|
EFFECTS OF WATER STRESS IMPOSED AT MID POD FILLING
UPON YIELD AND DRY MATTER PARTITIONING IN DRY BEANS
(Phaseolus vulgaris).
Thesis for the Degree of M. S.
MICHIGAN STATE UNIVERSITY
CATALINA SAMPER
1984
EFFECTS OF WATER STRESS IMPOSED AT MID-POD FILLING
UPON YIELD AND DRY MATTER PARTITIONING IN DRY BEANS
(Phaseolus vulgaris)
By
CATALINA SAMPER
A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Crop and Soil Science
1984
To Dad, Mom and Ivan
for their love, generosity and
encouragement of my growth
ACKNOWLEDGEMENTS
I cannot find the words to express my gratitude to my
major professor, Dr. M. Wayne Adams. During the course of my
studies he has been not only a major professor, but a friend
and a colleague. He was always at my side, sharing moments
of joy and sadness, giving me encouragement and
unconditional support. The challenging discussions that we
had, his encouragement of my independent thinking and his
deep honesty, will always be with me as an example to
follow. I had the privilege to work with a great scientist
and a true teacher but best of all, with a great human
being.
I want to express my gratitude to:
Dr. Andrew Hanson, a member of my committee, for his
encouragement of my development as an independent thinker,
his thorough and constructive criticism of my work, his
generosity and kind support. His conceptualization of
science and his fine example as a continues learner and
actively involved researcher are qualities that I greatly
admired.
Dr. Al Smucker, committee member, for reviewing this
manuscript and for his help on the progress of my research
project.
Dr. Peter Graham, Dr. Rogelio Lepiz, Dr. Ronald
Ferrera, Mr. Jorge Acosta and Mr. Abelardo NuHez, for
their help and interest in this project.
Special thanks to Greg, Nasrat, Sue, Joe, Rhea, Francisco,
Earl and to all the graduate students, the professors and
technicians of the "bean program", for their intellectual
stimulation, the hard work and the good times that we shared
together.
Last but not least, to my brothers Juan, Gordo and Bite,
my sister Cuca, and my friends Kim, Touran and Regina for
their love and unconditional support.
TABLE OF CONTENTS
LIST OF TABLES . . . . .
LIST OF FIGURES . . ..
INTRODUCTION . . ..
LITERATURE REVIEW . . . .
Yield Constraints . . .
Allocation of Assimilates .
Biological Nitrogen Fixation
Photosynthate Partitioning .
Drought Tolerance Mechanisms
MATERIALS AND METHODS . . .
RESULTS .
I. Water Effects . . .
A. Biological- Yield .
B. Economic Yield . .
C. Harvest Index . .
D. Seed Size . .
E. Seed Number . . .
F. Length of- Vegetative and
Reproductive Stages .
G. Leaf Dropping . .
Page
S. . vi
.ix
* 1
. 3
* 0 3
* 7
. 14
* 12
* 30
. 30
S . . 18
.* 30
S 37
* 37
0 37
. 40
.* 42
* 47
H. Plant Dry Weight at Physiological
Maturity .
I. Plant Dry Weight Changes:
Remobilization . . . .
J. Starch Analysis . . .
. 50
. 52
. 56
TABLE OF CONTENTS (Continued)
Page
K. 20 Upper and Lower Pods:
Seed number and Size. . . . 60
II. Nitrogen Effects . . . . . 64
A. Non-significant Effects . . . 66
B. Plant Dry Weight . . .66
C. Plant Dry Weight Changes:
Remobilization . . . . 69
DISCUSSION 74
1. Crop Growth Rate . . . . .75
2. Partitioning 79
3. The Filling Period . . . . 90
YIELD POTENTIAL AND DROUGHT SUSCEPTIBILITY . . 93
1. Drought Susceptibility Index . . . 93
2. Relationship between Control and
Stress Yields 99
3. Geometric Mean of Stress and Control
Yields as a Selection Criterion for
Drought Tolerance . . . . .
SUMMARY AND CONCLUSIONS . . . . .
LITERATURE CITED . . . . . .
APPENDIX I : Durango Experiment: Experimental Design
and Yield Data . . . .
APPENDIX II : Starch Analysis . . . .
. 103
. 112
. 115
. 122
LIST OF TABLES
TABLE 1.
TABLE 2.
TABLE 3.
TABLE 4.
TABLE 5.
TABLE 6.
TABLE 7.
TABLE 8.
TABLE 9.
TABLE 10.
TABLE 11.
TABLE 12.
Biological Yield (kg/ha) under two water
treatments. Iguala, 1982-3.. .
Economic Yield (kg/ha) under two water
treatments. Iguala, 1982-3. . . .
Harvest Index under two water treatments.
Iguala, 1982-3. .
Weight of 100 seeds (grs) under two water
treatments. Iguala, 1982-3. . . .
Seed number (seeds/mt2) under two water
treatments. Iguala, 1982-3. . . .
Days between flowering and physiological
maturity under two water treatments. Iguala,
1982-3. *
Leaf dropping under two water treatments.
Iguala, 1982-3. . .. .
Stem and Pod % of total dry weight at
physiological maturity under two water
treatments. Iguala, 1982-3. . . .
Mean values of starch (mgrs/gr dry wt) at
three different physiological stages under
two water treatments. Iguala, 1982-3. .
Seed number of 20 upper and lower pods
under two water treatments. Iguala, 1982-3.
Seed weight (mgrs/seed) of 20 upper and
lower pods under two water treatments.
Iguala, 1982-3. . .* *
Shoot:Root ratio under two Nitrogen
treatments at two physiological stages.
Iguala, 1982-3. .
. 33
S538
. 39
. 41
* 43
. 49
. 51
. 61
LIST OF TABLES (Continued)
TABLE 13.
TABLE 14.
TABLE 15.
TABLE 16.
TABLE 17.
TABLE 18.
TABLE 19.
TABLE 20.
TABLE 21.
TABLE 22.
TABLE 23.
TABLE 24.
TABLE 25.
Average Crop Growth Rates (kg/ha/day)
form planting to flowering. Iguala, 1982-3.
Average Crop Growth Rates (kg/ha/day)
from flowering to maturity under two water
treatments. Iguala, 1982-3. . . .
Average Fruit Growth Rate from flowering
to physiological maturity (kg/ha/day)
under two water treatments. Iguala, 1982-3.
Partitioning Factor under two water
treatments. Iguala, 1982-3. . .
Comparison between Grain yield, Fruit
Growth Rate, Seed number and Effective
seed filling period under two water
treatments. Iguala, 1982-3. . . .
Individual cultivar drought susceptibility
indices. Iguala 1982-3 and Durango 1983.
Group drought susceptibility indices -S-
Iguala 1982-3 and Durango 1983. . .
Group ranking by drought susceptibility
index (S). Iguala 1982-3 and Durango 1983
Ranking by drought susceptibility index (S)
of the eight cultivars planted in Iguala
and Durango . . . . . .
. 78
. 80
. 81
. 91
. 95
. 96
. 97
97
Yield differential, Arithmetic mean and
Geometric mean for the Iguala experiment
Yield differential, Arithmetic mean and
Geometric mean for the Durango experiment .
. 104
. 106
Cultivar ranking for the Iguala experiment
using four different selection criteria . 107
Cultivar ranking for the Durango experiment
using four different selection criteria
. 108
LIST OF TABLES (Continued)
TABLE 26. Mean yields of the selected top 20%
cultivars, using two different selection
criteria. Iguala experiment. . . . 110
TABLE 27. Mean yields of the selected top 20%
cultivars, using two different selection
criteria. Durango experiment. . . 111
viii
LIST OF FIGURES
FIGURE 1.
FIGURE 2.
FIGURE 3.
FIGURE 4.
FIGURE 5.
FIGURE 6.
FIGURE 7.
FIGURE 8.
FIGURE 9.
FIGURE 10.
FIGURE 11.
Maximum and minimum daily temperatures.
Iguala, 1982-3. . .
. 24
Flowering dates and maximum daily
temperatures. Iguala, 1982-3. . . 45
Biological yield under irrigation and
flowering dates. Iguala, 1982-3. . . 46
Economic yield under irrigation and
flowering dates. Iguala, 1982-3. . . 48
Stem, Root, Pod and Leaf dry weights
(grs/mt2) over three physiological
stages. Iguala, 1982-3. . . . 53
Changes in Stem-Starch contents (grs/mt2)
over three physiological stages under
two water treatments. Iguala, 1982-3.
Changes in Pod-Starch contents (grs/mt2)
over three physiological stages under
two water treatments. Iguala, 1982-3. . 59
Stem % of total plant dry weight at
flowering time. Iguala, 1982-3.
. 67
Stem, Root, Pod and Leaf weights at
flowering and at 15 dap. under two
different levels of added Nitrogen.
Iguala, 1982-3. . . . . 70
Proportion of fruit growth that can
be accounted for by post-anthesis
photosynthesis under irrigated
conditions. Iguala, 1982-3. . . 84
Proportion of fruit growth that can
be accounted for by post-anthesis
photosynthesis under stress conditions.
Iguala, 1982-3. . . . 85
. 58
LIST OF FIGURES (Continued)
FIGURE 12.
FIGURE 13.
FIGURE 14.
Relationship between change in stem
and leaf weight from anthesis to
maturity and grain yield under
irrigated conditions. Iguala, 1982-3. . 88
Relationship between change in stem
and leaf weight from anthesis to
maturity and grain yield under
stress conditions. Iguala, 1982-3. . 89
Relationship between control and
stress yield. Iguala, 1982-3. . . 100
FIGURE 15. Relationship between control and
stress yield. Durango, 1983. . . 101
INTRODUCTION
Varietal differences in the amount of starch present at
flowering time,grain filling and physiological maturity in the
dry bean (Phaseolus vulgaris) have been previously reported
(4,30). The capacity of certain genotypes to store and remobilize
starch to the grain may be an advantage when the plants are
subjected to stress and their photosynthetic activity is reduced.
The general objective of the project of which this thesis is
part of, was to study the relationship between photosynthate
partitioning, remobilization, and the seed filling processes in
several genotypes of P. vulgaris grown under different stress
conditions. Initial objectives were: 1) to determine the effect
of drought stress imposed during the latter part of the seed
filling period on the yield performance of 22 different bean
cultivars and to relate their performance under stress to their
ability to accumulate and remobilize non-structural
carbohydrates; 2) to compare the effect of nitrogen fertilizer
versus biologically fixed nitrogen (BNF) under stress conditions,
and subsequently to determine the relationship between total
amount of non-structural carbohydrates and their remobilization
with the plant's ability to buffer the adverse environmental
conditions; 3)to identify bean genotypes having tolerance to
drought and high BNF potentials and to relate their performance
under stress conditions with the patterns of accumulation and
remobilization of starch and soluble sugars; 4)to identify
specific traits or physiological characteristics that could be
associated with better cultivar performance under drought
conditions, 5)to use the information and genetic materials
obtained during the course of this research as sources of new
improved germplasm in the development of varieties with
resistance to drought and with the capacity to fix nitrogen under
water stressed conditions.
For these purposes an experiment was conducted in
Iguala,Mexico, from December of 1982 to April of 1983. This
experiment was intended to study and exploit the genetic
potential for differential storage and remobilization of non-
structural carbohydrates and the genetic capacity for higher
levels of BNF, and to relate this to the yield performance of
cultivars under conditions of drought and low soil nitrogen.
A second experiment was conducted in the summer of 1983 in
Durango, Mexico to provide further data on varietal performance
under stress.
LITERATURE REVIEW
Yield constraints
Over forty five per cent of the world production of dry
edible beans (Phaseolus vulgaris) is consumed in Latin America.
Nevertheless,low yields of this crop are limiting the traditional
role beans play as a staple food in the diets of poor and middle
income consumers of this region. Although bean yields of over
4000 kg./ha. have been reported from experimental plots at the
Centro Internacional de Agricultura Tropical (CIAT) in
Colombia,the average bean yield in Latin America remains near 600
kg./ha. (51). In the dryland production region of Mexico the long-
term yields are reported to average less than 300 kg./ha. (1). A
significant closing of the gap between current yields and
potential yields must be achieved if this crop is to fulfill its
role in meeting the nutritional needs of the population.
In large areas of Latin America and Africa, where beans
constitute major source of dietary protein, production is
limited mainly because beans are a crop of the small farmer and
the conditions under which the crop is usually grown are typified
by low soil fertility, and minimal technical inputs, such as
irrigation, fungicides and insecticides .
Approximately 20% of all potentially arable land in the
world is in arid and semiarid zones, and about 16% of the world's
population lives on these lands (40). Research and development in
arid and semi-arid agriculture has, therefore, global
significance. Arid and semi-arid lands have been defined in a
number of different ways. Their main characteristic is a low
and variable seasonal rainfall, a condition which is often
directly exacerbated by other variable elements of the climate,
such as temperature, sunshine, wind and humidity conditions.
Beans, among a few other crops, are dryland staples in many
developing countries, providing a major source of affordable
protein and carbohydrate. Bean breeding over the years has
focused on improving agronomic adaptation along with disease
resistance, with less direct emphasis upon yield itself. It is
acknowledged by bean plant breeders that there have been no
decisive breakthroughs in yield, excepting increases originating
from disease resistance or favorable maturity adjustments (2).
Increasing yield is imperative, but this objective must be
integrated with the genetic improvement of adaptation and
resistance to stresses brought about by diseases,insects and
physical causes. The improvement of both agronomic characters and
yield could maximize the responses of the bean plant to available
resources characteristic of the site and local production system.
This is especially important for the Latin American small
farmers,because the conditions under which the crop is usually
grown are typified by the lack of irrigation systems,little or no
use of fungicides and insecticides, and small amounts of
fertilizers. In Mexico, where 1.7 million hectares are planted
annually with beans, 24 out of the 30 states that produce beans
raise them under rainfed conditions. During 1970 to 1975
approximately 1.2 million hectares were planted annually with
beans,and the average yield was around 545 kg./ha (35). This was
enough for the internal demand, but from there on only during
1978 and 1980 was production considered to be at its normal
level. As a consequence, during 1980 Mexico had to import more
than 250,000 tons of beans to satisfy the internal demand.
This production shortfall originated basically because of
adverse climatic effects such as drought and early frost. In
Mexico, beans are planted twice a year, during the Spring-Summer
and the Autumn-Winter cycles. During the Spring-Summer cycle when
the majority of the total production is obtained, about 1.4
million ha. are planted and 530,000 tons are harvested. In this
cycle typical low yields of 387 kg/ha are caused by adverse
environmental factors such as drought -scarce or irregular
rainfall- and early frost in the northern part of the country.
During the Autumn-Winter cycle about 260,000 ha. are planted and
246,000 tons are harvested; this corresponds to 31% of the total
national production and 16% of the total planted area. It is
interesting to note that with only 16% of the total planted area
almost one third of the total national production is obtained,
with average yields being 933 kg./ha. In states such as Nayarit,
Sinaloa and Baja California where irrigation is widely practiced,
average yields are over 1100 kg./ha., while for the country as a
whole 88% of the area that produces beans is rainfed only and the
average yields are about 350 kg./ha. Of this, 88% or
approximately 1 million ha. are in the states of Aguascalientes,
San Luis Potosf, Zacatecas and Chihuahua. These areas are
frequently affected by either scarce or badly distributed rain
throughout the growing season. In the state of Durango, nearly
30% of the planted beans are lost annually due to insufficient
water,and in bad years such as 1979 the losses can reach up to
60% of the total planted area (1).
In Colombia, in the states of Huila, Narifo and Antioquia
where the average farm sizes are 29.5 9.2 and 4.4hectares,
respectively, the percentage of farms that use irrigation is 2,
0, and 0 and the average yields are 680, 467 and 533 kg./ha. (44).
On the other hand, in the state of Valle del Cauca, where the
average farm size is 48.0 ha., 45% of the farms use irrigation
and the average yield is 906 kg./ha.
Among factors other than drought tolerance that could
contribute to improved crop yields the availability of fixed
nitrogen to crops is probably one of the greatest importance. In
1974 40 x 106 tons of fertilizer nitrogen with an approximate
value of 8 billion dollars were used as opposed to the 3.5 x
106 tons that were used annually twenty five years ago (28). The
scarcity of nitrogen fertilizers and their increased selling
price has produced a tremendous interest in the search for
alternative technologies. Inoculation of legumes with Rhizobium
at the farm level appears to offer promise as a possible
substitute for nitrogen fertilizers. Recent reports from farm
trials performed in Colombia by CIAT show that in the absence of
any nitrogen amendments, inoculation of a local variety with a
mixture of Rhizobium strains gave yields that were not
significantly different from a farmers' technology treatment, in
which 20 kg./ha. of chemical nitrogen in the form of urea and 2
tons/ha. of chicken manure were applied. Substitution for
nitrogen fertilizers by a Rhizobium inoculant would reduce total
costs of production by 34%, while the net return per peso
invested would rise from 5.5 to 7.7 pesos.
A bean breeder who wants to develop a variety for the small
farmers of Latin America should be aware that increased yields
must be obtained with very limited cash inputs.
Allocation of Assimilates
In recent years breeders have been considering the
development of plant ideotypes (14). In dry beans an ideotype for
production under monoculture has been proposed by Adams (2), who
suggested that productivity increases in dry beans could be
obtained if a more efficient allocation of assimilates into the
economic sink is developed by breeding.
Two of the principal physiological processes that can be
considered for improvement of crop yields are photosynthate
production and photosynthate partitioning to the economically
important organs. However, the importance of transpiration as a
central factor in explaining the influence of water limitation on
productivity, as pointed out by Fischer and Turner (20) cannot be
overlooked. It depends on the inevitable association between
water loss and C02 assimilation. Dry matter production over a
given period of time is a function of the total transpiration for
the given period and the water use efficiency. The importance of
respiration rates in determining the net accumulation of dry
matter is commonly overlooked. As Gifford et. al. (22) pointed
out, in leaves the fixed carbon is partitioned between its
retention in the plant and its photorespiratory release. Over
long periods, a full understanding of productivity requires
consideration of how each increment of dry weight is allocated to
both vegetative and reproductive sinks.
The potential for increasing crop productivity by optimizing
canopy structure has been documented by experimental research,
modeling, and computer simulation (50). Assimilate partitioning
is a dynamic process and varies with the stage of plant
development. In the vegetative stage of the dry bean plant, the
distribution of assimilates is dominated by the proximity between
the source and the sink. After flowering, when the developing
pods become major sinks, there is a more complex pattern,
although the relationship between leaves and pods in their own
axils still predominates (3).
Use of 14C as a tracer, and changes in dry weight of
specific organs have been important techniques in helping to
understand assimilate distribution. However, many important
aspects of this process, like the mechanisms of regulation,
accumulation and remobilization of storage assimilates under
different conditions, still remain to be studied in order to
provide guidelines for the increase of yields by manipulation of
photosynthate partitioning.
Large varietal differences in the ability to translocate
14c- assimilates and a clear trend for varieties with the higher
translocation rates to have higher photosynthetic rates, were
reported by Adams and Reicosky (3). From data obtained on
carbohydrate translocation patterns in beans, they suggested that
the two facets upon which breeding studies might be focused were
rate of translocation and direction of partition to sinks of the
assimilate. They also suggested that these characters may be
under genetic control and might be used in a plant breeding
program.
In recent years, attention has been given to carbohydrate
production and partitioning in plants as factors related to crop
yield; carbohydrate mobilization may be especially important
under stress conditions (6). Varietal differences in carbohydrate
accumulation or partitioning may be related to maintenance of a
high rate of seed filling during periods of temporary
environmental stress when photosynthesis is adversely affected.
Evans (18) considers that whereas photosynthesis during the
storage phase can be an important determinant of yield,
photosynthesis prior to that contributes to the determination of
storage capacity and generates reserves that may be mobilized
during the storage phase.
Gifford et al. (22) reviewing the partition of
photoassimilates and crop productivity, examined the
photosynthetic basis for increasing yield of major field crops in
terms of improving the partitioning of photoassimilates to organs
of economic interest. Although little is known about the
regulation of carbohydrate partitioning between starch storage
(for later utilization) and sucrose synthesis (for immediate
export), they afirm that sink demand plays a very important role.
The partitioning of photosynthetically fixed carbon is important
for plant growth not only because the formation of sucrose
partially determines the carbon export from photosynthesizing
leaves, but also because leaf starch is mobilized to sucrose when
current photosynthesis is low relative to sink demand for
assimilates. In their discussion they suggest that photosynhtesis
and the mechanism of phloem loading determines the amount of
photosynthetic assimilate available fro translocation, while the
mechanism and kinetics of unloading into competing sinks
determines the partition of loaded materials.
Carbohydrates reach a maximum concentration in the plant's
vegetative parts around flowering time, after which they start to
decrease. Yoshida (58) found that stored carbohydrate could be
translocated into the rice grain, thus contributing to grain
filling, or it could be consumed as a substrate for respiration.
The carbohydrate loss from the vegetative parts during grain
filling provides only a maximum estimate of the contribution of
the stored carbohydrates to the grain.
Evidence that stored carbohydrate can be translocated into
the grain has been obtained for rice and wheat by labeling the
stored carbohydrate with 14c. Cock and Yoshida (11) showed that
under normal field conditions 60% of the stored carbohydrate was
translocated into the grain. When photosynthesis during the
ripening period is restricted by shading or defoliation, the
stored carbohydrate appears able to support the grain growth of
rice and corn at almost the normal rate for some time. Perhaps
the stored carbohydrate can serve as a buffer to support normal
grain growth despite weather fluctuations (58). Whether the yield
capacity or assimilate supply limits the grain yield is not
clear. However, defoliation and shading experiments in rice at or
after heading clearly demonstrate that impaired photosynthesis
during the ripening period can severely limit the grain yield
(48). Assimilate supply may limit grain yield under stress
conditions: if photosynthetic activity is limited by shading, or
if translocation of assimilates into the grain decreases, a
certain portion of the grains may remain unfilled (58).
Biological Nitrogen Fixation
Beans are a crop of small farmers in much of the third
world, and are often produced on marginal soils deficient in
nitrogen. In a world of rising fertilizer prices, the need for
cultivars with improved ability to fix nitrogen is especially
important. The identification of genetic variability for
biological nitrogen fixation (BNF) in beans has made selection
for enhanced BNF possible (5).
Graham (25) suggested that at least three factors could
contribute to the variability in N-fixation observed in
P.vulgaris : a) supply of carbohydrates to the nodules, b)
relative rates of nitrogen uptake from soil, and c) time to
flowering. Hardy and Havelka (29) indicated that the amount of
photosynthate available to the nodules may be the most
significant factor limiting N-fixation. They examined factors
that affect photosynthate availability to the nodule such as
light intensity, size of photosynthetic source, competitive
sinks, CO2 enrichment and photosynthate translocation. With
respect to the effect of variation of each of these parameters on
N-fixation in soybeans, N-fixation correlates directly with the
amount of photosynthate available to the nodule. Nodules in
general maintain low reserves of readily utilisable carbohydrates
relative to their requirements for fixation, so they probably
rely for their growth and functioning on photosynthetic products
currently translocated from the leaves, or on carbohydrate
reserves mobilized from other regions of the plant (41).
Experimental evidence is consistent with this view, since a very
close relationship has been observed between photosynthesis,
amount of photosynthate and N-fixation. Reducing light or
defoliation decreases fixation, while supplemental light
increases it (7,26,27,29,41,46,47,49) ; pod removal increases N-
fixation (7,27,28) presumably by leaving more photosynthate
available for the nodules. Lawn and Brun (33) indicated that the
decline in soybean nodule activity was associated with the
development of the pods as a competing assimilate sink. The fact
that the decline in nodule activity coincided with the time when
pod growth rate first exceeded total top growth rate is an
indication of mobilization of previously assimilated material
into the pods.
Factors, both genotypic and environmental, which tend to
lessen competitive effects by enhancing the photosynthetic
source-sink ratio,may be expected to minimize a decline in N-
fixation and should be considered in the future development of
higher yielding varieties with high BNF potential.
In order to understand how varietal differences in N-
fixation might relate to carbohydrate supply and availability in
the bean nodule, Graham and Halliday (23) planted fourteen
commercial varieties, inoculated and sampled at initiation of
fixation and at the beginning of decline in fixation rates.
Marked varietal differences were found, and a highly active N-
fixing variety (P590) showed a higher soluble carbohydrate
percentage in all organs and also partitioned more of its total
carbohydrate to the nodule as compared to an inactive N-fixing
line (P635). In this study, climbing varieties which had been
previously reported to be good N-fixers (24) were found to hold
more of their carbohydrate in the soluble form.
The ontogenetic development of four dry bean cultivars with
reference to the relationships that may exist between symbiotic
nitrogen fixation and the energy supply (in the form of
carbohydrates ) to the nodules was studied by Martinez (38). His
data are consistent with the hypothesis that carbohydrate supply
to the nodules limits fixation. He showed that an increase of
total photosynthate available to the symbiotic system, achieved
through C02 enrichment, resulted in higher rates of nitrogen
fixation. The nitrogen in the bean plant is stored temporarily in
the leaves, and it is suggested that mobilization of this
nitrogen to the seeds results primarily from leaf aging. Martinez
(38) showed a similar phenomenon of mobilization of carbohydrates
temporarily stored in the stems and leaves.
Wilson et al. (56) performed experiments to study the
nonstructural carbohydrates, the nitrogen content of plant
tissues and the nitrogenase activity throughout the development
of male sterile and male fertile soybean plants. Male sterile
plants set approximately 85% fewer pods than the male fertile
plants, and reduced pod set was found to increase carbohydrate
accumulation in the leaf and root systems. Although roots of male
sterile plants contained greater quantities of carbohydrate, a
decline in nitrogenase activity occurred after flowering. The low
percentage of soluble carbohydrates in roots of either type (male
sterile and male-fertile) during the pod filling stage might be
one of the many possible explanations for the similar trends
observed in male sterile and male fertile nodule activity.
In efforts to increase N-fixation it is not necessary to
restrict selection only to genetic factors that affect
nodulation, increase nitrogenase activity or generate larger
amounts of accumulated nitrogen. Certain genotypes may be
superior to others in their allocation of assimilatory resources
to the various plant parts (27,29,41) The functional economy of
whole plants and the interactions of their organs during growth
should be considered in order to determine the plant factors that
are responsible for the variation in nodulation and nitrogen
fixation (57).
Effective photosynthate partitioning
The selection of cultivars with more effective partitioning
of nitrogen and carbon assimilates to the reproductive organs
than older cultivars was thought to be the key factor for the
improvement of yield in other crops, namely rice (58), peanuts
(16) and cotton (55).
Genotypic variation in carbohydrate and nitrogen
remobilization during periods of environmental stress when
photosynthesis is adversely affected, may enable maintenance of a
high rate of seed filling and may buffer and stabilize yields.
Photosynthate partitioning has been shown to be under genetic
control in cereals (15), soybeans (31) and sugar beets (45). In
beans, Adams et. al. (4) showed genetic variation for
accumulation of starch during reproductive development. Izquierdo
(30) also showed that differences in sugars and starch (
total nonstructural carbohydrates ) and nitrogen were associated
with cultivars and physiological stages over the entire
reproductive growth period. Izquierdo (30) showed genetic
variation of seed filling parameters (rate and duration) in this
crop and the relationship of these parameters to patterns of
assimilate partitioning among genotypes. He concluded that yield
differences among cultivars are more associated with the length
of the seed filling period than with the rate of seed growth.
Constable and Hearn (12) performed a series of experiments
with sorghum and two soybean varieties (Ruse and Bragg) under two
different water treatments. Sorghum and Ruse soybean showed a
significant (17-25%) loss in stem dry weight during grain filling
under both treatments. In Bragg soybeans, only the stressed
plants had a loss in stem dry weight during grain filling. One
can infer that in sorghum and in Ruse, the significant loss in
stem dry weight during grain filling could have been a
consequence of relocation of dry matter from the stem to the
developing grain. This agrees with Yoshida's conclusion (58) that
the weight loss from vegetative parts during grain filling sets
an upper limit to the possible contribution of stored
carbohydrates to the grain. An apparently large difference
between soybean cultivars in the effect of water treatment on the
contribution of stem storage to yield was reported by Constable
and Hearn (12); in cultivar Ruse an estimated 25% of grain dry
weight could have come from the stem, while in Bragg only the
stressed plants appeared to use stem reserves. This suggests that
Bragg was sink limited and had little requirement for storage
carbohydrates, except during stress. Rawson et al. (42)
substantiate Constable and Hearn's conclusion that when water
deficits restrict current photosynthesis during grain
development, the plant may buffer yield by drawing heavily on
reserves. Also Egli and Leggett (17) have suggested that
soybean seed growth rates are not closely related to rates of
photosynthate production because storage carbohydrate acts as a
buffer between seed growth and photosynthesis.
Evidence supporting the idea that plant growth rate and seed
yield are not directly affected by total photosynthesis was
reported recently by Ford et al. (21). They used soybean lines
divergently selected for rates of 14C02 uptake per unit leaf area
and tested the effect of this divergent selection for leaf total
photosynthesis on crop growth rate and seed yield. Their data
showed that selection for improved photosynthesis per unit area
did not necessarily enhance seed yield.
The effects of drought on nodulation and nitrogen fixation
in field grown cowpeas were studied by Zablotowicz et al. (59).
The nodulation process was inhibited by drought and maximum
nodulation was observed at mid-pod fill in the drought regime
while plants from the well watered regime showed maximum
nodulation at early flowering. As the plants matured beyond mid-
pod fill,there was no significant difference in nodule mass
between water treatments. Droughted plants failed to form nodules
of high nitrogenase activity during the early stages of
development, and at the reproductive stages the N-fixation
capacity of the crop decreased, probably because there was
insufficient carbohydrate to support high activity at this stage.
Field canopies of two semi-dwarf wheat genotypes were
subjected to water stress that caused visible wilting during the
grain filling stage, and the distribution of photosynthesis
within canopies and the patterns of translocation of labeled
assimilates following 14c02 uptake were determined (32). In
stressed plants 24 hours after labeling, 46% of the 140 was
found in the grain compared to 35% in the control plants. Of the
total 14C recovered from the shoots at maturity, 83% was found in
the grain of stressed plants and 69% in control plants. The lower
percentage of 140 in grain of control plants at maturity was due
to its accumulation in stem segments, primarily in the form of
structural carbohydrates. Fischer and Turner (20) stated that
water stress during seed filling has its major effect upon
current assimilation through reductions in assimilatory activity
and assimilatory surface. They concluded that water stress not
only increases the proportion of current assimilate translocated
to the seed, but also may increase the contribution from
assimilate stored prior to seed filling.
In the broad bean Vicia faba N-fixation has been found
to be severely suppressed once flagging of the lower leaves has
commenced (46). If flagging of the lower leaves takes place,
photosynthesis is likely to be arrested and since these leaves
are likely to be the main providers of C to the nodules, it is
possible that the first reduction of N-fixation during drought
will be caused by reduction in assimilate supply (41) .
Drought tolerance mechanisms
Plant responses to water stress can be classified broadly
into escape, avoidance or tolerance mechanisms (36,53). Escape
can be achieved through more rapid development and through
developmental plasticity, whereby the coincidence of critical
developmental stages with periods of drought is avoided (34).
Water stress avoidance involves mechanisms either to reduce water
loss or increase water uptake. Water stress tolerance implies the
ability to survive large water deficits and may involve
mechanisms such as osmotic adjustment.
Lawn (34) evaluated the response of four different grain
legumes, soybean (Glycine max), black gram (Vigna mungo), green
gram (Vigna radiata) and cowpea (Vigna unguiculata) to water
stress under field conditions. These four legume species
responded to the stress in several ways, but the degree of
expression varied substantially between species. Each cultivar
exhibited some tendency to escape through faster development in
response to stress; the effect was small in soybean and large in
the Vigna cultivars, particular in the flowering to maturity
period. Each cultivar also exhibited to some degree two
mechanisms which served to avoid dehydration by reducing plant
water loss. The most important of these was stomatal control of
water loss in response to declining leaf water potentials, for
which there appeared to be substantial differences in response
between cultivars. Finally under stress each cultivar showed
some paraheliotropic leaf movement ; in these studies there was
some suggestion that paraheliotropy helps to lower leaf
temperatures under stress and presumably further restrict water
loss.
Developmental plasticity can be seen as a mechanism that
facilitates the matching of crop growth and development to the
constraints of the environment, especially in terms of minimizing
the occurrence of the critical reproductive phase during drought
periods. Faster development may allow the successful completion
of the plant's life cycle before the existing water supply is
exhausted. Turk et al. (52) growing cowpeas under water
stress, observed that drought resulted in earliness when present
at moderate levels, but severe drought delayed reproductive
activity. This provides the plant with two possible adaptive
responses. Under moderate drought the plant produces early pods
which may mature before the soil water is depleted. If there is
severe drought at early flowering, the plant remains in a
vegetative stage but has the ability to continue reproductive
activity if water is supplied. Determinate types flower, whether
water levels are optimum or not, while indeterminate types remain
in a vegetative stage under adverse conditions. Once rains start,
the latter enter into the reproductive phase while the former can
start a whole new cycle. It can be speculated that more
determinate cultivars may have less capacity for recovery after
mid-season drought.
Leaf movements which orient the leaf parallel to the sun's
rays, leaf flagging, and rolling are common features of response
in dry situations especially once leaf water potential begins to
fall (20). It is unknown whether these leaf movements are
beneficial to the plant. Shackel and Hall (43) considered that
leaf movements in cowpeas could substantially reduce heat load
and water deficits in cowpeas by minimizing transpiration. On the
other hand Lawn (34) states that paraheliotropic leaf movements
act to reduce total light interception by the canopy, implying a
reduction in photosynthesis. Recently, Ludlow and Bjorkman (37)
reported that the paraheliotropic movement of water stressed
Macroptilium atropurpureum cv. Siratro protected the primary
photosynthetic reactions from damage by excess light
(photoinhibition), heat, and the interactive effects of excess
light and high leaf temperatures. They concluded that even though
heat damage is more severe when it occurs, photoinhibition may be
a more common phenomenon during drought, unless paraheliotropic
leaf movements reduce the amount of solar radiation incident on
water stressed Siratro leaves.
The importance of root morphology for maintaining a supply
of water to the plant should not be overlooked. Under drought
conditions, an extensive root system is a characteristic that
enables the plant to exploit a higher proportion of the available
soil water without incurring severe plant water deficits (8).
Deeper root penetration of soybean was particularly evident in
the drought periods (Lawn, 34). He suggested that perhaps this
root system is related to the tendency of soybeans to keep
stomata open longer into the drought periods, thus maintaining a
supply of photosynthate for continued root growth.
No definite conclusions can be reached about the "best"
strategy to overcome drought stress. However one can conclude
that there is no absolute character to "drought resistance".
Rather, there are several alternative and perhaps inter-related
mechanisms, and their relative success depends on the seasonal
pattern of water availability, on soil type and depth, and on
other factors. The most appropriate strategy for a particular
environment presumably will be the one that simultaneously
maximizes production and minimizes risk in that environment.
Identifying the appropriate strategy requires assessment of the
probability of particular seasonal patterns of water availability
for that particular environment.
One effective approach to breeding for higher yield under
stress would be to identify physiological and morphological
components causing varietal differences in economic yield in the
presence and absence of stress, and to gain an understanding of
their genetic control. Evidence indicates that genetic
variability exists for all such components (54). If physiological
genetic data are used in selecting parents, it should be possible
to select directly for yield, using standard selection and
breeding procedures. Knowledge of the physiological genetics of
yield will improve the plant breeder's understanding of
desirable plant types and habit, and appropriate selection and
breeding methods can then be used. As world food demand
22
increases, production of drought tolerant beans may become
increasingly important to make optimal use of water-limited
lands.
MATERIALS AND METHODS
An experiment was conducted at the Campo Agricola
Experimental de Iguala-CAEIGUA-, Iguala, Mexico. The station is
located in the state of Guerrero, at the meridian 990 45'
longitude West and the parallel 180 30' latitude North. The
altitude at the station is 739 meters above sea level. The
average minimum temperature is 7C and the average maximum is 42
C, with an annual average of 24.5 The average annual
precipitation is 1155 millimeters, and the rainy season starts
during the last part of June. Eighty percent of the total annual
precipitation occurs between the months of May and October. The
experiment was planted in the second week of December, and the
final harvest was taken in the last week of March. Precipitation
and temperatures were recorded during the course of the
experiment, and are shown in Figure 1.
The experimental plots were on a silty clay soil, with a
high alkaline pH that varied between 8.25 and 8.75. The organic
matter content as well as total nitrogen were low the
percentage of organic matter being 1.05 and the total nitrogen
0.112 ppm. The levels of potassium, calcium and magnesium were
high, but the phosphorus content was relatively low (10.22 ppm).
Before planting, 40 kg. of phosphorus per hectare were applied to
all plots and 40 kgs. of nitrogen per ha. (in the form of Urea)
were applied to half of the plots. At planting time .all plots
were inoculated with a commercial granular Rhizobium inoculant,
NITRAGIN L x 441 for dry beans, obtained from the Nitragin
35 12a
25
Temp. C .= .
185
0 20 40 60 80 110
Days After Planting
Figure 1. Maximum and Minimum Dally Temperatures. Iguala, 1982-3.
a=Mllltmetere of rain
Company in Wisconsin ; 1.5 gms. of inoculant per meter of row
were applied. Twenty one dry bean genotypes were selected on the
basis of their performance under drought as well as on their
nitrogen fixation capabilities. They included:
a) three good nitrogen fixing lines from the University of
Wisconsin: 23-61, 21-58, and 21-54.
b) five CIAT lines reported to have some tolerance to drought:
BAT 332, BAT 85, BAT 47, A-162, and BAT 798.
c) seven Mexican lines with some tolerance to drought: Pinto
Nacional 1, Durango 222, Ojo de Cabra, Bayo Madero, C-5, 1213-2,
and LEF-2-RB.
d) two Michigan State varieties with good architecture and high
yielding potential: Neptune and 61065.
e) two Michigan State lines that showed leaf flagging under
severely dry conditions and were high yielders: 81017 and 800122.
f) two Michigan State lines that showed leaf flagging under
severe dry conditions and were poor yielders: 790131 and 800205.
A Tepary bean, Phaseolus acutifolius was also planted.
Each plot consisted of 6 rows 4 meters long; the distance
between rows was 75 cms. and the distance between plants within a
row was 10 cms. Two empty rows were always left between adjacent
plots in order to facilitate water management. The experimental
units were arranged in a split plot design with three
replications. The combination of nitrogen source and water level
was the whole plot factor and cultivars were the split plot
factor.
All plots were flood-irrigated every two weeks starting
before the planting day, until flowering time. Individual plots
of each cultivar were treated as separate units for water
management. After flowering, only the so-called "plus" water
plots continued to receive water.
A commercial micronutrient foliar spray was applied 42 and
50 days after planting. Insects were controlled by spraying once
a week with available commercial insecticides. Two center rows of
each plot were used for periodic sample collection, two were used
for final harvest, and the two outer rows were discarded.
Flowering notes were recorded and when 50% flowering was reached,
the first sample was taken. Ten of the twenty two planted
cultivars were selected for detailed sampling. This selection was
based on previous information regarding differences in N-fixation
potential and drought tolerance. The 10 cultivars chosen for more
detailed study included 8 drought tolerant lines ( 4 from Mexico,
2 from CIAT and 2 from MSU ), one good N-fixing line from
Wisconsin and one drought susceptible line from MSU.
Each sample consisted of five plants that had uniform
competition, they were dug up trying to get as much of the roots
as possible. Each sample was separated into stems, roots and
leaves; this material was placed in an oven at 800 C for one hour
and then was left out in the sun for completion of drying. After
dry weights were recorded, the tissue from each sample was ground
in a Wiley mill and saved for starch and soluble sugars
determinations. At the same time a 2-plant sample was taken
(plants were kept entire); after drying they were ground and
saved for total Kjeldahl nitrogen determinations. The second
sample was taken 15 days after flowering, the time when the
stress was expected to become effective. The third and last
sample was taken at physiological maturity. The sampling
procedures for the second and third samples were the same as for
the first sample, except that in the last two samples pods were
also separated.
Additional observations and notes such as occurrence of leaf
flagging, leaf dropping and leaf yellowness were recorded. A
leaf dropping scale from 1 to 5 was adapted, where 1 was no
defoliation and 5 was complete defoliation. Scores for each plot
were taken 85 days after flowering (before physiological
maturity was reached). At harvest time both economic and
biological yield were recorded and the Harvest Index was
calculated.
A random sample of 10 plants was taken at harvest time to observe
if there were any differences for seed weight and seed number
between the plant's upper and lower pods. For this purpose the
two lowest pods as well as the two highest pods of each sampled
plant were taken and their seed number and weight recorded.
Starch contents were determined with a colorimetric method with
perchloric acid. This technique involved 3 basic steps which are
described as follows.
1. Reagents.
a. Colorimetric Solution: 80 grs. of NaCl disolved in 250 ml.
of distilled water and 750 ml. of ethanol.
b. Diluted perchloric acid: 300 ml. of perchloric acid (70%)
and 224 ml. of distilled water.
c. Potassium Iodine: 20 grs. of KI dissolved in 20 ml. of
distilled water, and 2 grs. of Iodine diluted to 1 It. with
distilled water.
2. Calibration Curve.
One gram of starch is dissolved in 10 ml. of perchloric acid
solution and diluted to 100 ml. with distilled water. Aliquotes
of 2,3,4,5,6,7,8,9and 10 ml.are taken and 0.5 ml. of perchloric
acid solution are added and diluted to 100 ml. with distilled
water. The solutions will have 20,30,40,50,60,70,80,90 and 100
mgms./ml. of starch. To 5 ml. of each solution 4.5 ml.
of distilled water and 0.5 ml. of KI solution are added. The
final concentrations being attained are then
100,150,200,250,300,350,400,450 and 500 ,/gms. of starch in 10 ml.
of solution. After 20 minutes, absorbance is read at 600 nm.
with the spectrophotometer. The calibration curve is plotted
calculating the mgs. of starch that correspond to one unit of
absorbance (F).
3. Determination of starch in the sample.
Fifty mgs. of dry ground sample (duplicate samples) are taken;
after centrifugation, 12.5 mls. of the colorimetric solution are
added. The sample is placed in a water bath at 720C for 10
minutes, and centrifuged for 10 minutes at 2000 rpm. The
supernatant is discarded and to the residue 5 mls. of perchloric
acid solution are added. After letting the sample stand for 10
minutes, 5 mls. of distilled water are added and then it is
centrifuged at 15000 rpm for 20 minutes. An aliquot of the
supernatant is taken and the color is developed as in the
callibration curve. Then, absorbance is read at 600 nm. The
amount of starch is expressed as grs./100 grs. of sample, and is
calculated using the following formula:
% of starch =
A x F x V x D
a x m x 10
where A = absorbance
F = absorbance factor taken from the curve
V = volume
D = dilution
a = aliquot
m = sample weight
10 = conversion factor
RESULTS
I. Water Effects
The objective of this experiment was to determine the effect
of drought stress imposed during the latter part of the seed
filling period on the yield performance of 22 different
cultivars and to relate their performance under stress with the
ability to translocate non-structural carbohydrates. To
accomplish this objective, the different genotypes were irrigated
every two weeks from planting until anthesis. Working under the
assumption that with high temperatures and high solar radiation
the potential evaporation was high, we expected the irrigation
water to be depleted at about two weeks after it was added. Based
on these assumptions, withholding the water at anthesis
presumably would cause an effective stress in the middle of the
seed filling period, defined as 2 weeks after flowering. The
control plots were continuously irrigated every two weeks
throughout the entire growing season. Different cultivars were
treated independently, meaning that each plot was considered as a
separate unit for irrigation purposes.
The first evident symptom of water deficit was premature
defoliation; it started to occur two weeks after the plants were
expected to be under stress. A two-week lapse between the time
that we had intended to have the stress and the first visible
signs of stress might indicate that we did not actually impose
the stress at the physiological stage that we had originally
intended. However, we can not assure that the plants were not
under stress before this time because measurements that would
have indicated that, such as stomatal closure, osmotic adjustment
and photosynthesis reduction, were not taken. Another indication
of the presence of the water stress in the crop consisted in the
reduction of the length of the seed filling period ( days from
flowering to physiological maturity ), in the water stressed
plots as compared to the irrigated plots. Perhaps a faster
development allows the completion of the reproductive stage
before soil water is completely exhausted.
It is evident that we did have a water stress, but what we
can not assure is the degree of the stress or its precise timing.
The degree of correlation between control and stress yields has
been considered to be an indication of the severity of the stress
( 9,10). A mild drought stress reduces yield, but the grain yield
of the stressed plots is highly correlated with the yield
potential in the absence of the stress. Severe stress provokes
very different responses among genotypes with similar yield
potential, and the correlation between grain yield under stress
and yield potential is weaker. Since in this case the correlation
of control yield vs. stress yield was found to be positive and
highly significant ( calculated r= 0.895 ), we can infer that
the stress was moderate rather than severe.
Since there were not significant effects of N-treatment, the
water effects described in here are based on both the plus and
minus N treatments.
A. Biological Yield
A significant cultivar effect as well as a significant water
effect were indicated by the Analysis of Variance. Twelve of the
twenty two cultivars had a significant reduction of Biological
Yield under water stress, while only two cultivars, MSU 800122
and Mexico LEF-2-RB, showed a significant increase for this trait
under stress (Table 1). The other eight cultivars didn't show any
significant differences between treatments, but except for
cultivars MSU 61065 and CIAT BAT 332 Biological Yield was
reduced under water stress. In the case of 800122 and LEF-2-RB we
have no evidence that will allow a reasonable explanation.
The size of the biological yield reduction in some cultivars was
unexpectedly high, considering that the stress was not effective
until late in the season. In fact, in cultivar Bayo Madero this
reduction was more than 50%.
Differences in magnitude for Biological Yield were observed;
the high values of 81017, Ojo de Cabra and Bayo Madero contrast
with the relatively low values of Durango 222 and Pinto
Nacional. It is interesting to note that entry 22, the Tepary
bean, P. acutifolius known to be a drought tolerant line,
showed no difference in Biological Yield between the plus and
minus water treatments.
Table 1.
Biological yield (kg/ha) under two water
treatments. Iguala, 1982-3.
Entry No. Identification Irrigated Stress
Wisec 23-61
Wisec 21-58
Wisc 21-54
Neptune
61065
800122
81017
800205
790131
LEF-2-RB
1213-2
C-5
Bayo Madero
BAT 332
BAT 85
BAT 47
A-162
BAT 798
Pinto Nacional 1
Durango 222
Ojo de Cabra
Tepary
4945
4674
4464
5088
3908
4305
6050
4233
4021
4360
3853
3277
7592
3862
4049
4263
4551
3973
3187
3642
5819
3888
3443 **
3761 **
3536 **
4254 **
4012
5280 **
4308 **
3728
3457
5298 **
2951 **
2515 *
3371 **
4044
3482
3980
3761 *
3019 **
2316 **
3288
3827 **
3875
* = LSD at .10 (672)
** = LSD at .05 (802)
B. Economic Yield
The Analysis of Variance revealed a significant cultivar
effect and also a significant water effect at the 5% level. Only
six of the twenty two entries showed a significant yield
decrease under the water stress conditions, as compared to the
non stressed plots (Table 2). Of these six cultivars, four
were Mexican lines ( 1213-2, C-5, Bayo Madero and Pinto Nacional
1) and two were MSU lines ( 800122 and 800205 ). Among the other
sixteen entries, ten showed a non-significant decrease in
Economic Yield under stress conditions and six had a non-
significant yield increase.
Water stress, when imposed during the seed filling period,
seemed to have a greater effect on the economic yield of the
cultivars that retained more assimilates in the stems at
maturity. A significant negative correlation of 0.443 between
economic yield and stem % of dry weight at Physiological Maturity
under water stress supports this assertion. Inherently low
yielding genotypes such as 790131 and Durango 222 displayed a
decrease in stem weight only under stress at P.M.; under non-
stress stem weight was not reduced. The correlation between stem
% of total dry weight and economic yield for both cultivars was
non-significant under the plus water treatment, while under water
stress for both 790131 and Durango 222 there was a significant
negative correlation, with values of 0.850 and 0.975
respectively. Even though these are low yielding genotypes, their
economic yield was not significantly reduced under stress. This
Table 2.
Economic yield (kg/ha) under two water
treatments. Iguala, 1982-3.
Entry No. Identification Irrigated Stress
1 Wise 23-61 1536 1500
2 Wise 21-58 1887 1743
3 Wise 21-54 1330 1440
4 Neptune 1554 1673
5 61065 1815 1938
6 800122 1571 1178 **
7 81017 2086 2001
8 800205 1822 1487 **
9 790131 1225 1064
10 LEF-2-RB 1964 .1775
11 1213-2 1514 1200 **
12 C-5 1502 1181 **
13 Bayo Madero 1183 742 **
14 BAT 332 1779 1919
15 BAT 85 1966 1751
16 BAT 47 1843 1780
17 A-162 1833 1744
18 BAT 798 1081 1044
19 Pinto Nacional 1 1262 889 **
20 Durango 222 1102 985
21 Ojo de Cabra 628 693
22 Tepary 1521 1484
* = LSD at 0.10 (217)
** = LSD at 0.05 (259)
is consistent with the hypothesis that remobilization is
enhanced under stress conditions.
Bayo Madero and 800122 incurred significant reductions in
economic yield under stress as well as a significant increase in
stem dry weight suggesting either a poor remobilization and a
low capacity to buffer adverse environmental effects, or a weak
sink that does not have the ability to utilize the stored
assimilates. 800122, a late maturity cultivar, did not flower
until late in the season; for this reason as shown in Figure 2,
during the reproductive stage it was subjected to high
temperatures. I believe that the high temperatures during this
stage of development kept this particular variety from
remobilizing and senescing normally, and as a consequence yield
was significantly affected.
The three Wisconsin cultivars 23-61, 21-58 and 21-54, as
well as Neptune A-162 BAT 85 and Ojo de Cabra had a
significant reduction in biological yield under stress, however,
their economic yield was not significantly reduced This is
supportive of the hypothesis that in those genotypes that have
the ability to remobilize assimilates from stems ,storage
photosynthates act as a buffer between seed growth and
photosynthesis. However, we can not determine if the contribution
to seed yield is coming mainly from assimilates that were
produced before the plants were subjected to the water stress, or
if the seeds were filled with photosynthates produced after the
onset of stress.
C. Harvest Index
The AOV reveals a significant cultivar and water effect on
Harvest Index (H.I.) at the 1% level. Seven entries showed a
significant increase in H.I. under water stress as compared to
the non-stressed plots (Table 3). They include the Wisconsin
lines 23-61 and 21-54, the MSU lines Neptune and 81017 the
Mexican lines Ojo de Cabra and Bayo Madero and the CIAT line BAT
798. Significant reductions of biological yield under stress
account for differences in Harvest Index for these cultivars. In
the case of Bayo Madero, reductions in both economic and
biological yield occurred.
Two lines, MSU 800122 and Mexico LEF-2-RB, had a significant
reduction in HI under the minus water treatment. Nine out of the
thirteen remaining lines had a non-significant HI increase under
stress, while the other four had a non- significant decrease. It
is interesting to note that the Tepary bean (entry 22) had almost
the same value for HI under both stress and non stress
conditions.
D. Seed Size
The AOV for this trait, measured as weight of 100 seeds,
reveals a significant effect for cultivars and water treatment.
All entries incurred a reduction in single seed weight of about
one centigram due to stress (Table 4). However, only three out of
the twenty two entries had a statistically significant reduction
Table 3. Harvest Index under two
treatments. Iguala, 1982-3.
water
Entry No. Identification Irrigated Stress
1 Wisec 23-61 0.33 0.43 **
2 Wisec 21-58 0.41 0.46
3 Wisec 21-54 0.31 0.41 **
4 Neptune 0.32 0.40 **
5 61065 0.47 0.48 **
6 800122 0.39 0.22 **
7 81017 0.36 0.46 **
8 800205 0.42 0.42
9 790131 0.30 0.32
10 LEF-2-RB 0.44 0.37 **
11 1213-2 0.40 0.41
12 C-5 0.43 0.47
13 Bayo Madero 0.15 0.22
14 BAT 332 0.45 0.46
15 BAT 85 0.49 0.51
16 BAT 47 0.39 0.44
17 A-162 0.41 0.46
18 BAT 798 0.27 0.34 **
19 Pinto Nacional 1 0.39 0.38
20 Durango 222 0.30 0.30
21 Ojo de Cabra 0.15 0.24 **
22 Tepary 0.39 0.38
* = LSD at 0.10 (0.60)
** = LSD at 0.05 (0.70)
Table 4.
Weight of 100 seeds (grs) under
treatments. Iguala, 1982-3.
two water
Entry No. Identification Irrigated Stress
Wisec 23-61
Wisec 21-58
Wisc 21-54
Neptune
61065
800122
81017
800205
790131
LEF-2-RB
1213-2
C-5
Bayo Madero
BAT 332
BAT 85
BAT 47
A-162
BAT 798
Pinto Nacional 1
Durango 222
Ojo de Cabra
Tepary
18.25
20.93
20.05
16.42
18.28
16.36
20.60
16.68
23.11
25.10
35.96
34.65
43.60
16.88
20.78
26.78
19.51
20.38
35.63
50.58
39.55
11 .96
17.41
20.08
19.16
15.93
17.71
15.68
19.07
15.76
22.01
24.05
34.71
33.51
41.48 **
16.20
19.58
26.10
19.23
19.16
33.10 **
45.07 **
39.45
11.48
* = LSD at 0.10 (1.55)
** = LSD at 0.05 1.85)
in the weight of 100 seeds under the water stress treatment.
These three entries were all Mexican lines (Bayo Madero, Pinto
Nacional and Durango 222).
Variations in economic yield due to water stress in the
cultivars Bayo Madero and Pinto Nacional were due to reduction
of seed size as well as seed number. Smaller seeds under a low
water regime are the consequence of incomplete filling,
indicating lower photosynthesis and/or an inadequate reallocation
of carbohydrates during the seed filling process.
E. Seed Number
The AOV for the number of seeds per square meter indicated a
significant cultivar effect as well as a significant cultivar -
water interaction, but the water effect itself was not
statistically significant. Since the water stress did not become
effective until the late part of the growing season, when the
number of seeds had already been determined, no water treatment
effect is to be expected for this trait. Only two of the twenty
two entries had a significant reduction in the number of seeds
under the water stress treatment, while three had a significantly
larger number of seeds under stress (Table 5) The three
cultivars that had a significant increment in the number of seeds
(Neptune, 61065 and BAT 332) had a reduction of seed weight under
stress. This might be an indication of component compensation, in
which the reduction of seed weight is caused by the increment in
seed number. Although the reductions in seed number for Bayo
Table 5. Seed number (seeds/mt2) under two water
treatments. Iguala, 1982-3.
Entry No. Identification Irrigated Stress
1 Wisec 23-61 842 862
2 Wise 21-58 902 868
3 Wisec 21-54 663 752
4 Neptune 946 1051 *
5 61065 994 1108 *
6 800122 957 787 **
7 81017 1013 1049
8 800205 1093 944 *
9 790131 534 474
10 LEF-2-RB 732 735
11 1213-2 421 347
12 C-5 434 353
13 Bayo Madero 269 178
14 BAT 332 1052 1181 **
15 BAT 85 949 894
16 BAT 47 688 682
17 A-162 940 907
18 BAT 798 531 545
19 Pinto Nacional 1 354 269
20 Durango 222 221 216
21 Ojo de Cabra 159 176
22 Tepary 1272 1293
= LSD at 0.10 (101)
** = LSD at 0.05 (121)
42
Madero and Pinto Nacional were not statistically significant,
they can be considered large enough to explain the economic yield
loss observed under water stress.
F. Length of Vegetative and Reproductive Stages
The AOV for the length of the seed filling period, measured
as the number of days between 50% Flowering and Physiological
Maturity, showed a significant cultivar effect as well as a
significant water effect. All cultivars, without exceptions,
incurred a reduction in the length of the seed filling period
under the water stress treatment (Table 6). However, these
reductions turned out to be significant only for 15 of the 22
entries.
Variability in seed filling duration between cultivars and
between treatments within cultivars is shown in the data. Since
seeds compete for available assimilates, if the sink capacity at
the initiation of the seed filling stage is higher than the
source supply, extending the duration of seed filling under non-
stress conditions theoretically should provide more available
photosynthate to the seed which in turn will produce heavier
seeds and higher yields.
Earliness has been associated with improved adaptation in
crops subjected to drought probably as a mean of escaping
the stress through faster development. If this reduction in the
length of the reproductive stage under adverse conditions induces
an earlier partition of carbohydrates to the seeds, early
genotypes should be able to buffer adverse environmental effects
Table 6. Days between flowering and physiological maturity
under two water treatments. Iguala, 1982-3.
Entry No. Identification Irrigated Stress
1 Wisec 23-61 37 33 **
2 Wisc 21-58 38 34 **
3 Wisec 21-54 37 34 **
4 Neptune 37 34 **
5 61065 35 34
6 800122 34 34
7 81017 41 38 **
8 800205 37 36
9 790131 30 28 *
10 LEF-2-RB 39 36 **
11 1213-2 42 40 *
12 C-5 45 45
13 Bayo Madero 52 48 **
14 BAT 332 30 29
15 BAT 85 33 33
16 BAT 47 41 40
17 A-162 41 38 **
18 BAT 798 45 33 **
19 Pinto Nacional 1 41 39 *
20 Durango 222 50 47 **
21 Ojo de Cabra 52 49 **
22 Tepary 44 41 **
= LSD at 0.10 (2)
** = LSD at 0.05 (
in a more efficient manner. This is supported by the data shown
here where the cultivars Ojo de Cabra, Bayo Madero, Durango 222
and C-5 were among the lowest yielding lines and had the longest
seed filling period under stress. All the top yielding cultivars
81017 61065, LEF-2-RB, BAT 85 and BAT 332 had a relatively
short reproductive phase under water stress.
High temperatures during the reproductive stage as occurred
in Iguala (Figure 2), probably reduced the length of the period
from Flowering to P.Maturity by inducing an earlier partition
of carbohydrates that, in turn, hastened leaf senescence.
Developmental plasticity helps plants cope with an adverse
environment, especially in terms of delaying or accelerating the
onset of the reproductive phase so as to escape the more severe
periods of adversity. Faster development allows the completion of
the reproductive stage before the soil water is exhausted.
The length of the vegetative stage,expressed as the number of
days between planting and 50% flowering, is expected to be
closely associated with the accumulation of total dry matter. One
would also expect a positive correlation between Biological Yield
and days to flower in the irrigated cultivars, since the longer
they grow the greater the possibility for accumulation of dry
matter Figure 3 illustrates this point, and shows that only 2
of the 7 cultivars that flowered in less than 50 days after
planting had a high Biological Yield, while all the late
flowering cultivars had a high Biological Yield.
A greater accumulation of dry matter before the onset of
stress might act as a reserve pool of assimilates that can be
drawn upon when photosynthesis is reduced. If pre-anthesis
-M S..L
-'M h
u u
* i
U
Days After Planting
-- Indlcates 50% Flowering Date
3S5
Temp. C
S L
*
20
AI
110
A
M
Ii 11
.1
.~a, &
-a
'4~t~) *
11 ii I
Temp. C 21. 70 Gooo.-
35 "
1000- o
Is 1 4
,25 2
2000 .
0 20 40 60 80 110
Days After Planting
--* Indicates 50% Flowering Date
assimilates are being utilized to fill the seeds, the cultivars
that have a longer vegetative growth would be expected to have a
greater source of assimilates and maybe a greater potential to
buffer adverse environmental effects. Figure 4 shows that the top
yielding cultivars from this experiment are all within the group
that flowered 60 days or more after planting.
G. Leaf Dropping
Significant effects for the water treatment as well as
significant differences between cultivars were detected by the
AOV. All twenty two entries, without exception, had a significant
increase in the leaf dropping score under stress (Table 7), which
simply reveals that plants under stress had early defoliation.
However, it is important to note that different degrees of
defoliation were expressed. The leaf dropping scores under the
irrigated plots ranged between 2.3 and 3.8 (little to moderate
defoliation), while the scores for the water stress plots ranged
from 2.8 to 4.8 (moderate to almost complete defoliation). Under
water stress, varieties such as Ojo de Cabra, LEF-2-RB and
Neptune had corresponding scores of 2.8, 3.1 and 3.3 while BAT
85 1213-2 and C-5 had respective values of 4.8, 4.7 and 4.6.
Defoliation occurred in all cultivars under water stress and its
correlation with economic yield turned out to be a non
significant -0.161.
o s
101
1S eS
T a p U 0 5 0 U
10 0 90 or11
\ 1 Pa000--t
2100
-0-4
14 *5 e O
5 IAO 1000lat i
0 20 40 60 80 110
Days After Planting
-** Indicates 50% Flowering Date
Table 7.
Leaf dropping under two water
treatments. Iguala, 1982-3.
Entry No. Identification Irrigated Stress
1 Wisec 23-61 2.7 3.6 **
2 Wisec 21-58 2.8 3.5 **
3 Wisec 21-54 2.5 3.8 **
4 Neptune 2.5 3.3 **
5 61065 3.0 3.6 *
6 800122 2.0 3.5 **
7 81017 2.5 3.8 **
8 800205 3.0 3.6 **
9 790131 3.6 4.5 **
10 LEF-2-RB 2.6 3.1 *
11 1213-2 3.3 4.7 **
12 C-5 3.8 4.6 **
13 Bayo Madero 2.5 3.8 **
14 BAT 332 2.6 3.5 **
15 BAT 85 3.5 4.8 **
16 BAT 47 2.3 3.3 **
17 A-162 2.3 4.0 **
18 BAT 798 3.1 3.6 *
19 Pinto Nacional 1 3.6 4.8 **
20 Durango 222 3.8 4.5 **
21 Ojo de Cabra 2.3 2.8 *
22 Tepary 2.3 4.1 **
Based on a scale from 1 to 5, where 1= No defoliation and
5= Complete defoliation.
= LSD at 0.10 (0.5)
** = LSD at 0.05 (0.6
H. Plant Dry Weight at Physiological Maturity
Significant differences between cultivars were detected by
the AOV for total plant dry weight. When individual components of
total plant weight expressed as percent of the total dry weight
were examined, not only significant differences between cultivars
were evident, but also the water effects were significant for
stem and leaf % of total dry weight. One must be cautious in
interpreting these data. It is important to remember the
inexplicable increase of total dry matter in the cultivars 800122
and LEF-2-RB under stress (Table 1). One must be aware that the %
values may reveal some trends not shown by the total dry matter
data.
In Table 8, one can see that of the 10 genotypes sampled
only three had significant differences for stem % of total dry
weight between the stress and the non-stress treatments. LEF-2-RB
had a significant reduction of stem weight under water stress ,
while 800122 and Bayo Madero had a significant stem weight
increase under stress. The other seven cultivars sampled showed a
tendency towards decreasing stem weight under water stress, with
the exception of 790131 and BAT 85.
At physiological maturity, the stems of Bayo Madero and
800122 under normal irrigation constituted 22.3 and 27.4 % of
the total plant dry weight, indicating a high accumulation of dry
matter in the stems. Their corresponding values under stress
were 33.3 and 30.0 % ; these figures might indicate that these
two genotypes do not have the capacity to remobilize the stored
Table 8. Stem and Pod % of total dry weight at Physiological
Maturity under two water treatments. Iguala, 1982-3.
Stem %a Pod %b
Identification Irrigated Stress Irrigated Stress
Wisc 23-61 17.27 15.95 70.83 71.63
61065 17.35 17.94 71.90 71.00
800122 27.40 33.39 ** 56.93 51.61 **
790131 17.37 17.46 66.20 70.85 **
LEF-2-RB 18.12 15.31 68.55 74.09 **
1213-2 14.40 13.65 70.09 77.05 **
Bayo Madero 22.33 30.04 ** 65.48 54.60 **
BAT 332 18.10 16.53 69.39 72.61
BAT 85 15.19 16.12 70.96 71.67
Durango 222 16.12 15.84 70.75 72.67
a = LSD at 0.10 (1.86)
** = LSD at 0.05 (2.23)
a = LSD at 0.10 (3.78)
** = LSD at 0.05 (4.52)
assimilates from stems to the seeds. This inability to remobilize
then could be the cause of the significant reduction of economic
yield under stress for both Bayo Madero and 800122. However, this
could also mean that the reduction of seed number under stress,
although not statistically significant, resulted in an
insufficient sink demand to require remobilization.
The high yielding cultivar LEF-2-RB had a different behavior;
under no stress the stem weight corresponded to 18.1 % of the
total dry weight, while under stress it was only 15.3 %. The
changes in dry weight induced by the stress though not large by
these figures, may be enough to sustain the seed filling process
temporarily and as a consequence economic yield under stress was
not significantly reduced.
Pod % of total plant dry weight showed a significant
reduction under minus water in the low yielding cultivars 800122
and Bayo Madero, indicating that reduction in economic yield in
these two cultivars probably was due not only to the decrease in
single seed weight but also to the reduction of seed number.
I. Plant Dry Weight Changes: Remobilization
Figure 5 shows the changes in stem, root, leaf and pod dry
weights over the three different sampling times (Flowering, 15
days after Flowering and P.Maturity) for six different cultivars.
From Flowering to 15 d.a.f. stem weight increased in all
cultivars, however, the size of this increment varied among
cultivars. One can see fairly large increments in cultivars such
I = Flowering
0= Mid-pod-fill
I = Physiological Maturity
* = Stress
100
50-
10o f
Stem
Root
Pod
61065 790131 1213-2 B. Madero BAT 85 Dgo222
Cv.
Fig 5. Stem. Root. Pod and Leaf dry weights
three physiological stages. *lguala, 1982-3.
(grs/mt2) over
U
as 61065 and BAT 85, while 790131,1213-2 and Bayo Madero showed
only a relatively small increase. The changes from 15 d.a.f. to
P.Maturity differed in the sampled cultivars. BAT 85, 61065 and
790131 had a reduction in stem dry weight, while Durango 222,
1213-2 and Bayo Madero had an increase.
When comparing the stem dry weight reduction in the water
stress versus the non-stress plots, one can observe a general
tendency towards a greater reduction in dry weights under the
stress treatment. Durango 222 and 1213-2 incurred a reduction of
dry weight under the stress treatment, while no reduction
occurred under the plus water conditions. Bayo Madero showed an
increase in stem dry weight at P.Maturity with respect to 15
d.a.f., however, the increment was slightly smaller under stress.
BAT 85, as noted before, incurred a reduction in stem dry weight
from 15 d.a.f. to P.Maturity, the reduction being greater under
stress. 790131 had a small reduction from 15 d.a.f. to P.Maturity
but no differences were seen between the stress and the non-
stress treatments.
The high yielding lines BAT 85 and 61065 had the largest
stem weights at 15 d.a.f. Their respective losses from 15 d.a.f.
to P.Maturity might indicate that remobilization of stored
assimilates has taken place.
The low yielding line Bayo Madero, although it had high
values for stem weight at 15 d.a.f., apparently did not
remobilize its stored carbohydrates to the seeds. In the case of
Durango 222 remobilization occurred only under stress. The lack
of remobilization under non-stress conditions was probably due to
the lack of need to utilize the stored carbohydrates because
assimilate demand by the seeds was being satisfied by currently
produced photosynthates. Only when photosynthesis is adversely
affected would the seed filling process depend upon the stored
assimilates and their reallocation .
A significant correlation of 0.394 between root weight at
flowering time and economic yield points out the importance of
the root system in relation to yield. However, one must be
careful when interpreting these results, because a low
correlation even though statistically significant, still leaves a
great deal of yield variance unaccounted for that has to be
explained by other factors. The top yielding cultivars BAT 85 and
61065 had the highest values for root dry weight at flowering,
while the bottom yielding line Durango 222 had the smallest
value. No significant correlations were found between economic
yield and root weight at either 15 d.a.f. or P.Maturity. Large
values of leaf weight at flowering and at 15 d.a.f. represent a
large photosynthetic area and therefore a substantial
carbohydrate manufacturing site. The data show that the highest
values for leaf weight at these physiological stages were
produced by the top yielding lines BAT 85 and 61065. Changes in
leaf weight from 15 d.a.f. to P.Maturity show that under stress
the reduction of leaf weight is greater as compared to the non-
stress values. However, as pointed out before when describing the
leaf dropping results, economic yield and defoliation did not
show a significant correlation.
In general one can observe that the pod dry weight data presented
in Fig. 5, as expected, are in broad agreement with the economic
yield data given in Table 2. Small discrepancies such as the
higher pod weight in BAT 85 which was outyielded by cultivar
61065 (Table 2) might indicate a greater dry weight of the pod
walls in BAT 85 which are not included in the economic yield
data. Nevertheless, these discrepancies are small and not
statistically significant.
J. Starch Analysis
Significant cultivar differences for the amount of starch
present at flowering, 15 d.a.f. and physiological maturity in the
stems, roots and pods were detected by the AOV. Table 9 shows the
starch percentage ( mgrs of starch per gr of dry weight ) for the
different plant components at three physiological stages. The
estimated values are very low as compared to starch
determinations previously reported for dry beans (30,39). These
low values might be the result of the lack of sensitivity of
the method used, starch being determined by the colorimetric
method already described, or from high respiration rates caused
by high temperatures that prevailed during the growing season.
Figures 6 and 7 illustrate the changes in starch content in the
stems and pods over the 3 sampling times. One can see that the
starch content is always lower under the stress treatment as
compared to the irrigated plots. The AOV revealed significant
cultivar and treatment differences for the amount of starch
present in the stems at physiological maturity. Figure 6
illustrates the seasonal variation in starch content in the
stems. Bayo Madero was the only cultivar that had an increment in
Table 9. Mean values of starch ( mgrs / gr dry wt. ) at three
different physiological stages under two water
treatments. Iguala 1982-3.
Identification Stage* Stems
Irrigated Stress
Wisconsin 21-34
61065
800122
790131
LEF-2-RB
1213-2
Bayo Madero
BAT 332
BAT 85
Durango 222
F
MPF
PM
F
MPF
PM
F
MPF
PM
F
MPF
PM
F
MPF
PM
F
MPF
PM
F
MPF
PM
F
MPF
PM
F
MPF
PM
F
MPF
PM
20.8
8.6
21.4
25.0
7.8
18.7
8.4
32.8
27.9
20.5
39.0
19.9
36.8
124.5
40.0
21.4
63.7
56.4
36.7
12.5
40.5
49.5
7.3
18.6
58.1
41.4
7.1
4.4
8.4
7.4
13.7
24.5
50.2
7.5
4.9
25.7
* F= flowering
MPF= mid-pod filling
PM = physiological maturity
Irrigated
- -- Stress
3
grs/mt2
1
1 2 3
W21-54
grs/mt2
1 2 3
61065
1 2 3
800122
1 2 3
790131
1 2 3
LEF-RB
1 2 3 1 2 3
1213-2
8-Madero
1 2 3
BAT 332
1 2 3 1 2 3
BAT 85
DGO 222
1-Flowering
2 Mid-Pod-Fill
3-Physiological Maturity
Figure 6. Changes in Stem Starch Contents ( grs/mt2 ) Over Three Physiological
Stages Under Two Water Treatments. Iguala, 1982 3.
m Irrigated
V --- Stress
V
v
1 2 3
W21-54
gra/mt2
1 2 3
61065
1 2 3
800122
4%'
1 2 3
790131
1 2 3
LEF-RB
1 2 3 1 2 3 1 2 3
1213-2
B-Madero
BAT 332
1 2 3
BAT 85
1 2 3
DGO 222
1-Flowering
2-Mid-Pod-Fill
3-Physiological Maturity
Figure 7. Changes In Pod Starch Contents ( grs/mt2 ) Over Three Physiological
Stages Under Two Water Treatments. Iguala, 1982 3.
3
grs/mt2
i
the amount of starch in the stem from m.p.f. to PM, and this
increment was smaller under the water stress treatment. A greater
utilization of assimilates stored in the stems under water
stress may constitute a strategy by which the plants cope with
adverse environmental effects. A negative and significant
correlation between the starch present in the stems at P.M. and
seed yield supports this hypothesis. The calculated correlation
was -0.5 under water stress, while under irrigation the
corresponding value was -0.3. This might indicate that the seed
filling process, when photosynthesis is adversely reduced,
utilizes the stored carbohydrates, and that the greater the
capacity to remobilize them, the higher the seed yield would be.
Under irrigation, when photosynthesis is not drastically reduced,
there is a continues supply of assimilates so the plant does not
have to draw upon the reserves so heavily.
K. Twenty Upper and Lower Pods: Seed Number and Size
Significant differences were found among cultivars for
seed number in both the upper and lower pods, but no significant
water effect was indicated by the AOV. When the number of seeds
of the upper pods versus the lower pods was compared, no
significant differences were observed (Table 10).However, for
seed weight the AOV revealed a significant cultivar effect as
well as a significant water by cultivar interaction. Seed weight
in the upper and lower pods was reduced under the minus water
treatment (Tables 4 and 11) seeds from the upper pods being
Table 10. Seed number of 20 upper and lower pods under two
water treatments. Iguala, 1982-3.
20 lower podsa 20 upper podsb
Identification Irrigated Stress Irrigated Stress
Wise 23-61 102 113 111 119
61065 104 114 109 111
800122 109 107 108 112
790131 71 68 73 65
LEF-2-RB 87 106** 95 108
1213-2 61 56 70 61
Bayo Madero 57 61 62 65
BAT 332 96 108 84 99
BAT 85 106 107 113 106
Durango 222 50 45 52 52
a = LSD at 0.10 (17)
** = LSD at 0.05 (20)
b = LSD at 0.10 (16)
** = LSD at 0.05 (19)
Table 11. Seed weight (mgrs/seed) of 20 upper and lower pods
under two water treatments. Iguala, 1982-3.
20 lower podsa 20 upper podsb
Identification Irrigated Stress Irrigated Stress
Wisc 23-61 210.7 199.9 193.5 188.5
61065 190.5 186.7 174.2 165.6
800122 163.9 158.5 168.5 153.1
790131 230.1 231.1 218.1 217.5
LEF-2-RB 251.2 254.0 242.2 242.9
1213-2 360.6 341.2 347.0 354.1
Bayo Madero 446.0 391 .1** 409.7 373.3**
BAT 332 171.7 163.3 169.2 161.1
BAT 85 219.2 209.6 211.7 181.7**
Durango 222 519.4 422.8** 481.2 469.8
a = LSD at 0.10 (27.7)
** = LSD at 0.05 (33.1)
b = LSD at 0.10 (19.9)
** = LSD at 0.05 (23.8)
63
smaller than those from the lower pods. Since the lower pods are
the first ones to be formed during the plant's developmental
processes they were at a more advanced seed filling stage when
the stress became effective. This probably implies that either
they did not have to rely upon the stored carbohydrates to fill
their seeds because they were being filled with currently made
photosynthates, or due to closer proximity they were the first
ones to use the stored carbohydrates; 'they had essentially
completed filling before the stress became severe.
II.Nitrogen Effects
A secondary objective of this experiment was to determine
the nitrogen fixation potential of the 22 cultivars, and to
relate their potential with the effect of drought stress imposed
during the latter part of the seed filling period and with the
ability to translocate non-structural carbohydrates. To
accomplish this objective, the twenty two cultivars were planted
under two contrasting nitrogen levels and under two water
regimes. As described in the Materials and Methods section, the
experimental plots were on a Silty Clay soil in which the organic
matter and total N contents were low (%of organic matter = 1.05,
total N = 0.112 ppm). Before planting the so called plus N plots
were fertilized with 40 kgs of N per hectare, applied in the form
of urea, while the minus N plots did not receive any N
fertilizer. At the time of planting all plots, except 2 border
rows that ran the length the field, were inoculated with a
commercial granular Rhizobium inoculant.
We expected to see differences due to the N treatment, but
we could not observe any visual symptoms of N deficiency in the
non- fertilized plots; also, the observed nodulation throughout
the season was considered fairly poor for plants grown under the
two different N treatments. The only clear N deficiency
symptoms, such as severe yellowness and reduced growth, were
observed in the two border rows that had neither fertilizer nor
inoculant. We have no definite explanations for the lack of
difference between the plus and minus N treatments. Whether the
soil analysis was faulty and the actual content of N in the soil
was higher than shown by the analysis is unknown. We can't answer
this question now because after harvesting this experiment the
soil was plowed and uniformly fertilized for the next crop to be
planted. Another possible explanation, though it may be remote,
might be found in the levels of N03 present in the irrigation
water. If such levels were high enough to provide the plants with
sufficient N for their vegetative growth, it is possible that no
N treatment differences did in fact exist. The fact that only two
border rows, which were at the end of the field and therefore did
not receive as much water as the other plants did, supports this
assumption.
The quality of the applied inoculant turned out to be poor,
having only 5.8 x 104 Rhizobia per gram. A good quality inoculum
should have at least 108 Rhizobia per gram. Nevertheless, this
does not imply that there were not enough bacteria in the soil
sufficient to have established a symbiotic relationship with the
host plants. Countings of the native Rhizobia population existing
in the soil before inoculation ranged from 1.8 x 103 to 1.7 x
107 colonies per gram of soil.
Since the water stress was not effective until the late pod
filling stage and the data herein described refers to
earlier physiological stages, the results are based on only one
water treatment.
A.Non significant effects
The individual AOVs for Biological Yield, Economic Yield,
Harvest Index, Weight of 100 seeds, Length of seed filling
period, Leaf dropping, Seed number and weight from the 20 upper
and lower pods and % of Nitrogen did not show a significant
Nitrogen effect. However, some Nitrogen x Cultivar interactions
as well as Nitrogen x Water and Nitrogen x Water x Cultivars
interactions were significant. These interactions will be
referred to in the next section.
The effect of added Nitrogen on N-fixation depends on the
specific cultivar; different cultivars show different responses
to N fertilizer. In the section, Interaction Effects, the
differential response of the genotypes used in this study will be
discussed and I will attempt to reach some conclusions from this
experiment.
B. Plant dry weight
The AOV for total plant dry weight at the three sampling
times (Flowering, 15 d.a.f., and Physiological Maturity) did not
detect any significant differences due to N effect. However, with
respect to the individual components of plant dry weight over the
three different samples, a significant Nitrogen effect was given
for stem % of total plant dry weight at flowering time.
Figure 8 illustrates the differential responses of the 10
sampled cultivars over the two different levels of Nitrogen. In
genotypes such as BAT 332, 61065, and 800122, the stem
constituted over 42 % of their total plant dry weight, while in
- i ........l ~ J .1 ~ I &~ a a a .*~ I I a I a a ~. -' -. -. -.
Wisc 23-61 61065
800122 790131 LEF-2-RB 1213-2
Bayo BAT 332 BAT 85 Durango
Madero 222
* I= 40 kg/ha added N. CV.
E2 = no added N.
Figure 8. Stem % of total plant dry weight at flowering time. *Iguala 1982,3.
50 -
40-
% 030
68
Durango 222 the corresponding value was less than 30 %. Even
though the correlation between stem % of dry weight at flowering
and Economic yield was expressed as a significant r value of
0.408, the data show that in high yielding lines such as BAT 85
the stem % of total dry weight was approximately the same as in
the low yielding lines Bayo Madero and 790131.
Different cultivars showed different responses to N
fertilizer. A line previously selected for high BNF potential,
Wisconsin 23-61, showed an increase in stem weight when 1i was not
added; BAT 332, also previously reported by CIAT to be a good N-
fixer, showed an opposite response. No conclusions can be drawn
from these results except that if stem % of total dry weight is
positively correlated with yield, good N-fixers should have high
stem weights at flowering time. The carbohydrates that are stored
in the stems can be remobilized and utilized in the later stages
of plant development when photosynthetic activity is reduced,
particular in the lower (shaded) portion of the canopy where
carbohydrates might be needed for supporting N-fixation. In the
case of N-fixation, a great amount of photosynthate is required
by the nodules in order to maintain their growth and to
facilitate the organic binding of the fixed Nitrogen. Since the
nodules store very few reserves, they depend on the supply of
assimilates that is available to them. Genotypes like BAT 332,
61065 and 800122 should have a greater BNF potential than 1213-2
and Durango 222.
The roots constitute a potential site for carbohydrate
storage and a possible supplier of assimilates to the nodules. A
positive and significant correlation between the economic yield
and the root weight at flowering time (when N-fixation is
supposed to be at its maximum activity) was found, the calculated
r-value being a significant 0.394.
C. Plant dry weight changes: remobilization
Figure 9 shows the changes in stem, root, leaf and pod dry
weights that occurred from flowering to 15 d.a.f., the time that
we consider to be the middle of the pod filling stage. This
figure illustrates the differences in dry weights of six
genotypes under added N as well as under non added N.
Stem weight increased in all cultivars, from flowering to 15
d.a.f. but the s ze of the increment varied for the different
cultivars. The incremental changes were essentially the same for
N-fertilized and non-fertilized treatments. Root weights remained
approximately the same from flowering to 15 d.a.f.; only 61065
and Durango 222 showed a noticeable increase in root dry weight,
but no overall differences were seen between the two N
treatments. With respect to pod weight at 15 d.a.f., a general
tendency towards a greater pod weight under no added N was
observed. These data suggest that in this experiment N fixation
or soil N supply was sufficient to maintain a large number of
flowers which developed into pods. Leaf weight increased over
time, but again no differences due to N source were seen.
Since no differences were observed between N treatments for
the individual components of plant dry weight, a Shoot:Root ratio
(S:R) was calculated in order to try to understand the behavior
a L
2
100 1
40 *
so
-40 -
SO -
g1 1 2 1 2 I
ri 1
r-m -I
F'
- -- ruf-u
[I
j 4
I
*0 40 kthe alddedN. I 1 PNleAld
* a-MddedLN. I 1- AIMp.
41
Stems
II
61065 700131 1213-2 B. Madero BAT85 DOo222
Fig. 9 Stern, Root, Pod and Leaf weights at Flowering and at 15 dap.
under two different levels of added NItrogen. *Igiial 1982-3.
j
1 2
of the different genotypes. This ratio, as a measure of the
pattern of differential growth, can provide an index for the
performance of each plant organ under different growth
conditions. An increase in the S:R ratio might be the result of a
greater utilization of carbohydrates by the shoot at the expense
of the root, possibly bringing about a shortage in carbohydrate
supply to the nodules that will translate into poor or reduced N-
fixation. The effects of added N on S:R ratio as shown in Table
12 varied with plant genotype and stage of development. When
reading this table one must be aware that very high values of S:R
ratio probably reflect very incomplete harvest of the root
system.
At flowering time, the cultivars 61065, 1213-2 and Bayo Madero
had a higher S:R ratio under non-added N, while Durango 222 and
790131 remained the same. Only BAT 85 had a higher S:R ratio
under added N at flowering time.
At 15 d.a.f. the cultivars 61065, 1213-2, Bayo Madero and
Durango 222 showed no differences in S:R ratio due to added N.
However, 790131 which showed no differences at flowering time,
did show an increase of S:R ratio at 15 d.a.f. under non-added N.
Also, BAT 85 had an increase of S:R ratio at 15 d.a.f. under the
non fertilized treatment. Higher S:R ratios at flowering under
non-fertilized conditions, are caused by an increase in shoot dry
weight. Shoot growth was enhanced when no N was added and it was
diminished under added N. One possibility is that the applied
dosage of N fertilizer (40 kgs/ha) was enough to inhibit N
fixation but at the same time it was not enough to maintain a
Table 12. Shoot:Root ratio under two Nitrogen treatments
at two physiological stages. Iguala, 1982-3.
Identification Stage*1 Treatment*2 S:R ratio
61065 P + 6.7
8.0
MPF + 17.0
17.6
790131 F + 9.9
10.1
MPF + 25.9
29.6
1213-2 F + 14.9
17.5
MPF + 18.8
18.9
Bayo Madero P + 17.3
19.8
MPF + 16.9
16.2
BAT 85 F + 12.7
10.8
MPF + 32.3
41.6
Durango 222 P + 22.5
21.6
MPF + 16.6
17.0
*1 F= flowering
MPP= mid-pod filling
*2 + = added N
= non-added N
continued vigorous growth. On the other hand, the plants that
were grown under non-added N conditions were able to maintain
high levels of N-fixation which resulted in shoot growth. A
second possible explanation is that temperatures were too high
for maintaining high levels of the enzyme Nitrate Reductase,
whose presence and activity is necessary for the utilization of
N-fertilizer. In the case of BAT 85, the applied dosage of N
fertilizer was either enough to satisfy the plant's requirements
and promote optimum growth or it acted as a "starter" and
stimulated nodulation and plant growth at flowering time.
However, in a later stage of development, BAT 85 had a higher S:R
ratio under non added N as compared to the N-fertilized plots.
This illustrates that the effect of N fertilizer varies not only
among cultivars but also between stages of development within the
same cultivar.
The data presented herein suggest that N availability in
both N-fertilized and inoculated plants, was sufficient to
support vegetative growth. Economic yield was not affected by N
treatments for any of the sampled cultivars, nevertheless small
differences were seen between different combinations of genotype
and N treatment.
DISCUSSION
One of the main purposes of this experiment was to try to
identify physiological changes during the course of the growing
season that can be responsible for maintaining normal yields
under stress conditions. It seems reasonable to think that a
better understanding of the basis of differences among cultivars
and the relationship between these differences and their yield
potential should provide basic information that would be very
valuable in choosing a drought tolerance breeding strategy. As
suggested by Duncan et al. (16), three plausible explanations for
differences in yield between cultivars can be given. The first
one is a difference in photosynthetic efficiency of leaf
canopies, which would result in differences in the amount of
carbon fixed over the growing season. This efficiency could
result from better canopy geometry resulting in better light
interception, or from greater leaf area duration. A second reason
for yield differential could be the proportion of daily produced
assimilates that is partitioned to the economic sink. It is
likely that a higher yielding cultivar either partitions more of
the daily assimilate production to the seeds or is capable of
utilizing stored assimilates to fill the seeds. As a result, a
greater number of seeds (increased sink size) and heavier seeds
can be attained. The third reason could be the duration of the
vegetative and reproductive periods, the latter commonly known as
the seed filling period. Seed yield is the result of the rate and
duration of the filling period times the size of the economic
sink. With this experiment we sought to answer the following
questions: 1. Is there storage capacity in the stems of all
cultivars ? 2. When stem weight declines, does this correspond to
a non-structural carbohydrate loss? 35. What proportion of the
seed dry weight increase can be accounted for by changes in dry
weights, particular by stem dry weight loss? 4. Are there
cultivar and treatment differences in the contribution of storage
assimilates to seed yield? Having these questions in mind, the
following discussion is organized around the three possible
reasons for yield differences that were mentioned above.
1. Crop Growth Rate
Ground cover by the leaf canopy and rate of accumulation of
dry weight generally increase exponentially until light
interception is complete (16). In dry beans a fully closed
canopy is achieved at around flowering time, and this was the
case for the 22 cultivars planted in this experiment. Given that
after reaching a closed canopy full light interception is
attained, a further increase in LAI should not have any effect on
light interception. The data presented here show that total leaf
area (expressed as total leaf dry weight) continued to increase
after flowering. However, for the reasons stated above, no
further gains in light interception were expected after flowering
time.
Total dry matter accumulation or net photosynthesis,
expressed as kgs of dry matter per hectare, is simply the
difference between the total amount of carbon fixed by
photosynthesis and the respective carbon losses due to growth and
maintenance respiration. Net photosynthesis as well as average
crop growth rates for 10 different cultivars during the
vegetative stage are given in Table 13. It is evident that the
late flowering cultivars had both a greater accumulation of dry
matter and a higher growth rate during their vegetative phase of
development. However, the CIAT line BAT 85 stands out for its
high photosynthetic efficiency ( expressed as kgs. of dry matter
accumulated per hectare per day ) given that it was not included
among the late flowering cultivars, and that the cultivars that
flowered at the same time as BAT 85 (63 to 64 days after
planting) had lower crop growth rates. The efficiency of BAT 85
can not be explained further with our current data; we can not
determine whether high photosynthetic capability, low respiratory
losses or both are responsible for the high dry matter
accumulation that occurred during the vegetative phase. The
photosynthates accumulated during the reproductive stage of
development and the crop growth rates for that period for the 10
sampled cultivars are given in Table 14. Crop growth rates
increased in all cases in the reproductive stage as compared to
the vegetative stage. Cultivar responses to the stress differed:
BAT 332, LEF-2-RB, 800122 and 61065 had higher growth rates
under the minus water treatment as compared to the irrigated
plots. The other 6 cultivars had lower crop growth rates under
stress. It becomes evident from these data that not only do
cultivar differences for crop growth rate exist but also that
different cultivars react differently under water stress.
Table 13. Average Crop growth rates( kg/ha/day) from
planting to flowering. Iguala, 1982-3.
Identification
Days to
Flower
Total dry matter
at Flow.
(kg/ha)
Average Crop
growth rate from
planting to Flow.
Wisec 23-61 64 1198 18.7
61065 64 1286 20.1
800122 70 1969 28.1
790131 63 1309 20.7
LEF-2-RB 64 1369 21.3
1213-2 48 838 17.4
Bayo Madero 50 1022 20.4
BAT 332 70 1910 27.3
BAT 85 63 1816 28.8
Durango 222 43 688 16.0
Table 14. Average Crop growth rates ( kg/ha/day ) from
flowering to maturity under two water treatments.
Iguala, 1982-3.
Identification Total dry matter Average Crop growth
at PM (kg/ha) rates from Fl. to PM
Irrigated Stress Irrigated Stress
Wisc 23-61 4464 3536 88.2 68.7
61065 3908 4012 74.9 80.1
800122 4305 5280 68.7 97.3
790131 4021 3457 90.4 76.7
LEF-2-RB 4360 5298 76.6 98.2
1213-2 3853 2952 71.7 52.8
Bayo Madero 7592 3371 126.3 48.9
BAT 332 3862 4044 65.0 73.5
BAT 85 4049 3482 67.6 52.0
Durango 222 3642 3288 59.0 55.3
Net photosynthesis is the result of a biological input-
output system that has several constraints. Identifying and
quantifying the relevant constraints would help the plant breeder
achieve a maximization of photosynthetic production.
Physiological and morphological components which determine the
crops efficiency of light conversion in a particular environment,
such as rapid establishment of a closed leaf canopy, efficient
canopy photosynthesis and effective distribution of assimilates
to the relevant economic sinks for as long a period as possible,
and the genetic variation associated with them, should be the
focus of detailed study.
2. Partitioning
The most important determinant of economic yield, as Donald
and Hamblin (15) stated, is not total crop photosynthesis, but
the way in which assimilates are distributed within the plant,
either for continued vegetative growth or for accumulation in
storage organs, seeds or fruits. However, it is not clear how
this allocation is regulated. It can be regulated by the supply
of assimilates (source strength), by the ability of the sink to
make use of the assimilates (sink strength) or by the rate of
translocation. How far sink strength can influence photosynthetic
rate is still an unanwered question. The term partitioning, as
used here, indicates the allocation of assimilates between
reproductive and vegetative plant parts. It is a dynamic day-to-
day process, that differs with cultivars and with physiological
Table15. Average Fruit growth rate from flowering to
physiological maturity (kgs/ha/day) under two water
treatments. Iguala, 1982-3.
Identification Irrigated Stress
Wisc 23-61 80.6 109.8
61065 94.0 101.5
800122 84.2 86.4
790131 77.0 88.6
LEF-2-RB 77.5 116.9
1213-2 56.9 62.5
Bayo Madero 56.7 37.3
BAT 332 91.0 123.2
BAT 85 123.0 109.4
Durango 222 57.6 46.8
Table 16. Partitioning Factor under two water1
treatments. Iguala, 1982-3. Calculated %
Identification Irrigated Stress
Wisec 23-61 91.4 159.7
61065 125.5 126.6
800122 114.9 88.7
790131 85.2 115.5
LEF-2-RB 101.0 119.0
1213-2 79.3 118.3
Bayo Madero 44.9 76.2
BAT 332 139.8 167.4
BAT 85 181.7 210.1
Durango 222 97.5 76.4
*1 %=(fruit growth rate/crop growth rate)x 100
stages such as early or late pod filling. The partitioninig of
assimilates between new vegetative tissue and storage can be very
important for plant performance under environmental stresses such
as temperature or water stress.
Table 15 illustrates the fruit growth rates of the 10
sampled cultivars under irrigated and stress conditions.
Differences not only among cultivars but also among treatments
were obtained. This indicates that the daily partitioning of
assimilates to the fruits was determined by the genotype and the
water treatment. When comparing the data shown in Table 14 (Crop
gowth rates CGR-) with the corresponding values in Table 15,
one can see that Fruit growth rates -FGR- exceeded CGR in 7
cultivars under water stress, while under irrigation FGR exceeded
CGR in 5 cultivars.
One of the basic questions to be answered is whether or not
the water stress treatment induces a greater partitioning of
stored assimilates to the fruit. The division of FGR by CGR
during a given period of time gives the average fraction of net
photosynthate partitioned to fruit growth. If all fruit growth
can be explained by current photosynthesis, net accumulation of
dry matter should be greater than or equal to fruit growth; if
not, one can assume that fruit growth was sustained in part with
photosynthates that were fixed in an earlier developmental stage.
A calculated partitioning factor shown in Table 16, indicates
that in 7 out of the 10 sampled cultivars under stress, the
calculated partitioning ratio exceeded 100%, and under irrigation
' in 5 entries the ratio exceeded 100%. Whether this can be
extrapolated to the extent that we can be sure that water stress
induces a greater partition of assimilates to the fruit is not
clear, but it is clear that treatment and cultivar differences in
partitioning exist.
A greater partition of assimilates can be the result of a
greater fruit load, or what was called before, namely "sink
strength". We have no conclusive evidence to say that this in
fact is the case, but the data show a very consistent trend in
which the cultivars with a high partitioning factor such as BAT
332 and BAT 85 had between 149 and 231 grs of seed/mt2, while in
cultivars with a low partitioning factor such as Durango 222 and
Bayo Madero sink size varied from 80 to 111 gra of seed/mt2.
Considering the change in plant dry matter between anthesis
and maturity as an indicator of net plant photosynthesis during
this period, and comparing this increment with the corresponding
increment in fruit weight, seems to be a reasonable way of
illustrating the proportion of pre-anthesis and post-anthesis
assimilates that were utilized for fruit growth under the 2 water
treatments. In Figures 10 and 11 plus water and minus water
treatments respectively, the X-axis is the net post-anthesis
photosynthesis expressed in kgs/ha and the Y-axis is the fruit
growth from anthesis to maturity also expressed in kgs/ha. The
1:1 line shows the position where a cultivar would lie if all the
assimilate produced after anthesis had gone into the grain. A
cultivar that lies further from and below the line is fixing more
carbon and it is using it for fruit growth at a slower rate than
it is produced. On the other hand in a cultivar that is
positioned above the 1:1 line fruit growth exceeds total growth
5000
4500
4000
3500
3000
2500
2000
1500
1 = Wisc 23-61
2= 61065
3 = 800122
4 = 790131
5=LEF-2-RB
6=1213-2
7 = B. Madero
8 = BAT 332
9= BAT 85
10= Dgo222
*7
2000
3000 4000 5000 6000
Net post-anthesis photosynthesis (kgs/ha)
Fig. 10 Proportion of fruit growth that can be accounted for by post-
anthesis photosynthesis under irrigated conditions. Iguala,
1982-3.
1:1 line.
.*9
-
*2
5
10
*8 93
.*6
*4
7000
|
