F3 | ‘. I‘m-‘1“? r “3 '5‘ "rar- .v". - .‘V-r ”'7:— ‘ nun. .' q“), . . . 1| ' l , . (‘5
I . ‘ . . ‘ . _ ‘

x. W ‘ w :‘ ‘ a . ‘ . . '. ' ' - ' , ' ‘ -- «a
n ‘*,I' . | l ‘ ‘ > - ‘ ‘- ' ‘~ . ' ‘ . :

'. . . ‘. I . pg ‘ ' '

.. 1. I ‘ ‘ m‘ . g

‘u , V ' ‘ V ‘> I '

‘ .I . I ‘ . -' ' I '» . .
. ~ \ . »‘ " ‘3,

”"2"“ | ~ " ' I" ‘ ' . I" ‘ " I

n-s‘.» 'v-v' ‘n 1‘ . { - . .- .‘

,__ . H ‘ . A . . . .- y . . . , 1

" ‘ ‘ ‘ \I ‘ ‘ u ' .7

'.' v u x I n 4 .. .. "n H. ‘ ' . . , .H

.1 ‘ y ‘ ‘ . ll ‘ :i-‘_ l " I .; v y " y _'.-

_‘.| , . | ‘ .. y I‘ I‘ J ~ ' . .V . ‘2

‘ ‘ ' . ‘ . ,. U '. ' ‘ ".' " '. v "

u. ‘ . | ‘ . " ‘ '4 t‘ .- .7 ,

\ 1 ‘I V - u ' ' ‘ ‘ V - .

~ \ .‘ ' .‘ . “. _ . L,~ .. " .",

3‘
o
.
I
\.

~L_o"0“:l’a ,

'4

 

Date

0-7639

 

 

 

 

This is to certify that the

thesis entitled

Integrating Leaf and Seed Production Strategies
for Cowpea (Vigna unguiculata (L.) Walp.)

 

presented by

Robert Patrick Barrett

has been accepted towards fulfillment
of the requirements for

Master of Scienced Horticulture

egree in

 

 

 

Mi

,7
Major professor

2/»7/r6

MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

 

lV‘fSl.) BEIURNING MATERIAL§z

Place in book drop to
LJBRAfiJES remove this checkout from
.a-IIC!IIIL. your record. FINES will
be charged if book is
returned after the date
stamped below.

._—._.-——-—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

iii A- A #V_—_———-o¢

INTEGRATING LEAF AND SEED PRODUCTION STRATEGIES FOR COWPEA

(VLQNA UNQQICULATA (L.) WALP.)

BY

Robert Patrick Barrett

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Horticulture

1987

{ll/(\—

.JW
W.

(/ r- C‘

ABSTRACT

INTEGRATING LEAF AND SEED PRODUCTION STRATEGIES FOR COWPEA
(ELENA W (L.) WALP-)

BY

Robert Patrick Barrett

Cowpea is eaten both as a grain legume and a leaf vegetable
in much Of sub-saharan Africa. Three methods for harvesting
leaves at flowering time, with and without apex removal at 28
days after planting, were compared using 6 diverse African
cultivars. Apex removal did not change vegetative growth, and
rarely altered seed yield. All methods Of leaf harvest reduced
seed yield, but increased edible dry weight when harvested leaves
were added to seed weights. The average edible dry weight yields
were 136%, 118%, and 104% Of the control's seed weight for
multiple harvest, single harvest, and pruning, respectively. The
best treatment was 6 weekly harvests on 'Vita 7' with pinched
apex, which yielded 209% of the control. Edible dry weight yield
was higher in trailing cultivars than in bushy cultivars. The 2
traditional cultivars from Botswana yielded about the same seed
weight and less edible dry weight than the best of 4 improved

cultivars from Nigeria.

ii

ACKNOWLEDGEMENTS

I wish to thank Dr. Skip Bittenbender for introducing me to
the cowpea as a leaf vegetable, for serving as my major professor
through the data collection stage of my research, for Obtaining
the Nigerian cowpea seeds from his Old friends at IITA, and for
assistance beyond the call Of duty in hand-harvesting and
measuring vast numbers Of cowpea leaves in a greenhouse in
summer. I also want to express my appreciation to the remaining
members Of my committee for their encouragement and gOOd advice,
to Dr. Stan Ries for taking over as my major professor after Dr.
Bittenbender left Michigan State at the end of 1985 and for
helping me revise many long drafts Of this thesis, to Dr. Bernie
Zandstra, and to Dr. Patricia Barnes-McConnell for hand-carrying
the cowpea seeds here from Botswana. For computer assistance I
thank my wife, Linda, who took the time to read the directions

for the word-processing program.

iii

TABLE OF CONTENTS

PAGE

LIST OF TABLES..........................................Vi
INTRODUCTION. O O O O O I O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O I O O O O O O O 1
LITERATURE REVIEW. 0 O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 2

Multiple uses Of cowpea................ ........ .....2
Productivity for pods and hay.......................2
Productivity Of GOWpea for hay......................2
Protein productivity Of leaves and seeds compared...4
Cropping systems in Africa..........................5
Marketing Of cowPea leaves..........................6
Processing and storage..............................7
Effect Of sun drying on nutritional value...........8
Cooking methods classified by country...............9
Place of cowpea leaves in African diets.............9
Traditional African vegetables.....................13
Nutritional value of cowpea leaves.................13
Introduction to leaf and seed yield

in bean and cowpea ...........................17
Cowpea leaf initiation, expansion, and area........18
Photosynthesis and growth rates....................21
Photosynthesis Of pods.............................23
Effects Of heat and drought on photosynthesis......23
Photosynthesis rate, hormones, and

reversal Of senescence........................25

The self-destruction hypothesis............. ....... 29
Carbon and N metabolism and translocation.... ...... 30
Constraints to N fixation................ ...... ....42
Chemical reactions of N in the plant..... .......... 43
Influence of temperature on reproduction ........... 44

Seed growth and components Of yeild in cowpea......45
Effects of stress on seed growth and

components of yield ........................... 46
Effect Of amount and timing of defoliation ......... 49
Effect of the age Of the leaves removed ............ 54
Effects of plant size on seed yield ................ 59
Compensatory growth following defoliation .......... 61
Summary and conclusions ............................ 62

iv

MATERIALS AND METHODS ................................... 65

The cultivars...................... .............. ..65
Experimental design........... ..................... 65
Growing conditions.................................66
Leaf measurements........................... ....... 67
Weighing......................... .................. 68
Missing data........................ ............... 71
RESULTS...................................... ........... 71
Harvest index for leaves, seeds, and total
edible products........................ ....... 71
Edible products yield....................... ....... 76
Seed yield................................. ........ 76
Components of seed yield.................. ......... 81
Relationships between components of seed yield.....85
Leaf growth, abscission, and harvest.. ........ .....85
Weight of various above ground plant parts ...... ..93
Days to first flower, first dry pod,
and to ripen and dry all pods ................. 97
DISCUSSION AND CONCLUSIONS ................ ........... ..97
Effects of cultivars................. ........... ...97
Effects of treatments............ ................. 100
Practical applications............... ............. 102
LITERATURE CITED................ ....................... 105

LIST OF TABLES

TABLE PAGE

1 Cowpea leaf cooking methods classified

by country, with citation numbers ............ 10
2 Nutrient content/ 100g edible portion

of cowpea leaves....................... ...... 14
3 Proportions of seed N content from current

uptake and translocation.... ................. 31
4 Experimental conditions for N translocation

analysiSOO...OOOOOOOOOOOO0..0.0.0.0....0.IO..32

5 Percentage of N in parts of cowpea plants
at anthesis. ................................34

6 Percentage of N in parts of cowpea plants
at final harvest.....OOOOOOOOOOOOOOOOOOOOO0.035

7 N concentration and distribution in
components of cowpea 'K-2809' at 3 growth
stageSOOOOOOOOOOOIOOOO...OOOOOOOOOOOOOOO0.0.037

8 Yield as percent of control in 'Ica-guali'
and 'Porrillo sintetico' beans...............52

9 Seed yields expressed as percentage of the
control resulting from defoliation of zones
A (upper 1/3), B (middle 1/3), and C (lower
1/3) in bean cv. '373' at 2 growth stages....58

10 Defoliation percentages for each of the
multiple leaf harvests, and cumulative total
defoliation percentages for multiple, single
and pruning leaf harvest methods, as
measured for each of the plants in block 1...69

11 Estimated defoliation percentages for leaf
harvest methods, averaged from all blocks....70

12 The effect of cultivar, apex removal, and
leaf harvest method on the harvest index for
leaves (edible leaf wt/total wt, x 100) ...... 73
13 The effect of cultivar, apex removal, and

leaf harvest method on the harvest index
for seeds (seed wt/total wt, x 100) .......... 74

vi

14

15

16

17

18
19
20

21
22

23

24

25

26

The effect of cultivar and leaf harvest

method on the harvest index for edible
products (seed wt + harvested leaves

wt/total wt, x 100).................... ...... 75

The effect of cultivar and leaf harvest
method on the dry weight of edible products
(seeds and leaves combined) (g/plant)... ..... 77

The effect of cultivar and leaf harvest

method on the dry weight of edible products
(seeds and leaves combined)

(metric tons/hectare).................. ...... 78

The effect of cultivar and leaf harvest

method on the daily dry weight production

rate of edible products (seeds and leaves
combined) (g/mz/day).................. ....... 79

The effect of cultivar and leaf harvest
method on the estimated seed yield (9 dry
weight/plant)......................... ....... 80

The effect of cultivar and leaf harvest
method on the percent change in seed yield
with apex removal compared to apex intact....82

The effect of cultivar and leaf harvest
method on the number of pod-bearing
peduncles per plant............... ........... 82

The effect of cultivar and leaf harvest
method on the number of pods per plant. ...... 83

The effect of cultivar and leaf harvest
method on the number of seeds per pod ........ 83

The effect of cultivar and leaf harvest
method on the dry weight per 100 seeds
(g/plant) .................................... 84

The effect of cultivar on components of

yield for the controls, and on the

percentage of each cultivar's control

values for the leaf harvest methods .......... 86

The effect of cultivar and leaf harvest

method on the dry weight of all leaves
(harvested, abscised, and remaining)

(g/plant) .................................... 87

The effect of cultivar, apex removal, and

leaf harvest method on the dry weight of

senesced abscised leaves (g/plant) ........... 88
vii

27

28

29

3O

31

32

33

The effect of cultivar and leaf harvest

method on the mean number of leaves

harvested, with means for each of the

multiple harvests....................... ..... 90

The effect of cultivar and leaf harvest

method on the dry weight of harvested

leaves (g/plant), with means for each of

the multiple harvests................ ........ 91

The effect of cultivar and leaf harvest
method on the total area of harvested
leaves (mZ/plant)............. ..... .. ........ 92

The effect of cultivar and leaf harvest
method on the estimated dry weight of all
the above ground plant parts (g/plant).. ..... 94

The effect of cultivar and leaf harvest
method on the estimated dry weight of pod
hu8ks (g/plant)0.0.0.0000...00.0.0000... ..... 95

The effect of cultivar and leaf harvest
method on the dry weight of stems and
petioles (g/plant).................. ......... 96

The effect of cultivar (averaged across
treatments) and leaf harvest method (averaged
across cultivars) on number of days between
planting and anthesis, first dry pod, and

last dry pod............... ....... .. ......... 98

viii

INTRODUCTION

The cowpea is important in Africa as a grain legume and as a
leaf vegetable, and sometimes both products are harvested from
the same plants (14, 26). On a dry weight basis, the leaves are
as nutritious as the seeds (14). Though partial defoliation
usually reduces seed yield, increased seed yield has been
recorded following leaf removal (34, 39, 82). Because past
research has concentrated on the yield of only one product, this
research focused on the production of both leaves and seeds. It
was hypothesized that the total yield of edible products (dry
'weight) from cowpea could be increased by harvesting both leaves
and seeds. Three factors were investigated: methods of leaf
harvest, removal of apical bud at 28 days, and genotype. six
African cultivars representing a range of phenotypes were
selected. Four were developed for high seed yields, and two were
landraces consumed for leaves and seeds. The experiment was

conducted in a greenhouse at Michigan State University from June-

September, 1985.

LITERATURE REVIEW

u se 0 e

The cowpea is usually considered as strictly a grain legume,
but in the more humid parts of East Africa the crop is raised
more for the leaves than the seeds (1). In Zimbabwe, the leaf
and grain are given equal importance (79). The use of beans and
cowpeas as both leaf vegetables and grain legumes has been
documented in 14 African countries, especially in eastern and
southern Africa (14). The young pods, stem tips, and unripe
seeds are also eaten in the tropics, and even the roots are
consumed in Ethiopia and Sudan (117). In Botswana, researchers
have selected triple-purpose landraces for seed, forage, and leaf

vegetable production (26).

Ergdugtivity for pods and hay

In India, it has been noted that harvesting immature cowpea
pods for human consumption prevented leaf senescence, and the
plants remaining after pod harvest were valuable for hay,
containing 11-12% protein on a dry weight basis. Green fodder
including pods was compared with hay yield plus separate pod
harvest for 2 cultivars. Net profits showed that harvesting 2

products from the same plant can be more profitable (9).

Productivity of coypea for hay

Cowpea is an important forage and hay crop, and its
productivity under tropical conditions is well documented.
Because the coarse stems of cowpea hinder rapid drying of the

hay, cowpea is mainly grown for fodder where the shortage of
2

water makes production of other species, such as alfalfa, too
difficult (109). Cowpea plants continually produce new leaves if
out back regularly from an early stage. As early as 1907 it was
reported that, weather permitting, cowpea pastures will
regenerate vigorously if not severely damaged by grazing (30).
In New South Wales, Australia, cowpea is used for strip-grazed
pastures, and cattle graze the same crop 2 to 5 times (109).
Many cultivars of cowpea are suitable for multiple harvests,
which can be more productive on a kg/ha/day basis than when
harvested once at the end of the season. Cutting forage COWpea
'E.C. 4216' once or twice gave dry weight yields of 1630-2110
kg/ha and 2560-3970 kg/ha, respectively. Average production for
single harvests was 39.6 kg/ha/day, but 48.0 kg/ha/day if out
twice (80).

The highest reported cowpea fodder production is from India.
Comparing 5 cultivars under irrigation in the dry season, yields
ranged from 3557 to 4174 kg/ha of dry matter, with crude protein
yields of 692 to 889 kg/ha (19.3-23.4% protein). In the monsoon
(rainy) season, dry matter yields ranged from 3779 to 6249, and
crude protein from 618 to 1210 kg/ha (15.3-20.2% protein). The
highest yielder in the monsoon season (65 days) was 'Russian
Giant', which gave dry weight and protein yields of 96.1 and 18.5
kg/ha/day, respectively (102).

Also in India, 26 COWpea cultivars were tested for fodder
and nutrient production, but the number of days from sowing to
harvest in the "pod-initiation stage" was not reported . Fodder

dry weight ranged from 2090 to 6160 kg/ha, and crude protein from

4
374 to 998 kg/ha, averaging 4330 and 675 kg/ha, respectively.

Protein content varied between 13.2% and 18.8% of dry weight,
averaging 15.9% (121).

The effects of sowing date on fodder and seed production for
14 cultivars of cowpea were tested in Shika, Nigeria. The hay
was harvested when the first yellow pod began to dry in half the
plots, and the seeds when all pods were dry. Fodder dry weight
(1048-5044 kg/ha) was divided into the components of leaves (244-
1412 kg/ha), stems (413-2060 kg/ha), and pods (279-1715 kg/ha).
Average crude protein content (dry weight basis) was 24.4% in
leaves, 11.0% in stems, 21.7% in pods (averaging 17.5% for

fodder), and 27.1% in mature seeds (4).

W

The high productivity of leaf vegetables compared to grain
legumes has been emphasized, not only for volume of food but for
protein as well (8). Using U.S. yield averages, protein
contents, and amino acid profiles, the yield of essential amino
acids in cowpea was 3.5 times higher from forage than from seeds
(3).

On a dry weight basis, the protein content of COWpea leaves
has been measured at 28.5% (61) and 32.8-34.3%, with 88% retained
after cooking (64). This is higher than the 23.3-30.0% of dry
‘weight range for protein in seeds of 14 cultivars (4).

It was estimated that cowpea, under optimum conditions in a
60 day season, grown at 50 plants/m2, would produce .5-1.5 kg/m2
of fresh leaves, but only .06-.3 kg/m2 of seeds (56). Cowpea

seeds range from 19 to 35% crude protein (dry weight basis)(68),

5
giving .01-.09 kg/m2 of protein. With 4.2% protein in the leaves

(94), they would produce .02-.06 kg/m2 of protein. However,
season length must also be considered. It has been calculated
that 15 times as much protein would be produced per day from
cowpea leaves (using an average of 1.0 kg/m2 in 60 days) as from
seeds (using a world average of .0212 kg/m2 in 90 days) (14).

The above calculations consider seeds and leaves as mutually

exclusive products, but some cropping practices yield both.

WM

African farmers sometimes broadcast cowpea seeds into a
grain field when the plants are about 50 cm high. The seedlings
are thinned according to the availability of soil moisture, and
the thinnings are used as potherbs (69). In Uganda, cowpea plots
are sown broadcast, thinned gradually during growth, and the
thinnings boiled (125). In northern Cameroon, indeterminate
cultivars are usually intercropped with cereals. They are
harvested over a long period, first for leaves and green pods,
and later for dry pods, as time permits while the farmers are
primarily occupied with sorghum cultivation. The long vines are
also valued for forage (119). Some African farmers reportedly
believe that harvesting a moderate amount (about 2 tons/ha fresh
weight) of stem tips and leaves at flowering time increases seed
yield, but removing double that amount lowers it (94). In
Botswana, any removal of leaves is considered to lower the seed
yield, but long-season cultivars are valued for leaf harvesting
<mxring the vegetative stage. Short-season cultivars are

unsuitable for double-purpose production in Botswana, because

6

they do not produce excess leaves (27). In Kenya, COWpea plots
are harvested for leaves starting in about 3 weeks when 5-10
leaves are present, and continuing up to flowering. Sometimes
leaves are harvested during flowering as well, but not after pod
filling begins (14).

Cowpea is sown thickly if grown only for the leaves, and is
harvested in 3 weeks by uprooting the seedlings or cutting at
ground level. This system is used year-round in West and Central

Africa (94) and in western Kenya (14).

112W

In many parts of tropical Africa, GOWpea leaves are among
the top 4 leaf vegetables in terms of quantities eaten (14). The
low price compared to other vegetables in the markets favors high
consumption in southern Benin (127) and the highlands of Kenya,
where cowpea leaves are the least costly (94). In parts of Kenya
they are sold only in the rainy season (14, 64), but throughout
much of the country they are available all year, and have become
an important cash crop near urban areas (14).

Nutrient prices were calculated from market prices (surveyed
for 2 years) in Bumbuli, Usambara, Tanzania. Mchicha, translated
as "wild spinach", applies to 4 species of leaf vegetables,
ixuiluding cowpea, Sonchus oleraceus, and 2 species of Solanum.
Beans and maize were cheaper sources of protein than mchicha, but
mchicha was much cheaper than animal protein. Maize and beans
were also the most important sources of nutrients and energy, but
were deficient in vitamin A, which was best supplied by liver or

mchicha. Of the 28 foods considered, liver was the cheapest

7

source of all commonly deficient vitamins, but mchicha was a
close second for riboflavin, vitamin A and iron, and third for
vitamin C and niacin. For calcium, mchicha was 3 times cheaper
than the second place source, beans (107).

In addition to Benin, Kenya, and Tanzania, cowpea leaves
have been reported from markets in Ghana (137), Mali (29), almost
all regions of Cameroon (138), Ethiopia (138), Uganda (82, 101),

and Malawi (143).

WW9:

The Sukuma of Shinyanga district, northwest Tanzania, put
leaves in the sun near the house for about 4 hr before placing in
a pot with cold water, and cooking for about 3 hr, changing the
water halfway through. Cooking is usually done at night, and
then the pot is covered and left until morning when the leaves
are pounded and sun dried until the early afternoon. Storage is
in pots, for up to 1 year (48).

The Pedi of the northeast Transvaal, South Africa, prepare
cowpea and other leaf vegetables for storage by boiling for about
1.5 hr. The drained leaves are kneaded into golf ball-sized
pellets and set on a flat rock to dry in the sun. After 3 days
the lumps are scraped into sacks and stored (135).

Cowpea leaves are also preserved by boiling and sun drying
by the Venda, Ndebele, Shangaan (44), and the Kwena or Bakoni
(45), all of the northern Transvaal, by the Shona of eastern
Zimbabwe (17), and by the Tlokwa of eastern Botswana (55).

In Malawi, COWpea leaves picked in the morning are spread

out in the sun to wither for 2 or 3 hr, before being put back

8

into baskets. This reduces the bulk and preserves the flavor.
The next morning they are packed into a large (16 liter)
earthenware pot with a liter of cold water, and steamed 10-20
minutes until soft, and then spread on mats for drying in the sun
1-3 days. After 3-4 such batches have been dried, the leaves are
packed into 1.5-2.5 kg balls, covered with large masuku (Uapaca
kirkigna) leaves, bound tightly, and hung up in the hut until
needed (143).

Fresh cowpea leaves are dried in Zambia without cooking
(94). In Uganda, cowpea leaves may be dried in the sun with or
without steaming them first, and are sometimes stored in powdered
form (66). Storage as a dry powder is also recorded from
northern Nigeria (24) and East Africa (1). Storage Of dried
leaves has been reported in Ghana (137), Cameroon (139), and

Kenya (61).

WW

Excessive losses of beta-carotene often occur in sun drying
fresh vegetable leaves, and in subsequent storage. Smaller
losses occurred if the leaves were steamed first. Only 10-20%
losses were recorded from cowpea and cassava leaves, showing
remarkable stability of the B-carotene compared to other species
(49). Cowpea leaves were shown superior to other leaf vegetables
(cassava, sweet potato, and Amaranthus) in percentage and
absolute retention of vitamin C and carotene under various solar
drying treatments after minimal boiling. In full sun, cowpea

leaves lost 89% and 71% of vitamin C and total carotene,

9
respectively, but when dried in the shade they lost only 76% and

43%, respectively (76).

For human nutrition, the protein in fresh cowpea leaves was
shown to be limiting in the S-containing amino acids. Lysine was
found limiting for sun-dried leaves because a large proportion

was destroyed by drying (75).

WW
Reported methods of cowpea leaf preparation vary by country.

Boiling is the most popular, either as a potherb or alone as
”spinach" or ”greens". They may also be fried or combined with a
starchy staple food, but nowhere are cowpea leaves consumed
without cooking. Cowpea leaf consumption has been recorded from
24 countries in Africa, Asia, and the Pacific (table 1). It is
probable that cowpea leaves are traditionally eaten in many other
countries in the tropics, but they have not been reported in the

diets of New World populations.

'ets

Maize ugali, a stiff porridge,is an important staple in
Kenya. Since most of the population consumes little food from
animal products, leaf vegetables are crucial for many nutrients.
Cowpea leaves are eaten as a side dish with ugali, and to a
lesser extent cooked together with maize kernels, legumes, and
eithem'potatoes or green bananas (64). A Kenyan recipe for irio
is given as follows: boil 1 kg each of maize and beans slowly for
2-4 hru When almost ready, add 1 kg green bananas and .5 kg

chopped.cowpea leaves and boil for 20 minutes more (94).

10

Table 1. Cowpea leaf cooking methods classified by country, with
citation numbers.

Mixed in a

Sonntzx A§_§nina§hz A§_22£herby starshx_f22d fried
West Africa (24)
Senegal (122) (122)

Ghana (137)x

Benin (127)
Mali (88)

Nigeria (24)

Cameroon (139)

Central Africa

Zaire, Central African Republic, and Sudan
(Azanae)w (28)

Sudan (41)

East Africa

Ethiopia (Chako) (138) (138)

Kenya (14,64) (14,64,94) (14,49,64,94) (14)
(KikUYU) (59) (94)
Uganda (66,101) (101,125)

(Ganda) (14) (14) (14)

(Northwestern) (120) (120) (120)
Tanzania (48,75,81,101) (48)
(Sandawe) (91)

continued on next page

Table 1- continued-

quntrx A§_§pinashz A§_22£herby starshx_fggd
Southern Africa

Malawi (143) (14,143)

Zambia (94) (40.87)

Zimbabwe (17,79)

Botswana (26,55) (26)

South Africa (44,45) (44)

(Pedi) (135) (135)
Asia

Hawaii (83)

Philippines (15)

Indonesia (92)

Bangladesh (7) (89)

India (123) (89)

Pakistan (89)

z

11

Mixed in a

fried

(94)

"Spinach" is cooked alone by boiling or steaming, is

consumed as a single item, and includes references to use as
a "vegetable" or "greens".

y A "potherb" is boiled in a soup, stew, or sauce with other

foods.

parentheses.

Method of preparation not reported.

Specific tribes or regions within a country shown in

12

The Sandawe of central Tanzania consumed a diet based on
ugali, supplemented with a wide variety of protein-rich relishes.
Relishes from animal and vegetable sources were consumed with
equal frequency. Gathered plants were eaten at 45% of all meals,
but cultivated vegetables at only 9%. Leaves of crop plants
(cowpea, bean, cassava, and sweet potato) were eaten at 4% of
meals. Leaves of cowpea and other cultivated plants were dried
and saved for the dry season, when they became more prominent in
the diet (91).

In the northeastern Transvaal, the Pedi people have a
proverb: "Meat is a visitor, but spinach is a daily food". Their
normal diet consists of maize porridge, cowpeas, some cucurbits,
and a wide variety of leaf vegetables. The most common spinaches
are morogo, translated as "wild green leaves", and GOWpea leaves.
Morogo can include wild plants and cultivated COWpea and
cucurbits. During the dry season, dried morogo is crushed and
sprinkled over the porridge (135).

The Kikuyu of Kenya consume various grain legumes and
starchy staple foods, but rarely eat meat. The most frequently
eaten vegetables are the leaves of cowpea, pumpkin, cabbage, and
kale (59) .

A large number of leaves and fruits are eaten by several
ethniczgroups in northwestern Uganda. Leaves of cultivated
cowpea were commonly eaten, and less Often the wild cowpea (ssp.
deijuitiana). Cowpea leaves, dried for storage, were found equal
in protein content (22.6%) and "overall food value" to dry COWpea

seeds, except the leaves had more vitamin C (120). In addition

13

to northwestern Uganda, leaves from wild GOWpeas are gathered for

food in Kenya (67), Tanzania (78, 106, 107), and South Africa

(135).

I ill. 1 EE . ! l]

The displacement of traditional African vegetables, both
gathered and cultivated, by "exotic" or European vegetables has
been noted by many authors. This trend has alarmed some, who
point out that the traditional vegetables are generally more
nutritious, better adapted to the local environment, better able
to compete with weeds or in mixed culture, and less subject to
pests and diseases (78). With respect to vitamin A, lettuce,
cabbage, Swiss chard, and even carrots were shown inferior to
traditional East African leaf vegetables such as cowpea, bean,
cassava, and Amaranthng (49, 78). Cowpea leaves had over 4 times
the protein of cabbage grown in Tanzania (78).

The usefulness of annual leaf vegetables is limited in most
areas of Tanzania. Because of a shortage Of water, they can be
grown only during the rainy season when alternate sources of
leaves are most abundant. Rural families gather nutritious

leaves of weeds and crop plants to eat, including cowpea (129).

.Nutritienal_xalue of cowpea learse

The nutritional value of cowpea compares favorably with

 

other'leaf vegetables commonly eaten in Africa. It is a good
source of minerals, especially iron, calcium, phosphorus, and
zinc (table 2). Available iron was measured at 14% of total iron

(64).

14

Table 2. Nutrient content/ 1009 edible portion of cowpea leaves.

 

Preparation method

EI2§D:HDQQQE§Q
figulssz 1321 1§§1 1121
calories (KCAL) 30 - 44
water (9) - 85.7 85.0
protein (9) 4.8 - 4.7
vitamin A (IU) 712 13500 4000
vitamin C (mg) 36 - 56
niacin (mg) 1.1 - 2.1
riboflavin (ug) 175 - 370
folacin (ug) - 288 -
thiamine (ug) 354 - 200
calcium (mg Ca) 73 217 256
potassium (mg K) - 446 -
phosphorus (mg P) 106 88 63
iron (mg Fe) 2.2 5.5 5.7
zinc (mg Zn) - 2.8 -

Fresh-Cooked
(68) (65)
86.8 89.3
333 10833
12 5.8
43 -

- 73
.04 -

- 133

- 128

- 42

- 4.7

- 2.1

Dried

1191
227
10.6
22.6
45000

86

1556

348

Reference cited.

15

The fluoride content of common leaf vegetables grown in
different parts of Kenya was highest in cowpea, from 10 to 115 mg
E/kg (96).

Vitamin C content in cowpea leaves was reduced by 76% by
boiling for 15-20 minutes, from 91 to 19 mg/100 9. An extreme
range of 56-123 mg/100 g was recorded for fresh leaves. Boiling
with trona, a crude hydrated sodium carbonate formula commonly
used by Kenyans to soften fibrous leaf vegetables, reduced the
vitamin C content to "a mere trace" (50).

Vitamin C content in cowpea leaves was reduced 90% by
continued cooking for 30 minutes after boiling, from 59 down to 6
mg/100 9 (not counting an additional 2 mg/lOOg leached into the
cooking water) (64, 65). While Kenyans usually consume the
cooking water from vegetables, they use trona for its flavor even
with young tender cowpea leaves that need no softening (14).

Thus it appears that even more than 90% of the vitamin C would be
destroyed when trona is used.

Vitamin C content in cowpea leaves was reduced by 91% when
boiled for 50 seconds and dried in full sun, but only by 76% when
the drying was done under shade (76).

An international unit (IU) of vitamin A equals 0.6
micrograms of B-carotene or 0.3 micrograms of retinol. Vitamin A
occurs as retinol only in animal products, but the B-carotene
eaten in vegetables is converted to retinol at the ratio of 6:1.
Vitamin A and carotene are not soluble in water or destroyed by
cooking, but rapidly decompose when exposed to air (142). Drying

COWpea leaves greatly reduces the availability of vitamin A, but

16

the amount remaining may still be large. The wide range of
concentrations reported for B-carotene in cowpea leaf samples
probably reflects differences in freshness, water content,
cooking, and analytical techniques. Differing cultivars and
growing conditions in the various locations could also be
involved.

B-carotene concentration in cowpea leaves from the highlands
of Kenya ranged between 332-345 IU/IOOg (67).

When the nutrient content of various common foods of
northeast Tanzania was determined, vitamin A was shown to be one
of the most deficient nutrients in the diet. Yellow maize was
presumed to supply about 50% of the vitamin A requirement. It
had the second-highest concentration of all foods analyzed, with
about 400 IU/100g. Cowpea leaves were highest, with 3000 IU/100g
(6).

The carotene content of fresh cowpea leaves declined by 47%
when boiled for 1 hr, from 9500 IU to 5000 IU Because eye
diseases from vitamin A deficiency were common in Tanzania, more
consumption of leaf vegetables and fish livers was recommended
(79).

Total carotene in COWpea leaves was measured at 13,500
IU/lOOg before and 10,833 IU/100g after cooking 30 minutes, for a
decline of only 20% (64, 65).

A mean of 15,693 IU/lOOg B-carotene was recorded in fresh
cowpea leaves, with a range of 13,750-17,500 IU/100g. When
measured on a dry weight basis, B-carotene content increased 70%

with 15 minutes of boiling, from 91,845 to 156,322 IU/100g. In

17

other leaf vegetables, B-carotene increases of 24-88% were
observed under the same treatment, presumably due to better
extraction by the acetone solvent. For Kenya, dried cooked
cowpea leaf proved economical and culturally acceptable as a
vitamin A supplement. As little as 10-20 g of dried leaf per day
would give an adequate intake (51). Cowpea leaves were still an
excellent source of vitamin A, even after the concentration had
been reduced by drying.

On a dry weight basis, fresh cowpea leaves contained 114,000
IU/lOOg, which was reduced 58% to 48,000 IU/100g by boiling 50

seconds and sun-drying (76).

I at- or 0 :e‘ are =“! ‘ Q I 0‘: - d Co oea
In many parts of Africa, bean (Pnaaaglaa yalgagis) and
cowpea (Elana gngnigglaga) plantings are harvested for 2, 3, or

even 4 edible products. Tender pods with immature seeds and the
full-sized "shell beans" or "green peas" are eaten as vegetables,
and the dried seeds are soaked and boiled before eating. The
young leaves are also cooked and eaten (14). Harvesting leaves
is not often seriously considered by western scientists, as it is
believed to reduce seed yields, erroneously considered to be the
only economic product. However, yield increases following leaf
removal have been reported (34, 39, 82). Most of the reports
reviewed here cover defoliation in the laboratory as part of
basic research, or try to quantify insect damage in the field.

Many factors are involved in the effect of leaf removal on
seed yield in nitrogen-fixing legume crops. The following

discussion will examine the role of the leaves in providing,

18
through photosynthesis and translocation, the carbohydrates and

nitrogen (N) required for seed production, and will examine the
processes that defoliation affects. An empirical formula for
relating yield loss to defoliation percentage is worthless if it
does not consider the age of the foliage and the plant's stage of
development (60). Because cultivars differ in the time required
to mature and weather is so variable, reporting treatments
relative to the plant's growth stage is much more useful than the

number of days from planting.

W

In cowpea, many processes that affect leaf area are
influenced by temperature. High or low temperatures during
vegetative growth can greatly accelerate or retard leaf area and
reproductive growth (71, 116).

The final size of each leaf and rates of leaf appearance and
expansion increase with temperature, while the duration of
expansion for each leaf decreases, because a constant number of
leaves on each apex are expanding at any given time (71). Others
also found cowpea leaf appearance rate highly dependent on
temperature (12).

Cowpea leaves are initiated about twice as fast at 30 C as
at 20 C average temperatures (116). The base temperature for
leaf appearance is estimated at 16 C and for leaf expansion at 21
C (71). Since cowpea leaves expand mostly at night, daily
minimum rather than average temperatures would be critical for
leaf area expansion (116). Higher daytime temperatures following

partial defoliation would therefore be expected to speed up leaf

19

initiation (but not leaf expansion) on branches below the canopy,
speeding up replacement of the lost leaves.

Leaf initiation is slowed by cool temperatures, moisture
stress, and dense planting (which can promote moisture stress)
(116). Dense plantings of 50-100 plants/m2 reduced the leaf
appearance rate (71). Leaf appearance rate was not related to
leaf water potential, but was closely related to stomatal
conductance (r=.80**). Stomatal conductance declined as vapor
pressure deficit increased (141).

Leaf dry weight/area tended to decrease with rising
temperature (in controlled environments but not in the field) and
decrease with time. Leaves continued to increase in dry weight
after full expansion (73).

In cowpea, as in fava bean (yigia faya) and soybean (Glycine
max), final leaf size increases with node number for a time, and
then decreases in the last leaves that develop. The variation is
mainly due to the rate, rather than duration of expansion. This
can be explained by source-sink relations. The earliest, lowest
leaves expand more slowly because so little active leaf surface
supports the new growth, but this constraint gradually
diminishes. The latest leaves become ever smaller as assimilates
are diverted to growing seeds (71). For this reason, harvested
leaves would be replaced most easily when they are largest, at or
before anthesis.

In cowpea, nutrient translocation, leaf senescence, and
death begin sooner at higher temperatures. Corrected for

temperature, the time of the start of leaf death divided by the

20
time of the end of leaf growth gives a constant of 0.86. These

appear to be 2 ends of a single process, which is accelerated by
higher mean temperatures, but is not influenced at all by time of
flowering. Apparently, flowering is influenced by minimum,
rather than mean temperature. Once leaf senescence starts, the
overall rate of leaf area loss depends on the pod-filling pattern
rather than temperature (71).

Leaf Area Index (LAI) is the ratio of a plant's leaf area
divided by the area of ground the plant shades at noon. The LAI
depends on the balance between the rates of expansion in young
leaves and abscission in senescent old leaves. Maximum light
interception is reached at a LAI of about 3 in cowpea, regardless
of growth habit or leaf shape. At LAI values below 3, the dry
matter accumulation rate increases linearly with LAI. Peak LAI
is normally reached in cowpea at flowering time, but the number
of days required and the LAI reached depend on genotype,
temperature, and water supply (116). The greatest LAI was
observed during the 2 weeks after first flower in 3 cowpea
cultivars (116), and 18 days after blooming in 'Caloona' (58).

Peak LAI values above 5 are typical for indeterminate
cowpeas. Determinate cultivars frequently have a maximum LAI of
3, but the time needed to develop it has ranged from 30 to 60
days for various cultivars, spacings, locations, and weather
conditions. Generally, cowpeas attain a larger LAI and reach a
given value sooner at closer spacings (116).

In a determinate and an indeterminate cultivar of cowpea at

2 planting densities (12/m2 and 24/m2), plants at the low density

21

had greater leaf areas and leaf, stem, and total dry weights per
plant. The difference was relatively greater for the
indeterminate cultivar which had double the leaf area/plant at
half the density, compared with only 50% greater for the
determinate cultivar (19).

The canopy temperatures of well-watered plants were 2-7 C
below ambient air temperatures. But drought-stressed leaves
could not maintain lower temperatures in the middle of the day by
evaporative cooling. The stressed plants adapted by limiting the
expansion of new leaves, so that maximum LAI was 3, compared with

5 in the control (141).

WWW

Leaf photosynthesis rates are known to vary between cowpea
cultivars, but a wide range of measurements is often reported for
the same cultivar. The maximum photosynthetic rate for each
cowpea leaf occurs at full expansion, when light-saturation is
reached at about 66% of full sunlight in the tropics. As leaves
age the rate declines gradually, but plunges suddenly as they
begin to senesce (116).

Maximum photosynthetic rates/area for leaves increase during
leaf expansion. High maximum rates were recorded for a span Of
about 20 days in each leaf, with the peak usually reached just
after expansion ceased. Photosynthetic rates and leaf areas
declined thereafter, as the leaves senesced. The earliest leaves
have lower maximum rates, which increase with leaf number up to
about the fifth leaf, and then remain constant. The mean values

for maximum photosynthetic rate were about 1.2mg COZ/mZ/sec for

22
the third leaf, and about 1.5mg COZ/mz/sec for leaves 5, 7, and 9

(72).

The net photosynthesis rate of 'Caloona' cowpea plants rose
to a peak at 70 days after planting (10 days after anthesis), but
before the greatest leaf area was measured at 78 days. Net
photosynthesis then declined rapidly as leaves abscised (58).

In a field experiment, the crop growth rates (mg/cmz/day) of
3 cowpea cultivars were linearly related to the leaf area index
(which never exceeded 3 because of wide spacing). The mean net
assimilation rate (crop growth rate) for 3 cultivars in the field
and greenhouse varied only between 4.6 and 5.2 g/mz/day. The
maximum leaf area and crop growth rate occurred during the first
2 weeks after first flower (93).

The crop growth rate (g/mz/week) of cowpea cv. 'Morod' from
4-6 weeks after planting was directly related to LAI (r=1.00) in
the range of 0.2-2.0. The relationship was also close for 6-8
weeks with LAI below 2.0. A peak growth rate of about
85g/m2/week (or 12g/day) was recorded at a LAI of about 1.6 (36).

Maximum growth rates of 98g/m2/week (14g/day) were measured
at a LAI of about 2.5, with no differences between broad and
narrow leaf COWpea types (140).

During early growth in cowpea, dry weight increase was
proportional to intercepted radiation. Weight differences
between trials at different seasons were due equally to
variations in sunlight, the proportion intercepted, and net
photosynthetic efficiency (73). The mean radiation conversion

efficiency has been measured at 3.8% (73) and 3.4% (126).

23
Some have concluded that declining photosynthesis with age

limits seed yield in cowpea. During rapid seed growth,
carbohydrate use was greater than production (19).

In the dark, cowpea leaves respire and give off CO2 at rates
equivalent to about 5-10% of the C02 uptake during maximum
photosynthesis (116).

The C02 compensation point for cowpea has been measured at

62 ppm C02, at a temperature of 22-25 C (62).

WW
Cowpea pods and peduncles did not show a net export of

carbohydrates, but were shown to contribute 80-85% of their

respiration requirements (72). In beans under field conditions,
76-97% of 14002 fixed by the pods was retained in the pod 24 hr
later. At earlier stages of pod growth, a higher percentage of
labeled material was exported, but the proportion sent to other

pods steadily increased with time (74).

on on ot s nthesis
Even when adequately watered, fruits and leaves exposed to

the sun commonly reach temperatures above 30 C that inhibit
photosynthesis. Shading can reduce surface temperatures by up to
4 C (116). Canopy temperatures of well-watered cowpea plants
were 2-7 C below ambient air temperatures. Although moisture
stress reduced the water potential in cowpea leaves, stomatal
conductance showed a close negative correlation with vapor
pressure deficit (VPD). Despite stomata closing as VPD rose,

evaporative cooling of leaves increased with VPD (141).

24

Gas exchange measurements indicated that photosynthesis is
not seriously impaired by moisture stress at constant temperature
(27% reduction at full irradiance). But, drought-stressed cowpea
leaves could not stay below the ambient air temperature by
evaporative cooling in the middle of the day. Less
photosynthesis would occur in the short term from high
temperatures, and in the long term due to limited expansion of
new leaves. (115, 143).

In field-grown 'Michelite' beans, there was little relation
between LAI and seed yield. For various planting dates, the
effect of a large leaf area could be either helpful or harmful,
depending on the weather at anthesis. When the weather was hot
and dry, a large leaf area was a liability (25).

Thus, if leaves become too hot and less productive when
moisture is limiting, removing a number of leaves could make the
available water supply adequate to cool the remaining leaves,
allowing full production. Under such conditions, partial
defoliation may not reduce net photosynthesis. Because most
plants grow rapidly early in the season when competition for
water and light is minimal, they often acquire more leaf area
than can be utilized in hot dry weather later. Since net
photosynthesis declines more or less linearly with reduced turgor
pressure, the loss of a few leaves under drougOv conditions
(which occurs naturally through more rapid senescence) would
benefit the plant by allowing greater water supplies to the
remaining leaves, and greater photosynthesis (57). The labor

cost of trimming leaves during drought may or may not be

25
justified by the value of the additional seeds produced, but when

the harvested leaves are assigned a fair market value as

vegetables, this practice can be profitable.

v sal s nesc nce

Photosynthetic rates in cowpea leaves decrease with age (71,
60). Chlorophyll, protein, carotenoid, and RNA levels and leaf
weight/area decline as leaves senesce. Plant hormones including
auxins, gibberellins, and cytokinins have been shown to delay
these symptoms of senescence (128). Greater photosynthetic
efficiency and reversal of senescence have been documented in
bean and cowpea following defoliation (5, 108, 110, 128, 132).

In a series of defoliation experiments with urd bean (gigaa
mangg), photosynthetic rates increased in remaining leaves, which
varied directly with defoliation severity (98).

During defoliation experiments with beans, the remaining
leaves in the defoliated plots were much darker green than in the
control plots. This suggests that chlorophyll content increased
to compensate for the leaves removed (33).

In mung bean (yigaa LQQiQLQ), pod number was not affected by
applied NAA, GA3, or the cytokinin 6-benzylaminopurine (BAP)
during bud swelling, blooming, or pod setting, though BAP applied
through all 3 stages significantly increased pod number. It was
hypothesized that pod set and development are limited by
photosynthate supply, either directly to the pods, or indirectly.
Because the root system was a major sink, especially during

flowering and pod setting, it is possible that root-synthesized

 

in
“V.

I:'

5"
O

26

hormones (either gibberellins or cytokinins) regulate
reproductive yield (22).

Removing the upper leaves from a cowpea plant significantly
increased photosynthesis and dry weight accumulation in the
primary leaves and the second trifoliate, reversing the symptoms
of senescence. Continued removal of new leaves maintained a high
photosynthetic rate in remaining lower leaves (108).

Treating bean plants with the synthetic cytokinin
benzyladenine (BA) or decapitating the plants above the primary
leaves produced marked increases in DNA, RNA, protein,
carotenoid, and chlorophyll levels. Senescence was reversible by
either treatment even after the primary leaves had begun to
yellow and the carotenoids and protein had been largely depleted
35 days after planting, but severing the upper part of the plant
was shown to be more effective than BA. This was attributed to a
continuous supply of endogenous cytokinins to the leaves with no
competition from other plant parts (128).

Decapitation of bean seedlings induced additional
chlorophyll formation, greater expansion, and a longer life span
in the primary leaves (110). In partially defoliated bean plants
(all younger leaves and buds above the second trifoliate
removed), the remaining leaves became darker green, thicker, and
leathery. Compared with measurements (per area of leaf) on the
control plants, which showed only minor fluctuations, the
photosynthesis rate and the in-vitro ribulose bisphosphate
carboxylase activity both increased in the treated plants until 8

or 10 days after defoliation, and then leveled off. Total

 

00
u

‘1

[.3

3.]

27

chlorophyll, however, continued to increase for the duration of
the experiment (132).

In defoliated bean plants (all leaves above the first
trifoliate removed as they emerged), the photosynthetic rate of
the first trifoliate slowly increased. It was always above the
control, but the difference was not significant until the final
measurement at 45 days. These results place doubt upon the
hypothesis that photosynthetic capacity of leaves is proportional
to the demand placed upon them by the rest of the plant (the
source-sink ratio). The photosynthetic rate would be expected to
stay below rather than above the rate for the control. But if
the photosynthetic rate and chlorophyll content vary with
endogenous hormone levels, they could be increased by defoliation
or other damage which removes the competitive demand of other
plant parts (5). This hypothesis fits the present evidence, and
is supported by other results (16).

In another experiment, 3 treatments on bean seedlings left
the primary leaves intact. Some plants were decapitated and
totally debudded, others were heat-girdled above the primary
leaves to seal off the phloem and then debudded below the wound,
and the third group was defoliated (all leaves were pinched off
before they could expand) but not debudded. Heat-girdled plants
continued normal vegetative growth and defoliated plants
continued growing stems and branches (16).

The primary leaf area, thickness, chlorophyll content, and
photosynthesis rate increased similarly after the decapitation

and defoliation treatments. Decapitated plants also had 80-85%

 

1"

ti

In
I

 

y

'-

 

-
\
r

3.!

‘l

.‘

  

28
greater ribulose bisphosphate carboxylase activity than the heat-

girdled plants and the control. In the control and the heat-
girdled plants, primary leaf area and thickness did not increase
with time. In the decapitated plants, photosynthetic activity
did not increase in the first 4 days, but did so within 7-8 days,
and was even higher 16 days after treatment. Photosynthetic
activity did not increase in the primary leaves of the heat-
girdled plants where translocation of assimilates downward was
blocked, even though the primary leaves obtained no assimilates
from above just as in the decapitated group. The defoliated
plants had a much greater demand placed on the primary leaves to
support continued stem growth, yet the measured increases were
equal to those in the decapitated plants (16).

Thus, it is unlikely that photosynthesis is regulated by
source-sink relationships. Neither is it likely that factors
regulating photosynthesis are transmitted from younger leaves
through the phloem or against the flow through the xylem, nor
from the buds (16). To maintain photosynthesis in primary
leaves, it is not even necessary to remove the leaves above.
Bean primary leaves that have started to senesce can be
rejuvenated by preventing transpiration from the upper leaves, as
by greasing them or covering with a plastic bag (144).

An explanation for these observations has been suggested.
Increased photosynthesis in leaves remaining after partial
defoliation may result from an increased supply of endogenous
cytokinins, which are synthesized in the root and transported in

the xylem (136). Applying cytokinins to bean primary leaves has

29

increased chlorophyll content, leaf area and thickness, ribulose
bisphosphate carboxylase activity, and photosynthesis rate. It
has been further hypothesized that the flow rate of water through
the roots and xylem does not change the production rate for the
regulating factor. Thus, the endogenous regulation of each
leaf's photosynthetic capacity depends on the amount of the
hormone arriving in the transpiration stream. Defoliation
increases the amount received by the remaining leaves (16).

There is further evidence that cytokinins are involved in
promoting photosynthesis and delaying leaf senescence.
Cytokinins are mainly synthesized in roots, and are carried
through the xylem to the leaves in the transpiration stream (46,
122, 144). Senescence in detatched leaves can be reversed if the
leaves are induced to form roots (144). Since removing buds or
fruits (and presumably other organs) increases the cytokinin
concentration in the remaining parts, endogenous cytokinins
probably travel to all parts of the shoot (124).

To confirm the cytokinin hypothesis would require that
endogenous cytokinin levels be correlated with the physical
changes in the leaves that resulted from the treatments. This
was not done in any of the experiments on bean and cowpea

reviewed here.

Tha self-destruction hypothesis

In most annual legume crops (including cowpea but not
peanut, Arrachis hypogaea) the N demand by growing seeds is
greater than can be fulfilled by current uptake, requiring

translocation from vegetative parts. Metabolizing proteins to

30
export N from the vegetative tissues eventually halts

photosynthesis, induces senescence, terminates the seed-filling
period and restricts yield. Seed yield in such self-destructive
species depends on the N uptake rate during pod-filling, since
the deficit must be met by internal translocation. A low
assimilation rate means more rapid senescence, a shorter duration
of seed growth, and lower yield (111).

Later researchers concluded that the decline in
photosynthesis following early fruit development limited seed
yield in cowpea, since carbohydrate consumption during rapid seed
growth outpaced production. It was hypothesized that
indeterminate cultivars which continue vegetative growth while
producing pods would be higher-yielding, as would those with a
longer duration of pod development (19).

Others claim that more recent data do not support the
hypothesis of self-destruction, but do not elaborate (116). The
evidence presented in the next section supports the self-

destruction hypothesis.

b 's t slocat'on
The proportions of seed N content derived from current
uptake (from soil and symbiotic fixation) and translocated from
leaves and stems varied in different experiments (table 3),
influenced by the cultivar and experimental conditions (table 4).
Generally, about 50% of seed N came from current uptake, 35% from

leaves, and 15% from stems.

31

Table 3. Proportions of seed N content from current uptake and

 

 

 

translocation.
_______&_Iran§lecated

§_Qurrent Ergm_leaf Er2m_§tem§ Tetal Reference
56 25 19 44 (58)
50z - - 50 (73)
47 40 13 53 (32)
78 13 10 23 (90)Y
50 34 16 50 (32)
47 37 16 53 (115)
53 33 14 47 (85)x
31 36 33 69 (85)w

Estimated values.

Y Numbers given add up to 101%.

X

Nodulated: no added N.

w Not nodulated: fertilized with nitrate.

32

Table 4. Experimental conditions for N translocation analysis.

 

92112131: Def? 1!! 81112.25 EEK EX ____Reference
Caloona No None Yes 62 120 (58)
TVu-4552 Yes 30kg/ha Yes 30-37 83-87? (73)
TVu-4552 Yes 30kg/ha Yes 35-40 56-61 (32)
TVu-1503 ? None Yes 35? 87 (90)
K-2809 Yes 25ppm Yes 48 90 (32)
K-2809 Yes 30ppm Yes 45-47 88-94 (115)
K-2809 Yes None Yes 37 77 (85)
K-2809 Yes 200ppm No 37 77 (85)

Determinate cultivar.

Y Applied N.

x Rhizobial nodules present.
w Days to first flower.
v

Days to final harvest.

 

'l

.-

‘3

It

 

‘ a

5“

Q
7‘03
I l'

I
In

‘5

 

\
n
l...
l
.._‘
‘-
<
;.
. .
..‘
‘\
’n
H.
‘
‘\
\~‘:
I
n
..
4 1
s

33

The percentages Of N in plant parts at anthesis were also
measured. Leaves contained 3-6%, stems 1-3%, roots 1-2%, and
rhizobial nodules 6-7% (table 5).

The percentages of N in plant parts at final harvest was
lower than at anthesis for leaves and stems, and about the same
for roots and nodules. Seeds contained 3.5-4.3% N, and pod husks
0.9-1.5% (table 6). Other authors reported 0.9% N in husks of
mature pods (105), and an average figure of 4.0% N for mature
cowpea seeds (117). The N contents of leaves, stems, and pods
(with seeds) were 3.9%, 1.8%, and 3.5%, respectively, when
harvested for fodder at near maturity (4).

In annual legumes the leaves abscise after exporting much of
their nutrient reserves to the seeds. The dynamics of C and N
metabolism and transfer during the cowpea life cycle have been
reported in detail for a determinate cultivar 'K-2809' (32, 85,
115, 118) and for an indeterminate fodder cultivar, 'Caloona'
(58), which exhibits the "self-destruction syndrome" (90).

The net photosynthesis rate of 'Caloona' plants rose to a
peak during early flowering (70 days), and then declined rapidly
as leaves abscised. The greatest leaf area was at 78 days.
Maximum N fixation was measured shortly before anthesis (60
days), but it also dropped sharply as nodules abscised so that
fixation ceased by day 92. During the seed-filling stage (79-120
days), all vegetative parts lost C, N, and dry weight. All the N
given up was assumed to have been derived from protein breakdown,
but no evidence was offered. Before leaflets abscised, 31% of

their N content was exported, and 32% was mobilized from

 

La.)

34

Table 5. Percentage of N in parts of cowpea plants at anthesis.

Laayaa 53am§ Ragga Nodules Reference
5.2-5.6 1.7-1.8 1.7-1.9 6.8-7.0 (86)
4.1-4.4 1.3-1.5 1.2-1.3 6.2-7.1 (84)
4.5 1.5 0.8 6.1 (32)
4.0-4.6 1.7-2.0 1.0-1.2 5.9-6.8 (118)2
3.2-4.4 1.2-2.6 1.1-1.5 --- (118)Y

z Nodulated: no N added.

Y Not nodulated: nitrate fertilizer added.

35
Table 6. Percentage of N in parts of cowpea plants at final

 

2 Reported in (86).

Reported in (118).

harvest.
Npfiple dependant N_ depepdent

Elamdrf .5. _1_ _1

Green leaves - 2.8-3.0 2.6-2.9
Dead leaves 2.1-2.2 1.7-1.8 1.7-2.0
Main stem 0.6-0.9 0.6-0.8 0.6-1.1
Side branches 1.4-1.7 1.0-1.3 0.9-1.3
Peduncles - 0.8-1.2 1.1-1.6
Roots - 1.1-1.2 0.9-1.3
Nodules 7.2 5.6-5.8 -

Pod husks 0.9-1.3 1.3-1.5 1.3-1.5
Seeds 3.8-4.3 3.5-3.6 3.5-3.8

36

senescing stems. Assuming 100% efficient transfer of these
reserves to the growing pods, 25% of the N needed for the seeds
would have been met by leaflets and 19% by the stems. The
remaining 56% would have come from senescing roots, rhizobial
fixation and uptake from the soil. Current photosynthesis
provided 80% of the seeds' carbohydrate requirement; only 20% was
mobilized from stems and leaves. The conversion efficiency of
imported organic substances to seed dry matter was calculated at
67% (58).

Plant N was accumulated at 28 mg/plant/day during both
vegetative and late reproductive stages. During early
reproductive growth over half of total plant N was accumulated,
and the rate rose to 111 mg/plant/day. Translocation from the
stems and leaves formed before anthesis accounted for 30% of the
N content of developing fruits at mid pod-fill, and for 12% of
final content in seeds. Translocation from all vegetative parts
after mid pod-fill contributed 62% of the increase in seed N
content (44% of final content), with 37% from leaves, 18% from
stems and peduncles, and 7% from nodules, with the remaining 38%
of the increase from current uptake (table 7). Source of total N
in the seeds at harvest is shown in table 4. At the final
harvest on day 90, seeds comprised 48% of total dry weight and
contained 63% of total plant N. Redistribution and N proportions
very close to these results, despite differences in growing
period, plant size, and total N content, were obtained with "TVu

4552" cowpeas in the field in Nigeria (32).

 

.A

'1‘

9

44.:

 

 

p.
.Q
.-

 

37

Table 7. N concentration and distribution in components of
cowpea 'K-2809' at 3 growth stages. (32)

 

Writing 48 67 90
gppyph_§pag§ Appaaaia Mid podfill Matupity
92mm in Letaii 3.1! i_e__f an 3.3. 3__of all
.LQQYEE
PIG-bloom 4.49 69 2.81 19 2.48 4
Post-bloom 4.51 26 2.72 8
Total green 69 45 12
Abcised 1.61 9
Total 69 45 21
Elena
Pre-bloom 1.46 23 0.76 6 0.53 2
Post-bloom 1.73 9 0.84 4
Peduncles 1.85 5 0.96 3
Total 23 20 9
Roots 0.75 2 1.18 1 0.97 1
Nodules 6.07 6 7.04 6 5.27 2
Vegetative total 100 72 33

continued on next page

38
Table 7- continued-

 

48 days 67 days 90 days
Appaaaia Mid podfill Maturity
Consensus; 3N LLAlo a M Lf__o all 3N 3__of all
Pod husks 1.40 2 1.22 4
Seeds 3.60 22 3.33 63
Young pods 4.35 4
Reproductive total 28 67
Total N content (g) 1.329 3.664 4.318
% from each stage 31 54 15

N gain mg/plant/day 28 111 28

39

The cultivar "K-2809" is normally determinate, but under
long day (13-16 hr) conditions vegetative growth continued after
flowering. Over 76% of plant dry weight at maturity (including
dead leaves) was accumulated after flowering began at 48 days.
Between first flower and mid pod-fill at 67 days, fruits (pod
husks and seeds) composed only 32% of the weight increment, but
by final harvest at 90 days they comprised 63% of the post-bloom
weight increase (32).

Between mid pod-fill and maturity, the total dry weight of
vegetative components declined by 7%. Almost all of the decline
was accounted for by the weight loss in stems and branches
produced before flowering, reflecting the mobilization of
reserves for peduncle and fruit growth. Total leaf weight (green
+ senesced) declined only 2% between 67 and 90 days, showing that
the export of nutrients from senescing leaves was nearly balanced
by new leaf growth at the youngest nodes. Total peduncle weight
increased only 18% after day 67, while their average weight
decreased by 52%, due to translocation to fruits rather than a
change in morphology. At the final harvest on day 90, seeds
comprised 48% of total dry weight and contained 63% of total
plant N (32).

"K-2809" cowpea was tested in an adverse fluctuating
temperature regime known to diminish seed yields. Plants were
either nodule-dependent or were fed 200 ppm inorganic N which
suppressed nodulation. All plants first bloomed 37 days from
sowing, matured the first pods 17 days later, and were ready for

harvest at 77 days. The C and N contents of leaves, stems, and

O...

'1

.v
..

Isl
‘4

40

roots were the highest at 49 days in non-nodulated plants, and
peaked about a week later in the nodulated plants.
Remobilization of N from leaves and stems was detected a week
earlier in the non-nodulated plants than the rest, but for both
treatments the rate of N translocation to the pods increased
rapidly until day 63, and then declined as the seeds ripened
(85).

The N and total dry weights Of cowpea seeds were negligible
for 7 days after anthesis, but increased continuously until
maturity at 19 days. The dry weight of the pod husks increased
steadily, then leveled off. Pod husk N content increased, then
declined, suggesting translocation to the growing seeds.
Nitrogen analysis showed that doubling the density from 12 to 24
plants/m2 also doubled the proportion of total plant N found in
the seeds at harvest from about 15% to about 30%, but seed yield
was not changed (19).

Cowpea plants with and without rhizobial nodules were given
different concentrations of nitrate N. Nodulated plants had
higher N concentrations above ground at flowering time, compared
to non-nodulated plants over all fertilizer levels: 1.8% vs. 1.4%
in stems and branches, and 4.4% vs. 3.8% in leaves.
Concentrations declined as the seeds grew, and all differences
were minor by harvest time. Senesced leaves averaged only 1.8%
N, regardless of nodulation. At harvest, N concentrations for
pod husks and seeds averaged 1.4% and 3.6%, respectively, again

unaltered by N nutrition. In all treatments, about 1/3 of the

1')

(I)

‘4‘

‘A

’1'!

‘5
'4

41

plant's total protein N content remained in vegetative components
at harvest, with 2/3 in the pods (118).

Moisture stress for 2 weeks altered the N distribution
pattern in cowpea. At final harvest (63 days) stressed plants
put higher proportions of dry weight and N (55% and 70%) into
fruits than did controls (45% and 64%). Even though total dry
weight was reduced by moisture stress, sometimes significantly,
cowpea seed yield was not altered by any of the treatments. Pod
number decreased by 7-19%, seeds per pod remained unchanged, and
seed weight increased by 9-15%. (141).

Nitrogen fixation after anthesis was inhibited by 76-79% in
nodule-dependent 'Vita 3' cowpea plants by flushing the root
systems daily with 100% oxygen. This induced N starvation,
resulting in significantly higher seed abortion and lower seed
size and total weight, and more efficient mobilization of N from
vegetative parts to seeds. Seed weight/plant declined by 33%
and seed N weight/plant declined by 52%, giving a lower N
concentration in seeds. However, seeds accounted for a similar
proportion of dry weight at harvest (32-33%) regardless of
treatment, but 59% of plant N was in the seeds at harvest in the
starved plants compared to 55% in the control (99).

In summary, the N and dry weight harvest indexes have been
increased in cowpea by increasing planting density or through
moisture stress, but not by adding nitrate fertilizer. Actual N

harvest index ranged from 15% to 65-70%.

42
: l . ! l N El !°

Some have concluded that breeding or agronomic practices
which prolong symbiotic fixation during reproductive growth, and
thus delay senescence, could potentially improve the yield of
field-grown cowpeas (32, 90). Photosynthate availability is the
major limiting factor on symbiotic N fixation in cultivated
legumes (20).

Maintaining leaves on the plant for as long as possible
after anthesis is crucial for high seed yields because the seeds
need both the carbohydrates they produce and the N contained in
the leaves. There is also an indirect effect of leaf removal,
since reducing the supply of photosynthate to the root nodules
hinders symbiotic N fixation, cutting off the other main source
of N (58) .

Carbohydrate availability to nodules will often limit N
fixation in bean and cowpea. In beans, climbing cultivars appear
to supply a greater proportion of their carbohydrates to nodules,
mainly at the expense of the roots. Nodule growth and N fixation
are normally reduced as photosynthate is diverted to growing pods
(52).

In mungbean (yigpa radiata) a continuous 33% defoliation
significantly reduced nodule fresh weight per plant. This
resulted in significant reductions in per plant N fixation,
nodule H evolution, and nodule O uptake (respiration).
Defoliation did not change the relative efficiency of N fixation
expressed as: 1 - (H evoluted / acetylene reduced). The

respiratory cost of fixing a unit of N was calculated to be

43
significantly higher with 33% defoliation, though the cost

declined by 60-70% from the vegetative stage to pod filling
regardless of treatment (20).

In 'K-2809' cowpea the N source (nodules or inorganic
fertilizer) made no difference in the reduction of dry weight of
the total plant (by 50%) and the seeds (by 48%) due to
waterlogging during vegetative growth. Waterlogging after
flowering was less damaging to seed yield. All waterlogging
treatments delayed leaf senescence in the nodule-dependent
plants, but senescence was unaffected in the non-nodulated group.
Nitrogen concentration in green leaves measured much lower in
waterlogged plants (3.2%) than in the control (3.7%), but dead
leaves were very similar (2.1-2.2%). Nitrogen source differences
did not affect the N content of green or dead leaves (86).

Various authorities agree that yield of the symbiotic
combination cowpea cv. 'K-2809' and Rhizobium 'CB 756' can not be
increased by higher levels of inorganic N (32, 86, 118).
Vegetative growth, except for roots, was markedly stimulated by
added nitrate, but the number of branches was unaffected, and
each peduncle produced fewer open flowers and mature pods than on
the smaller nodule-dependent plants (86). Thus, other strategies

must be used to raise production.

Qhamicai peactions of N in the plant

Nitrate reductase activity in cowpea roots is very much
lower than in leaves (86). In 'K-2809' cowpea plants, nitrate
reductase activity in roots was low and declined throughout

growth, amounting to only 9% of the total for leaves. In the

 

44

stems, nitrate reductase activity was detected only in the
extreme apical portions. Though it remained relatively constant,
it was also minor compared to leaves. Leaf activities were very
high in young plants, but plunged after 21 days to about 25% of
the former level around flowering time (35-40 days). Foliar
activities later recovered to about 50% of the maximum, before
suddenly declining during senescence (85).

Evidence suggests intense metabolic conversion of root-
derived N (nitrates and ureides) to amino acids by vegetative
plant parts before translocation to pods. Severe defoliation
might hinder the metabolic conversion through a shortage of sites

for it to take place (99).

W

According to genotype, cowpea fruits may be borne above or
below the leaf canopy. Those that are shaded by the canopy can
remain 4 C cooler than fruits exposed to direct sun (116).
Because many processes are influenced by temperature, the sudden
change in sunlight penetration through the canopy due to
defoliation may be cause for concern. Within normal temperature
ranges, however, the minimum (night) temperature, which is
probably not affected by shading, appears more important than
maximum (day) temperature. Too low night temperatures result in
delayed growth, flowering, and pod development, while excessively
high night temperatures 5-7 days before anthesis will cause
pollen sterility and flower abortion (73, 129).

In cowpea, translocation of nutrients out of leaves, and

leaf senescence and death begin earlier at higher temperatures

45
(71). The timing of the start of pod filling depends on mean

temperature, rather than on time from first flower. The rate and
duration of pod-filling were not related to temperature, though
the duration of filling of individual pods at particular nodes
was found related to temperature. Pods tended to fill
simultaneously, rather than sequentially, and pod weight was a
relatively constant proportion of total dry weight, unaffected by

temperature variations (73).

W

In cowpeas, the number of leaves, peduncles, and pods are
related to the number of nodes on the plant, and all vary
directly with total plant weight. Within a single genotype, seed
yield is strongly correlated with number of mature pods, while
number of seeds per pod and weight of individual seeds seldom
contribute to yield variations. The effects on yield from
differing locations, weather conditions, soil factors, crop
mixtures, pest attacks, and dependency on rhizobial or inorganic
N supply have all been strongly correlated with pod number (116).

In extensive testing, pod weight was shown to be a
relatively constant proportion of total dry weight, unaffected by
temperature variations. Final pod number was reached after
considerable pod abortion in most trials, and was proportional to
total plant dry weight. Seed weight per pod did not change (73).

Components of yield were analyzed for 5 cowpea cultivars
with rhizobial inoculation and nitrate fertilization (100 kg/ha
CaNO3). Among the 3 major seed yield components (pods/plant,

seeds/pod, weight/seed), only pods/plant was influenced by all

46

treatments. Regression analysis showed that pods/plant accounted
for the greatest proportion of the variability in seed yield
among inoculated plants of the same cultivar. This component has
low heritability and is greatly influenced by individual plant
physiology and morphology and environmental factors, so is highly
unstable. There were no differences observed in seeds/pod or
weight/seed, which were quite stable and cultivar specific.

These components are known to have high heritability in COWpea
(42).

In the cowpea cultivars 'Adzuki', 'New Era', and 'Mala' in
the greenhouse and the field, 70-87% of flower buds abscised
before anthesis, and 45-64% of fruits set aborted before
maturity. The proportion of ovules producing seeds in each fruit
was inversely related to the proportion of flowers producing
fruits. This negative correlation between seeds/pod and
pods/plant was not always significant. Mean weight/seed has also
been found to decline as both seed number/pod and pod
number/plant increased in various cowpea cultivars (93). Because
environmental conditions were not mentioned, it may be that the

negative correlations were due to stress, rather than genetics.

Efifagpa pf appasg 9n sead growth and components of yield

It remains to be seen exactly what causes flower and pod
abortion in bean (11), yigpa radiata (21, 22), and other annual
legumes. Failure of some blossoms to produce seeds may be due to
a lack of photosynthate, a lack of other nutrients, hormonal
influences, or a combination of these factors. Much evidence

suggests that photosynthate supply determines pod number, either

47
directly, or indirectly through the roots. The root system is a

strong sink for photosynthate, especially during the flowering
and pod setting stages. It is possible that seed growth in the
earliest pods diverts assimilate from the root system,
diminishing hormone synthesis there, and that a shortage of the
necessary hormones (such as cytokinnins or gibberellins) curtails
pod growth (11).

The growth rate, duration of growth, and final weight of
soybean seeds were influenced by variation in the assimilate
supply at a critical time. This was at the linear stage of seed
development during the pod-filling stage, when existing cells
expand at a constant rate. However, rather large variations in
source-sink ratios were required to disrupt seed growth (35).

This suggests that, for legume crops, assimilate shortage
for an individual seed is not normally a constraint in the field.
Plants tend to maintain a relatively constant source-sink ratio
by aborting or abscising flowers, fruits, and immature seeds in
response to reduced assimilate supplies. Also, changes in
assimilate levels are buffered by the storage capacity in
vegetative plant parts. Both processes work to minimize
fluctuations in the growth and final weight of seeds (35).

Bean seed yield involves negative correlations between pod
number, seeds per pod, and seed size, so that one increases at
the expense of the others. To adjust to a limited nutrient
supply, reproductive growth is regulated according to a
hierarchy. The most recently formed pods abscise first, then

unopened flowers. This is followed by abortion of some ovules in

48
the next older pods, until the pods with expanding seeds are

reached. Next the open flowers and finally flower buds abscise,
preserving the pods with the most resources already invested in
seeds. In operation, this hierarchy means that the components of
yield are allowed to vary at different times. Pod number is
adjusted at the start of the reproductive phase, but after all
buds have bloomed or aborted, it can not increase. At the end of
the reproductive stage, only seed weight can be changed. In
between and overlapping the other components, seed number is
adjusted according to nutrient availability. Fertilized ovules
may abort at any time until near maturity (2).

Seed yield and pod number were found closely related
(r=0.98), as were seed yield and 100 seed weight (r=0.72), after
repeatedly defoliating bean plants. Reduced leaf area was blamed
for a shortage of carbohydrates which increased pod abortion and
decreased seed weight. Successive defoliations appear to disrupt
individual seed growth and final weight more than single
defoliations (33).

When various defoliation and decapitation treatments were
performed on 2 bean cultivars before anthesis, yield on
inflorescences at the second mainstem node was mainly determined
by pod number (10).

A fruiting cowpea plant responds to a shortage of
metabolites (carbohydrates and/or N) by abscising flowers to keep
the source-sink ratio from worsening, by abscising immature pods
to restore a balance, and if that fails, by aborting seeds in the

remaining pods (98).

49
Although moisture stress reduced total dry weight in cowpea,

seed yield was not significantly altered. Pod number decreased
significantly by 7-19%, seeds per pod remained unchanged, and
seed weight increased significantly by 9-15% (141).

Waterlogging stress during the vegetative stage results in a
smaller plant, so the reproductive load would be proportional to
the number of nodes (86).

In summary, stresses on annual legume plants may affect
different components of yield depending on the stage of
reproductive growth. Since flowers and immature fruits are
aborted first under stress, most observed yield reductions are
due primarily to fewer pods. If the stress continues to restrict
the nutrient supply to growing pods, some ovules will abort.
Ultimately, the weight of individual seeds may be reduced, but
this is rare. Weight per seed may increase if the plant recovers

after all remaining flowers have aborted.

'n O e i tio

Defoliation beyond some threshold level reduces total
photosynthesis, allowing fewer pods and seeds to form. With all
leaves removed, a plant must depend on the limited photosynthetic
surface of green stems, pods, and petioles to regenerate new
leaves and to produce seeds (98). The threshold for damage
depends on the developmental stage of the plant.

Complete leaf removal at 7, 14, 21, and 35 days after first
flowering reduced seed yield in S-5269 COWpea by 57%, 50%, 46%,
and a non-significant 3%, respectively. Pandey gave considerable

additional evidence that seed yield in COWpea and many other

50

grain legumes depended mainly on photosynthesis after flowering.
Defoliation later in the reproductive phase reduced yield less
and less: removing leaves that have already exported the bulk of
their substance depressed yield slightly (98).

Defoliation of bean plants was most damaging to yield during
anthesis and podsetting in 'McCaslan 42' (133) and 'S-182-N'
(130). Removing leaves 10-20 days after anthesis (during pod-
filling) gave the lowest yield in 'Carioca'. Treatment later
when the seeds were large, or earlier before blooming, had less
effect (130).

Removing 1 of the 2 primary leaves at the seedling stage did
not significantly influence the yield of pods in 'McCaslan 42'
pole bean. Removing both primary leaves (100% defoliation)
reduced yield by 59% and 71% in separate plantings. The loss of
all trifoliate leaves when the plants were 1 week old did not
influence yield, but yield was reduced significantly by various
levels of defoliation 3, 5, and 7 weeks after emergence. At 5
weeks, during anthesis and podsetting, the loss was the most
severe. Continuous weekly removal of 50% of the leaves (in
relation to control plants) reduced pod yield by approximately
40% in the fall and 28% in the spring (133).

By removing a portion of each leaf with scissors, 2
cultivars of beans were defoliated at 4 growth stages at 5 levels
of severity. The determinate cultivar 'ICA-Guali' has large
leaves, while the semi-indeterminate cultivar 'Porrillo
sintetico' has medium-sized leaves. The timing of defoliation

was at the stages of 3 trifoliate leaves (15 days from planting),

51
first flower (30 days), the start of pod-filling (45 days), and

maturity of the seeds (60 days). The seeds reached full
physiological maturity after 75 days. The semi-indeterminate
'Porrillo sintetico' cultivar showed a greater ability to
regenerate lost leaf area, thus reducing yield loss, with 20 to
60% defoliation. In both cultivars, regrowth was vigorous before
flowering, intermediate after flowering, and very slight after
pod-filling began (table 8) (47).

It was shown that 'Harvester' and 'Early Harvester'
snapbeans could lose 66% of their leaflets prior to anthesis with
no significant reduction in the harvest of green pods (53). At
flowering time, over 33% of the leaflets could be removed without
reducing green pod yield (54). Based on these results, it was
recommended that snapbeans be sprayed with insecticide only if
pests consume over 20% of the leaf area before blooming, or 10%
after (134).

Two determinate bean cultivars were defoliated at 3
intensities at 20 days (vegetative), 30 days (anthesis) and 40
days (pod formation). In 'Manteigao 977', 33% defoliation
lowered yield by 19% at 40 days, 66% defoliation lowered yield by
35% at 30 and 40 days, and removing all leaves reduced yield by
71-85% at each growth stage. In 'Manteigao Fosco II', only 100%
defoliation gave large yield reductions, of 78-85% at each growth
stage. 'Manteigao 977' requires about 70 days to mature, while
'Manteigao Fosco II' has very large leaflets and needs about 80
days. The difference in response between the 2 cultivars was

attributed to the larger leaflets of 'Manteigao Fosco II', which

52

Table 8. Seed yield as percent of control (intact) in 'Ica-
guali' and 'Porrillo sintetico' beans with various
levels of defoliation (47).

W 2.03 .493 9.01 as 113999

Isa-£19911
Growth stage
3 leaves 80 80 75 73 51
First flower 84 76 72 61 41
Pod-filling 86 72 67 52 23
Mature seed 99 95 88 92 78

Growth stage

3 leaves 94 88 76 71 66
First flower 88 80 72 62 40
Pod-filling 87 82 72 52 27
Mature seed 95 92 90 87 82

No indication of the statistical significance of these
figures was given, except for the comment that the yield
reduction of "around 20%" Observed for both cultivars at the
mature seed stage with 100% defoliation was not significant.
More severe reductions may have been significant.

53

permit less light to reach the lower leaves. Assuming that the
lower leaves were not functional, their loss would have less
effect (18).

Bean cultivars 'Jalo EEP 558' and 'Rico-23' were defoliated
by 33% and 66% at 30, 45, and 60 days from planting using
planting densities of 100,000 and 300,000 plants/ha. 'Jalo EEP
558' had fewer and larger leaves compared to 'Rico-23', and
showed a greater reduction in yield from defoliation at 45 days
because the seeds were almost mature at 60 days. On the average,
33% and 66% leaf removal gave 83% and 79% of the control seed
yield, respectively, in 'Jalo EEP 558', and 95% and 93%
respectively, in 'Rico-23', where the reduction in yield was
linear with advancing age. At 100,000 plants/ha, 33% and 66%
defoliations averaged 94% and 88% of the control seed yield,
respectively. At 300,000 plants/ha, 33% and 66% defoliations
both gave 87% of control (104).

In 4 legume species defoliated by 50% and 100% at various
growth stages, the greatest reductions in seed yield occurred
during the early podding stage. At this stage in cowpea, peanut,
soybean, and green gram (yigpa aureus), 50% defoliation resulted
in yields 86%, 57%, 58%, and 54% of control, respectively.
Complete leaf removal gave yields 21%, 40%, 14%, and 5%, of the
control, respectively. With yields reduced only 3-25% depending
on growth stage, COWpea 'Iran gray' tolerated the loss of half
its leaf area much better than the other species. This was
attributed to better light penetration into the canopy. Total

defoliation in the vegetative stage or at the start of flowering

54
gave 33% and 55% of the control seed yield, respectively, and

increased the number and weight of pods per peduncle. Complete
defoliation during early and late podfilling, and near maturity
gave yields of 21%, 40%, and 64%, respectively, showing that
damage decreases as the seeds mature (37).

In a series of defoliation experiments with urd bean (yigpa
pppgp), complete defoliation during the reproductive period at 7,
14, and 35 days after anthesis reduced seed yield to 21%, 10% and

64% of control yield, respectively (98).

W

In urd bean, partial defoliations of 33%, 67%, and "basal
leaves removed" (which was not described) gave seed yields of
74%, 41%, and 95% of control, respectively, when performed weekly
from 15 to 55 days after sowing (98).

Removing 60% of the leaf surface from the upper or lower
portions of bean plants decreased yield by 20% and 6%,
respectively, during vegetative growth, and reduced yield by 27%
and 13%, respectively, during pod formation (40 DAP). There was
no difference between the treatments at 60 days, but the effect
on yield was not reported (47).

Determinate '600/1' beans were subjected to 3 levels of
defoliation at 3 fertilizer levels over 2 seasons. The 3 top
fully expanded leaves were removed by pinching the petiole by
hand, starting at 21 days. Leaves were removed 0, 1, 2, and 3
times at weekly intervals. The last defoliation at 35 days
corresponded with full bloom. New leaves formed after the first

and second defoliations were smaller. Average leaf areas of the

55

3 leaves from the second and third defoliations were 63% and 25%
of the first defoliation. No new leaves grew after the third
defoliation. Remaining leaves on the defoliated plants were much
darker green than on the control plants. In 1974 heavy rains
promoted a higher incidence of halo blight in the defoliated
treatments, which did not occur in the drier growing season of
1975. Averaged over N levels, removing 3 leaves 1, 2, or 3 times
reduced yields by 8%, 50%, and 63% in 1974 and by 3%, 10%, andst12vp10
43% in 1975, respectively. Reduced leaf area was blamed for a
shortage of carbohydrates which increased pod abortion and
decreased individual seed weight, lowering seed yield (33).

Other researchers pinched all buds from 'K-2809' cowpea
plants with 6 trifoliate leaves, the youngest 2 not fully
expanded, and removed various leaves. Removing the 4 Oldest
leaves, which reduced leaf area by 86%, reduced subsequent weight
gain by only 45%, and expansion of new leaf area by only 3%.
Selective removal of young leaves had a much more drastic effect.
Taking off the 2 youngest leaves decreased the new leaf area by
fully 90%. Removing only the youngest leaf reduced new leaf area
by just 34% because the second youngest leaf was able to
compensate. Weight gain was also seriously affected. Removing
the 1, 2, 3, or 4 youngest leaves, which reduced the existing
leaf area by only 10%, 17%, 39% and 59% respectively, decreased
the weight increment compared to the control by 5%, 45%, 54%, and
81%. The loss of the 2 youngest leaves affected plant weight as
much as the loss of the 4 oldest leaves, although the loss of

leaf area was 17% compared with 86%. Fully expanded leaves which

56

were only 2-3 weeks old made very little contribution to growth,
implying that the photosynthetic capacity of the cowpea leaf is
rather short-lived. They hypothesized that leaves would endure
less time in the field with the added stresses from insects, wind
damage, and tropical heat. They further speculated that under
certain conditions removing old leaves could increase growth and
yield by improving the distribution of endogenous hormones (60).

Defoliating 'K-2809' cowpeas at first flower by removing the
oldest 80% of trifoliate leaves gave the same yield reduction
(60%) as taking Off the youngest 50% of the leaves. On the
average, 50% defoliation was twice as deleterious to yield when
biased to young leaves rather than old leaves (115).

Removing cowpea leaves or parts of leaves that have been
fully expanded for about 2 weeks was seen as unlikely to
significantly affect growth or yield. Losing parts of young
leaves would likewise have no major effect since compensating
expansion would occur. However, selectively removing several
young leaves from a single shoot would drastically reduce future
growth, impairing the seed production by stunting the plant (60).

In 'Mezed' COWpea, removing all young leaves produced after
flowering lowered yield by 16%. Taking off the 4 or 5 oldest
leaves (which subtended branches) raised yield by 22%. The
presence of old leaves after flowering appeared to depress yield,
since their removal raised yield. The net contribution of such
leaves may be minimal or negative (39).

Leaves of annuals show declining photosynthetic rates with

age, so the oldest ones would contribute little as they continued

57

to respire carbohydrates. Leaves in warm climates reach maturity
and senesce more rapidly than in cooler environments. Finally,
cowpeas typically have a dense canopy which shades the lower
older leaves, further retarding photosynthesis (39).

In 3-week-old bean plants, it made no difference if the
upper or lower leaves were selectively removed. Later in the
season, yields were reduced less when Older lower leaves were
removed than when younger leaves were taken (133).

Photosynthetic efficiency of different zones of the bean
plant was analyzed by selectively defoliating zones A (top 1/3,
youngest), B (middle 1/3), and C (bottom 1/3, Oldest) in all
possible combinations (table 9). Cultivar '373' was grown in
Malawi and treated by pinching off the petiole by hand, at 21 and
at 35 days from planting (34).

Zone B was the most efficient single zone and zone C was the
least efficient. This was attributed to age and the fact that
zone C had the least light. Plants treated at 35 days produced
new leaves at a slower rate and had much less time to recover
before leaf growth ceased (34).

Defoliations at 21 days removed fewer leaves than at 35
days, and the height and number of leaves would have increased
proportionally more after the early treatment. All defoliations
at 21 days removed leaves that probably would have been located
in zone C at maturity. This would give an added advantage by
reducing the number of parasitic leaves in zone C. The only
treatment at 35 days which surpassed the effect at 21 days was

the defoliation of zone C alone, which left fewer leaves in zone

58

Table 9. Seed yields expressed as percentage of the control
resulting from defoliation of zones A (upper 1/3), B
(middle 1/3), and C (lower 1/3) in bean cv. '373' at 2
growth stages (34).

 

 

Zonea, ' d s e cent e O cont 01
W latest we ma 5 ean
A BC 115% 89% 102%

AC 103% 78% 91%

C AB 109% 116% 113%

AB C 91% 50% 70%

AC B 125% 101% 114%

BC A 95% 68% 81%

ABC None 54% 30% 42%
Mean 99% 76% --—-

59

C at maturity than the earlier treatment. The only combinations
that augmented yield at both dates (and gave the highest average
increases) involved removing zone C and leaving the most
efficient zone B intact (table 9) (34).

All researchers found the lowest (oldest) leaves to be the
least valuable to the plant. Actual increases in yield following
removal of the oldest leaves have been reported (34, 39). When
old leaves lose so much of their photosynthetic ability or become
so shaded that the amount of carbohydrate they produce is less
than that needed to keep them alive, they become parasitic.
Apparently, the only functions they provide are the storage of N
for future translocation to the seeds, and acting as spare leaves
in case something destroys the upper foliage. When the top of
the plant is removed, remaining old leaves can rejuvenate and
return to peak production (107, 128, 132). If the hormones
responsible for rejuvenating old leaves act to stimulate
photosynthesis in leaves generally, then per leaf production Of
carbohydrates should increase after partial defoliation,
regardless of age. Under normal conditions, it appears that
removing the correct number of old leaves at the right time will

increase yield.

Effagpa pf piapt size on seed yield

Larger plants, whether influenced by growing conditions or
heredity, should have a higher threshold level for damage before
seed yield decreases since they have greater reserves and leaf
areas. For example, indeterminate bean cultivars may tolerate

more defoliation than determinate cultivars. The same leaf

60

removal treatment (not described) lowered yield in a determinate
bean by 22% and 41% when done once or twice. In an indeterminate
cultivar, defoliation 1, 2, or 3 times reduced yield by 12%, 20%,
and 34%, respectively (33). However, after comparing yield
losses from identical treatments in 2 determinate and 2
indeterminate bean cultivars, it was concluded that defoliation
tolerance bears no relation to growth habit. Instead, leaf size,
number, and orientation appeared to determine the level of
tolerance to defoliation (130).

In 'Michelite' beans the correlation between seed weight and
"straw" weight was highly significant in 14 of 18 trials. The
components of "straw" were not indicated, nor how dry it was when
weighed, but these results support the view that seed yield is
related to plant size (25).

Bean yields are often limited, especially in determinate
cultivars, by too few nodes on the plant, and poor leaf area
development and duration. These attributes result from early
flowering and curtailment of vegetative growth (51). In COWpeas,
the number of leaves, peduncles, and pods are related to the
number of nodes on the plant, which all vary directly with total
plant weight (116).

In 20 determinate cowpea accessions, growth rate and size
before flowering had little effect on seed yield. Seed yield per
pod, and harvest index, did not vary with plant size and varied
little within genotypes. Thus, seed yield is related to above-

ground biomass at maturity, but not during vegetative stages.

61

This explains the much lower sensitivity to defoliation in

younger plants (63).

w n e o ' t'o

Compensatory leaf expansion has been observed in defoliated
bean plants. All leaves above the first trifoliate were removed
as they emerged, so that only 1 leaf was allowed to expand. The
control plants were not defoliated. All measurements were taken
on the center leaflet of the first trifoliate. Leaf area
expanded throughout the sampling period giving a final area about
50% greater than the control. The dry weight increase was
continuous but even more dramatic, so that the weight per unit of
area was about 2/3 greater. At the end of the experiment (45
days) the leaflets' mean dry weights were 150% above the
control's. Defoliation also increased the photosynthetic rate in
this experiment (5).

Defoliated yigpa mppgp plants compensated for foliar loss by
growing new leaves, increasing areas of remaining leaflets, and
increasing photosynthetic rates in remaining leaves (98).

Pruning the side shoots from GOWpea plants before they
produced leaves increased leaf size and number on the main stem.
Total yield decreased slightly, but greater reproductive
efficiency resulted (38).

When 'Harzgruss' bean seedlings were decapitated by removing
all buds above the primary leaves 15 days after sowing, the
primary leaf area and chlorophyll concentration continued

increasing until 24 days, 15 days longer than in the control

(110).

62
W

The effect of defoliation on seed yield in annual legumes
such as bean and cowpea depends on the growth stage, and the
quantity and age of the leaves. Minor defoliation does not
cause significant yield reduction, but the threshold level is
lower after anthesis. Lost leaf area can be replaced during
vegetative growth by expansion of both remaining and new leaves,
but after flowering the necessary resources are diverted to
reproduction, preventing compensatory growth. Increased rates of
photosynthesis in leaves remaining after defoliation result in
compensation even without restored leaf area.

Continued production of carbohydrates during the
reproductive stage is vital for high seed yields, to feed the
symbiotic N-fixing bacteria in the root nodules and also to fill
the seeds. Leaf removal while pods and seeds are forming is the
most damaging to yield, but as the seeds approach maturity they
become less vulnerable. Stored carbohydrates and N are exported
from stems and leaves to the growing seeds, so defoliation late
in the reproductive stage is relatively harmless; the nutrients
are already in the seeds and the leaves will soon abscise.

When nutrients are insufficient for all the potential seeds
to mature, the source-sink ratio is adjusted by aborting flowers,
young pods, and young seeds. Individual seed weight is reduced
only if these actions fail to restore a balance.

High planting densities and/or moisture stress will reduce

plant size but increase harvest index. Adding N fertilizer to

63

nodulated cowpea plants can increase size but not harvest index,
and may even promote vegetative growth at the expense of seeds.

Larger plant size, which increases storage capacity and leaf
area, generally increases seed yield. However, not all leaves
make a positive contribution to yield. Photosynthetic capacity
declines with age and shading reduces illumination so that lower
leaves become parasitic. Removing the oldest leaves can increase
seed yield by reducing carbohydrate losses to respiration, and
possibly also by improving efficiency of fuctional leaves through
increased hormone supplies. Endogenous hormones, possibly
cytokinins produced in the roots, may regulate photosynthesis and
senescence in bean and cowpea leaves. Leaf functions apparently
depend on the total hormone supply per leaf, since both
defoliation and applied hormones rejuvenate old leaves.

Additional leaves beyond a leaf area index of about 3 are
surplus in cowpea. Their cost in carbohydrates and hormones may
exceed their value as storage organs. Under drought conditions
they promote wilting and hinder photosynthesis until shedding
restores a balance between uptake and transpiration. If excess
leaves were harvested the moisture stress could be relieved more
rapidly.

To maximize the harvest of edible bean or cowpea leaves
without significantly reducing the yield of seeds, defoliation
must be performed early and lightly enough (generally 33% or
less) for the plant to recover before vegetative growth ceases.
The older lower leaves should be selected because they will be

shaded later and will contribute less to yield. Although a

64

greater number of Old inactive leaves could be picked later in
the life cycle, which would probably increase the seed yield as

well, such leaves would be too fibrous for eating.

MATERIALS AND METHODS

W

'Vita 7' is a hybrid from Nigeria, and has narrow leaves
typical of many traditional cultivars in Nigeria, where leaves
are not eaten except in the far north. Leaves are commonly eaten
in Botswana, where cultivars generally have wide leaves. The 2
traditional Botswanan cultivars tested, 'B-138' and 'B-162', have
unusually large leaves, which speed the harvest of single leaves
by hand. An improved cultivar from Nigeria with equally large
leaves, 'TVu-1948', was also tested. The other 2 improved
Nigerian selections, 'TVu-3662' and 'Vita 5', have small leaves
on smaller plants and a more compact habit than the others, which
are trailing types. All the Nigerian types were bred at the
International Institute for Tropical Agriculture, while the
Botswanan cultivars were collected from peasant farmers for a

germplasm project assisted by Colorado State University.

Expepimeppal Qesign

The treatments were applied as a 6 x 2 x 4 factorial in a
split plot design with 6 blocks. Cultivars were the main plots;
apex removal at 28 days and leaf harvest method were the
subplots. The experimental unit was 1 plant. Leaf harvest
methods were: the control, with no leaves removed; multiple
harvest with weekly removal of 1 or 2 leaves from each stem (the
3rd and 4th fully expanded leaves counted from the apex) starting

at 35 days and continuing for 4-6 weeks depending on the
65

66

cultivar; single harvest with removal of all leaves from all
stems down to the 3rd or if possible 4th fully expanded leaf at
or soon after anthesis for each cultivar (46-60 days after
planting); pruning, which was identical to the single harvest
except that it also removed the stem just below the lowest leaf
picked, to represent a more rapid harvesting process. Leaves on
stems with only 1 or 2 fully expanded leaves were not harvested
under any treatment. The multiple harvest procedure was based on

the method reported by Mehta (82).

W

The experiment was conducted in a greenhouse at Michigan
State University. Day temperatures were usually 28-38°C and
night temperatures 18-22°C, with an extreme range of 13-48°C.
Seeds were sown May 31, 1985, in Bacto potting mix in flats with
compartments 2.5 by 5.0 cm. Seedlings emerged in 3 days after
planting. At the 2-3 trifoliate leaf stage (21 days), seedlings
were transplanted to 25 cm diameter 8 liter black plastic pots.
The media was 60% peat moss, 20% vermiculite, and 20% sterilized
sand.

Before transplanting, the seedlings grew under the natural
photoperiod of 15 hr. After transplanting, day length was
limited to 13 hr by shading with black cloth.

Each plant was allowed .18 m2 of space, or 5.7 plants/m2.
Trailing branches that grew over other plants or beyond the edge

of the greenhouse bench were tied to vertical stakes.

67

The cowpeas were not inoculated with rhizobial culture, but
were uniformly given 20-20-20 and 20-5-30 N-P-K fertilizer. The
total amounts per plant were 1.80 g N, 0.46 g P, and 1.91 g K.
These amounts are equivalent to 102 kg/ha N, 26 kg/ha P, and 108
kg/ha K.

The plants were treated with Malathion, nicotine smoke,
Orthene, and Pentac, about every 2 weeks to control thrips and
two-spotted spider mites.

Least significant differences at 1% and 5% levels were
calculated with the analysis of variance in the "Factor"

subroutine of the MSTAT microcomputer software program.

Leaf_msa§urements

Changes in leaf area were calculated weekly by recording
length and width for each leaflet on each trifoliate on each of
the 48 plants in block 1. The areas of all leaves in block 1
were measured on a LI-COR leaf area meter, either when harvested,
upon abscission, or at the end of the season when the plants were
dismembered for weighing.

The product of the length and maximum width measurements of
the center leaflet (LxW) of the trifoliate COWpea leaf is a
useful indicator of leaf area (95). The LxW products were added
together for each plant, and the total was divided by the leaf
area measured on the LI-COR meter for those same trifoliates.
This gave a ratio (generally between 0.4 and 0.6) of LxW to
actual area, which was unique for each plant. Then LxW for each
leaf was divided by this ratio to give an estimate of each

trifoliate leaf's area. From the weekly measurements the number

68

and estimated areas of leaves on each plant could be calculated
for all leaf harvest dates.

The actual leaf area recorded from each harvest was divided
by the estimated area of the mature leaves on the plant at the
time, giving a defoliation percentage (DP)(table 10). With
multiple harvests, each plant had 4-6 separate DPs, plus a
cumulative DP (representing the proportion of leaf area removed
in all harvests, not counting leaves produced after the last
harvest). In cases where the single and pruning harvests were
conducted 3 or more days after the newest mature leaves were last
measured, the estimated areas for the leaves measured the next
week (if any) were assumed to have been present at harvest time,
and were included in the calculations. Leaflet length and width
measurements were always recorded the same or next day after the
multiple harvests, providing more accurate estimates for this
treatment.

Because of the variability of the small samples, DP averages
for all 6 blocks were estimated (table 11). Harvested leaf areas
from the other blocks were assigned DPs according to the
proportions in block 1, and the 6 DPs were averaged. It was
assumed that all plants of the same cultivar and harvest
treatment had the same total leaf area at the same time. Because
single harvest and pruning treatments used the same procedure to
select leaves for harvest, at the same time, their estimated

average DPs were combined.

69
1191911189

Leaf and stem samples were sealed in plastic bags and
refrigerated upon harvest to avoid weight losses. After fresh
weight and leaf area were measured, samples were placed in a
Table 10. Defoliation percentages for each of the multiple leaf

harvests, and cumulative total defoliation percentages

for multiple, single and pruning leaf harvest methods,
as measured for each Of the plants in block 1.

flaxyagp ppppa; Qumpiative tptals
sultixar.Anex.1 .2 :1 A. 5 .6 MELLL§1£QL§ EZBLQQ
8138 + 26 12 31 45 64 86 66
8138 - 20 9 45 o 45 92 95
8162 + 38 31 38 37 3o 76 58 53
8162 - 18 13 42 44 36 7o 59 72
TVu3662 + 19 32 39 56 79 61 6O
TVu3662 - 10 3o 34 50 73 61 64
TVu1948 + 29 34 15 47 53 41 85 66 64
TVu1948 - 28 21 50 32 49 54 88 -- --
Vita 5 + 15 19 49 34 77 81 71
Vita 5 - 9 3o 58 46 78 73 78
Vita 7 + 24 22 33 4o 20 20 72 55 45
Vlta 7 - 19 22 3o 35 31 16 77 66 40

+ = apex intact, - = apex removed at 28 days.

70

Table 11. Estimated defoliation percentages for leaf harvest
methods, averaged from all blocks.

names; number W _tot._a.. ls

291121181: 1 2 3 4 5 e Mist—22
8138 21 14 29 29 56 85
8162 29 22 32 37 35 69 60
TVu3662 16 29 36 46 71 60
TVu1948 30 20 44 35 37 45 80 70
Vita 5 14 19 39 51 73 75
Vita 7 22 27 27 33 27 25 77 50

2

Single and pruning harvest methods had the same procedure
for selecting the leaves to be harvested, so they have the
same estimated average defoliation percentages for each
cultivar.

4,

oh

2‘
\a.

drying oven at 60°C for at least 4 days until constant weight was
reached. At the end of the season when all pods on a plant were
dry, the seeds were removed. The husks, abscised parts, and all
remaining vegetative parts were then oven-dried. In order to
maintain viability, the seed was air-dried. Based on samples
that had been pulverized and oven-dried, the dry weight Of the
seeds was calculated at 90% of the air-dry weight. A factor of
56.5 was used to convert g/plant weights to kg/ha.

Seed weight (yield) and pod number per plant were measured
for all 6 blocks. Weight per seed was determined for 4 blocks by

counting and weighing 50 seeds from each plant.

Magnum

Chipmunks ate some seeds and pods at maturity. To estimate
the weights of the missing seeds and pod husks, the average
weights of seeds and husks were calculated for intact plants in
each of the 24 cultivar by harvest method combinations. The
numbers of missing pods on the plants were multiplied by these
averages, and the products were added to the weights of the

remaining pod husks and seeds.

RESULTS

Harvest index for leaves. seeds. and total edible products
Harvest index as discussed here is the percentage of the

plant's above-ground dry weight (not including the parts that

abscised before harvest) that was harvested as edible. Although

the stems removed in the pruning treatment and the harvested

71

72

leaves were absent at the end of the experiment, they are
included in the plant's total weight. The harvest index does not
indicate that the remainder of the plant is useless or is never
harvested for fodder, but only that it is not harvested for human
food.

For edible leaves, harvest index values ranged from 14-30%,
and both extremes were reached by 'Vita 7' (table 12). Apex
removal generally caused no change or gave a higher harvest
index, but caused decreases in 'TVu-3662'. Multiple harvest gave
the highest efficiency of leaf production in most cultivars, but
the lowest efficiency in 'B-138'. 'TVu-3662' showed no response
to leaf harvest method.

The average harvest index for seeds was 36% for the
controls, 27% for multiple and single harvests, and 22% for
pruning. 'TVu-3662' had the highest efficiency, with 52% in the
control (table 13). 'Vita 7' had the lowest measurement, with
15% for pruning. 'Vita 5' required a longer time to mature the
pods on the new growth produced after pruning, resulting in its
improved efficiency for that treatment.

The harvest index for both products was increased by
multiple and single harvest treatments in all cultivars (table
14). The effect of pruning was either positive or negative,
depending on the cultivar. The greatest responses were seen in
'Vita 5' and the smallest in 'TVu-3662'. Cultivars with a low
harvest index in the control group showed the most improvement
from leaf harvesting. The differences between the harvest

treatments were unusually large for 'Vita 7'. 'TVu—3662'

73

Table 12. The effect of cultivar, apex removal, and leaf harvest
method on the harvest index for leaves (edible leaf
wt/total wt, x 100).

 

 

9111111229
flatbed 31.3.8. 3.1.62. 111116.63. W V TAS VITA?
Multi 15 24 23 25 20 30
Apex cut 15 24 21 27 21 28
Single 19 18 25 21 18 15
Apex cut 19 25 20 23 18 17
Pruned 14 17 23 18 18 14
Apex cut 21 20 22 21 18 16

LSD for means in the same row or column
5% 3

1% 4

74

Table 13. The effect of cultivar, apex removal, and leaf harvest
method on the harvest index for seeds (seed wt/total
wt, x 100).

Qultixars
m 3.1.3.8. m2 M3662 M1948 VITA5 VITA7
Control 35 40 51 38 23 28
Apex cut 37 48 52 33 27 27
JMulti 30 29 36 29 20 20
Apex cut 29 29 38 25 20 22
Single 27 32 33 28 21 24
Apex cut 28 29 34 22 24 23
Pruned 27 24 23 18 27 15
Apex cut 23 21 26 20 23 16

LSD for means in the same row or column
5% 5

1% 6

75

Table 14. The effect of cultivar and leaf harvest method on the
harvest index for edible products (seed wt + harvested
leaves wt/total wt, x 100).

 

Qultixars
HQEDQQ Bl§§ filfil lyflééél lyfllflié VITA5 XLIAZ
Control 36 44 51 36 25 27
Multi 44 53 59 53 41 ' 50
Single 46 51 55 47 40 4o
Pruned 42 41 47 38 44 30

LSD for means in the same row or column
5% 5

1% 6

76
had the highest harvest index in all treatments, but had the

lowest above-ground dry weight in the control.

Wm

All leaf harvesting treatments yielded more than the
control, except in TVu-3662 with single harvest and pruning, and
TVu-1948 with pruning (table 15). The dry weight of edible
products was highest with multiple harvests in all cultivars.
All leaf harvest methods lowered seed yield, but increased the
yield of edible products an average of 4% for pruning, 18% for
single harvest, and 36% for multiple harvest. In 'Vita 7',
multiple harvest almost doubled the control yield. The seed
yields in the controls ranged from 700-1220 kg/ha, while the
maximum seed and leaf yield was 1690 kg/ha (table 16).

The yield per square meter per day is the total seed and
leaf yield divided by the length of the growing season. It is an
average rate for the entire season and not the same figure for
each day. The season length was recorded for each individual
plant, being the number of days required to dry all the pods.
Multiple harvests gave the highest yield in all cultivars, with
'TVu-1948' and 'Vita 7' yielding above 2.0 g/m2/day (table 17).
Pruning gave the lowest yields in 4 cultivars, and the control
was lowest in 'Vita 5' and 'Vita 7'. The low values in the

pruning treatment were due to both reduced yields and more days.

Seed yield

All leaf harvest treatments reduced seed yield (table 18).

Multiple harvest decreased seed yield the least for 3 cultivars,

77

 

Table 15. The effect of cultivar and leaf harvest method on the
dry weight of edible products (seeds and leaves
combined) (g/plant).

Qultilare

MQLDQQ £115 .Eléz IXHQQQZ.I¥Q12£§. Eliéé ELIAZ

Control 21.1 20.9 21.6 19.8 12.4 15.6

Mnlti 25.8 26.4 22.1 28.9 18.5 30.0

Single 24.0 25.0 20.9 22.9 15.4 23.1

Pruned 21.9 22.2 19.1 19.7 16.0 18.0

LSD for means in the same row or column

5%
1%

78

 

 

Table 16. The effect of cultivar and leaf harvest method on the
dry weight of edible products (seeds and leaves
combined) (metric tons/hectare).

antigens

Mgtngfi filgfi Big; Typaaaa Iygi948 V TAS VITA7

Control 1.19 1.18 1.22 1.12 0.70 0.88

Multi 1.46 1.49 1.25 1.63 1.05 1.69

Single 1.36 1.41 1.18 1.29 0.87 1.31

Pruned 1.23 1.26 1.08 1.11 0.90 1.01

LSD for means in the same row or column
5% 0.12

1% 0.16

79

 

 

 

 

Table 17. The effect of cultivar and leaf harvest method on the
daily dry weight production rate2 of edible products
(seeds and leaves combined) (g/m2 /day).

Mm

MEL-DEG m @162 1293662 TVU1948 VITA5 VITA7

Control 1.57 1.53 1.56 1.43 0.93 1.11

Multi 1.92 1.92 1.62 2.14 1.35 2.12

Single 1.74 1.82 1.53 1.68 1.10 1.65

Pruned 1.46 1.48 1.41 1.29 1.07 1.20

LSD for means in the same row or column
5% .15

1% .20

80

Table 18. The effect of cultivar and leaf harvest method on the
estimated seed yield (g dry weight/plant).

 

 

 

Culpivars
M £1}; M; 1293662 TVU1948 $121135 VITA7
Control 21.1 20.9 21.6 19.8 12.4 15.6
Multi 17.1 14.4 13.9 14.6 9.1 12.7
Single 14.3 14.8 12.6 12.3 8.6 13.6
Pruned 12.9 12.4 10.0 9.6 9.5 9.3

LSD for means in the same row or column
5% 1.9

1% 2.5

81

and was close behind another treatment for the other 3. Apex
removal had a minor effect on seed yield on the average, but gave
some large changes in yield depending on the cultivar and harvest
treatment (table 19). Because of chipmunk damage, the averages

include some estimates.

WM

Generally, the controls had the highest numbers of peduncles
with pods (table 20). Pruning reduced seed yield due to fewer
peduncles on the stems that remained. Usually all peduncles on
the pruned plants had mature pods, while in the other treatments
all pods aborted on the younger outermost peduncles which bloomed
last.

All treatments reduced the number of pods per plant (table
21). Generally, pruning gave the greatest losses and multiple
harvest the least.

The harvest treatments generally increased seed number per
pod, except for 'TVu-l948' (table 22). Extreme increases of over
10% were associated with the largest reductions in pod number.
This inverse relationship was not Observed in 'B-138', where
seeds per pod was remarkably stable, nor in 'TVu-1948', where
both pod and seed numbers declined.

Multiple and single harvest always reduced the weight per
seed, but pruning increased it in 3 of the cultivars (table 23).
The weight per seed declined an average of 9% in response to

multiple and single harvests, and dropped 1% with pruning.

82

 

 

 

 

 

 

Table 19. The effect of cultivar and leaf harvest method on the
percent change in seed yield with apex removal
compared to apex intact.

QultiYa—rs

napaga B138 pig; TVU§§62 TVU1948 VITA5 VITA7

Control 1 21 0 -14 21 -11

Multi 2 9 7 -14 2 15

Single 3 -25 16 -22 20 -2

Pruned -16 -9 10 13 -23 10

Table 20. The effect of cultivar and leaf harvest method on the
number of pod-bearing peduncles per plant.

Qaitivaps

Methgd £138 £162 Tyyaaa; TVU1948 VITA5 VITA7

Control 12.9 12.5 15.8 10.6 9.3 10.4

Multi 11.9 10.8 12.4 10.1 8.3 9.3

Single 11.2 9.3 10.9 10.4 6.0 10.5

Pruned 6.9 6.1 8.0 6.1 6.3 4.9

LSD for means in the

5%
1%

83

Table 21. The effect of cultivar and leaf harvest method on the
number of pods per plant.

 

 

Qultixars
Hfilhgd Bllfi £152 IYHQ§§2.IEHIQA§ EllAé VITA7
Control 13.6 20.3 25.8 14.1 14.6 11.9
Multi 12.4 15.0 18.9 12.1 12.8 10.9
Single 11.6 13.1 16.4 12.8 8.9 11.3
Pruned 8.6 9.3 13.3 8.2 9.4 6.3

LSD for means in the same row or column
5% 2.0

1% 2.7

Table 22. The effect of cultivar and leaf harvest method on the
number of seeds per pod.

 

 

 

gaitivars
Mpg B118 M TVU3662 TVU1948 ELTA_5 VITA7
Control 14.2 6.9 8.7 11.7 7.8 9.1
Multi 14.8 7.1 9.0 11.6 7.0 9.4
Single 14.1 8.6 9.0 9.8 8.3 9.6
Pruned 14.3 9.5 10.1 9.9 8.6 10.1

LSD for means in the same row or column
5% 1.6

1% 2.2

84

Table 23. The effect of cultivar and leaf harvest method on the
dry weight per 100 seeds (g/plant).

 

Cpitivara
NEEDQQ B122 2152 Iflfléééz IXQLQAQ VITA5 21151
Control 10.63 15.12 9.26 11.99 11.49 14.62
Multi 9.70 13.76 8.57 10.73 11.20 12.80
Single 9.12 13.80 8.45 10.36 11.38 13.13
Pruned 10.17 15.48 8.42 12.40 11.73 14.03

LSD for means in the same row or column
5% 0.75

1% 1.04

85

w 0 nt seed ie d

The seed yield losses resulting from all 3 harvest
treatments were mainly due to a reduction in pod number. Seed
number per pod was usually the same or higher than the control,
except in 'TVu-1948'. Weight per seed usually changed little,
and always declined except in the pruning treatment. Yield
losses in 'B-138' and 'Vita 7' with multiple and single harvests
were about equally due to fewer pods and to reduced weight per
seed (table 24). The percentages of control yield shown in table
21 were based on all 6 blocks, and do not always agree with the

products of the 3 components which are the means of 4 blocks.

8 est

Except for 'Vita 5', all cultivars produced new leaves to
replace those harvested (table 25). 'Vita 5' and 'TVu-3662' both
have a distinctly determinate habit, so their growth type is a
poor indicator of suitability for leaf harvest.

Apex removal resulted in 20% higher dry weight of abscised
leaves, but it was 65% and 69% higher for 'B-138' and 'B-162',
respectively (table 26). The trend was reversed for 'Vita 5'
(16% lower). The control group had more abscised leaves than any
of the treatments, except in 'TVu-l948' with apex removed, where
the weight was 77% higher in the pruned treatment. After the
control, pruned plants dropped the most leaves, which was due to
delayed blooming and ripening of the last pods. During the
longer season, more leaves senesced. Removing the ends of the
stems in this treatment also deprived the plants of an important

reserve of mobilizable nutrients, so more had to be translocated

86

 

 

 

Table 24. The effect of cultivar on components of yield for the
controls, and on the percentage of each cultivar's
control values for the leaf harvest methods.

Qaitiyars

ggntzgl 8118 (pig; Typagaz Typi948 VI A VITA7

Yield (g/plant) 21.1 20.9 21.6 19.8 12.4 15.6

Pods/plant 13.6 20.3 25.9 14.1 14.6 11.9

Seeds/pod 14.2 6.9 8.7 11.7 7.8 9.1

100 s Wt(g/plant) 10.63 15.12 9.26 11.99 11.49 14.62

H9111 co 0

Yield 81 69 64 74 73 82

Pods/plant 91 74 73 86 87 92

Seeds/pod 104 102 103 98 90 103

100 Seed Wt 91 91 93 89 97 88

single 1_ef_228tr21

Yield 68 71 58 62 69 89

Pods/plant 85 65 64 91 61 94

Seeds/pod 99 123 103 83 106 105

100 Seed Wt 86 91 91 86 99 9O

Pruned % of control

Yield 61 59 46 48 76 6O

Pods/plant 63 46 52 58 65 52

Seeds/pod 101 136 116 84 111 111

100 Seed Wt 96 102 91 103 102 96

87

 

Table 25. The effect of cultivar and leaf harvest method on the
dry weight of all leaves (harvested, abscised, and
remaining) (g/plant).

Qultixare

fleshed £118 £192 Iflfllééz.122121§ VI A5 VIT 7

Control 12.54 10.63 9.70 13.86 15.17 16.82

Multi 17.06 17.03 12.09 19.03 16.92 23.26

Single 17.01 17.41 14.22 17.98 14.06 19.28

Pruned 17.06 18.20 14.81 20.47 13.43 21.41

LSD for means in the same row or column
5% 1.45

1% 1.92

88

Table 26. The effect of cultivar, apex removal, and leaf harvest
method on the dry weight of senesced abscised leaves

(q/plant)-

 

ti r
M91829 £118 £162 Iflfllééz 1221248 21282 YIEAZ
Control 2.62 4.79 6.59 5.80 1.85 6.80
Apex cut 3.75 8.64 6.57 3.97 1.68 7.87
Multi 0.88 1.18 2.08 0.99 0.26 1.83
Apex cut 0.80 1.60 2.14 0.72 0.40 1.64
Single 0.29 2.19 3.21 1.80 0.92 3.90
Apex cut 1.32 3.40 3.48 1.66 0.27 3.58
Pruned 1.09 2.86 1.76 4.35 1.24 4.30
Apex cut 2.16 4.98 2.09 7.04 1.25 5.25

5% 1.72

1% 2.27

89

out of leaves, also increasing senescence. The dry weight of
abscised leaves was generally least in the multiple harvest
treatment because fewer leaves remained on the plants and
proportionally fewer senesced during podfilling.

'Vita 5' dropped a very small proportion of its leaves and
remained vigorous after the pods ripened, which matches previous
observations of this cultivar in the greenhouse. In a
preliminary trial of 130 days, 'Vita 5' plants produced 3
progressively smaller crops of pods with very little senescence
of the oldest leaves. New growth after each crop was
overwhelmingly reproductive, with only a few small leaves. The
plants may have produced more crops of pods, but they had to be
disposed of because of spider mite infestation.

On the average, apex removal increased both the number
(table 27) and dry weight (table 28) of harvested leaves by 7%.
In general, multi-harvest was most productive and pruning was
least productive, with the yield directly related to the amount
of labor required for picking. The decision of which harvesting
practice to use thus depends on local economic factors. The
number and/or dry weight of harvested leaves in the multi-harvest
treatment reached a peak 2 weeks before vegetative growth ceased
in each of the cultivars. This characteristic may be useful in
scheduling farming operations, as it could predict how much
longer the leaf harvesting season will continue.

The harvested leaf area was 9% greater on the average with
apex removal, but increased 22% in 'B-l62' and declined 8% in

'TVu-3662' (table 30). Multiple harvest averaged 24% above the

90

 

Table 27. The effect of cultivar and leaf harvest method on the
mean number of leaves harvested, with means for each
of the multiple harvests.

Qultixare

HELEQQ £118 Eléz Iflfllééz IEngifi 11155 11281

Control 0 0 0 0 0 0

Apex cut 0 0 0 0 0 0

Multi 12.5 15.5 26.7 26.5 20.5 40.3

Apex cut 12.7 19.5 25.7 30.8 22.3 40.8

Single 14.3 12.2 22.5 20.5 17.5 22.3

Apex cut 16.7 14.5 19.8 20.5 19.8 26.2

Pruned 11.7 11.0 24.2 18.8 19.5 23.2

Apex cut 16.8 15.5 21.2 21.2 16.0 24.7

LSD for means in the same row or column

5% 2.7

1% 3.6

W

lst 2.0 2.0 2.2 2.0 2.1 2.0

2nd 1.4 1.6 7.3 2.8 3.8 5.2

3rd 5.2 4.8 9.2 6.4 7.8 7.7

4th 4.0 4.8 7.6 5.8 6.3 8.3

5th 0 4.3 0 6.1 0 8.9

6th 0 0 0 5.6 0 5.3

91

 

Table 28. The effect of cultivar and leaf harvest method on the
dry weight of harvested leaves (g/plant), with means
for each of the multiple harvests.

9911.11.15

MQEDQQ Bllfi Blflz Iflfléééz IXngié 11152 11251

Control 0 0 0 0 0 0

Multi 8.70 12.01 8.18 14.27 9.42 17.24

Single 9.66 10.20 8.34 10.61 6.74 9.26

Pruned 8.94 9.86 9.10 10.14 6.54 8.68

5% 1.10

1% 1.46

W

lst 0.83 0.78 0.49 0.94 0.56 0.72
2nd 1.24 1.50 1.54 1.41 1.54 1.83
3rd 3.28 3.02 3.19 2.90 3 75 3.12
4th 3.36 3.40 2.96 3.21 3.57 4.22
5th 0 3.31 0 3.48 0 4.38
6th 0 0 0 2.33 0 2.95

92

Table 29. The effect of cultivar and leaf harvest method on the
total area of harvested leaves (m /plant).
Qulfixars
Methed £115 5152 1993592 2121255. 22255 22251
Single .257 .234 .216 .326 .191 .261
Apex cut .289 .301 .206 .357 .201 .319
Pruned .220 .216 .245 .322 .202 .265
Apex cut .302 .310 .211 .361 .171 .292
Multi .221 .343 .246 .428 .217 .466
Apex cut .208 .355 .237 .470 .237 .476
LSD for means in the same row or column
5% .037
1% .049
W
.024 .029 .015 .033 .017 .025
.032 .045 .050 .047 .045 .058
.086 .105 .100 .106 .093 .103
.073 .096 .077 .099 .072 .111
0 .074 0 .093 0 .101
0 0 0 .070 0 .073

93

mean of the other 2 harvest methods, which were very close, but

was 66% higher in 'Vita 7' and 20% lower in 'B-138'.

v v t a s

Because pod and seed weights had to be reconstructed for
some plants, the averages for total above-ground plant dry weight
also include estimates. Only in 'B 162' and 'Vita 7' did a
treatment (pruning) give a higher weight than the control (table
30). The differences between treatments were less for these 2
cultivars. Beyond the general superiority of the control, there
was no clear pattern for harvest treatments. 'Vita 7' had the
greatest weight in every treatment, which resulted from a longer
period of vegetative growth. 'TVu-3662' and 'Vita 5' had the
least weight, due to their compact bushy growth habit and shorter
period of vegetative growth.

Pod husk weight was estimated for some plants because of
chipmunk damage, but was highest in the control and lowest with
pruning for all cultivars (table 31). Weight was higher for
multiple harvest than for single harvest except with 'Vita 7'.

All cultivars except 'B-138' and 'Vita 5' produced new stems
to compensate for the pruned branches. Stem and petiole weight
was higher in the control than for multiple and single harvest
treatments, because of impaired vegetative growth and/or because
of greater translocation of reserves from stems to compensate for

fewer remaining leaves (table 32).

94

 

 

 

Table 30. The effect of cultivar and leaf harvest method on the
estimated dry weight of all the above ground plant
parts (g/plant).

maritime

m m 3.1.6.2. M2. 120 1248 VITA5 VITA7

Control 62.6 54.5 49.3 60.3 52.4 64.4

Multi 59.5 51.7 40.2 55.2 46.0 62.1

Single 53.1 52.1 41.5 50.9 39.3 62.4

Pruned 53.6 58.3 42.8 57.5 38.0 64.8

LSD for means in the same row or column

5%
1%

95

Table 31. The effect of cultivar and leaf harvest method on the
estimated dry weight of pod husks (g/plant).

 

 

Mrs
W 11.2.8. 319.2 M w VITA5 VITA7
Control 7.2 5.4 4.3 6.4 3.7 5.0
Multi 6.2 4.2 3.1 5.0 2.9 4.3
Single 5.1 3.9 2.8 4.3 2.7 4.6
Pruned 4.8 3.2 2.5 3.5 2.9 3.4

LSD for means in the same row or column
5% 0.6

1% 0.8

96

 

 

Table 32. The effect of cultivar and leaf harvest method on the
dry weight of stems and petioles (g/plant).
Quipivars
MELLQQ 5115. Eli; 2222552.2¥Hl25§. 22252 VITA7
Control 21.67 17.40 13.44 20.26 20.98 27.07
Multi 19.13 15.92 10.76 16.64 16.94 21.66
Single 16.56 15.75 11.49 16.30 13.87 24.60
Pruned 18.84 24.48 15.37 23.92 12.19 30.65

5%
1%

naya pp firs; flowerl first dry pad. and to ripen and drv all
9.9.45

 

The overall average date of flowering was 50.9 days after
planting, with a range of 45-71 days. Most cultivars averaged
49.4-50.9 days, except 'Vita 7' with 55.0 days (table 33). The
harvest treatments made little difference.

The pruning treatment delayed the first dry pod by 4.2 days,
but multiple and single harvests had no effect (table 33). The
overall average first dry pod date was 69.5 days after planting,
with a range of 62-92 days.

While the other harvest treatments varied from the control by
less than 1 day, pruning delayed ripening of the last pod by 6.6
days (table 33). This was because some of the pruned plants
produced new branches, on which flowers formed. The resulting
pods ripened much later. The overall average was 79.2 days after

planting, with a range of 73-99 days.

DISCUSSION AND CONCLUSIONS

Effacps of cuitivars

The dry weights of all plant parts were affected by leaf
harvest methods and cultivar differences. Apex removal affected
harvested leaf weight, number, and harvest index, and weight of
abscised leaves. The cultivars did not react similarly to

treatments, except for apex removal.

97

98

Table 33. The effect of cultivar (averaged across treatments)
and leaf harvest method (averaged across cultivars) on
number of days between planting and anthesis, first
dry pod, and last dry pod.

r a n
M hss M M
3138 50.1 67.3 78.8
8162 50.9 69.4 79.3
TVu3662 50.5 69.6 77.2
TVu1948 49.5 70.9 79.7
Vita 5 49.4 67.4 79.3
Vita 7 55.0 72.8 80.9
Willem
5% 1.3 1.4 2.3
;% 1.7 1.9 3.1
Laaf aaryesp mepaod
Control 50.7 68.5 77.4
Multi 50.7 68.5 77.4
Single 50.3 68.5 78.1
Pruned 51.9 72.7 84.0
LSD for leaf harvest methods
5% 0.8 1.1 1.5

ii 1.1 1.4 2.0

99

Of the 6 cultivars tested, 'Vita 5' was the least responsive
to treatments. It had a shorter period of vegetative growth, did
not produce more leaf or stem material to replace what had been
removed, and did not allow many leaves to abscise. 'Vita 5' had
the lowest seed yield, seed harvest index, and edible harvest
index in the control and multiple and single harvest treatments.
It also had the lowest leaf yield with pruning and single harvest
treatments, and the lowest edible products yield in all cases.
However, this indeterminate bush type cultivar has the potential
to produce multiple crops of pods. Most 'Vita 5' plants had
flowered a second time before being cut for weighing.

'TVu-3662' had the highest seed yield in the control and the
highest harvest index in all cases, but had the greatest yield
loss due to defoliation. The 2 Botswanan cultivars, 'B-138' and
'B-162', had relatively high seed yields from all treatments.
They would probably yield more than the other cultivars at wider
spacings because their trailing vines would rapidly cover the
ground. 'TVu-3662' is smaller and more determinate, and would
probably give the highest yield at higher planting densities.
Since it had the smallest seeds of all cultivars tested, a higher
planting density would not necessarily require more total seed
weight.

The multiple harvest treatment of 'Vita 7' resulted in the
highest yield of leaves. 'TVu-l948' was second highest for leaf
yield with multiple harvests and highest with pruning and single
harvest treatments. These 2 cultivars had the longest periods of

vegetative growth, allowing 6 weekly leaf harvests, and had large

100
leaves which facilitate picking by hand. The 2 bushy cultivars,

'Vita 5' and 'TVu-3662', had the smallest leaves and the shortest
leaf production seasons, though 'TVu-3662' actually had the
highest harvest index for leaves with pruning and single
harvests.

For total yield of edible products, multiple leaf harvests
were most productive in all cultivars, and 'Vita 7' had the
highest yield, closely followed by 'TVu-1948'. 'TVu-3662' had
the highest harvest index in every treatment, but also had the
lowest average above-ground weight. Therefore, if space is not
limiting, larger cultivars like 'Vita 7' and 'TVu-1948' would
appear to be ideal. When limited by space rather than labor
availability, small cultivars like 'TVu-3662' with a high harvest
index would probably be most productive provided the seeds were
planted close together.

The origin of the cultivar does not appear useful for
predicting yield levels or efficiency for leaves or seeds.
Contrary to expectations, some Nigerian high-yield cultivars
produced more edible leaves and less weight of seeds than the
traditional cultivars from Botswana (in a greenhouse
environment). Growth habit was a poor indicator of seed yield
level and harvest index efficiency. The larger trailing
cultivars produced more weight of harvested leaves than the
smaller bushy cultivars, but a bushy cultivar sometimes had the

highest harvest index.

101

f e s

This study showed that apex removal at 28 days after
planting had little effect on seed yield, but tended to increase
leaf production and accelerated leaf senescence. All leaf
harvesting methods reduced seed yield. Lower yields were mainly
due to fewer pods, as noted by others (10, 33, 86, 97). Because
single harvest and pruning treatments were performed near the
time of first flowering, pod number was usually reduced with
little effect on other components of yield. In some cultivars
the pruning treatment greatly increased seeds/pod, indicating
that the number of pods had been reduced excessively by removing
the ends of the branches. Multiple leaf harvests were shown to
reduce weight/seed in bean (33). Yield reductions in 3 of the 4
large-leafed trailing cultivars under both multiple and single
harvest treatments were about equally due to fewer pods and
smaller seeds. Nutrients may have been limiting at both the
start and the end of reproductive growth (2, 35). In the other
large-leafed cultivar, 'B-162', and both small-leafed cultivars
the magnitude of the reduction in pod number was much greater
than the reduction in weight/seed (table 24).

All leaf harvest methods lowered seed yield, but increased
the total yield of edible products an average of 4% for pruning,
18% for single harvest, and 36% for multiple harvest. In 'Vita
7' with apex removal, multiple harvest yielded 209% of the
control.

In this study the defoliations ranged from 50-85% depending

on cultivar, and significant reductions in seed yield resulted.

102

Less severe defoliations would give higher yields of seeds and
lower yields of leaves. Although lower defoliation levels were
not tested, earlier researchers have observed that large
reductions in seed yield rarely occur with 33% defoliation, but
are usually obtained with 50-67% leaf removal (18, 34, 37, 47,
54). If a 30-50% defoliation does not reduce a cultivar's seed
production, the total yield of edible products could possibly be
greater than measured in this experiment. More testing is needed

to clarify this matter.

E ll 1 J' !'

Since the increases in food production were related to the
amount of labor required, recommendations for farmers would
depend on local economic factors, such as labor availability and
market prices of leaves and seeds. Harvesting and drying cowpea
leaves does not require great strength or any tools other than
baskets, so it can be done by children while the adults
concentrate on more strenuous farming tasks. In addition, leaf
harvesting has the advantage of producing food earlier in the
season than mature seeds or even immature pods would be
available. Kenyan farmers harvest cowpea leaves from 3 weeks
after planting up to the start of podfilling (14).

Cowpea leaves are nutritious fresh or dried, are easily
stored when dry, and are already marketed and consumed in many
parts of Africa. It appears that intensive cultivation of cowpea
for both leaves and seeds would become more important as the

population and food requirements of Africa continue to increase.

103

Before new cowpea cultivars are distributed to farmers in
regions where leaf harvesting is practiced, their responses to
leaf removal should be determined, because this is as much a part
of the local crop environment as the soil, daylength, and
prevalent diseases. Ideally, such testing would occur during
selective breeding, with the goal of developing multi-purpose
cultivars.

Because of the variability of local environments, improved
alien cultivars often do not perform as well as traditional local
cultivars which are better adapted. In such places, a better
strategy might be to cross the local cultivars with improved
introduced cultivars, and to select the hybrids with qualities
that the farmers desire (119).

Insecticides may be used safely on cowpea if reasonable
precautions are taken (100). Pyrethrum daisies (gapyaanthemum
pyraphrpm) are grown in great quantities in East Africa, to
produce the insecticide pyrethrin, which is not poisonous to
humans or livestock (1). This locally produced insecticide is
likely to be safer and cheaper for African farmers than imported
oil-based products. Breeding for resistance to leaf-eating pests
might make cowpea leaves unpalatable to people as well.

Generations of African farmers have managed cowpea plantings
for multiple products. However, harvesting edible leaves
requires proper timing and minimal damage so that seed production
is not greatly reduced. Leaves are replaced most readily during
the vegetative stage when plants are small. Plants are larger in

the reproductive stage, but defoliation during pod growth is

104

harmful to seed yield, and after the seeds are grown the leaves
tend to wither and abscise. The time period at anthesis appears
to be optimal for leaf harvesting, but the harvest season can be
extended by removing a few leaves at a time.

Other tropical regions with inadequate food resources may
also benefit by combining seed and leaf production. Over 185
legume species are consumed as leaf vegetables. In addition to
cowpea, annual legume crops which have edible leaves include
bean, lima bean (Enaaapip§_iapapp§), winged bean (Esophpcappus
patpagppplppgfi), lablab bean (Dalipapa_iaplap), fennugreek
(11W). and pea (ELL—sum ativum)(14).

10.

LIST OF REFERENCES
Acland, J. D. 1971. East African crops. Longman, London.

Adams, M. W. 1967. Basis of yield component compensation in
crop plants with special reference to the field bean,
Phaseolus vulgaris. Crop Science 7:505-510.

Akeson, W. R., and M. A. Stahmann. 1966. Leaf protein
concentrates: a comparison of protein production per acre of
forage with that from seed and animal crops. Economic Botany
20:244-250.

Akinola, J. 0., and J. H. Davies. 1978. Effects of sowing
date on forage and seed production of 14 varieties of cowpea
(Vigna unguiculata). Experimental Agriculture 14:197-203.

Alderfer, R. G. and C. F. Eagles. 1976. The effect of
partial defoliation on the growth and photosynthetic
efficiency of been leaves. Botanical Gazette 137:351-355.

Attems, M. G. 1969. The Shambala system of agriculture
(Usambara). In: Kraut, H., and H. D. Cremer, (eds.)
Investigations into health and nutrition in East Africa.
Weltforum Verlag, Munich.

Attfield, H. D. 1979. Gardening with the seasons. V.I.T.A.
technical bulletin 46, Mt. Ranier, MD, USA.

Berry, R. E. 1981. Tropical fruits and vegetables as
potential protein sources. Food Technology 35 (11):45-49.

Bhaid, M. U., and S. K. Talapatra. 1965. Cowpea as a dual
purpose crop. Indian Journal of Dairy Science 18:153-155.

Binnie, R. C., and P. E. Clifford. 1980. Effects of some
defoliation and decapitation treatments on the productivity
of French beans. Annals of Botany 46:811-813.

105

11.

12.

13.

14.

15.

16.

17.

18.

19.

106

Binnie, R. C., and P. E. Clifford. 1981. Flower and pod
production in Phaseolus vulgaris. J. Agric. Sci. Cambridge
97:397-402.

Bishnoi, O. P. and R. H. Sedgley. 1985. Effect of
temperature on the rate of emergence of leaves of maize and
cowpea. Indian Journal of Agricultural Sciences 55:99-103.

Bittenbender, H. C. 1983. Handbook of tropical vegetables.
Department of Horticulture, Michigan State University, East
Lansing, MI, USA.

Bittenbender, H. C., R. P. Barrett, and B. M. Indire-
Lavusa. 1984. Beans and cowpeas as leaf vegetables and grain
legumes. Monograph no. 1, Bean/Cowpea CRSP, East Lansing,
MI, USA.

Brown, W. H. 1954. Useful plants of the Philippines.
Republic of the Philippines Department of Agriculture and
Natural Resources technical bulletin 10, vol. 2, Manila.

Carmi, A. and D. Koller. 1979. Regulation of photosynthetic
activity in the primary leaves of bean (Phaseolus vulgaris
L.) by materials moving in the water-conducting system.
Plant Physiology 64:285-288.

Carr, W. R. 1956. The preparation and analysis of some
African foodstuffs. The Central African Journal of Medicine
2:334-339.

Chagas, J. M., C. Vieira, M. Maestri, and A. A. Cardoso.
1978. Response of two bean cultivars to artificial
defoliation. Annual Report of the Bean Improvement
Cooperative 21:15-18. New York Agricultural Experiment
Station, Geneva, New York, USA.

Chaturvedi, G. S., P. K. Aggarwal, and S. K. Sinha. 1980.
Growth and yield of determinate and indeterminate COWpeas in
dryland agriculture. Journal of Agricultural Science,
Cambridge 94:137-144.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

107

Chen, C. L., and F. J. M. Sung. 1982. Effect of source-sink
manipulation on nitrogen fixation in mungbeans. Field Crops
Research 5:225-231.

Clifford, P. E. 1979. Source limitation of sink yield in
mung beans. Annals of Botany 43:397-399.

Clifford, P. E. 1980. Control of reproductive sink yield in
mung beans. Zeitschrift fur Pflanzenphysiologie 102:173-181.

Dalziel, J. M. 1916. A Hausa botanical vocabulary. T.
Fisher Unwin, London.

Dalziel, J. M. 1937. The useful plants of West Tropical
Africa. Crown Agents for the Colonies, London.

Davis, J. F. 1945. The effect of some environmental factors
on the set of pods and yield of white pea beans. Journal of
agricultural research 70:237-249.

De Mooy, B. E. 1985. Germplasm evaluation for Botswana
cowpea (Vigna unguiculata (L.) Walp.) Landraces. M.S.
Thesis, Michigan State University, East Lansing, Michigan,
UOSOA.

De Mooy, C. J., and B. E. De Mooy, 1985. Search for more
suitable cowpea varieties for semiarid conditions in
Botswana. Bean/COWpea CRSP Research Highlights Vol.2 No.1,
Bean/Cowpea CRSP, East Lansing, MI, USA.

De Schlippe, P. 1956. Shifting cultivation in Africa: the
Zande system of agriculture. Routledge and Kegan Paul,
London.

Diarra, N. 1975. Le jardinage urbain et suburbain au Mali -
le cas de Bamako. Journal D'Agriculture Tropicale et de
Botanique Appliquee 22:359-364.

Dickson, R. H. B. 1907. Cowpeas. The Rhodesian Agricultural
Journal 5:15-27.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

108

Duke, J. A. 1981. Handbook of legumes of world economic
importance. Plenum Press, New York.

Eaglesham, A. R. J., F. R. Minchin, R. J. Summerfield, P.
J. Dart, P. A. Huxley, and J. M. Day. 1977. Nitrogen
nutrition of cowpea (Vigna unguiculata) III. distribution of
nitrogen within effectively nodulated plants. Experimental
Agriculture 13:369-380.

Edje, O. T. 1981. Effects of nitrogen and leaf removal on
Phaseolus bean yield. Luso: Journal of Science and
Technology (Malawi) 2:39-51.

Edje, O. T. and L. K. Mughogho. 1976. Photosynthetic
efficiency of the different zones of the bean plant. Annual
Report of the Bean Improvement Cooperative 19:26-29. New
York Agricultural Experiment Station, Geneva, New York, USA.

Egli, D. B., R. D. Guffy, L. W. Meckel, and J. E. Leggett.
1985. The effect of source-sink alterations on soybean seed
growth. Annals of Botany 55:395-402.

Enyi, B. A. C. 1972. A spacing/time of planting trial with
cowpea (Vigna unguiculata (L.) Walp.) Ghana Journal of
Science 13:78-85.

Enyi, B. A. C. 1975. Effects of defoliation on growth and
yield in groundnut (Arachis hypogea), COWpeas (Vigna
unguiculata), soyabean (Glycine max) and green gram (Vigna
aureus). Annals of Applied Biology 79:55-66.

Ezedinma, F. O. C. 1965. Some effects of removal of
axillary branches on semi-upright cowpeas (Vigna unguiculata
(L.) Walp.) Horticultural Research 5:113-121.

Ezedinma, F. O. C. 1973. Effects of defoliation and topping
on semi-upright cowpeas (Vigna unguiculata (L.) Walp.) in a
humid tropical environment. Experimental Agriculture 9:203-
207.

F.A.O. 1979. A Zambian handbook of pasture and fodder
crops. F.A.O., Rome.

109
41.

Ferguson, H. 1954. The food crops of the Sudan and their
relation to environment. Philosophical Society of the Sudan,
Khartoum.

42.

Fernandez, G. C. J., and J. C. Miller, Jr. 1985. Yield
component analysis in five cowpea cultivars.
Hort. Sci. 110:553-559.

J. Amer. Soc.

43.

Fleuret, A. 1979. The role of wild foliage plants in the
diet, a case study from Lushoto, Tanzania. Ecology of Food
and Nutrition 8:87-93.

44. Fox, F. W.,

and M. E. N.
Delta Books,

Young. 1982. Food from the veld.

45. Franz, H. C.

1971. The traditional diet of the Bantu in the
Pietersburg district. South African Medical Journal 45:1232-
1235.

46. Galston, A. W., P. J. Davies, and R. L. Satter. 1980. The
life of the green plant. Prentice-Hall, Englewood Cliffs,
NJ, USA.

47. Galvez, G. E., J. J. Galindo, and G. Alvarez. 1977.
Defoliacion artificial para estimar perdidas por danos
foliares en frijol (Phaseolus vulgaris L.). Turrialba
27:143-146.

48.

Glegg, C. G. 1945. Native foodstuffs in Tanganyika.
Tropical Agriculture 22:32-38.

49. Gomez, M. I. 1981.

Carotene content of some green leafy
vegetables of Kenya and effects of dehydration and storage
on carotene retention.

Journal of Plant Foods 3:231-244.

50. Gomez, M. I. 1982. Sources of vitamin C in the Kenya diet
and their stability to cooking and processing. Ecology of
Food and Nutrition 12:179-184.

51.

Graham, P. H.

1978. Some problems and potentials of field
beans (Phaseolus vulgaris L.) in Latin America.
Research 1:295—317.

Field Crops

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

110

Graham, P. H. 1981. Some problems of nodulation and
symbiotic nitrogen fixation in Phaseolus vulgaris L.: a
review. Field Crops Research 4:93-112.

Greene, G. L. 1971. Economic damage levels of bean leaf
roller populations on snap beans. Journal of Economic
Entomology 64:673-674.

Greene, G. L., and D. R. Minnick. 1967. Snap bean yields
following simulated insect defoliation. Proceedings Florida
State Horticultural Society 80:132-134.

Grivetti, L. E. 1978. Nutritional success in a semi-arid
land: examination of Tswana agro-pastoralists of the eastern
Kalahari, Botswana. American Journal of Clinical Nutrition
31:1204-1220.

Grubben, G. J. H. 1977. Tropical vegetables and their
genetic resources. International Board for Plant Genetic
Research, Rome.

Harris, P. 1974. A possible explanation of plant yield
increases following insect damage. Agro-ecosystems 1:219-
225.

Herridge, D. F. and J. S. Pate. 1977. Utilization of net
photosynthate for nitrogen fixation and protein production
in an annual legume. Plant Physiology 60:759-764.

Hoorweg, J., and R. Niemeyer. 1980. Preliminary studies on
some aspects of Kikuyu food habits. Ecology of Food and
Nutrition 9:139-150.

Huxley, P. A. and R. J. Summerfield. 1976. Leaf area
manipulation with vegetative cowpea plants (Vigna
unguiculata (L.) Walp.). Journal of Experimental Botany
27:1223-1232.

Imbamba, S. K. 1973. Leaf protein content of some Kenya
vegetables. East African Agricultural and Forestry Journal
38:246-251.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

111

Imbamba, S. K., and L. L. Tieszen. 1977. Influence of light
and temperature on photosynthesis and transpiration in some
C3 and C4 vegetable plants from Kenya. Physiologia Plantarum
39:311-316.

Imrie, B. C., and K. L. Butler. 1983. Joint contribution of
individual plant attributes to seed yield of cowpea (Vigna
unguiculata L.) in small plots. Field Crops Research 6:161-
170.

Imungi, J. K., and N. N. Potter. 1983. Nutritional contents
of raw and cooked cowpea leaves. Journal of Food Science
48:1252-1254.

Imungi, J. K., and N. N. Potter. 1985. Nutritional and
sensory properties of canned and stored cowpea leaves.
Nutrition Reports International 31:21-34.

Jameson, J. D. 1970. Agriculture in Uganda. Oxford, London.

Johnson, R. M., and W. D. Raymond. 1964. The chemical
composition of some tropical food plants: II. pigeon peas
and cowPeas. Tropical Science 6:68-73.

Keller, W., E. Muskat, and E. Valder. 1969. Some
observations regarding economy, diet, and nutritional status
of Kikuyu farmers in Kenya. In: Kraut, H., and H. D. Cremer,
(eds.) Investigations into health and nutrition in East
Africa. Weltforum Verlag, Munich.

Leakey, C. L. A., and J. B. Wills. 1977. Food crops of the
lowland tropics. Oxford, London.

Leung, W. T. W. 1968. Food composition table for use in
Africa. FAO, Rome.

Littleton, E. J., M. D. Dennett, J. Elston, and J. L.
Monteith. 1979. The growth and development of COWpeas (Vigna
unguiculata) under tropical field conditions. Journal of
Agricultural Science, Cambridge 93:291-307.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

112

Littleton, E. J., M. D. Dennett, J. Elston, and J. L.
Monteith. 1981. The growth and development of cowpeas (Vigna
unguiculata) under tropical field conditions. Journal of
Agricultural Science, Cambridge 97:539-550.

Littleton, E. J., M. D. Dennett, J. L. Monteith, and J.
Elston. 1979. The growth and development of cowpeas (Vigna
unguiculata) under tropical field conditions 2. accumulation
and partition of dry weight. Journal of Agricultural
Science, Cambridge 93:309-320.

Lucas, E. O., G. M. Milbourne, and P. N. Whiteford. 1976.
The translocation of 14C photosynthate from leaves and pods
in Phaseolus vulgaris. Annals of applied Biology 83:285-290.

Maeda, E. E. 1985. Effect of solar dehyration on amino acid
pattern and available lysine content in four tropical leafy
vegetables. Ecology of Food and Nutrition 16:273-279.

Maeda, E. E., and D. K. Salunkhe. 1981. Retention of
ascorbic acid and total carotene in solar dried vegetables.
Journal of Food Science 46:1288-1290.

Maheshwari, M.L., and S. K. Ranjhan. 1969. Yield, chemical
composition, and outturn of nutrients of various varieties
of cowpea fodder. Indian Journal of Dairy Science 22:200-
201.

Manyafu, J. L. 1971. The role of wild fruits and vegetables
in Tanzania. Acta Horticulturae 21:93-96.

Mariga, I. K., D. Giga, and P. Marmba. 1985. Cowpea
production constraints and research in Zimbabwe. Tropical
Grain Legume Bulletin 30:9-14.

Matai, S., D. K. Bagchi, and S. Chanda. 1978. Yield of
extracted leaf protein from BC. 4216, a fodder variety of
cowpea. Indian Journal of Agronomy 23:163-166.

McLaren, D. S. 1961. Sources of carotene and vitamin A in
Lake Province, Tanganyika. Acta Tropica 18:78-80.

82.

83.

84.

85.

86.

87.

88.

89.

90.

113

Mehta, P. N. 1971. The effect of defoliation on seed yield
of cowpeas (Vigna unguiculata (L.) Walp.) and analysis of
the leaf harvest for dry matter and nitrogen content. Acta
Horticulturae 21:167-171.

Miller, C. D., and B. Branthoover. 1957. Nutritive values
of some Hawaii foods. Hawaii Agricultural Experiment Station
circular 52, Honolulu, HA, USA.

Minchin, F. R., and R. J. Summerfield. 1976. Symbiotic
fixation and vegetative growth of cowpea (Vigna unguiculata
(L.) Walp.) in waterlogged conditions. Plant and Soil
45:113-127.

Minchin, F. R., R. J. Summerfield and M. C. P. Neves. 1980.
Carbon metabolism, nitrogen assimilation, and seed yield of
cowpea (Vigna unguiculata (L.) Walp.) grown in an adverse
temperature regime. Journal of Experimental Botany 31:1327-
1345.

Minchin, F. R., R. J. Summerfield, A. R. J. Eaglesham and
K. A. Stewart. 1978. Effects of short-term waterlogging on
growth and yield of cowpea (Vigna unguiculata). Journal of
Agricultural Science, Cambridge 90:355-366.

Miracle, M. P. 1967. Agriculture in the Congo Basin.
University of Wisconsin Press, Madison, WI, USA.

Mission Socio-economique du Soudan. 1956-1958.
L'Alimentation des populations rurales du Delta Vif du Niger
et de L'Office du Niger. Republique du Mali.

Murray, J. A. 1881. The plants and drugs of Sind. Indian
Book Gallery, Delhi.

Neves, M. C. P., R. J. Summerfield, and F. R. Minchin.
1982. Effects of complete leaf shading during the late
reproductive period on carbon and nitrogen distribution and
seed production by nodule-dependent cowpea (Vigna
unguiculata) plants. Tropical Agriculture (Trinidad) 59:248-
253.

91.

92.

93.

94.

95.

.96

97.

98.

99.

100.

114

Newman, J. L. 1975. Dimensions of Sandawe diet. Ecology of
Food and Nutrition 4:33-39.

Ochse, J. J. 1931. Vegetables of the Dutch East Indies.
Archipel Drukkerij, Buitenzorg, Java, Dutch East Indies.

Ojehomon, O. O. 1970. A comparison of the vegetative
growth, development and seed yield of three varieties of
cowpea, Vigna unguiculata (L.) Walp. Journal of Agricultural
Science, Cambridge 74:363-374.

Oomen, H. A. P. C., and G. J. H. Grubben. 1977. Tropical
leaf vegetables in human nutrition. Department of
Agricultural Research communication 69, Koninklijk Instituut
voor de Tropen, Amsterdam.

Osei-Yeboah, S., J. I. Lindsay, and F. A. Gumbs. 1983.
Estimating leaf area of cowpea (Vigna unguiculata (L.)
Walp.) from linear measurements of terminal leaflets.
Tropical Agriculture (Trinidad) 60:149-150.

Owuor, P. O. 1985. Fluoride content of common vegetables
from different parts of Kenya. Food Chemistry 18:283-289.

Pandey, R. K. 1983. Influence of defoliation on seed yield
in cowpea (Vigna unguiculata (L.) Walp.) in a sub-tropical
environment. Field Crops Research 7:249-256.

Pandey, R. K. and V. B. Singh. 1981. Influence of partial
defoliation on dry-matter production and seed yield of urd
bean (Vigna mungo L. Hepper). Journal of Agricultural
Science, Cambridge 97:437-443.

Peoples, M. B., J. S. Pate, and C. A. Atkins. 1985. The
effect of nitrogen source on transport and metabolism in
fruiting plants of cowpea (Vigna unguiculata (L.) Walp.).
Journal of experimental Botany 36:567-582.

Price, M., and W. R. Dunstan. 1982. The effect of four
insecticides on leaf miner damage of COWpea in Tanzania.
Tropical Grain Legume Bulletin 27:23-26.

101.

102.

103.

104.

105.

106.

107.

108.

109.

110.

115

Purseglove, J. W. 1943. Some Uganda vegetables. East
African Agricultural and Forestry Journal 9:98-101.

Ranjhan, S. K., A. N. Singh and M. L. Maheshwari. 1972.
Chemical composition, yield and outturn of nutrients of
Australian varieties of cowpeas in comparison to Russian
Giant cowpea (Vigna species). Indian Journal of Dairy
Science 25:299-301.

Ranjhan, S. K., S. K. Talapatra and A. C. Kala. 1967. Yield
and nutritive value of dual purpose crop cowpea (Vigna
catjung) - COWpea hay as a growth production ration. Indian
Journal of Dairy Science 20:146-149.

Santa Cecilia, F. C., M. A. P. Ramalho, and M. B. Gomide.
1980. Efeito do desfolhamento sobre a producao de feijao
(Phaseolus vulgaris L.). Cienc. Prat. Lavras 4:66-72.

Saxena, J. S., R. B. Upadhyaya, and C. B. Johri. 1971.
Chemical composition and nutritive value of empty COWpea
pods. The Indian Veterinary Journal 48:513-516.

Schlage, C. 1969. Analysis of some important foodstuffs of
Usambara. In: Kraut, H., and H. D. Cremer, (eds.)
Investigations into health and nutrition in East Africa.
Weltforum Verlag, Munich.

Schlage, C. 1969. Observations in a local market in
Usambara. In: Kraut, H., and H. D. Cremer, eds.
Investigations into health and nutrition in East Africa.
Weltforum Verlag, Munich.

Schoch, P. G. 1974. Reprise d'activitie des premieres
feulles de Vigna sinensis L. a la suite de defoliations.
Physiologie Vegetale 12:289-298.

Sellschop, J. P. F. 1962. Cowpeas, Vigna unguiculata (L.)
Walp. Field Crop Abstracts 15:259-266.

Siffel, P., N. N. Lebedev, and Z. Sestak. 1985. Changes in
chloroplast absorption and fluorescence spectra during
ontogeny of primary bean leaves. Photosynthetica 19:127-137.

111.

112.

113.

114.

115.

116.

117.

118.

119.

120.

116

Sinclair, T. R. and C. T. de Wit. 1975. Photosynthate and
nitrogen requirements for seed production by various crops.
Science 189:565-567.

Singh, R., P. C. Gupta and K. Pradhan. 1974. Variability in
protein, cell-wall constituents and IN VITRO dry matter and
cell-wall digestibility of some promising cowpea strains
(Vigna sinensis). HAU Journal of Research 4:309-313.

Smartt, J. 1985. Evolution of grain legumes. III. Pulses in
the genus Vigna. Experimental Agriculture 21:87-100.

Sreeramulu, N. 1982. Chemical composition of some green
leafy vegetables grown in Tanzania. Journal of Plant Foods
4:139-141.

Stewart, K. A., R. J. Summerfield, and B. J. Ndunguru.
1978. Effects of source-sink manipulations on seed yield of
cowpea (Vigna unguiculata (L.) Walp.). 1 - Defoliation.
Tropical Agriculture (Trinidad) 55:117-125.

Summerfield, R. J., F. R. Minchin, E. H. Roberts and P.
Hadley. 1983. Cowpea. In: Smith, W. H., and S. Yoshida,
(eds.) Potential Productivity of Field Crops under Different
Environments. IRRI, Los Banos, Philippines. p.249-280.

Summerfield, R. J., P. A. Huxley and W. Steele. 1974.

Cowpea (Vigna unguiculata (L.) Walp.). Field Crop Abstracts
27:301-312.

Summerfield, R. J., P. J. Dart, P. A. Huxley, A. R. J.
Eaglesham, F. R. Minchin and J. M. Day. 1977. Nitrogen
nutrition of cowpea (Vigna unguiculata). I. effects of
applied nitrogen and symbiotic nitrogen fixation on growth
and seed yield. Experimental Agriculture 13:129-142.

Ta'Ama, M. 1985. A new look at the importance of cultivars
in cowpea research: evidence from Northern Cameroon.
Research Highlights, Michigan State University Bean/Cowpea
CRSP vol.2, no.5.

Tallantire, A. C., and P. M. Goode. 1975. A preliminary
study of the food plants of the West Nile and Madi districts

121.

122.

123.

124.

125.

126.

127.

128.

129.

117

of Uganda. East African Agricultural and Forestry Journal
40:233-255.

Tamboliya, T. S., C. A. Patel and A. S. Patel. 1975.
Nutritional evaluation of cowpea varieties. The Indian
Journal of Nutrition and Dietetics 12:366-371.

Tardieu, M., and D. Sene. 1966. Le haricot niebe (Vigna
unguiculata Walpers) au Senegal. Agronomie Tropicale 21:918-
926.

Terra, G. J. A. 197. Horticulture's contribution to human
nutrition in the tropics. 17th International Horticultural
Congress 83-95.

Thimann, K. V. 1980. The senescence of leaves. Ch. 5 in:
Senescence in Plants. CRC Press, USA.

Tothill, J. D. 1940. Agriculture in Uganda. Milford,
London.

Varlet Grancher, C., and R. Bonhomme. 1974. Utilisation de
l'energie solaire par une culture de Vigna sinensis. Ann.
Agron. 25:811-819.

Vander Zon, A. P. M., and G. J. H. Grubben. 1976. Les
legumes-feuilles spontanes et cultives du Sud-Dahomey.
Department of Agricultural Research communication 65,
Koninklijk Instituut voor de Tropen, Amsterdam.

Venkatarayappa, T., R. A. Fletcher, and J. E. Thompson.
1984. Retardation and reversal of senescence in bean leaves
by benzyladenine and decapitation. Plant and Cell Physiology
25:407-418.

Verheij, E. W. M. 1982. Horticulture in Tanzania - position
and perspective. Department of Agricultural Research
communication 74, Koninklijk Instituut voor de Tropen,
Amsterdam.

130.

131.

132.

133.

134.

135.

136.

137.

138.

139.

118

Vieira, C. 1981. Effect of artificial defoliation on the
yield of two indeterminate bean (Phaseolus vulgaris L.)
cultivars. Turrialba 31:383-385.

Viglierchio, D. R., and F. W. Went. 1957. Plant growth
under controlled conditions. IX. Growth and fruiting of the
Kentucky Wonder bean (Phaseolus vulgaris). American Journal
of Botany 44:449-453.

Von Caemmerer, S., and G. D. Farquhar. 1984. Effects of
partial defoliation, changes of irradiance during growth,
short-term water stress and growth at enhanced p(C02) on the
photosynthetic capacity of leaves of Phaseolus vulgaris L.
Planta 160:320-329.

Waddill, V., K. Pohronezny, R. McSorley, and H. H. Bryan.
1984. Effect of manual defoliation on pole bean yield.
Journal of Economic Entomology 77:1019-1023.

Waddill, V., R. McSorley, and K. Pohronezny. 1981. Field
monitoring: basis for integrated management of pests on snap
beans. Tropical Agriculture (Trinidad) 58:157-169.

Waldmann, E. 1975. The ecology of the nutrition of the
Bapedi, Sekhukuniland. Ecology of Food and Nutrition 4:139-
151.

Wareing, P. F., M. M. Khalifa, and K. J. Treharne. 1968.
Rate-limiting processes in photosynthesis at saturating
light intensities. Nature 220:453-457.

Watson, J. D. 1976. Ascorbic acid content of plant foods in
Ghana and the effects of cooking and storage on vitamin
content. Ecology of Food and Nutrition 4:207-213.

Westphal, E. 1975. Agricultural systems in Ethiopia. Centre
for Agricultural Publishing and Documentation, Wageningen,
Netherlands.

Westphal, E. 1981. L'Agriculture autochtone au Cameroun.
Miscellaneous papers no. 20, Landbouwhogeschool, Wageningen,
Netherlands.

140.

141.

142.

143.

144.

119

Wien, H. C. 1982. Dry matter production, leaf area
development, and light interception of cowpea lines with
broad and narrow leaf shape. Crop Science 22:733-737.

Wien, H. C., E. J. Littleton and A. Ayanaba. 1977. Drought
stress of cowpea and soybean under tropical conditions.
Paper presented at the International Conference on Stress
Physiology in Important Crop Plants. Boyce Thompson
Institute, Yonkers, NY, USA. June 28-30, 1977.

Williams, E. R., and M. A. Caliendo. 1984. Nutrition:
principles, issues, and applications. McGraw-Hill, New York.

Williamson, J. 1975. Useful plants of Malawi. University of
Malawi, Zomba, Malawi.

Woolhouse, H. W. 1978. Senescence processes in the life
cycle of flowering plants. BioScience 28:25-31.