVITY AND SPROUTING OF TARO, COLOCASIA
ESCULENTA L. SCHOTT DURING STORAGE.
Abstract
The surface morphology of starch grains changes during storage. Scanning electron
microscopy was used to examine the changes that occur to the surface morphology of
starch granules of cormels of Dumbe-dumbe, Mgingqeni and Pitshi stored at ambient
temperature. Alpha-amylase activity and sprouting were determined to establish
relationships between physical changes in the starch granules and alpha-amylase
activity as well as sprouting. The surface of the Dumbe-dumbe and Pitshi starch
granules were smooth at harvest but during storage the surface of the granules became
progressively more pitted. The surface of starch granules of Mgingqeni was not
smooth at harvest and during storage degradation was more pronounced for
Mgingqeni compared to the other two taro cultivars. The degradation of starch
granules, alpha amylase activity and sprouting increased with storage time. The
results also show that alpha-amylase activity and sprouting increase with storage
temperature. Cormels stored in polyethylene bags and Mgingqeni showed highest
alpha-amylase activity and sprouting and hence starch degradation.
Key Words: alpha-amylase, Colocasia esculenta, sprouting, starch granules, storage.
5.1 Introduction
The surface morphology of starch granules changes during storage (Cottrell et al.,
1993). This is observed by various patterns of erosion and it is due to starch
degradation. Extensive surface erosion was shown to indicate a high degree of
hydrolysis, whereas less surface erosion indicated less degradation (Zhang and Oates,
1999). The normal visible response to degradation in vivo appeared to be varying
degrees of roughness of the granule surface (Silva and Luh, 1978, Wetzstein and
Wetzstein, 1981) or holes bored through the granule (Dronzek et al, 1972). The same
results were found by Cottrell et al. (1993) who reported pitting and indentations in
potatoes, Hoover and Sosulki (1991) who reported surface pitting and erosion in
104
legumes and Valetudie et al. (1993) who reported that roughening of the surface with
pitting and fissures along the edges of the planes of contact of the starch granules in
tannia, sweet potato and cassava and external pitting on only one face of the granule
to produce an internal cavity in yams. In chickpea, on the other hand, no difference in
character of starch granules could be seen during starch degradation (Fernandez and
Berry, 1989).
The erosion is believed to be brought about by alpha-amylase because cereal starch

granules showed holes in the surface and corroded channels into the interior when
subjected to in vitro digestion with alpha amylase (Dronzek et al., 1972, Macgregor
and Balance, 1980; Sreenath 1992). The activity of the enzyme was reported to
increase during storage (Cochrane et al., 1991; Cottrell et al., 1993). Fincher (1989)
previously stated that degradation is initiated by synthesis of new alpha amylase in
wheat and barley. The surface of potato granules showed resistance to enzyme attack
with a few shallow pits and ridges initially but it degrades rapidly once the enzyme
reaches the interior of the granule (Gallant et al, 1972, 1973).
According to Biemelt et al. (2000) starch degradation is not a prerequisite for the
initiation of sprouting. They observed no change in starch degrading enzymes prior
visible sprout growth and stimulated alpha-amylase activity by approximately 30%
was observed after the onset of sprouting in potatoes. Lorenz et al. (1981), on the
other hand, found that the production of modified starches was possible by sprouting
corn, barley and triticale grains. They also reported that the starches susceptibility to
the enzyme increased with sprouting. Sun and Henson (1990) indicated that alpha-
amylase is synthesized during germination in barley and this might be the case in taro,
that alpha-amylase is synthesized during sprouting. Fernandez and Berry (1989) and
Frias et al. (1998) reported that germination sharply increased the enzymatic
digestibility of starch granules from the chickpea and lentil, but no change in the
appearance of scanning electron micrographs could be attributed to germination
(Fernandez and Berry, 1989).
The present study was undertaken to study the differences in the changes in the
surface morphology of three taro starches and the relationship between these changes
105
and alpha-amylase activity and sprouting of corms, and the effect of storage
temperature and packaging on sprouting and alpha-amylase of taro corms.
5.2.1 Plant material used
Cormel materials used in this study were of the three taro cultivars Dumbe-dumbe,
Mgingqeni and Pitshi which were planted in October without fertiliser at Ukulinga.
The post harvest storage experiment was arranged in a split-split plot design with
temperatures (12oC and ambient temperature) as main plots, packaging (polyethylene
bag, box and mesh bag) as sub-plots and cultivars as sub-sub-plots. Each storage
temperature was replicated three times and the cultivars were randomised within
packaging within storage temperatures. Each packaging contained 7, 11 and 13
cormels of Dumbe-dumbe, Mgingqeni and Pitshi respectively. One cormel was
sampled, peeled, cut into slices, immediately frozen in liquid nitrogen and freeze-
dried monthly for four months starting one month after storage for chemical analysis.
5.2.2 Scanning electron microscopy
Scanning electron microscopy was performed at harvest and monthly for four months
starting one month after storage for cormels of the three taro cultivars which were
stored at ambient temperature. The freeze-dried cormel materials were mounted on
the brass stubs using double-sided cellotape, and then coated with gold palladium,
using a Polaron E5100 sputter coater. Samples were viewed at 10kW using a Hitachi
S-570 scanning electron microscope (SEM). Starch grains were spot viewed and the
area for starch grain viewing was selected randomly. The micrographs were used to
compare the morphology (form and structure) of the starch granules.
5.2.3 Alpha-amylase activity
Alpha-amylase activity was determined at harvest and monthly for four months
starting one month after storage according to Modi and Cairns (1994) with
modifications. Only alpha amylase activity of corms stored at ambient temperature
was used to study the relationship between the changes in the surface morphology of
106
three taro starch granules, alpha-amylase activity and sprouting of cormels. Freeze-
dried corm material was ground to pass through a 0.5mm motor mill sieve.
Approximately 0.1g was weighed into a 100mL Erlenmeyer flask and incubated in
10mL of the extraction buffer [5g NaCl: 0.2g Ca (C2H3O2)2, mass/mass]. Alpha-
amylase was extracted by shaking the contents for 20 min in a water bath at 30oC.
Contents were then filtered through a Whatman No.1 filter paper into a test tube. One
millilitre of the supernatant was diluted 10 times in the extraction buffer and one
Phadebas tablet (Pharmacia Diagnostics AB, Uppsala, Sweden) was introduced into
the test tube. Phadebas tablet is composed of a substrate made made by cross-linking
partially hydrolysed starch, using 1.4 butandiol-diglycidether as a cross-linking agent.
The soluble starch is transformed into a three-dimensional insoluble lattice network
which expands in water. The number of cross-linking bridges regulate the degree of
expansion and susceptibility of the substrate to alpha-amylase attack. The substrate is
labeled with Cibachron Blue by covalent bonds. Each tablet contains 45 mg of blue
starch polymer, and 25 mg buffer salt to give 0.2 mol dm-3 phosphate buffer, pH 7 and
0.05 mol dm-3 NaCl when appropriately dissolved. The tablets also contain bovine
serum albumin. The contents of the test tube were mixed for 30 sec on a vortex mixer
and then incubated in the shaking water bath at 50oC for 10 min. The reaction was
then terminated by adding 1mL of 0.5M NaOH and the contents were filtered through
Whatman No.1 filter paper. Absorbance of the supernatant was determined at 620 nm
in a spectrophotometer.
5.2.4 Sprouting
Number of sprouted cormsel was counted monthly for four months starting one month
after storage and percentages were calculated. Cormels were considered sprouted
when they had developed either shoot or roots =/>3.0 mm.
5.2.5 Statistical analysis
Analysis of variance was performed using GenStat Release 11.1 (2008). Least
107
5.3 Results
5.3.1 Scanning electron microscopy
The appearance of starch granules changed considerably during storage. The starch
granules of Dumbe-dumbe and Pitshi appeared smooth at harvest (Figs 5.1 and 5.3).
For Mgingqeni, some granules were smooth while some were rough (Fig. 5.2). After
one month in storage all cultivars showed signs of degradation, although the severity
varied between them. Dumbe-dumbe (Fig.5.4 1a) and Mgingqeni (Fig.5.4 1b) were
showing some indentations whereas Pitshi (Fig. 5.4 1c) was showing some
depressions. The indentations appeared shallow and did not penetrate deep into the
body of the granules.
By the second sampling after two months, granules from all cultivars showed
evidence of erosion. Granules from Dumbe-dumbe (Fig.5.4 2a) and Pitshi (Fig.5.4 2c)
showed deep indentations but they were not as deep as those showed by Mgingqeni
(Fig.5.4 2b). After three months in storage, Mgingqeni (Fig.5.4 3b), like after two
months showed more degradation compared to Dumbe-dumbe (Fig.5.4 3a) and Pitshi
(Fig.5.4 3c) which were still showing some indentations though they seemed to be
deeper than after two months. By the final sampling after four months, further more
degradation was evident and it was more pronounced for Mgingqeni (Fig.5.4 4b)
followed by Pitshi (Fig.5.4 4c) and Dumbe-dumbe (Fig.5.4 4a) respectively. The
smaller pieces were starting to break and fall from the starch granules.
108
Figure 5.1 Scanning electron micrographs showing starch granules of Dumbe-dumbe
at harvest.
Figure 5.2 Scanning electron micrographs showing starch granules of Mgingqeni at

harvest.
109
Figure 5.3 Scanning electron micrographs showing starch granules of Pitshi
at harvest.
110
1a 1b 1c
2a 2b 2c
3a 3b 3c
4a 4b 4c
Figure 5.4 Scanning electron micrographs showing starch granules of taro cultivars
stored in polyethylene bags for four months at ambient temperature. (a, b and c
represents Dumbe-dumbe, Mgingqeni and Pitshi respectively and 1, 2, 3 and 4
represents the sampling time in months (The whitish substances are fragments of cell
walls).
111
5.3.2 Alpha-amylase activity
Alpha-amylase activity of cormels planted without fertiliser in October at Ukulinga

generally increased with time in storage (Fig. 5.5). After one month in storage alpha-
amylase increased slightly for all cultivars with Dumbe-dumbe having the highest
alpha-amylase activity followed by Pitshi and Mgingqeni respectively. A decline was
observed after two months for both Dumbe-dumbe and Mgingqeni whereas Pitshi
increased slightly until three months after storage. Mgingqeni displayed the lowest
alpha-amylase activity from one to four months after storage. Alpha-amylase activity
increased rapidly from three to four months after storage.
0.18
Alpha-amylase activity (E.U. 0.1g )
-1
Dumbe-dumbe
0.16
Mgingqeni
0.14 Pitshi
0.12
0.1
0.08
0.06
0.04
0.02
0
0 1 2 3 4
Months after storage
Figure 5.5 Alpha-amylase activity in taro cormels at harvest up to four months after
storage in polyethylene bags at ambient temperature.
The interaction between storage temperature, packaging method and months after
storage significantly affected the alpha-amylase activity of taro cultivars (P < 0.001).
Alpha-amylase activity generally increased with time in storage. The cultivars were
not different from each other in alpha-amylase activity at harvest (Table 5.1).
Ambient temperature generally showed higher alpha-amylase activity than 12oC.
Polyethylene packaging and Pitshi also displayed higher alpha-amylase activity
compared to mesh bag and box and Mgingqeni and Dumbe-dumbe respectively.
112
Table 5.1 Alpha-amylase activity of taro cultivars stored at different temperatures in different packaging materials. Means of the interaction
effect within columns and rows (LSD (0.05) = 0.04) followed by the same letter are not significantly different.
Temperature Packaging Landrace 0 1 2 3 4
o
12 C Polyethylene Dumbe-dumbe 0.05 a 0.08 b 0.10 c 0.08 a 0.09 abc
Mgingqeni 0.04 a 0.08 b 0.09 bc 0.08 a 0.11 cd
Pitshi 0.03 a 0.08 b 0.09 bc 0.10 a 0.08abc
Box Dumbe-dumbe 0.05 a 0.08 b 0.07 bc 0.07a 0.07 abc
Mgingqeni 0.04 a 0.07 ab 0.06 abc 0.02 a 0.06 ab
Pitshi 0.03 a 0.05 ab 0.05 ab 0.07a 0.09 abc
Mesh bag Dumbe-dumbe 0.05 a 0.06 ab 0.08 bc 0.07a 0.05 a
Mgingqeni 0.04 a 0.08 b 0.07 abc 0.08 a 0.10 bd
Pitshi 0.03 a 0.06 ab 0.06 abc 0.07 a 0.05 a
Ambient Polyethylene Dumbe-dumbe 0.05 a 0.08b 0.08 bc 0.08 a 0.17 e
Temperature Mgingqeni 0.04 a 0.06 ab 0.06 abc 0.07 a 0.11 cd
Pitshi 0.03 a 0.07 ab 0.08 bc 0.08 a 0.14 de
Box Dumbe-dumbe 0.05 a 0.07 ab 0.07 bc 0.07 a 0.08abc
Mgingqeni 0.04 a 0.08 b 0.08 bc 0.08 a 0.08 abc
Pitshi 0.03 a 0.07 ab 0.06 abc 0.08 a 0.14 de
Mesh bag Dumbe-dumbe 0.05 a 0.06 ab 0.07 bc 0.07 a 0.07 abc
Mgingqeni 0.04 a 0.08 b 0.03 a 0.06 a 0.05 a
Pitshi 0.03 a 0.03 a 0.08 bc 0.08 a 0.11 cd
113
5.3.3 Sprouting
The three cultivars did not sprout in polyethylene bags after one month in storage at
ambient temperature (Fig. 5.6). After one month in storage the cormels started
sprouting slowly with Mgingqeni showing a significantly higher sprouting percentage
followed by Dumbe-dumbe which was not different from Pitshi. The cultivars started
sprouting rapidly after two months in storage with Pitshi sprouting more rapidly than
Dumbe-dumbe. At four months in storage, all cormels of Mgingqeni had sprouted
followed by Pitshi and Dumbe-dumbe respectively.
100
Dumbe-dumbe
Mgingqeni
80
Pitshi
Sprouting (%)
60
40
20
0
1 2 3 4
Figure 5.6 Sprouting in taro cormels at harvest up to four months after storage in
polyethylene bags at ambient temperature.
The interaction between storage temperature, packaging method and months after
storage significantly affected the sprouting of taro cultivars (P < 0.001). Sprouting
increased with time in storage (Table 5.2). At 12oC, the cormels did not sprout until
after three months in storage in all packaging methods. After four months the cormels
of all cultivars had sprouted with polyethylene bags showing significantly highest
sprouting followed by those in boxes and mesh bags which were also significantly
different from each other. Mgingqeni and Pitshi in mesh bags did not show
significantly different sprouting after four months in storage at 12oC. At ambient
temperature, sprouting started after one month for polyethylene packaged cormels
114
only with Mgingqeni showing significantly highest sprouting than the other cultivars.
After two months in storage sprouting had started for all packaging methods. Cormels
in mesh bags showed significantly lowest sprouting after three months in storage with
Dumbe-dumbe showing more sprouting than the other cultivars in mesh bags and
boxes. Mgingqeni showed the highest sprouting in polyethylene bags. After four
months, Dumbe-dumbe displayed the highest sprouting than other cultivars except in
polyethylene bags where all Mgingqeni cormels had sprouted. Ambient temperature
generally showed rapid sprouting with highest number of sprouted cormels compared
to the colder temperature.
Table 5.2 Sprouting of taro cormels stored at different temperatures in different

packaging materials. Means of the interaction effect within columns (LSD (0.05) =
5.57) followed by the same letter are not significantly different.
Temperature Packaging Cultivar 1 2 3 4
o
12 C Polyethylene Dumbe-dumbe 0 0 0 36.36de
Mgingqeni 0 0 0 60.61h
Pitshi 0 0 0 50.51f
Box Dumbe-dumbe 0 0 0 54.55fg
Mgingqeni 0 0 0 33.33cd
Pitshi 0 0 0 21.21b
Mesh bag Dumbe-dumbe 0 0 0 16.16b
Mgingqeni 0 0 0 10.10a
Pitshi 0 0 0 7.07a
Ambient Polyethylene Dumbe-dumbe 0 5.56a 36.36d 58.84gh
Temperature Mgingqeni 0 19.44b 66.67f 100.00k
Pitshi 0 2.22a 57.07e 79.49i
Box Dumbe-dumbe 0 0 79.49g 87.18j
Mgingqeni 0 0 40.61d 55.57fgh
Pitshi 0 0 30.77c 40.61e
Mesh bag Dumbe-dumbe 0 0 25.15b 30.77c
Mgingqeni 0 0 10.00a 20.51b
Pitshi 0 0 11.62a 15.81b
5.4 Discussion
Scanning electron micrographs showing the changes in surface morphology of starch

granules of taro cultivars during storage demonstrate that starch break down occurs
early in storage but the extent varies with cultivars. The starch granules breakdown
starts with roughening of the surface followed by development of small pits and
indentations which becomes deeper with time and cause the granules to open up and
breakdown onto smaller granules. The same results were previously reported in
115
legumes, potatoes, tannia, sweet potato and cassava (Hoover and Sosulki, 1991;
Cottrell et al., 1993; Valetudie et al., 1993). Extensive surface erosion was displayed
by Mgingqeni and according to Zhang and Oates (1999) that might suggests a higher
degree of hydrolysis compared to the other cultivars.
The alpha-amylase increase during the first month of cormel storage and decrease
thereafter was previously reported by Modi (2004) though it increased again after two
months in this study. This increase was also reported by Panneerselvam et al. (2009)
in Dioscorea esculenta that alpha-amylase activity was lower up to 35 days after
harvest and afterwards it increased to a rapid phase. The activity of the enzyme was
also reported to increase in potatoes during storage (Cochrane et al., 1991; Cottrell et
al., 1993). Modi (2004) reported that corm sprouting is associated with alpha-
amylase activity which enhanced starch mobilisation. This is possible considering the
increase in alpha-amylase activity which correlates with the starch granules break
down (Fig. 5.4) and sprouting (Fig. 5.6). This suggests that the enzyme is responsible
for the break down of starch granules and sprouting, moreover because of a report that
at the time of sprouting alpha-amylase activity increased to a higher level when
compared to early storage period (Panneerselvam et al., 2009). The increase in alpha-
amylase was positively correlated with germination in cherry tomato (Modi and
White, 2004).
The starch granules breakdown was only monitored at ambient temperature but it is
believed that the same process takes place at a lower temperature though it might be
slower since alpha-amylase activity and sprouting were lower at lower temperature.
This was in contrast with what was earlier reported by Nielsen et al. (1997) that
alpha-amylase was not affected by low storage temperature. Findings of the previous
study (Modi, 2004) suggested that air and temperature conditions could have
influenced corm sprouting.
The present work demonstrates the changes in surface morphology of starch granules
during storage. The degradation of starch granules, alpha-amylase activity and
sprouting increased with storage time. This might suggest that the starch granules
degradation and sprouting are as a result of alpha-amylase activity. The results also
116
show that alpha amylase activity and sprouting increase with storage temperature and
this suggests that starch degradation increases with increase in storage temperature.
Corms of Mgingqeni stored in polyethylene bags showed highest alpha-amylase
activity as well as sprouting and hence starch degradation.
117
References
Biemelt, S., Hajirezaei, M., Hentschel, E. and Sonnewald, U. 2000. Comparative

analysis of abscisic acid content and starch degradation during storage of
tubers harvested from different potato varieties. Potato
Research 43 (4): 297-393.
Cochrane, M.P., Duffers, C.M., Allison, M.J. and Mackay, G.R. 1991. Amylolytic
activity in stored potato tubers. 2. The effect of low temperature storage on
the activities of α- and β-amylase and α-glucosidase in potato tubers.
Potato Research 34: 333–341.
Cottrell, J.E., Duffus, C.M., Mackay, G.R. and Allison, M.J. 1993. Changes in the
surface morphology of starch granules of the cultivated potato, Solanum
tuberosum L. during storage. Potato Research 36 (2): 119-125.
Dronzek, B.C., Huwang, P. and Bushuk, W. 1972. Scanning electron microscopy of
starch from spouted wheat. Cereal Chemistry 49: 232-239.
Fernandez, M.L. and Berry, J.W. 1989. The effect of germination on chickpea starch.
Starch/Staerke 41: 17–21.
Fincher, G.B. 1989. Molecular and cellular biology associated with endosperm
mobilization in germinating cereal grains. Annual Review of Plant
Physiology and Plant Molecular Biology 40: 305-346.
Frias, J., Fornal, J., Ring, S.G. and Vidal-Valverde, C., 1998. Effect of germination on
physicochemical properties of lentil starch and its components.
Lebensmittel-Wissenschaft und-Technologie 30: 178–183.
Gallant, D., Mercier, C. and Guilbot, A. 1972. Electron microscopy of starch granules
modified by bacterial alpha-amylase. Cereal Chemistry 49: 354-365.
Gallant, P. D., Derrien, A., Aumaitre, A., and Massy, A. 1973. Dégradation in vitro de
l’amidon par le suc pancréatique. Starch/Staerke 25:5 6-64.
Hoover, R. and Sosulki, F.W. 1991. Effect of cross- linking on functional properties
of legume starches. Starch/Stärke 38: 149-155.
Lorenz, K, Collins, F. and Kulp, K. 1981. Sprouting of cereal grains. Effects on starch
characteristics. Starch/Stärke 33: 183–187.
118
MacGregor, A.W. and Balance, D.L. 1980. Hydrolysis of large and small starch
granules from normal and waxy barley cultivars by alpha-amylases from
barley malt, Cereal Chemistry 57: 397–402.
Modi, A.T. and Cairns, A.L.P. 1994. Molybdenum deficiency in wheat results in
lower dormancy levels via reduced ABA. Seed Science Research 4: 329-
333.
Modi, A.T. 2004. Short-term preservation of maize landrace seed and taro propagules
using indigenous storage methods. South African Journal of botany 70: 16-
23.
Modi, A.T and White, B.J. 2004.Water potential of cherry tomato (Lycopersicon
esculentum Mill.) placenta and seed germination in response to desiccation
during fruit development. Seed Science Research 14: 249-257.
Nielsen, T.H., Deiting, U. and Stitt, M. 1997. A β-amylase in potato tubers is induced
by storage at low temperature. Plant Physiology 113: 503-510.
Panneerselvam, R., Abdul Jaleel, C., Somasundaram, R., Sridharan, R. and
Gomathinayagam, M. 2007. Carbohydrate metabolism in Dioscorea
esculenta (Lour.) Burk. Tubers and Curcuma longa L. rhizomes during two
phases of dormancy. Colloids and Surfaces B: Biointerfaces 59: 59-66.
Silva, H.C. and Luh, B.S. 1978. Scanning electron microscopy studies on starch
granules of red kidney beans and bean sprouts. Journal of Food Science
43 (5): 1405-1408.
Sreenath, H.K. 1992. Studies on starch granules digestion by alpha-amylase.

Starch/Stärke 44 (2)61–83.
Sun, Z. and Henson, C. A. 1990. Degradation of Native Starch Granules by Barley
alpha-Glucosidases. Plant Physiology 94 (1): 320–327.
Valetudie, J.C., Colonna, P., Bouchet, B. and Gallant, D.J. 1993. Hydrolysis of
tropical tuber starches by bacteria and pancreatic α-amylases. Starch/Stärke
45: 270–276.
Wetzstein, H.Y. and Wetzstein, M.E. 1981. Scanning electron microscopy and size
analysis of temperature induced in vivo potato starch breakdown. Journal of
the American Society for Horticultural Science 106: 688-690.
119
Zhang, T. and Oates, C.G. 1999. Relationship between α-amylase degradation and
physico-chemical properties of sweet potato starches. Food Chemistry 65
(2): 157-163.
120
CHAPTER 6
EFFECTS OF PRE- AND POST-HARVEST PRACTICES ON

CARBOHYDRATES OF TARO (COLOCASIA ESCULENTA).
Abstract
The effect of four planting date (October, November, December and January) and
three fertiliser application rates (0, 160 N and 320 N kg ha-1) on starch and reducing
sugars content was investigated for the three taro cultivars from KwaZulu-Natal at
Ukulinga and Umbumbulu. The effect of two storage temperatures (12oC and ambient
temperature) and three packaging methods (polyethylene bags, mesh bags and open
boxes) on starch and reducing sugars following storage was also investigated for three
cultivars. At Ukulinga, delaying planting significantly decreased starch content
whereas at Umbumbulu starch content was significantly decreased by application of
fertiliser. Reducing sugars were reduced by delay in planting when no fertiliser was
applied but increased when 320 N kg ha-1 was applied at Ukulinga. Fertilisation had
no effect on reducing sugars of Dumbe-dumbe but increased reducing sugars of
Mgingqeni. For Pitshi reducing sugars were increased when 320 N kg ha-1 of fertiliser
was applied. At Umbumbulu, delaying planting generally increased reducing sugars
for Dumbe-dumbe and Pitshi and decreased reducing sugars for Mgingqeni. Ambient
temperature and mesh bag showed the best combination for taro cormel storage.
Key Words: Colocasia esculenta, planting date, starch, storage, sugars.
6.1 Introduction
Taro [Colocasia esculenta (L.) Schott] commonly known as amadumbe in Zulu, has
high potential for alleviation of food insecurity in rural areas of KwaZulu-Natal.
Recent developments indicated that taro also has a potential to compete with potato
(Solanum tuberosum) as a processing crop to produce crisp chips. However, no
research has been done on the pre- and post-harvest physiology of taro. Starch and
sugar contents are important quality parameters in crisp making as texture and colour
of crisps depend on them (Kita, 2002; Lisińska and Laszczyński, 1989; Pavlista
1997). There is lack of knowledge on the effects of pre- and post-harvest practices on
121
the carbohydrates of the crop. It is important to understand the factors that influence
the starch and sugar contents especially reducing sugars of taro so as to reduce losses
during storage. In potatoes, higher starch content is required for processing (Mitch,
1984). Starch and sugar properties vary with potato cultivars (Lisińska and
Laszczyński, 1989). Starch and sugar contents are also affected by cultivar,
temperature, moisture and fertilisation (Lisińska and Laszczyński, 1989; Wolfe et al,
1991; Feltran et al, 2004). Starch quality is influenced by environmental conditions
during crop growth and storage. Potato cultivars differ in their sensitivity to storage
temperature. Reducing sugar content of tubers was reported to have increased sharply
during the first weeks of storage at 4°C whereas at 10°C, reducing sugar content
remained constant or increased only slightly (Cottrell et al., 1993).
Proietti et al. (2005) reported large variation in total soluble sugars (glucose, fructose
and sucrose) between cultivars and the sowing times in potatoes. Reduced
incorporation of assimilates into starch in the tubers at high temperatures and
increased amount of labelled sucrose was reported by Wolfe et al. (1991). Nitrogen
has a more pronounced effect than phosphorus or potassium fertilisers. High doses of
nitrogen (200 kg ha-1) results in low starch content and high reducing sugars in
potatoes due to prolonged growing season which results in immature tubers at harvest
(Lisińska and Leszczyński, 1989; Sowokinos, 1990). Kyriacou et al. (2009), on the
contrary, reported that tuber sucrose and reducing sugars content were not affected by
nitrogen fertilisation rate when 0, 100, 200 and 300 kg ha-1 were applied in spring
potatoes. Water deficit during growth was found to increase reducing sugars
(Sowokinos, 1990; Nadler and Heuer, 1995). High nitrogen fertilisation was also
reported to reduce storability of potato tubers (Casa et al., 2005).
Planting taro at appropriate times, using suitable fertilisation and cultivars and proper
storage can enhance taro carbohydrates properties required for processing. The
objective of this study was to evaluate carbohydrate changes in three taro cultivar of
KwaZulu-Natal, South Africa, as affected by planting date and fertilisation during
growth and temperature and packaging during storage.
122
Taro cormels harvested from trials that were planted at Ukulinga (29o 37′S 30o 16′E)
and Umbumbulu (29o 36′S 30o 25′E) were used to assess the effect of planting date
and fertilisation on reducing sugars and starch content of three taro cultivars. The
experiment was designed at each site as a split-split-plot design with four planting
dates: October 2007, November 2007, December 2007 and January 2008 as main
plots, three taro cultivars (Dumbe-dumbe, Mgingqeni and Pitshi) as sub-plots and
three organic fertiliser (Gromor Accelerator) application rates (0, 160 N and 320 N
kg ha-1) as sub-sub-plots. Each planting date was replicated three times. Plot size was
4 m2 containing 16 plants spaced 0.5m between and within rows. Sowing was done by
hand on ploughed and harrowed fields. Planting holes were opened with a hand-hoe
and organic fertiliser was mixed with soil before one cormel was planted per hole.
Weeds were controlled by hand hoeing at 30, 60, 90 and 120 days after planting.
Harvesting was done eight months after planting.
Cormels harvested from Ukulinga trial that was planted without fertiliser in October
were also used to evaluate the effect of storage temperature and packaging method on
reducing sugars and starch content. The post-harvest storage experiment was arranged
in a split-split plot design with temperatures (12oC and ambient temperature) as main
plots, packaging (polyethylene bag, box and mesh bag) as sub-plots and cultivars
(Dumbe-dumbe, Mgingqeni and Pitshi) as sub-sub-plots. Each storage temperature
was replicated three times and the cultivars were randomised within packaging within
storage temperatures. Each packaging contained 7, 11 and 13 cormels of Dumbe-
dumbe, Mgingqeni and Pitshi respectively. One cormel was sampled monthly for four
months starting one month after storage.
The cormels were then peeled, cut into slices, immediately frozen in liquid nitrogen,
freeze-dried and milled for reducing sugars and starch content analysis.
6.2.1 Reducing sugar determination
Sugars were determined at harvest and monthly for four months starting one month
after storage according to Matsuura-Endo et al. (2006) with modifications. 0.2g of
123
peeled freeze-dried and ground cormel material was homogenized for 30 seconds
using Ultraturrax in 10 ml of 80% ethanol. The sugars in the homogenate were
extracted at 80oC for 1 hour and then stored overnight at 4oC. They were centrifuged
at 10,000 rpm for 15 minutes at 4oC. The supernatant was filtered through glass wool
and dried overnight in a Savant Vacuum drier. Dried samples were resuspended in 2
ml ultra pure water and then centrifuged for 15 minutes. 0.4 micron nylon syringe
filters were used to filter into HPLC sample vials. The concentrations of sugars in the
filtrate were then determined using HPLC (Shimadza) (Kontron Autosampler HPLC
360, 515 HPLC pump) with a Ca2+- column. Reducing sugars content was calculated
by adding glucose and fructose content together.
6.2.2 Starch content determination
Starch content was determined as glucose equivalents resulting from enzymatic

hydrolysis of the centrifugate from sugar determination using method used by
Rasmussen and Henry (1990).
6.2.3 Statistical analysis
Analysis of variance was performed using GenStat Version 11.1 (2008). Least
6.3 Results
6.3.1 Reducing sugar content
Reducing sugars content was significantly affected by the interaction of planting date
and fertilisation (P = 0.001) at Ukulinga. When no fertiliser was applied, reducing
sugars significantly decreased when planting was delayed by one month from October
(Fig. 6.1). Planting date had no effect on reducing sugars when 160 N kg ha-1 of
fertiliser was applied. Delaying planting until January increased reducing sugar
content of cormels when 320 N kg ha-1 of fertiliser was applied. There was no
significant difference between fertiliser rates when planting was done in October and
November. Fertiliser application of 160 N kg ha-1 showed higher reducing sugar
124
content than when no fertiliser was applied but not higher than when 320 N kg ha-1
was applied which was also not significantly different from fertilizer was not applied
for December planting. 320 N kg ha-1 displayed significantly higher reducing sugars
than both 0 kg ha-1 and 160 N kg ha-1 when planting was done in January.
0.8
0 N kg ha-1 LSD(0.05) = 0.15
Reducing sugar content (%)
0.7 160 N kg ha-1 d

0.6 320 N kg ha-1
0.5 c bc
abc abc abc
0.4 abc abc
ab a ab ab
0.3
0.2
0.1
0
Planting date
averaged across cultivar at Ukulinga.
At Umbumbulu, planting date significantly (P = 0.018) affected reducing sugar

content of taro cultivars. Delaying planting by one month from October significantly
increased reducing sugar content (Fig. 6.2). Further delay in planting until January
decreased reducing sugar content to level not significantly different from that that was
obtained with October planting.
125
0.7
LSD(0.05) = 0.19
Reducing sugar content (%)
b
0.6 b
0.5
0.4 a
a
0.3
0.2
0.1
0
Planting date
averaged across cultivar and Gromor Accelerator application rate at Umbumbulu.
In the storage experiment, reducing sugars were significantly (P = 0.002) affected by

the interaction of temperature, packaging, cultivar and sampling month. Reducing
sugars generally increased with time in storage. Taro cormels stored at 12oC generally
showed higher reducing sugars than those stored at ambient temperature especially
after 4 months after storage (Table 6.1). Cormels of Dumbe-dumbe and Mgingqeni
packaged in boxes displayed higher reducing sugars compared to those packaged in
polyethylene and mesh bags.
126
Table 6.1 Reducing sugar content of taro cultivars stored at different temperatures in
different packaging materials. Means of the interaction effect within columns (LSD
(0.05) = 0.12) followed by the same letter are not significantly different.
o
12 C Polyethylene Dumbe-dumbe 0.09 abc 0.09 ab 0.13 a 0.36d
Mgingqeni 0.08 abc 0.10 abc 0.28 bc 0.34 cd
Pitshi 0.03 a 0.10abc 0.37 c 0.39 d
Box Dumbe-dumbe 0.07 abc 0.11 abc 0.16 ab 0.65e
Mgingqeni 0.11 abc 0.18 bc 0.41 c 0.68 e
Pitshi 0.09 abc 0.10 abc 0.11 a 0.34 cd
Mesh bag Dumbe-dumbe 0.15 bc 0.21 c 0.31 c 0.43 d
Mgingqeni 0.08 abc 0.10 abc 0.15 a 0.36 d
Pitshi 0.04 ab 0.08 ab 0.12 a 0.14 ab
Ambient Polyethylene Dumbe-dumbe 0.05 abc 0.15 abc 0.15 a 0.17 ab
Temperature Mgingqeni 0.17c 0.11 abc 0.11 a 0.13 ab
Pitshi 0.06 abc 0.07 ab 0.12 a 0.17 ab
Box Dumbe-dumbe 0.09 abc 0.10 abc 0.18 ab 0.20 ab
Mgingqeni 0.08 abc 0.09 ab 0.16 a 0.20 ab
Pitshi 0.06 abc 0.09 ab 0.11 a 0.14 ab
Mesh bag Dumbe-dumbe 0.05 ab 0.07ab 0.10 a 0.23 bc
Mgingqeni 0.04 ab 0.04 a 0.09 a 0.10 a
Pitshi 0.10 abc 0.12 abc 0.14 a 0.16 ab
6.3.2 Starch content
Starch content was significantly (P = 0.014) affected by the interaction of planting

date and cultivars at Ukulinga. Starch content significantly decreased for Dumbe-
dumbe with delay in planting from November to December whereas for Pitshi the
decrease was observed with delay in planting from December to January (Fig. 6.3).
Planting date had no effect on starch content of Mgingqeni. Dumbe-dumbe displayed
significantly highest starch content when planting was done in November. There was
no significant difference between cultivars when they were planted in October,
127
60 Dumbe-dumbe
LSD(0.05) = 4.95
Mgingqeni
50 Pitshi
Starch content (%)
e
40 de abcd cde
abcd bcd
abcd abc abcd
abcd ab a
30
20
10
0
Planting date
Figure 6.3 Starch content of taro cultivars planted at different planting dates averaged
across Gromor Accelerator application rates at Ukulinga.
At Umbumbulu, planting date (P = 0.045) and the interaction of cultivar and

fertilisation (P = 0.01) had a significant effect on starch content of taro cormels.
Starch content was significantly increased when planting was delayed by two months
from October (Fig. 6.4). Application of 320 N kg ha-1 of fertiliser decreased starch
content of Dumbe-dumbe to the level not significantly different from that obtained
with no fertiliser (Fig. 6.5). Fertiliser application had no effect on starch content of
Mgingqeni, whereas it decreased starch content of Pitshi. When fertiliser was not
applied, Pitshi showed higher starch content than Mgingqeni but not Dumbe-dumbe
which was also not significantly different from Mgingqeni. Pitshi showed lower starch
content than Dumbe-dumbe and Mgingqeni which were not different from each other
when 160 N kg ha-1 of fertiliser was applied. There was no significant difference
between starch content of the cultivars when 320 N kg ha-1 of fertiliser was applied.
128
50
LSD(0.05) = 6.49
b b
40
Starch content (%)
ab
a
30
20
10
0
Planting date
Figure 6.4 Starch content of taro cultivar planted at different planting dates at
Umbumbulu.
60 Dumbe-dumbe
LSD(0.05) = 5.77
Mgingqeni
50
Pitshi
Starch content (%)
d
abcd cd bcd
40 abcd
ab abc a
a
30
20
10
0
0 160 320
-1
Figure 6.5 Starch content of taro cultivars planted with different Gromor Accelerator
application rates averaged across planting dates at Umbumbulu.
In the storage experiment, starch content was significantly (P < 0.001) affected by the
interaction of temperature, packaging, cultivar and sampling month. Starch content
generally decreased with time in storage. Ambient temperature generally showed
higher starch content than 12oC (Table 6.2). Cormels packaged in mesh bags and
129
Dumbe-dumbe also displayed higher starch content compared to those packaged in
polyethylene bags and boxes; and Mgingqeni and Pitshi respectively.
Table 6.2 Starch content of taro cultivars stored at different temperatures in different
packaging materials. Means of the interaction effect within columns (LSD (0.05) =
0.37) followed by the same letter are not significantly different.
o
12 C Polyethylene Dumbe-dumbe 36.44 i 33.52 j 30.42 i 29.06 k
Mgingqeni 32.25 fg 31.22 g 27.08 f 26.39 g
Pitshi 30.65 d 29.67 e 23.21 b 25.54 f
Box Dumbe-dumbe 36.40 hi 27.88c 27.63 g 26.82 h
Mgingqeni 32.39 fg 31.72h 26.86 ef 24.36 e
Pitshi 25.65 a 24.89 a 17.12 a 14.64 a
Mesh bag Dumbe-dumbe 36.21 hi 34.99 l 36.67 l 32.73 m
Mgingqeni 32.09 f 31.45gh 30.88 j 27.76 j
Pitshi 30.14 c 25.24 a 23.41 bc 22.34 d
Ambient Polyethylene Dumbe-dumbe 37.14 j 33.13 i 28.40 h 27.24 i
Temperature Mgingqeni 32.47 g 31.44 gh 30.98 j 26.37 g
Pitshi 31.51 e 30.85 f 27.79 g 26.30 g
Box Dumbe-dumbe 36.06 h 30.55 f 30.31 i 27.24 i
Mgingqeni 28.78 b 27.44 b 26.21 d 19.98 c
Pitshi 30.32 cd 29.93 e 23.72 c 18.36 b
Mesh bag Dumbe-dumbe 36.18 hi 33.94 k 32.35 k 31.86 l
Mgingqeni 32.21 fg 30.72 f 27.98 g 27.58 ij
Pitshi 31.26 e 29.24 d 26.51 de 25.90 f
6.4 Discussion
Application of fertiliser had no effect on reducing sugars when taro was planted early
in the season in October, November and December at Ukulinga. The findings are in
line with what was also found by Sharma and Arora (1988) that applied nitrogen
significantly did not affect the sugar content of potato tubers. A negative impact of
applying fertiliser on reducing sugars was evident when taro was planted in January.
According to Long et al. (2004) genotype is the major factor that influenced sugar
content in potatoes. This might also be true for taro since application of fertiliser had
no effect on reducing sugars of Dumbe-dumbe and increased reducing sugars of
Mgingqeni, whereas they were increased for Pitshi when 10660 kg ha-1 of fertiliser
was applied. This is also in line with what was reported that sugar content is highly
varied between cultivars (Lisińska and Laszczyński, 1989). Water deficiency lead to
high sugar content (Davies et al., 1989; Ilin et al., 1997). This was evident at
130
Umbumbulu where reducing sugars generally increased with delay in planting since it
was observed that lower rainfall was experienced late in the season.
Starch content at Ukulinga was negatively impacted by delay in planting. This may be
related to the fact that earlier planted taro experienced higher rainfall during the first
months of corm bulking and lower temperatures during maturation as mentioned by
Smith (1987) that adequate moisture results in high starch content in potatoes. Starch
content of Dumbe-dumbe significantly decreased when planting was delayed by two
months whereas in Pitshi it decreased to the level not significantly different from
those of October and November planting and Mgingqeni was not affected by planting
date. The different response of the taro cultivars confirmed what was reported by
Willis et al. (1983) that starch properties in taro are influenced by genetic variation.
Applying fertiliser negatively affected starch content of Pitshi at Umbumbulu. This
seems to suggest that increasing nitrogen fertiliser rates significantly decreases starch
content as stated by Sharma and Arora (1988) and Shan et al. (2004). Mgingqeni was
however not affected by increasing nitrogen.
Reducing sugars seemed to have lowered before storage and this might have been a
result of keeping corms in mesh bags at ambient temperature before storage as it was
reported by Heinze et al. (1955) and Kirkpatrick et al. (1976) that reducing sugars can
be lowered by holding tubers at 21oC. Starch decreased progressively over storage
time while sugars increased (Rivero et al., 2003). This is confirmed by the findings of
the present study that also shows decreasing starch content and increasing reducing
sugars with time in storage. The shelf life of root crops is said to be influenced by
genotype, storage temperature and packaging (Kay, 1987; Brown et al., 1990; Yosuke
et al., 2000). .This is also with taro. Starch content of taro corms was found to be
higher at ambient temperature compared to 12oC whereas reducing sugars were higher
at 12oC as compared to ambient temperature. This was also reported by Williams and
Cobb (1993), Wiltshire and Cobb (1996), Wong Yen Cheong and Govinden (1998),
Yosuke et al. (2000) and Pal et al. (2008) that low temperature storage causes tubers
to develop reducing sugars and decrease starch content hence become unsuitable for
processing into crisps.
131
It was also found in this study that cormels stored in the mesh bag had the higher
starch content whereas those in open box had the higher reducing sugars. This might
be due to the fact that the mesh back somehow warmed the taro cormels stored in
them and the corms stored in the box were more exposed to lower temperature
especially at 12oC. The higher starch content and reducing sugars displayed by
Dumbe-dumbe compared to other cultivars might be because the cultivar naturally has
higher starch content than Mgingqeni and Pitshi, and even if it develops more
reducing sugars a large proportion of starch remains at the end.
The study shows that delaying planting reduced starch content for Dumbe-dumbe and
Pitshi at Ukulinga. Delaying planting decreased reducing sugars when no fertiliser
was applied and application of 320 N kg ha-1 increased reducing sugar content.
Dumbe-dumbe had high reducing sugars when no fertiliser was applied. Mgingqeni
reducing sugars were increased by fertilisation. At Umbumbulu, fertilisation decreases
starch content of Pitshi only and delay in planting increases sugar content for Dumbe-
dumbe and decrease it for Mgingqeni and Pitshi. Starch of taro corms decreases and
reducing sugars increase with time in storage. Low temperature decreases starch
content while increases reducing sugars. Mesh bags had higher starch content while
open boxes had higher reducing sugars. Dumbe-dumbe had higher starch content and
higher reducing sugars.
It is then recommended that Dumbe-dumbe be planted in October and November

without fertiliser at Ukulinga and the same cultivar be planted without fertiliser in
October at Umbumbulu. It is also recommended that taro cormels be stored at 12oC
for two months and at ambient temperature for four months to maintain the acceptable
level of reducing sugars for crisp making.
132
References
219 - 227.
Davies, H.V., Jefferies, R.A. and Scobie, L. 1989. Hexose accumulation in cold-
stored tubers of potato (Solanum tuberosum L.): the effects of water stress.
Journal of Plant Physiology 134 (4): 471-475.
Feltran, J.C., Lemos, L.B. and Vieites, R.L. 2004. Technological quality and
utilization of potato tubers. Scientia Agricola 61 (6): 598-603.
Heinze, P.H., Kilpatrick, M.E. and Dochterman, E.F., 1955. Cooking quality and
compositional factors of potatoes of different varieties from several
commercial locations. USDA Technology Bulletin 1106, p. 69.
Ilin, Ž., Durovka, M. and Marković. V. 1997. Effect of fertility and irrigation on sugar
content in potato tubers. Acta Horticulturae 462: 303-309.
Kirkpatrick, M., Heinze, P.H., Craft, C.C., Mountjay, B.M. and Falatko, C.F., 1956.
French frying quality of potatoes as influenced by cooking methods storage
conditions and specific gravity of tubers. USDA Technology Bulletin 1142, p.
46.
Kyriacou, M.C., Siomos, A.S., Ioannides, I.M. and Gerasopoulos, D. 2009. Cold
storage, reconditioning potential and chip processing quality of spring potato
(Solanum tuberosum L. cv. Hermes) tubers in response to differential
nitrogen fertilization. Journal of the Science of Food and Agriculture 89 (11):
1955-1962.
133
Chemistry 76 (2): 173-179.
Lisińska, G. and Laszczyński, W. 1989. Potato Science and technology. London UK:
Elsevier Science Publishers Ltd, England.
Long, C.M., Snapp, S.S., Douches, D.S. and Chase, R.W. 2004. Tuber yield,
storability and quality of Michigan cultivars in response to nitrogen
management and seed spacing. American Journal of Potato Research 81:
347-357.
Matsuura-Endo, C., Ohara-Takada, A., Chuda, Y., Ono, H. Yada, H., Yoshida, M.,
Kobahashi, A., Tsuda, S., Takigawa, S., Noda, T., Yamauchi, H. and Mori,
M. 2006. Effects of storage temperature on the contents of sugars and free
amino acids in tubers from different potato cultivars and acrylamide in chips.
Bioscience, Biotechnology and Biochemistry 70 (5): 1173-1180.
Mitch, E.L. 1984. Potato starch: Production and uses. In: R.L. Whistler, J.N.
BeMiller and E.F. Paschall, Editors, Starch: Chemistry and technology,
Academic Press, New York pp. 479–490.
Nadler, A. and Heuer, B. 1995. Effect of saline irrigation and water deficit on tuber
quality. Potato Research 38: 119–123.
Pal, S., Bhattacharya, A., Konar, A., Mazumdar, A. and Das, A.K. 2008. Chemical
composition of potato at harvest and after cold storage. International
Journal of Vegetable Science 14 (2): 162-176.
Pavlista, A.D. 1997. , Potato types: Their characteristics and uses. The American
Biology Teacher 59: 26–34.
Proietti, S., Moscatello, S., Battistelli, A., Mauromicale, G. and Ierna, A. 2005.
Quality of early potato in Sicily as affected by cultivar, sowing time and
irrigation. Acta Horticulturae 684: 171-177.
Rasmussen, T.S. and Henry, R.J. 1990. Starch determination in horticultural plant
material by an enzymic colorimetric procedure. Journal of Food and
Agricultural Science 52:159-70.
Rivero, R.C., Rodriquez, E.R and Romero, C.D. 2003. Effects of current storage
conditions on nutrient retention in several varieties of potatoes from
Tenerife. Food Chemistry 80: 445-450.
134
Shan, L., Sattelmacher, B., Kutzmutz, E., Mühling, K. H. and Dittert, K.
2004. Influence of nitrogen nutrition on tuber quality of potato
with special reference to the pathway of nitrate transport into
tubers. Journal of Plant Nutrition 27 (2): 341-350.
Sharma, U.C. and Arora, B.R. 1988. Effect of applied nutrients on the starch, proteins
and sugars in potatoes. Food Chemistry 30 (4): 313-317.
Smith, O. 1987. Potato chips. In: W. Talburt and O. Smith, Editors, Potato
processing, Van Nostrand Reinhold Company Inc, New York.
Sowokinos, J.S. 1990. Stress-induced alterations in carbohydrate metabolism. In:
Vayda, ME, Park WD (eds) The molecular and cellular biology of the
potato. CAB International, Wallingford, UK, pp 137–158.
Williams, R.O. and Cobb, A.H. 1993. The relationship between storage temperature,
reconditioning and tuber storage. Potato Research 36: 373-392.
Wiltshire, J.J.J. and Cobb, A.H. 1996. A review of the physiology of potato tuber
dormancy. Annual Applied Biology 129: 553-569.
Wolf, S., Marani, A. and Rudich, J. 1991. Effect of temperature on carbohydrate
metabolism in potato plants. Journal of Experimental Botany 42: 619-
625.
Wong Yen Cheong, J.K.C and Govinden, N. 1998. Quality of potato during storage at
three temperatures. AMAS 1998. Food and Agricultural Research
Council, Réduit, Mauritius.
Katsumi, T. 2000. Property changes in potato tubers (Solanum
tuberosum L.) during cold storage at 0 and 10oC. Food Preservation
Science 26 (3): 153-160.
135
CHAPTER 7
GENERAL DISCUSSION AND CONCLUSIONS
Taro quality for crisping includes yield, chemical composition and storability of
corms (Eskin, 1989; Kader and Rolle, 2004; Carputo et al., 2005; Gebhardt et al.,
2005; O`Keefe et al., 2005). Taro cultivars that combine high corm yield with good
chemical composition and storability are required for processing. Yield and chemical
composition are affected by genotype, planting date and fertilisation (Casa et al.,
2005; Martin et al., 2005; Proietti et al. (2005); Mazid, 1993) while storability is
influenced by storage temperature and packaging method (Kay, 1987; Brown et al.,
1990; Yosuke et al., 2000). Findings reported in Chapter 3 of this study support this
and indicate that the number of cormels plant-1 and fresh cormel mass plant-1 were
negatively affected by delayed planting at both sites and positively affected by
fertilisation at Umbumbulu only. All three taro cultivars performed best in terms of
the number of cormels plant-1 when they were planted in October at both sites, with
Dumbe-dumbe and Mgingqeni also performing best when planted in November at
Ukulinga (Fig. 3.17 and 3.18). The cultivars showed differences in the number of
cormels plant-1 when planted early in the season and the differences became less or
disappeared completely when planting was delayed further. Yield varies among
cultivars (Babaji et al., 2009). Dumbe-dumbe yielded less cormels followed by
Mgingqeni and Pitshi respectively when they were planted in October, but for
November planting Mgingqeni and Pitshi were not different from each other at both
sites (Fig. 3.17 and 3.18). Fresh cormel mass plant-1 was best for all cultivars when
planted in October with Dumbe-dumbe also performing best when planted in
November at Ukulinga (Fig. 3.29). Application of Gromor Accelerator enhanced the
number of cormels plant-1 and fresh cormel mass plant-1, especially when planting was
done in October at Umbumbulu (Fig. 3.19 and 3.30). Fertiliser application enhanced
the number of cormels plant-1 for all cultivars though number of cormels plant-1 was
increased for Mgingqeni and Pitshi when 160 N kg ha-1 of the fertiliser was applied,
whereas for Dumbe-dumbe the increase was observed with addition of 320 N kg ha-1
of the fertiliser (Fig. 3.20).
The high yielding in response to early planting dates may be attributed to earlier
emergence (Table 3.5 and Fig. 3.6) and vigorous leaf growth (Figs. 3.8, 3.9, 3.14, 3.15
136
and 3.16) which were due to adequate rainfall and temperature (Table 3.1) during
planting, vegetative growth and corm initiation stage of taro (Sangakkara, 1993; Lu et
al., 2001; Miyasaka et al. 2003; Scheffer et al., 2005). Tubers of about 50-350 g in
weight are preferred because they produce the crisp of the right size and pass through
the processing line with less need for hand trimming (Panhwar, 2005). Dumbe-dumbe
yielded less and more fresh cormel mass plant-1 than both Mgingqeni and Pitshi when
cultivars were planted in October and November, respectively (Fig. 3.29) but higher
number of cormels in the usable size category (Table 7.1, 7.2, 7.3, 7.4, 7.5, 7.6, 7.7
and 7.8). Mgingqeni and Pitshi might have produced significantly higher fresh cormel
mass plant-1 and number of cormels plant-1, but the cormels were too small for
crisping (Fig. 3.21, 3.22, 3.23, 3.24, 3.25, 3.26, 3.27 and 3.28).
Dry matter content of taro was negatively affected by application of fertiliser. It has
been established that excessive nitrogen fertilisation reduces dry matter allocated to
the corms in taro (Manrique, 1994). This was evident at Ukulinga where the decrease
in dry matter content was more pronounced when 320 N kg ha-1 of Gromor
Accelerator was applied (Fig. 4.1). Dry matter content was positively affected by
delaying planting date. Dry matter was optimum for November and December
plantings for Dumbe-dumbe, December and January plantings for Mgingqeni and
December planting for Pitshi (Fig. 4.2). And when it decreased again it decreased to
initial levels realised in October. It can also be postulated that application of fertiliser
did not affect dry matter content at Umbumbulu. Dry matter content yielded by all
planting dates, fertiliser rates and cultivars (Figs. 4.1 and 4.2) was above the minimum
acceptable for crisping (Smith., 1987; Kita, 2002; Panhwar, 2005).
The response of specific gravity to planting date was reported to be variable

depending on cultivar (Mohamed, 1985). This was evident with decreasing specific
gravity as planting date was delayed for Dumbe-dumbe but increasing for Mgingqeni
while with no effect for Pitshi at Ukulinga (Fig.4.3). However, at Umbumbulu, it
decreased for Dumbe-dumbe and Pitshi but without effect for Mgingqeni (Fig. 4.5).
Taro cultivars responded differently at the two experimental sites in respect to specific
gravity. They showed no response to fertiliser application when planting was done
early in the season at Ukulinga. The later the cultivars were planted the more they
137
responded to fertiliser application. This is displayed in Fig. 4.4 where application of
320 N kg ha-1 of the fertiliser decreased and increased specific gravity to levels
equivalent to those obtained when no fertiliser was applied at Ukulinga. At
Umbumbulu, on the other hand, fertiliser application decreased specific gravity when
cultivars were planted in October and applying 320 N kg ha-1 of the fertiliser
decreased specific gravity when planting was done in December (Fig. 4.6).
Planting date and fertilization had no effect on total protein when planting was done
late in the season (Figs. 4.11 and 4.12). Protein and nitrogen contents were increased
by late planting while phosphorus, potassium, calcium and copper content were
decreased by the delay in planting at Ukulinga. El-Sirafy et al. (2008) found
application of farm yard manure to increase protein content, as well as nitrogen,
phosphorus and potassium content of potato tubers. This was evident at Ukulinga
where protein content was increased by application of 160 N kg ha-1 Gromor
Accelerator, and at Umbumbulu where phosphorus and potassium content was
increased by fertiliser application.
The results of this study can be used to determine the usefulness of taro cormels for
processing purposes using information that is already in the literature as a basis to
create discriminative scores. The arbitrary scoring of taro cultivars (Appendix 7.1)
displays Dumbe-dumbe planted with 160 N kg ha-1 of organic fertiliser in October and
320 N kg ha-1 in November at Ukulinga to be the best for crisping. This is due to the
highest total scores (5.6) of combination of the crisping quality parameters assessed
(Table 7.1 and 7.2). This might also imply that the later Dumbe-dumbe is planted the
higher amount of fertiliser it requires to be suitable for crisping. Although Pitshi
shows the moderately high total score (Fig. 7.1, 7.2 and 7.3), it had the smallest
number of corms that are suitable for crisping. The high total score is due to the
lowest reducing sugar content that was found in Pitshi compared to the other
cultivars. The number of corms weighing 50-350g, which are suitable for crisping
decreased for Dumbe-dumbe as planting was delayed (Fig.7.1, 7.2, 7.3 and 7.3).
At Umbumbulu, Dumbe-dumbe planted in October with 320 N kg ha-1 of fertiliser,

shows the highest total score of 5.6 compared to the other combinations of taro
cultivar, planting date and fertiliser application rate (Table 7.5, 7.6, 7.7 and 7.8). This
138
makes Dumbe-dumbe, 320 N kg ha-1 Gromor accelerator and October the best
combination for crisping at Umbumbulu.
139
Table 7.1 Arbitrary scoring of crisping quality of taro cultivars planted in October at Ukulinga.
Taro Gromor Number of cormels Dry matter Specific Starch Reducing Total
cultivars Accelerator weighing 50-350g content gravity content sugars score
rate (N kg ha-1) (Score) (Score) (Score) (Score)
Dumbe- 0 0.9 1.1 1.0 1.4 0.2 4.6
dumbe 160 1.1 1.1 1.1 1.5 0.8 5.6
320 1.1 1.0 0.7 1.3 0.6 4.7
Mgingqeni 0 0.7 1.0 0.5 1.3 0.7 4.2
160 0.8 1.0 0.7 1.3 0.8 4.6
320 1.0 1.0 0.5 1.3 0.9 4.7
Pitshi 0 0.6 1.2 0.9 1.3 1.0 5
160 0.9 1.1 0.4 1.3 0.9 4.6
320 0.8 1.0 0.4 1.3 0.7 4.2
Scores are interpreted in appendix 7.1
Table 7.2 Arbitrary scoring of crisping quality of taro cultivars planted in November at Ukulinga.
cultivars Accelerator weighing 50-350g content gravity content sugars Score
rate (N kg ha-1) (Score) (Score) (Score) (Score) (Score)
Dumbe- 0 1.2 1.5 0.7 1.5 0.6 5.5
dumbe 160 1.3 1.3 0.7 1.5 0.7 5.5
320 1.6 1.1 0.6 1.5 0.8 5.6
Mgingqeni 0 0.6 1.1 0.7 1.3 0.9 4.6
160 0.5 1.1 0.7 1.3 0.9 4.5
320 0.6 1.0 0.7 1.3 0.7 4.3
Pitshi 0 0.6 1.2 0.6 1.3 0.9 4.6
160 0.6 1.1 1.1 1.3 0.9 5.0
320 0.5 1.1 0.7 1.3 0.9 4.5
140
Table 7.3 Arbitrary scoring of crisping quality of taro cultivars planted in December at Ukulinga.
cultivars Accelerator weighing 50-350g content gravity content sugars Score
Dumbe- 0 0.5 1.4 0.8 1.3 0.8 4.8
dumbe 160 0.6 1.2 1.2 1.2 0.6 4.8
320 0.5 1.1 0.5 1.3 0.7 4.1
Mgingqeni 0 0.5 1.2 0.7 1.3 1.0 4.7
160 0.5 1.2 0.7 1.4 0.5 4.3
320 0.7 1.2 0.3 1.4 0.6 4.2
Pitshi 0 0.5 1.4 0.7 1.4 0.9 4.9
160 0.6 1.4 0.7 1.3 0.8 4.8
320 0.5 1.2 0.7 1.4 0.7 4.5
Table 7.4 Arbitrary scoring of crisping quality of taro cultivars planted in January at Ukulinga. Note: * = missing data.
Dumbe- 0 0.5 1.2 0.5 1.3 0.6 4.1
dumbe 160 1.0 1.0 0.0 1.3 0.9 4.2
320 0.5 1.0 0.7 1.2 0.3 3.7
Mgingqeni 0 0.5 1.2 0.7 1.2 1.0 4.6
160 0.5 1.2 0.6 1.4 0.8 3.5
320 * * * * * *
Pitshi 0 0.5 1.2 0.8 1.2 0.9 4.6
160 * * * * * *
320 0.5 1.1 0.9 1.2 0.1 2.8
Scores are interpreten in appendix 7.1
141
Table 7.5 Arbitrary scoring of crisping quality of taro cultivars planted in October at Umbumbulu.
Dumbe- 0 0.7 1.2 0.9 1.3 0.7 4.8
dumbe 160 1.6 1.2 0.6 1.3 0.7 5.4
320 1.9 1.1 0.5 1.3 0.8 5.6
Mgingqeni 0 0.7 1.1 1.1 1.2 0.9 5.0
160 0.9 1.0 0.7 1.3 0.8 4.7
320 1.1 1.1 0.6 1.3 0.8 4.9
Pitshi 0 0.6 1.1 1.2 1.3 0.6 4.8
160 0.6 1.1 0.6 1.2 0.7 4.2
320 0.9 1.0 0.6 1.3 0.6 4.4
Table 7.6 Arbitrary scoring of crisping quality of taro cultivars planted in November at Umbumbulu.
Dumbe- 0 0.5 1.1 0.8 1.3 0.4 5.1
dumbe 160 0.8 1.1 0.8 1.4 0.6 4.7
320 0.8 1.2 0.8 1.4 0.2 4.4
Mgingqeni 0 0.7 1.1 0.8 1.2 0.3 4.1
160 0.7 1.2 0.7 1.4 1.1 5.1
320 0.6 1.4 0.8 1.4 0.5 4.7
Pitshi 0 0.6 1.2 0.8 1.4 0.6 4.6
160 0.6 1.2 0.9 1.3 0.6 4.6
320 0.8 1.1 0.9 1.3 0.3 4.4
142
Table 7.7 Arbitrary scoring of crisping quality of taro cultivars planted in December at Umbumbulu.
Dumbe- 0 0.5 1.1 0.5 1.6 0.8 4.5
dumbe 160 0.5 1.1 0.9 1.5 0.8 4.8
320 0.6 1.1 0.3 1.4 0.6 4.0
Mgingqeni 0 0.5 1.1 0.8 1.4 0.6 4.4
160 0.5 1.1 0.8 1.4 0.7 4.5
320 0.5 1.0 0.6 1.5 0.6 4.2
Pitshi 0 0.5 0.9 1.0 1.5 -0.1 2.8
160 * * * * * *
320 0.5 1.0 -0.2 1.4 0.0 2.7
Table 7.8 Arbitrary scoring of crisping quality of taro cultivars planted in January at Umbumbulu. Note: * = missing data.
cultivars Accelerator weighing 50-350g content Gravity Content Sugars score
Dumbe- 0 0.5 1.2 0.5 1.5 1.0 4.7
dumbe 160 0.5 1.5 0.5 1.7 0.7 4.9
320 0.5 1.4 0.5 1.3 0.8 4.5
Mgingqeni 0 0.5 1.2 0.9 1.4 1.0 5.0
160 * * * * * *
320 * * * * * *
Pitshi 0 0.5 1.6 0.7 1.5 0.9 5.2
160 0.8 1.1 0.6 1.3 0.7 4.5
320 0.5 0.9 0.6 1.3 0.4 3.7
143
It has been reported that the surface morphology of starch granules changes during
storage (Cottrell et al., 1993). This is supported by findings reported in Chapter 5 that
showed that starch granules of Dumbe-dumbe and Pitshi appeared smooth at harvest
and some of Mgingqeni appeared smooth while some were rough (Figs 5.1, 5.2 and
5.3) and that they showed signs of indentations, depressions and pitting which
deepened with time in storage and ending up in granules breaking into smaller pieces.
The severity of the change in surface morphology of starch granules varied between
landraces being more pronounced for Mgingqeni followed by Pitshi and Dumbe-
dumbe respectively (Fig. 5.4). Alpha amylase activity and sprouting also increased
with time in storage with Dumbe-dumbe displaying highest alpha amylase activity
than Pitshi and Mgingqeni respectively and Mgingqeni displaying highest sprouting
than Pitshi and Dumbe-dumbe respectively. The response of reducing sugars and
starch content to planting date and fertilisation was variable depending on landrace.
Delayed planting positively affected reducing sugars and negatively affected starch
content of Dumbe-dumbe, and negatively affected reducing sugars of Pitshi at
Ukulinga (Fig 6.1). Fertiliser application had no effect on reducing sugars of Dumbe-
dumbe but negatively affected those of Mgingqeni and Pitshi at Umbumbulu (Fig.
6.2) Fertilisation decreased starch content of Pitshi, delayed planting increased sugar
content for Dumbe-dumbe and decreased it for Mgingqeni and Pitshi at Umbumbulu.
Dumbe-dumbe had higher starch content and higher reducing sugars. Reducing sugars
increased and starch content decreased with time and decrease in temperature in
storage. Mesh bags had higher starch content while open boxes had higher reducing
sugars. Dumbe-dumbe had higher starch content and higher reducing sugars.
From the findings of this study it can be recommended that Dumbe-dumbe is the best
taro cultivar for crisping and the best time to plant it is October with 160 N kg ha-1 of
organic fertiliser and November with 320 N kg ha-1 whereas at Umbumbulu the best
time to plant Dumbe-dumbe is October with 320 N kg ha-1 of the fertiliser. This
conclusion needs to be tested in a food processing study. It is also recommended that
taro corms be stored at 12oC for two months and at ambient temperature for four
months to maintain the acceptable level of reducing sugars for crisp making.
144
From the agronomic perspective, the results of this study can be used to advise
farmers on selection of taro cultivars in terms of planting date, yield performance and
possible food processing characteristics.
145
References
Babaji, B.A., Amans, E.B., Chiezey, U.F., Falak, A.M., Tanimu, B. and Mukhtar,
A.A. 2009. Unmarketable tuber yield and other agronomic parameters of four
Varieties of Irish potato (Solanum tuberosum L.) as influenced by NPK
fertilizer rate and type of seed tuber at Samaru, Nigeria. Asian Journal of
Crop Science 1(1): 26-33.
219 - 227.
Carputo, D., Aversano, R. and Frusciante, L. 2005. Breeding potato for quality traits.
El-Sirafy, Z.M., Abbady, K.A., El-Ghamry, A.M and El-Dissoky, R.A. 2008. Potato
yield quality, quantity and profitability as affected by soil and foliar
potassium application. Research Journal of Agriculture and Biological
Sciences 4 (6): 912-922.
Eskin, N.A.M. 1989. Quality and preservation of vegetables. CRS Press, Florida,
USA.
Gebhardt, C., Menendez, C., Chen, X., Li, L. Schäfer-Pregl, R. and Salamini, F. 2005.
Bulletin 152.
Chemistry 76 (2): 173-179.
146
Lu, H.Y., Lu, C.T., Chan, L.F. and Wei, M.L. 2001. Seasonal Variation in linear
increase of taro harvest index explained by growing degree days. Agronomy
Journal 93: 1136-1141.
Manrique, L. A. 1994. Nitrogen requirements of taro. Journal of plant nutrition 17(8):
1429-1441.
Martin, R.J., Scheffer, J.J.C., Deo, B. and Halloy, S.R.P. 2005. Effect of planting date
on Ulluco yield. Acta Horticulturae 670: 181-187.
Mazid, M.A. 1993. Sulfur and nitrogen for sustainable rainfed lowland rice. PhD
Thesis, University Philippines at Los Banos. 1-232.
Miyasaka, S.C., Ogoshi, R.M., Tsuji, G.Y. and Kodani, L.S. 2003. Site and planting
date effects on taro growth: Comparison with aroid model predictions.
Agronomy Journal 95: 545–557.
Mohamed, A.I. 1985. Dry matter, protein content and yield of potato as affected by
cultivar and planting date in Sudan. Acta Horticulturae 158: 139- 141.
O`Keefe, O., Mason, G., Willis, A. and Bell, G. 2005. Commercial taro chip
development using agri-chain partnerships: A processing, marketing and
financial analysis. RIRDC Publication No 05/144. RIRDC Project No
DAQ-296A.
Panhwar, F. 2005. Potato (Solanum tuberosum L.) – Cultural practice in Sindh
Pakistan. Digitalverlag GmbH, Germany.
Proietti, S., Moscatello, S., Battistelli, A., Mauromicale, G. and Ierna, A. 2005.
Quality of early potato in Sicily as affected by cultivar, sowing time and
irrigation. Acta Horticulturae 684: 171.
Sangakkara, U.R. 1993. Response of sweet potato (Ipomoea batatas L.) to different
planting times. Journal of Agronomy and Crop Science 172 (2): 113 – 118.
Scheffer, J.J.C., Douglas, J.A. and C.M. Triggs, C.M. 2005. Factors
affecting the production and quality of Japanese taro cormels.
Smith, O. 1987. Potato chips. In: W. Talburt and O. Smith, Editors, Potato
processing, Van Nostrand Reinhold Company Inc, New York , p. 373.
Katsumi, T. 2000. Property changes in potato tubers (Solanum tuberosum
L.) during cold storage at 0 and 10oC. Food Preservation Science 26 (3):
153-160.
147
APPENDICES
Appendix 3.1 Analysis of variance of the growth and yield parameters of taro
obtained at Ukulinga.
A. Final_Emergence
Source of variation d.f. (m.v.) s.s. m.s. v.r. F pr.
Replication stratum 2 212.52 106.26 0.39
Replication.Planting_date stratum
Planting_date 3 104918.26 34972.75 127.94 <.001
Residual 6 1640.08 273.35 1.18
Replication.Planting_date.Cultivar stratum
Cultivar 2 13563.14 6781.57 29.25 <.001
Planting_date.Cultivar 6 4211.67 701.95 3.03 0.036
Residual 16 3709.59 231.85 3.83
Replication.Planting_date.Cultivar. Fertiliser stratum

Fertiliser 2 675.97 337.99 5.58 0.007
Planting_date. Fertiliser
6 661.00 110.17 1.82 0.121
Cultivar. Fertiliser 4 534.26 133.56 2.21 0.087
Planting_date.Cultivar. Fertiliser
11 (1) 2003.58 182.14 3.01 0.006
Residual 38 (10) 2300.44 60.54
Total 96 (11) 115054.49

*NB – Numbers in brackets represent missing values.
-1
B. Number of leaves plant
Planting_date 3 1771.82 590.61 47.79 <.001
Residual 6 74.15 12.36 0.80
Cultivar 2 30.64 15.32 0.99 0.395
Residual 16 248.57 15.54 0.97

Fertiliser 2 3.67 1.84 0.11 0.892
Planting_date. Fertiliser 6 64.09 10.68 0.67 0.678
Planting_date.Cultivar. Fertiliser 11 (1) 243.78 22.16 1.38 0.222
Residual 38 (10) 610.27 16.06 0.99
Replication.Planting_date.Cultivar. Fertiliser.*Units* stratum

268 (56) 4343.69 16.21
Total 364 (67) 6936.01

148
-1
C. Plant_height plant
Planting_date 3 75101.3 25033.8 581.88 <.001
Residual 6 258.1 43.0 0.36
Cultivar 2 877.0 438.5 3.68 0.048
Residual 16 1905.3 119.1 1.27

Fertiliser 2 450.3 225.1 2.40 0.104
Planting_date.Cultivar. Fertiliser 11 (1) 3612.7 328.4 3.50 0.002
Residual 38 (10) 3561.0 93.7 0.25

275 (49) 102351.8 372.2
Total 371 (60) 176357.0

-1
D. Leaf_area plant
Replication stratum 2 2642233. 1321117. 1.20
Planting_date 3 236956431. 78985477. 71.54 <.001
Residual 6 6624562. 1104094. 1.72
Cultivar 2 400778. 200389. 0.31 0.737
Planting_date.Cultivar 6 12993682. 2165614. 3.37 0.024
Residual 16 10291099. 643194. 0.92

Fertiliser 2 3390895. 1695447. 2.42 0.102
Planting_date. Fertiliser 6 7372226. 1228704. 1.76 0.135
Cultivar. Fertiliser 4 4169143. 1042286. 1.49 0.225
Planting_date.Cultivar. Fertiliser 11 (1) 13853417. 1259402. 1.80 0.089
Residual 38 (10) 26600910. 700024. 0.48

268 (56) 394813217. 1473184.
Total 364 (67) 661451548.
149
-1
E. Number_of_cormels plant
Planting_date 3 3184.84 1061.61 332.18 <.001
Residual 6 19.18 3.20 0.30
Cultivar 2 487.71 243.86 23.01 <.001
Residual 14 (2) 148.34 10.60 1.04

Fertiliser 2 7.78 3.89 0.38 0.686
6 45.19 7.53 0.74 0.622
10 (2) 50.80 5.08 0.50 0.880
Residual 36 (12) 367.17 10.20
Total 91 (16) 4091.20

-1
F. Fresh_cormel_mass plant
Planting_date 3 2182468. 727489. 76.85 <.001
Residual 6 56801. 9467. 1.59
Cultivar 2 5711. 2856. 0.48 0.628
Residual 14 (2) 83158. 5940. 0.69

Fertiliser 2 2244. 1122. 0.13 0.878
6 5611. 935. 0.11 0.995
10 (2) 41470. 4147. 0.48 0.890
Residual 36 (12) 309537. 8598.
Total 91 (16) 2496335.
150
Appendix 3.2 Analysis of variance of the growth and yield parameters of taro
obtained at Umbumbulu.
A. Final_emergence
Planting_date 3 29461.16 9820.39 80.79 <.001
Residual 6 729.29 121.55 0.61
Cultivar 2 20472.58 10236.29 51.03 <.001
Planting_date.Cultivar
6 34440.91 5740.15 28.62 <.001
Residual 16 3209.50 200.59 2.45

Fertiliser 2 134.67 67.33 0.82 0.446
6 1157.79 192.97 2.36 0.047
Cultivar. Fertiliser
4 403.05 100.76 1.23 0.312
12 940.71 78.39 0.96 0.502
Residual 43 (5) 3520.62 81.87
Total 102 (5) 82982.70

-1
B. Number of leaves plant
Planting_date 3 393.426 131.142 10.09 0.009
Residual 6 77.997 12.999 1.67
Cultivar 2 159.406 79.703 10.24 0.001
Residual 16 124.517 7.782 1.41

Fertiliser 2 34.811 17.406 3.15 0.053
Planting_date.Cultivar. Fertiliser 12 55.234 4.603 0.83 0.617
Residual 43 (5) 237.484 5.523 0.84

296 (28) 1940.500 6.556
Total 398 (33) 3039.674

151
-1
C. Plant_height plant
Planting_date 3 13322.5 4440.8 9.38 0.011
Residual 6 2840.0 473.3 9.33
Cultivar 2 2325.8 1162.9 22.92 <.001
Planting_date.Cultivar 6 6339.2 1056.5 20.83 <.001
Residual 16 811.7 50.7 0.52

Fertiliser 2 6018.2 3009.1 31.12 <.001
Planting_date.Cultivar. Fertiliser 12 984.0 82.0 0.85 0.603
Residual 43 (5) 4157.3 96.7 0.52

298 (26) 55517.4 186.3
Total 400 (31) 90981.9

-1
D. Leaf area plant
Planting_date 3 50167969. 16722656. 24.57 <.001
Residual 6 4082934. 680489. 7.83
Cultivar 2 4565423. 2282712. 26.25 <.001
Planting_date.Cultivar 6 5426177. 904363. 10.40 <.001
Residual 16 1391395. 86962. 0.43

Fertiliser 2 10517229. 5258614. 25.87 <.001
Planting_date. Fertiliser 6 12067641. 2011273. 9.90 <.001
Planting_date.Cultivar. Fertiliser 12 2470940. 205912. 1.01 0.454
Residual 43 (5) 8739752. 203250. 0.43

296 (28) 140898620. 476009.
Total 398 (33) 235894673.

152
-1
E. Number_of_cormels plant
Planting_date 3 980.186 326.729 105.17 <.001
Residual 6 18.641 3.107 0.67
Cultivar 2 170.859 85.429 18.42 <.001
6 91.333 15.222 3.28 0.034
Residual 13 (3) 60.279 4.637 0.99

Fertiliser 2 343.554 171.777 36.77 <.001
6 330.202 55.034 11.78 <.001
4 61.932 15.483 3.31 0.020
9 (3) 45.996 5.111 1.09 0.390
Residual 38 (10) 177.507 4.671
Total 91 (16) 2205.547

-1
F. Fresh_cormel_mass plant
Planting_date 3 897200. 299067. 87.37 <.001
Residual 6 20538. 3423. 0.40
Cultivar 2 32393. 16197. 1.89 0.191
6 50544. 8424. 0.98 0.476
Residual 13 (3) 111673. 8590. 3.46

Fertiliser 2 208028. 104014. 41.95 <.001
6 206012. 34335. 13.85 <.001
4 24176. 6044. 2.44 0.064
9 (3) 25183. 2798. 1.13 0.367
Residual 38 (10) 94222. 2480.
Total 91 (16) 1548590.

153
Appendix 3.3 Regression analysis of growth and yield parameters
A. Ukulinga
*** Summary of analysis ***
d.f. s.s. m.s. v.r. F pr.

Regression 5 726904. 145381. 50.36 <.001
Residual 30 86602. 2887.
Total 35 813505. 23243.
Percentage variance accounted for 87.6
*** Estimates of parameters ***
estimate s.e. t(30) t pr.

Constant 23.4 49.2 0.47 0.638
Emergence 1.016 0.617 1.65 0.110
Number_of_leaves_plant_1
-14.6 10.5 -1.38 0.176
Plant_height_plant_1
2.15 2.62 0.82 0.419
Leaf_area_plant_1
0.0881 0.0646 1.36 0.183
Number_of_corms_plant_1
6.94 3.68 1.89 0.069
*** Accumulated analysis of variance ***
Change d.f. s.s. m.s. v.r. F pr.

+ Emergence 1 600418. 600418. 207.99 <.001
+ Number_of_leaves_plant_1
1 22253. 22253. 7.71 0.009
+ Plant_height_plant_1
1 70195. 70195. 24.32 <.001
+ Leaf_area_plant_1
1 23767. 23767. 8.23 0.007
+ Number_of_corms_plant_1
1 10270. 10270. 3.56 0.069
Residual 30 86602. 2887.
Total 35 813505. 23243.
154
B. Umbumbulu
*** Summary of analysis ***
d.f. s.s. m.s. v.r. F pr.

Regression 5 354817. 70963. 17.60 <.001
Residual 30 120947. 4032.
Total 35 475764. 13593.
Percentage variance accounted for 70.3
*** Estimates of parameters ***
estimate s.e. t(30) t pr.

Constant -72.1 60.6 -1.19 0.243
Emergence 1.283 0.445 2.88 0.007
Leaf_n 3.2 13.4 0.24 0.815
Pt_ht 2.04 2.77 0.74 0.468
Leaf_are -0.0802 0.0756 -1.06 0.297
Corm_n 21.27 3.33 6.40 <.001
*** Accumulated analysis of variance ***
Change d.f. s.s. m.s. v.r. F pr.

+ Emerge 1 136243. 136243. 33.79 <.001
+ Leaf_n 1 19637. 19637. 4.87 0.035
+ Pt_ht 1 28234. 28234. 7.00 0.013
+ Leaf_area 1 5806. 5806. 1.44 0.239
+ Corm_n 1 164897. 164897. 40.90 <.001
Residual 30 120947. 4032.
Total 35 475764. 13593.
155
Appendix 3.4 Net returns hectare-1 of taro in KwaZulu-Natal.
Item Unit Quantity Price/Unit (Zar) Total (Zar)
Output kg 8000 7 56000
(A) Variable costs
Fertiliser kg 5330 0.4 2132
Transport km 20 1 20
Total 2152
(B) Labour inputs (Mandays)
Propagule preparation MD 2 40 80
Planting MD 6 40 240
Weeding MD 6x3 40 720
Fertilisation MD 6 40 240
Harvesting MD 6 40 240
Cleaning and sorting MD 6 40 240
Transport to market 20 x 6 1 120
Total 1880
Total costs (TC) 4032
Net returns (NR) 51968
Benefit cost ratio (NR/TC) 12.89 : 1.00
156
Appendix 4.1 Analysis of variance of dry matter content of taro cormels obtained
at Ukulinga.
Planting_date 3 334.84 111.61 2.43 0.163
Residual 6 275.42 45.90 1.87
cultivar 2 37.65 18.83 0.77 0.484
Residual 14 (2) 344.29 24.59 1.32

Fertiliser 2 163.73 81.86 4.39 0.020
6 115.39 19.23 1.03 0.422
10 (2) 163.35 16.33 0.88 0.564
Residual 36 (12) 671.81 18.66
Total 91 (16) 2492.24

Appendix 4.2 Analysis of variance of dry matter content of taro cormels obtained
at Umbumbulu.
Planting_date 3 704.14 234.71 1.67 0.272
Residual 6 844.18 140.70 1.83
Cultivar 2 109.27 54.63 0.71 0.509
6 282.07 47.01 0.61 0.717
Residual 13 (3) 998.53 76.81 1.60

Fertiliser 2 8.21 4.11 0.09 0.918
6 192.03 32.00 0.67 0.677
4 151.29 37.82 0.79 0.540
9 (3) 531.13 59.01 1.23 0.307
Residual 38 (10) 1824.36 48.01
Total 91 (16) 4855.13

157
Ukulinga.
Planting_date 3 0.005959 0.001986 0.27 0.843
Residual 6 0.043556 0.007259 3.01
Cultivar 2 0.003064 0.001532 0.63 0.545
Residual 14 (2) 0.033801 0.002414 0.47

Fertiliser 2 0.000847 0.000423 0.08 0.922
6 0.099410 0.016568 3.20 0.013
10 (2) 0.067963 0.006796 1.31 0.261
Residual 36 (12) 0.186307 0.005175
Total 91 (16) 0.451354


Umbumbulu.
Replications stratum 2 0.006355 0.003177 0.37
Planting_date 3 0.070256 0.023419 2.71 0.138
Residual 6 0.051790 0.008632 2.93
Cultivar 2 0.032539 0.016269 5.53 0.018
6 0.058566 0.009761 3.32 0.033
Residual 13 (3) 0.038233 0.002941 1.47

Fertiliser 2 0.018817 0.009408 4.71 0.015
6 0.053055 0.008843 4.43 0.002
4 0.015109 0.003777 1.89 0.132
9 (3) 0.008654 0.000962 0.48 0.878
Residual 38 (10) 0.075870 0.001997
Total 91 (16) 0.311286

158
Ukulinga.
Planting_date 3 359.798 119.933 30.39 <.001
Residual 6 23.681 3.947 0.91
Cultivar 2 0.379 0.190 0.04 0.958
Residual 14 (2) 60.949 4.354 1.87

Fertiliser 2 17.650 8.825 3.79 0.032
6 13.316 2.219 0.95 0.471
10 (2) 34.178 3.418 1.47 0.194
Residual 35 (13) 81.587 2.331
Total 90 (17) 502.891


Umbumbulu.
Planting_date 3 56.969 18.990 5.73 0.034
Residual 6 19.882 3.314 1.69
Cultivar 2 103.581 51.790 26.42 <.001
6 77.771 12.962 6.61 0.002
Residual 13 (3) 25.484 1.960 1.43

Fertiliser 2 115.946 57.973 42.24 <.001
6 8.520 1.420 1.03 0.418
4 5.209 1.302 0.95 0.447
9 (3) 10.462 1.162 0.85 0.579
Residual 38 (10) 52.160 1.373
Total 91 (16) 321.166

159
Ukulinga.
Source of variation d.f.(m.v.) s.s. m.s. v.r. F pr.
Planting_date 3 1525.65 508.55 23.83 <.001
Residual 6 128.05 21.34 2.73
Cultivar 2 25.03 12.51 1.60 0.237
Residual 14(2) 109.59 7.83 0.73
Replication.Planting_date.Cultivar.Fertiliser stratum
Fertiliser 2 2.77 1.38 0.13 0.879
Planting_date.Fertiliser
6 17.30 2.88 0.27 0.947
Cultivar.Fertiliser 4 26.03 6.51 0.61 0.658
Planting_date.Cultivar.Fertiliser
10(2) 44.41 4.44 0.42 0.929
Residual 35(13) 373.19 10.66
Total 90(17) 2256.84


Umbumbulu.
Source of variation d.f.(m.v.) s.s. m.s. v.r. F pr.
Planting_date 3 350.770 116.923 42.71 <.001
Residual 6 16.425 2.738 0.77
Cultivar 2 17.525 8.762 2.45 0.125
Residual 13(3) 46.411 3.570 2.21
Fertiliser 2 63.204 31.602 19.53 <.001
6 74.143 12.357 7.64 <.001
9(3) 13.171 1.463 0.90 0.531
Residual 38(10) 61.476 1.618
Total 91(16) 630.661

160
Ukulinga.
A. Nitrogen content
Planting_date 3 9.21082 3.07027 30.39 <.001
Residual 6 0.60623 0.10104 0.91
Cultivar 2 0.00971 0.00485 0.04 0.958
Residual 14 (2) 1.56030 0.11145 1.87

Fertiliser 2 0.45184 0.22592 3.79 0.032
6 0.34089 0.05681 0.95 0.471
10 (2) 0.87496 0.08750 1.47 0.194
Residual 35 (13) 2.08864 0.05968
Total 90 (17) 12.87401

B. Phosphorus content
Planting_date 3 0.125719 0.041906 20.91 0.001
Residual 6 0.012026 0.002004 1.13
Cultivar 2 0.209736 0.104868 59.36 <.001
Residual 14 (2) 0.024734 0.001767 0.71

Fertiliser 2 0.004138 0.002069 0.83 0.445
6 0.019022 0.003170 1.27 0.297
10 (2) 0.020734 0.002073 0.83 0.604
Residual 35 (13) 0.087494 0.002500
Total 90 (17) 0.443503

161
C. Potassium content
Planting_date 3 11.8562 3.9521 47.49 <.001
Residual 6 0.4993 0.0832 0.90
Cultivar 2 0.2123 0.1061 1.14 0.347
Residual 14 (2) 1.2994 0.0928 0.85

Fertiliser 2 0.3530 0.1765 1.62 0.212
6 0.5004 0.0834 0.77 0.602
10 (2) 0.9831 0.0983 0.90 0.541
Residual 35 (13) 3.8113 0.1089
Total 90 (17) 20.7341

D. Calcium content
Planting_date 3 0.0094658 0.0031553 1.76 0.254
Residual 6 0.0107408 0.0017901 1.30
Cultivar 2 0.0062883 0.0031442 2.28 0.139
Residual 14 (2) 0.0193182 0.0013799 1.52

Fertiliser 2 0.0011212 0.0005606 0.62 0.545
6 0.0041626 0.0006938 0.76 0.603
10 (2) 0.0041693 0.0004169 0.46 0.905
Residual 35 (13) 0.0317845 0.0009081
Total 90 (17) 0.0906615

162
E. Magnesium content
Planting_date 3 0.003315 0.001105 1.41 0.329
Residual 6 0.004706 0.000784 0.74
Cultivar 2 0.055776 0.027888 26.27 <.001
Residual 14 (2) 0.014863 0.001062 0.96

Fertiliser 2 0.000054 0.000027 0.02 0.976
6 0.008374 0.001396 1.26 0.300
10 (2) 0.012960 0.001296 1.17 0.342
Residual 35 (13) 0.038710 0.001106
Total 90 (17) 0.119213

F. Zinc content
Planting_date 3 12958. 4319. 5.24 0.041
Residual 6 4943. 824. 0.74
Cultivar 2 5962. 2981. 2.67 0.104
Residual 14 (2) 15657. 1118. 0.75

Fertiliser 2 436. 218. 0.15 0.865
6 6527. 1088. 0.73 0.631
10 (2) 13213. 1321. 0.88 0.558
Residual 35 (13) 52402. 1497.
Total 90 (17) 133872.

163
G. Copper content
Planting_date 3 31.029 10.343 3.03 0.115
Residual 6 20.498 3.416 0.90
Cultivar 2 1.179 0.590 0.15 0.858
Residual 14 (2) 53.274 3.805 1.54

Fertiliser 2 0.431 0.216 0.09 0.917
6 22.553 3.759 1.52 0.200
10 (2) 29.930 2.993 1.21 0.317
Residual 35 (13) 86.380 2.468
Total 90 (17) 219.630

H. Iron content
Planting_date 3 598.80 199.60 3.95 0.072
Residual 6 302.99 50.50 0.39
Cultivar 2 530.28 265.14 2.05 0.166
Residual 14 (2) 1812.82 129.49 1.36

Fertiliser 2 28.17 14.08 0.15 0.863
6 625.61 104.27 1.10 0.384
10 (2) 819.44 81.94 0.86 0.576
Residual 35 (13) 3329.11 95.12
Total 90 (17) 8889.85

164
I. Manganese content
Planting_date 3 554.35 184.78 2.26 0.182
Residual 6 490.42 81.74 6.10
Cultivar 2 86.69 43.35 3.24 0.070
Residual 14 (2) 187.50 13.39 0.75

Fertiliser 2 27.73 13.87 0.78 0.468
6 98.74 16.46 0.92 0.492
10 (2) 47.39 4.74 0.27 0.985
Residual 35 (13) 625.62 17.87
Total 90 (17) 2252.99

J. Sodium content
Planting_date 3 132127. 44042. 1.20 0.387
Residual 6 220145. 36691. 3.05
Cultivar 2 10654. 5327. 0.44 0.651
Residual 14 (2) 168445. 12032. 2.11

Fertiliser 2 33976. 16988. 2.97 0.064
6 76304. 12717. 2.23 0.064
10 (2) 116696. 11670. 2.04 0.058
Residual 35 (13) 200014. 5715.
Total 90 (17) 1006048.

165
Appendix 4.10 Analysis of variance of mineral content of taro cormels obtained
at Umbumbulu.
A. Nitrogen content
Planting_date 3 1.45842 0.48614 5.73 0.034
Residual 6 0.50898 0.08483 1.69
Cultivar 2 2.65167 1.32583 26.42 <.001
6 1.99094 0.33182 6.61 0.002
Residual 13 (3) 0.65240 0.05018 1.43

Fertiliser 2 2.96821 1.48410 42.24 <.001
6 0.21812 0.03635 1.03 0.418
4 0.13334 0.03334 0.95 0.447
9 (3) 0.26784 0.02976 0.85 0.579
Residual 38 (10) 1.33529 0.03514
Total 91 (16) 8.22184

B. Phosphorus content
Planting_date 3 0.0048505 0.0016168 0.92 0.487
Residual 6 0.0105881 0.0017647 1.34
Cultivar 2 0.0057016 0.0028508 2.16 0.155
6 0.0114571 0.0019095 1.45 0.270
Residual 13 (3) 0.0171402 0.0013185 1.41

Fertiliser 2 0.1858619 0.0929309 99.26 <.001
6 0.0151422 0.0025237 2.70 0.028
4 0.0009827 0.0002457 0.26 0.900
9 (3) 0.0231217 0.0025691 2.74 0.014
Residual 38 (10) 0.0355779 0.0009363
Total 91 (16) 0.2572989

166
C. Potassium content
Planting_date 3 2.2790 0.7597 0.91 0.488
Residual 6 4.9816 0.8303 8.54
Cultivar 2 2.5927 1.2963 13.34 <.001
6 1.0820 0.1803 1.86 0.165
Residual 13 (3) 1.2636 0.0972 0.86

Fertiliser 2 2.5935 1.2968 11.53 <.001
6 1.2199 0.2033 1.81 0.124
4 0.3285 0.0821 0.73 0.577
9 (3) 0.7815 0.0868 0.77 0.642
Residual 38 (10) 4.2730 0.1124
Total 91 (16) 16.5702

D. Calcium content
Planting_date 3 0.0057354 0.0019118 1.07 0.428
Residual 6 0.0106837 0.0017806 1.48
Cultivar 2 0.0256716 0.0128358 10.67 0.002
6 0.0121805 0.0020301 1.69 0.202
Residual 13 (3) 0.0156414 0.0012032 1.23

Fertiliser 2 0.0036160 0.0018080 1.85 0.171
6 0.0087116 0.0014519 1.48 0.210
4 0.0037541 0.0009385 0.96 0.441
9 (3) 0.0043296 0.0004811 0.49 0.871
Residual 38 (10) 0.0371863 0.0009786
Total 91 (16) 0.1185250

167
E. Magnesium content
Planting_date 3 0.0047294 0.0015765 4.52 0.055
Residual 6 0.0020936 0.0003489 1.03
Cultivar 2 0.0015535 0.0007768 2.30 0.140
6 0.0049195 0.0008199 2.43 0.085
Residual 13 (3) 0.0043938 0.0003380 1.73

Fertiliser 2 0.0003670 0.0001835 0.94 0.399
6 0.0026931 0.0004488 2.30 0.054
4 0.0011338 0.0002834 1.45 0.236
9 (3) 0.0020468 0.0002274 1.17 0.344
Residual 38 (10) 0.0074167 0.0001952
Total 91 (16) 0.0279685

F. Zinc content
Planting_date 3 865.45 288.48 6.44 0.026
Residual 6 268.81 44.80 0.57
Cultivar 2 61.02 30.51 0.39 0.688
6 527.63 87.94 1.11 0.408
Residual 13 (3) 1029.95 79.23 1.49

Fertiliser 2 394.93 197.46 3.72 0.034
6 738.10 123.02 2.32 0.053
4 328.87 82.22 1.55 0.208
9 (3) 344.54 38.28 0.72 0.687
Residual 38 (10) 2018.23 53.11
Total 91 (16) 5332.74

168
G. Copper content
Planting_date 3 7.692 2.564 0.68 0.594
Residual 6 22.539 3.756 1.87
Cultivar 2 174.885 87.442 43.57 <.001
6 17.888 2.981 1.49 0.258
Residual 13 (3) 26.090 2.007 1.96

Fertiliser 2 11.890 5.945 5.81 0.006
6 7.694 1.282 1.25 0.301
4 11.145 2.786 2.72 0.043
9 (3) 25.922 2.880 2.82 0.012
Residual 38 (10) 38.857 1.023
Total 91 (16) 293.257

H. Iron content
Planting_date 3 339.9 113.3 1.45 0.318
Residual 6 467.5 77.9 0.42
Cultivar 2 235.8 117.9 0.63 0.548
6 2530.6 421.8 2.26 0.103
Residual 13 (3) 2430.5 187.0 1.15

Fertiliser 2 619.7 309.8 1.91 0.162
6 1454.8 242.5 1.49 0.207
Cultivar.Fertiliser
4 406.3 101.6 0.63 0.647
9 (3) 1489.2 165.5 1.02 0.443
Residual 38 (10) 6173.0 162.4
Total 91 (16) 12424.3

169
I. Manganese content
Planting_date 3 1009.10 336.37 3.54 0.088
Residual 6 570.26 95.04 3.15
Cultivar 2 861.95 430.97 14.31 <.001
6 149.24 24.87 0.83 0.570
Residual 13 (3) 391.66 30.13 0.87
Fertiliser 2 396.05 198.02 5.71 0.007
6 561.75 93.63 2.70 0.028
Cultivar.Fertiliser
4 154.53 38.63 1.11 0.364
9 (3) 170.88 18.99 0.55 0.830
Residual 38 (10) 1317.06 34.66
Total 91 (16) 5250.61

J. Sodium content
Planting_date 3 1239848. 413283. 1.20 0.386
Residual 6 2063338. 343890. 3.08
Cultivar 2 26045. 13022. 0.12 0.891
6 341054. 56842. 0.51 0.791
Residual 13 (3) 1449369. 111490. 2.05
Fertiliser 2 108713. 54356. 1.00 0.378
6 99940. 16657. 0.31 0.930
Cultivar.Fertiliser
4 49778. 12445. 0.23 0.921
9 (3) 793080. 88120. 1.62 0.145
Residual 38 (10) 2068092. 54423.
Total 91 (16) 8486333.

170
Appendix 5.1 Analysis of variance of alpha-amylase activity of taro cormels of
three taro cultivars (Dumbe-dumbe, Mgingqeni and Pitshi) stored at 12oC and
ambient temperature in three packagings (Polyethylene bag, open box and mesh
bag) for four months.
Replication.Wholeplot stratum
Storage_temperature 1 0.0073872 0.0073872 6.87 0.120
Residual 2 0.0021500 0.0010750 1.39
Replication.Wholeplot.Subplot stratum
Packaging 2 0.0076523 0.0038261 4.94 0.040
Storage_temperature.Packaging
2 0.0080367 0.0040184 5.19 0.036
Residual 8 0.0061906 0.0007738 0.89
Replication.Wholeplot.Subplot.Subsubplot stratum
Cultivar 2 0.0118370 0.0059185 6.78 0.005
Storage_temperature.Cultivar
2 0.0233011 0.0116506 13.34 <.001
Packaging.Cultivar 4 0.0033910 0.0008478 0.97 0.442
Storage_temperature.Packaging.Cultivar
4 0.0424579 0.0106145 12.15 <.001
Residual 24 0.0209627 0.0008734 1.29
Replication.Wholeplot.Subplot.Subsubplot.*Units* stratum
Sampling_Month 4 0.0933289 0.0233322 34.52 <.001
Storage_temperature.Sampling_Month
4 0.0465050 0.0116263 17.20 <.001
Packaging.Sampling_Month
8 0.0372076 0.0046509 6.88 <.001
Cultivar.Sampling_Month 8 0.0956707 0.0119588 17.69 <.001
Storage_temperature.Packaging.Sampling_Month
7 (1) 0.0423961 0.0060566 8.96 <.001
Storage_temperature.Cultivar.Sampling_Month
8 0.0924113 0.0115514 17.09 <.001
Packaging.Cultivar.Sampling_Month
16 0.0760198 0.0047512 7.03 <.001
Storage_temperature.Packaging.Cultivar.Sampling_Month
5 (11) 0.0428136 0.0085627 12.67 <.001
Residual 120 (24) 0.0811100 0.0006759
Total 233 (36) 0.6572201

171
Appendix 5.2 Analysis of variance of sprouting of taro cormels of three taro
cultivars (Dumbe-dumbe, Mgingqeni and Pitshi) stored at 12oC and ambient
temperature in three packagings (Polyethylene bag, open box and mesh bag) for
four months.
Source of variation d.f. s.s. m.s. v.r. F pr.
Temperature 1 11362.09 11362.09 1020.23 <.001
Residual 2 22.27 11.14 0.69
Packaging 2 9531.73 4765.87 294.29 <.001
Temperature.Packaging
2 2092.92 1046.46 64.62 <.001
Residual 8 129.56 16.19 1.12
Cultivar 2 776.18 388.09 26.72 <.001
Temperature.Cultivar
2 195.24 97.62 6.72 0.005
Packaging.Cultivar 4 3902.56 975.64 67.18 <.001
Temperature.Packaging.Cultivar 4 1091.69 272.92 18.79 <.001
Residual 24 348.52 14.52 1.30
Month 4 77374.28 19343.57 1737.98 <.001
Temperature.Month 4 16682.30 4170.57 374.72 <.001
Packaging.Month 8 15847.03 1980.88 177.98 <.001
Cultivar.Month 8 1254.93 156.87 14.09 <.001
Temperature.Packaging.Month 8 2977.55 372.19 33.44 <.001
Temperature.Cultivar.Month 8 387.15 48.39 4.35 <.001
Packaging.Cultivar.Month 16 5853.95 365.87 32.87 <.001
Temperature.Packaging.Cultivar.Month
16 1886.08 117.88 10.59 <.001
Residual 144 1602.71 11.13
Total 269 153328.13
172
Appendix 6.1 Analysis of variance of reducing sugar content of taro cormels
obtained at Ukulinga.
Planting_date 3 0.09443 0.03148 2.31 0.176
Residual 6 0.08188 0.01365 0.32
Cultivar 2 0.21542 0.10771 2.55 0.116
Residual 13 (3) 0.54847 0.04219 1.70
Fertiliser 2 0.16537 0.08269 3.34 0.052
6 0.76659 0.12777 5.15 0.001
9 (3) 0.14405 0.01601 0.65 0.748
Residual 25 (23) 0.61973 0.02479
Total 78 (29) 2.13752

Appendix 6.2 Analysis of variance of reducing sugar content of taro cormels

obtained at Umbumbulu.
Planting_date 3 1.82616 0.60872 7.55 0.018
Residual 6 0.48351 0.08058 0.53
Cultivar 2 0.27082 0.13541 0.89 0.434
Residual 13 (3) 1.97847 0.15219 5.15
Fertiliser 2 0.00459 0.00229 0.08 0.926
6 0.24753 0.04125 1.40 0.267
8 (4) 0.26711 0.03339 1.13 0.388
Residual 19 (29) 0.56167 0.02956
Total 71 (36) 3.03246

173
Appendix 6.3 Analysis of variance of reducing sugar content of taro cormels of
three taro cultivars (Dumbe-dumbe, Mgingqeni and Pitshi) stored at 12oC and
ambient temperature in three packagings (Polyethylene bag, open box and mesh
bag) for four months.
Temperature 1 0.395134 0.395134 134.35 0.007
Residual 2 0.005882 0.002941 0.31
Packaging 2 0.122810 0.061405 6.48 0.021
Temperature.Packaging 2 0.011125 0.005563 0.59 0.578
Residual 8 0.075764 0.009470 0.98
Cultivar 2 0.050205 0.025102 2.59 0.096
Temperature.Cultivar 2 0.102922 0.051461 5.31 0.012
Packaging.Cultivar 4 0.126182 0.031545 3.26 0.029
Temperature.Packaging.Cultivar
4 0.179729 0.044932 4.64 0.006
Residual 24 0.232438 0.009685 2.37
Month 3 1.344432 0.448144 109.89 <.001
Packaging.Month 6 0.111313 0.018552 4.55 <.001
Cultivar.Month 6 0.113996 0.018999 4.66 <.001
Temperature.Packaging.Month
6 0.165271 0.027545 6.75 <.001
Temperature.Cultivar.Month
6 0.065013 0.010836 2.66 0.020
Packaging.Cultivar.Month
12 0.214129 0.017844 4.38 <.001
8 (4) 0.107757 0.013470 3.30 0.002
Residual 100 (8) 0.407804 0.004078
Total 203 (12) 3.402170

174
Ukulinga.
Planting_date 3 246.16 82.05 1.41 0.328
Residual 6 348.76 58.13 2.45
Cultivar 2 98.88 49.44 2.08 0.161
Residual 14 (2) 332.31 23.74 0.58
Fertiliser 2 2.86 1.43 0.04 0.966
6 100.02 16.67 0.41 0.869
10 (2) 202.12 20.21 0.49 0.882
Residual 36 (12) 1470.12 40.84
Total 91 (16) 3154.31


Umbumbulu.
Variate: Starch_content
Planting_date 3 1431.07 477.02 5.02 0.045
Residual 6 570.20 95.03 1.44
Cultivar 2 241.06 120.53 1.83 0.200
Residual 13 (3) 857.52 65.96 1.35
Fertiliser 2 142.07 71.03 1.46 0.245
6 679.90 113.32 2.33 0.052
9 (3) 259.15 28.79 0.59 0.796
Residual 38 (10) 1850.62 48.70
Total 91 (16) 5822.49

175
Appendix 6.6 Analysis of variance of starch content of taro cormels of three taro
cultivars (Dumbe-dumbe, Mgingqeni and Pitshi) stored at 12oC and ambient
temperature in three packagings (Polyethylene bag, open box and mesh bag) for
four months.
Temperature 1 21.09262 21.09262 443.94 0.002
Residual 2 0.09502 0.04751 1.07
Packaging 2 528.39979 264.19989 5950.40 <.001
Temperature.Packaging 2 7.26627 3.63313 81.83 <.001
Residual 8 0.35520 0.04440 0.99
Cultivar 2 1387.63707 693.81854 15467.41 <.001
Temperature.Cultivar 2 192.28082 96.14041 2143.27 <.001
Packaging.Cultivar 4 119.71358 29.92840 667.20 <.001
Temperature.Packaging.Cultivar
4 107.51135 26.87784 599.19 <.001
Residual 24 1.07656 0.04486 0.82
Month 3 1570.26964 523.42321 9605.13 <.001
Packaging.Month 6 137.43789 22.90632 420.34 <.001
Cultivar.Month 6 100.61177 16.76863 307.71 <.001
Temperature.Packaging.Month
6 44.51222 7.41870 136.14 <.001
Temperature.Cultivar.Month
5 (1) 23.94719 4.78944 87.89 <.001
Packaging.Cultivar.Month
11 (1) 126.48009 11.49819 211.00 <.001
6 (6) 23.93457 3.98909 73.20 <.001
Residual 91 (17) 4.95896 0.05449
Total 190 (25) 3051.94502

176
Appendix 7.1 Arbitrary scores for taro crisping quality parameters
A. Number of cormels plant-1 weighing 50-350g

Cormel number ranges Score
0.00-0.25 0.5
0.26-0.50 0.6
0.51-0.75 0.7
0.76-1.00 0.8
1.01-1.25 0.9
1.26-1.50 1.0
1.51-1.75 1.1
1.76-2.00 1.2
2.01-2.25 1.3
2.26-2.50 1.4
2.51-2.75 1.5
2.76-3.00 1.6
3.01-3.25 1.7
3.26-3.50 1.8
3.51-3.75 1.9
B. Dry matter content

Dry matter content ranges (%) Scores
17.0-19.9 0.9
20.0-22.9 1.0
23.0-25.9 1.1
26.0-28.9 1.2
29.0-31.9 1.3
32.0-34.9 1.4
35.0-37.9 1.5
38.0-40.9 1.6
C. Specific gravity
Specific gravity ranges Scores
0.850-0.869 -0.2
0.870-0.889 -0.1
0.890-0.909 0.0
0.910-0.929 0.1
0.930-0.949 0.2
0.950-0.969 0.3
0.970-0.989 0.4
0.990-1.009 0.5
1.010-1.029 0.6
1.030-1.049 0.7
1.050-1.069 0.8
1.070-1.089 0.9
1.090-1.109 1.0
1.110-1.129 1.1
1.130-1.149 1.2
177
1.150-1.169 1.3
D. Starch content
Starch content ranges (%) Scores
15.00-19.99 1.0
20.00-24.99 1.1
25.00-29.99 1.2
30.00-34.99 1.3
35.00-39.99 1.4
40.00-44.99 1.5
45.00-49.99 1.6
50.00-54.99 1.7
E. Reducing sugars content

Reducing sugars ranges (%) Scores
0.860-0.919 -0.1
0.800-0.859 0.0
0.740-0.799 0.1
0.680-0.739 0.2
0.620-0.679 0.3
0.560-0.619 0.4
0.500-0.559 0.5
0.440-0.499 0.6
0.380-0.439 0.7
0.320-0.379 0.8
0.260-0.319 0.9
0.200-0.259 1.0
0.140-0.199 1.1
178
179
180
181
182
183

Taro

Uploaded by

Copyright:

Available Formats

Taro

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Taro

Uploaded by

Copyright:

Available Formats

TARO (COLOCASIA ESCULENTA L.

SCHOTT) YIELD AND QUALITY RESPONSE

Submitted in fulfillment of the requirements for the degree of

I declare that the above statement is correct.

I express my gratitude to my supervisor, Professor Albert T. Modi for his determined

This study was funded by SANPAD (South Africa-Netherlands Research Programme on

LIST OF FIGURES xii

LIST OF TABLES xviii

2.1 Quality of taro cormels for crisp making 6

3 GROWTH, DEVELOPMENT AND YIELD OF TARO IN RESPONSE TO

3.2 Materials and methods 38

3.2.1 Planting material 38

3.2.2 Site description 38

3.2.3 Experimental design and agronomic practices 42

3.2.4 Data collection and analysis 42

3.3.2 Growth parameters 45

3.3.2.1 Number of leaves 45

3.3.2.2 Plant height 48

3.3.2.3 Leaf area 51

3.3.3.1 Number of cormels 53

3.3.3.2 Grading of cormels 56

3.3.3.3 Fresh cormel mass 61

3.5 Conclusions and recommendations 64

4 INFLUENCE OF PLANTING DATE AND ORGANIC FERTILISER ON

SPECIFIC GRAVITY, DRY MATTER CONTENT AND MINERAL CONTENT

4.2 Materials and methods 72

4.2.1 Planting material 72

4.2.2 Site description 72

4.2.3 Experimental design 73

4.2.4 Data collection and analysis 73

4.2.4.1 Dry matter content 74

4.2.4.2 Specific gravity 74

4.2.4.3 Protein content and total protein 74

4.2.4.4 Mineral content 74

4.3.1 Dry matter content 75

4.3.3 Protein content and total protein 79

4.3.3.1 Protein content 79

4.3.3.2 Total protein 82

4.3.4 Mineral content of taro cormels 84

4.3.4.1 Nitrogen content 84

4.3.4.2 Phosphorus content 87

4.3.4.3 Potassium content 89

4.3.4.4 Calcium content 91

4.3.4.5 Magnesium content 92

4.3.4.6 Zinc content 93

4.3.4.7 Copper content 95

4.3.4.8 Manganese content 96

4.5 Conclusions and recommendations 100

5 CHANGES IN THE SURFACE MORPHOLOGY OF STARCH GRANULES,

ALPHA-AMYLASE ACTIVITY AND SPROUTING OF TARO DURING

5.1 Introduction 104

5.2 Materials and methods 106

5.2.1 Plant material used 106

5.2.2 Scanning electron microscopy 106

5.2.4 Sprouting 107

5.3 Results 108

5.3.1 Scanning electron microscopy 108

5.3.2 Alpha-amylase activity 112

5.3.3 Sprouting 114