Effect of excess and limited soil moisture on nitrogen fixation of several leguminous crops

MISSING IMAGE

Material Information

Title:
Effect of excess and limited soil moisture on nitrogen fixation of several leguminous crops
Series Title:
Florida Water Resources Research Center Publication Number 69
Physical Description:
Book
Creator:
Bennett, Jerry M.
Albrecht, S. L.
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Notes

Abstract:
Experimental greenhouse work was conducted to examine the effects of legume crop species and soil moisture content on leaf and nodule water status, stomatal resistance, root respiration, and nitrogen fixation. Data were collected on forage and grain crops that are important in Florida and increasing in importance in other semi-tropical and tropical climates. The effect of both flooding and drought on the capacity of Aeschynomene americana, Desmodium heterocarpon, cowpea (Vigna unuiculata L.), alfalfa (Medicago sativa L.), soybean (Glycine max L.), and bean (Phaseolus vulgaris L.) to fix nitrogen was examined in a series of experiments. Evaluations of 28 accessions of Aeschynomene suggested a tolerance to flooding. The capacity to fix nitrogen in flooded plants remained about the same as the well-watered controls, and the number of nodules was found to increase on lateral roots which initiated after the flooding was imposed. In contrast, flooded Desmodium, bean, and alfalfa showed a marked decrease in nitrogenase activity, with a concomitant decline in plant water status and root respiration. Drought stressed Aeschynomene, Desmodium, and alfalfa plants showed a decrease in the ability to fix nitrogen which was associated with reductions in leaf water potentials. In drought stressed soybeans, large reductions in leaf and nodule water potentials, stomatal conductance, and root respiration occurred as gravimetric soil water content decreased below about 1%. In general, nitrogenase activity declined along with other physiological processes, however, there was an indication that nitrogen fixation was more severely inhibited by drought. In an adjunct field study, nitrogen fixation was reduced by soil water deficits more than photosynthesis or stomatal conductance, suggesting that reductions in nodule water status had a direct effect on nitrogen fixation. Rewatering previously drought stressed plants restored control levels of the measured physiological parameters within 24 hours. Flooded soybean plants increased rates of nitrogen fixation during the flooding period, while other plant water stress indicators remained similar to those of well-watered plants. The increase in nitrogenase activity was probably due to young nodules which initiated on newly formed lateral roots. In cowpeas, flooding had little effect on.either water status or nitrogen fixation activity. Bean plants were sensitive to the flooding treatments and even though plant and nodule water potentials varied little from normally watered plants, nitrogenase activity was severely inhibited. While drought stress was detrimental to nitrogenase activity in all species examined, differential responses to flooding were observed. Desmodium, alfalfa, and bean were quite susceptible to flooding while soybean and cowpea were moderately tolerant and Aeschynomene was very tolerant.

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution and holding location.
System ID:
AA00001545:00001


This item is only available as the following downloads:


Full Text















Publication No. 69


THE EFFECT OF EXCESS AND LIMITED SOIL MOISTURE ON
NITROGEN FIXATION OF SEVERAL LEGUMINOUS CROPS

By

Jerry M. Bennett
Agronomy Department
University of Florida
Gainesville

and


S. L. Albrecht
USDA-ARS
Gainesville


'


.~.a~it













TABLE OF CONTENTS Page

ACKNOWLEDGEMENTS............................................ 3

ABSTRACT....................................................4

CHAPTER

I. INTRODUCTION...........................................5

II. GREENHOUSE EXPERIMENTS...... ........................ 9

A. SOYBEANS, EXPERIMENTS I, II, AND III.............. 9

B. AESCHYNOMENE AND DESMODIUM, EXPERIMENTS
I, II, AND III ................................. 24

C. COWPEA AND BEAN.................................27

D. ALFALFA.........................................41

III. FIELD EXPERIMENT, SOYBEANS............................ 48

IV. SUMMARY..... ........................ .. .............58

LITERATURE CITED.............. .......... ... .......... .. 59

PUBLISHED PAPERS .......................................... 62


- 2 -












ACKNOWLEDGEMENTS


We are indebted to Carmen Harris for the many hours of technical
assistance provided in this study. We also appreciate the assistance
of Ann Newitt, Gary Moore, Mark Moser, and Andrea Givens. Thanks are
also expressed to Dr. K. J. Boote for allowing us to utilize soybean
field plots for an adjunct experiment and for providing data on carbon
exchange rates. We also thank Dr. K. H. Quesenberry for his coopera-
tion and for providing greenhouse space for the Aeschynomene and
Desmodium experiments. Our appreciation is extended to Carolyn Meyer
for final assembly and typing of this report.


- 3-










ABSTRACT


Experimental greenhouse work was conducted to examine the
effects of legume crop species and soil moisture content on leaf
and nodule water status, stomatal resistance, root respiration,
and nitrogen fixation. Data were collected on forage and grain
crops that are important in Florida and increasing in importance
in other semi-tropical and tropical climates. The effect of both
flooding and drought on the capacity of Aeschynomene americana,
Desmodium heterocarpon, cowpea (Vigna unuiculata L.), alfalfa
(Medicago sativa L.), soybean (Glycine max L.), and bean (Phaseolus
vulgaris L.) to fix nitrogen was examined in a series of experiments.

Evaluations of 28 accessions of Aeschynomene suggested a
tolerance to flooding. The capacity to fix nitrogen in flooded
plants remained about the same as the well-watered controls, and the
number of nodules was found to increase on lateral roots which ini-
tiated after the flooding was imposed. In contrast, flooded Desmodium,
bean, and alfalfa showed a marked decrease in nitrogenase activity,
with a concomitant decline in plant water status and root respiration.
Drought stressed Aeschynomene, Desmodium, and alfalfa plants showed
a decrease in the ability to fix nitrogen which was associated with
reductions in leaf water potentials.

In drought stressed soybeans, large reductions in leaf and nodule
water potentials, stomatal conductance, and root respiration occurred
as gravimetric soil water content decreased below about 1%. In
general, nitrogenase activity declined along with other physiological
processes, however, there was an indication that nitrogen fixation was
more severely inhibited by drought. In an adjunct field study, nitrogen
fixation was reduced by soil water deficits more than photosynthesis or
stomatal conductance, suggesting that reductions in nodule water status
had a direct effect on nitrogen fixation. Rewatering previously drought
stressed plants restored control levels of the measured physiological
parameters within 24 hours.

Flooded soybean plants increased rates of nitrogen fixation during
the flooding period, while other plant water stress indicators remained
similar to those of well-watered plants. The increase in nitrogenase
activity was probably due to young nodules which initiated on newly
formed lateral roots. In cowpeas, flooding had little effect on either
water status or nitrogen fixation activity. Bean plants were sensitive
to the flooding treatments and even though plant and nodule water poten-
tials varied little from normally watered plants, nitrogenase activity
was severely inhibited.

While drought stress was detrimental to nitrogenase activity in all
species examined, differential responses to flooding were observed.
Desmodium, alfalfa, and bean were quite susceptible to flooding while
soybean and cowpea were moderately tolerant and Aeschynomene was very
tolerant.


- 4 -












CHAPTER I. INTRODUCTION


Next to photosynthesis, the reduction of atmospheric nitrogen to
ammonia is probably the most important reduction reaction on earth.
The ability to reduce atmospheric nitrogen to ammonia is confined to
certain prokaryotic organisms, including the genus Rhizobia, which
may live symbiotically with plants. Agriculturally, legumes are the
most important of the symbiotic systems involving Rhizobia. Before
1966, nitrogen fixation was determined by either total nitrogen analysis
or measuring 15N2 incorporation (Burris and Wilson, 1957). These methods
are time consuming and expensive, however, the discovery that the nitro-
genase enzyme complex can also reduce acetylene to ethylene (Dilworth,
1966; Schollhorn and Burris, 1966) provided the basis for a rapid,
inexpensive assay, using gas chromatography to separate the two
hydrocarbons.

Soil moisture has long been known to be a factor that will in-
fluence the growth of plants. Members of the Leguminosae family are
generally intolerant of either an excess or a deficiency of water in
their root environment, and there are numerous reports that water stress
will affect both yield and quality of several legume crops (Kilmer et
al., 1960; Bourget and Carson, 1962; Mack, 1973). The effect of soil
moisture on nitrogen fixation has received relatively little attention
until the past decade, but there are several reports in the literature
which suggest that symbiotic nitrogen fixation by legume root nodules
is very sensitive to either desiccated or flooded soils. Nitrogen
fixation by root nodules requires that the soil be able to supply
optimum amounts of both water and certain gases, notably nitrogen and
oxygen (Evans and Russell, 1971). The simultaneous need for the soil
to be wet enough for adequate water supply to the root system and dry
enough for gaseous exchange allows a very narrow optimum for nodule
nitrogenase activity (Huang et al., 1975a). There have been examina-
tions of water stress effects on nitrogenase in a number of representa-
tive legumes including Trifolium repens (Engin and Sprent, 1973),
Glycine max (Sprent, 1971, 1972; Huang et al., 1975a, 1975b), Vicia faba
(Sprent, 1972), Phaseolus vulgaris (Sprent, 1976a), Lupinus arboreus
(Sprent, 1973), and Aeschynomene americana (Albrecht et al., 1981).

The pattern of response to drought stress is similar in most legume
species. Nitrogenase activity falls to undetectable levels at about 40
percent of maximum fresh nodule weight (Sprent, 1972). Nodule activity
can be restored to that of desiccated nodules by watering, full activity
being resumed in one or more hours (Sprent, 1972). Plants with meriste-
matic nodules can recover from more damaging kinds of stress by regrowth
of existing nodules, which can occur two to three days after watering
(Engin and Sprent, 1973). On plants with spherical nodules severe
stress causes nodule loss (Sprent, 1973). Recovery is slower and


-5-









involves the formation of new nodules. Huang et al. (1975b) found
that inbibition of nitrogenase activity at low water potentials could
be partially reversed by exposing the shoots to high concentrations of
C02.

In addition to reducing the nitrogenase activity of existing
nodules, water stress affects the growth of young nodules and the
formation of new nodules. Sprent (1976a) reports that in young P.
vulgaris plants, stressed to the point of wilting, nitrogenase activity
was reduced by 90 percent. Nodule number and size were also depressed
by this treatment. It is not known if this depression of nodulation
is caused by fewer Rhizobia being present in the rhizosphere and
available for nodulation, or if the stress affects the infection process.

Water stress usually occurs gradually and progressively as moisture
is lost from the soil. Sprent (1972) has shown that under these condi-
tions the wilting of the lower leaves of a legume is a good indication
of suboptimal nitrogenase activity. However, under simulated wind
conditions, if the soil remains moist, the shoots of both G. max and P.
vulgaris will wilt, without a noticeable effect on nitrogenase activity
in the nodules (Gallacher and Sprent, 1978). In contrast to Huang and
co-workers (1975a, b) who suggest that the inhibition of photosynthesis
activity accounted for the inhibition of nitrogenase at low water
potentials, Sprent (1976b) suggests that nodule stress occurs when the
root systems cannot supply sufficient water to export materials from
the nodules and replace the water lost from the nodule to the drying
soil.

Sensitivity to excessive soil moisture stress has been reported
for several legumes (Minchin and Pate, 1975; Minchin and Summerfield,
1976; Sprent, 1972). Mague and Burris (1972), Schwinghamer et al.
(1970), and Sprent (1969) have reported that waterlogging depresses
nitrogen fixation, largely as a result of oxygen deficiency. Sprent
(1976a) has shown that nodule number, size, and water content are also
affected by waterlogging. Minchin and Summerfield (1976) report that
total dry weight of Vigna unguiculata nodules was reduced by 60 percent
after only 8 days of waterlogging. Prolonged waterlogging did not
affect the percentage nitrogen content of the various plant parts, but
they found that the plant dry weight could be reduced by as much as
60 percent.

There does appear to be some resistance to the effects of excessive
soil moisture. Sprent (1976a) found in P. vulgaris that there was some
variation among nodules formed with different strains of rhizobia in
their ability to withstand waterlogging. There is evidence that some
tropical forage legumes are flood tolerant (Brolmann, 1978; Mclvor, 1976;
Quesenberry et al., 1982), and A. americana cultivars have been shown
to have good rates of nitrogen fixation activity in flooded conditions
(Albrecht et al., 1981).

There can be an increase in nodule activity as the soil dries from
flooded conditions and this is usually explained by increased gas






- 7 -


diffusion through the soil (Fishbeck et al., 1973; Sprent, 1972).
Although inhibitory desiccation, that beyond what is required for
maximum nodule activity has been explained in terms of reduced nodule
respiration (Engin and Sprent, 1973; Sprent, 1972), the reduction in
respiration was not as severe as that displayed by nitrogenase activity.

Pankhurst and Sprent (1975) attempted to measure the water potential
of nodules by the Shardakov dye method and by measuring the water poten-
tial of sand equilibrated with stressed nodulated root systems using a
dewpoint psychrometer. They found that loss of up to about 25 percent
in fresh weight (equivalent to -8 to -10 bars) results in reversible
effects on nitrogen fixation, but beyond this, the effects became
progressively more severe.

The relationship between carbon fixation, nitrogen fixation, and
water potential has been investigated to some extent. Minchin and Pate
(1975) found the ratio of fixed nitrogen to C02 absorbed decreased as
soil water suction was increased. They also found the effects of drought
stress were considerably smaller than those for waterlogging. These
observations are in agreement with the hypothesis that the nitrogen-
fixing process .is more sensitive than carbon fixation to stress, even
though the nodules are closer to the normal water supply than the leaves.
However, Huang et al. (1975a, b) describe a series of experiments using
intact soybeans which suggest that the reduced supplies of photosynthate
were responsible for the depression of nitrogenase activity under drought
stress. Legumes are notorious for their extravagent use of water, as
shown by Ludlow and Wilson (1972) who found for the tropical grasses
Pennisetum purpureum and Sorghum album a transpiration ratio (g H20
transpired/g C02 fixed) of just under 80 compared to 180 for the legumes
G. wightii and Calopogonium mucunoides.

There are many reports of moisture stress reducing the yields of
legumes, and the responses may vary among varieties (Mederski and
Jeffers, 1973). This may be related to cessation of root growth at
flowering (Salter and Drew, 1965). After flowering, and into the pod-
filling stage, nitrogen fixation decreases in many species (Pate, 1958)
and may reflect competition for photosynthate between nodules and
developing seeds (Lawn and Brun, 1974). This competition, linked to no
new growth by the roots, and hence impaired water uptake would make the
symbiotic plant especially sensitive to water stress at this time.

Although numerous experiments have been conducted to elucidate some
of the effects that soil moisture deficits can have on nitrogenase
activity, relatively few experiments have been concerned with soil
waterlogging. Very few researchers have included more than one crop
species in their experiments for comparative evaluations. A large
portion of the experiments reported in the literature concerning water
stress effects on nitrogen fixation were conducted with crops, soil
types, and environmental conditions which are not encountered in
Florida. The purpose of this research was to utilize several crop







-8-


and forage legumes used extensively in Florida, and to emphasize sandy
soils, which are commonly found in Florida, to examine the effects of
water stresses on nitrogen fixation as related to other physiological
changes. In addition to the greenhouse work done for this project,
an adjunct project was carried out in the field, utilizing field condi-
tions.












CHAPTER II. GREENHOUSE EXPERIMENTS


A series of greenhouse studies were conducted during 1981 and
1982 utilizing Aeschynomene americana, Desmodium heterocarpon, alfalfa
(Medicago sativa L.), bean (Phaseolus vulgaris L.), cowpea (Vigna
unguiculata L.), and soybean (Glycine max L.). Although several ex-
periments were conducted at various time intervals, procedural details
were very similar for all. greenhouse experiments with the exception of
crop species evaluated, durations of stress treatments, inoculants
used, and other obvious variations. A complete description will only
be given for the first soybean experiments. These details generally
apply to all other greenhouse experiments as well.

A. Soybeans, Experiments I, II, and III

Three greenhouse experiments designed to examine the effects of
water stresses on nitrogen fixation of soybeans in relation to other
physiological processes were conducted in the spring of 1981 and 1982
and the fall of 1982. In all experiments, 'Cobb' soybeans were grown
in 25 cm diameter black plastic pots containing a freely-draining
washed, coarse sand. Before planting, seeds were coated with a com-
mercial preparation of Rhizobium japonicum using gum arabic as an
adhesive. Pots were arranged in a randomized complete block design.
During the experimental periods, greenhouse temperatures generally
ranged between a minimum of 12 C and a maximum of 35 C. On clear days,
photosynthetically active radiation (PAR) at midday inside the green-
house was approximately 1500 uE m-2 s-. After germination, the photo-
period was extended to prevent floral induction by providing additional
light from 1800 to 2400 h with incandescent lamps. Plants were thinned
to two per pot after seedling establishment.

Except during the water stress treatment periods, plants were
watered as needed with a nitrogen-free, half-strength Hoagland's
nutrient solution. The soil was leached weekly with water to prevent
buildup of salts. Flooding treatments were imposed by placing the
treatment pot inside a dark plastic container filled with water. The
plastic containers were 27 cm diameter, which allowed complete submer-
sion of the smaller pot, but little room for mixing of water and
aeration between the two container walls. The water level in the
flooded pots was maintained approximately 5 cm above the soil surface.
To retard algal growth, aluminum foil was secured over the top surface
of the pot to exclude light. Drought cycles were imposed by simply
withholding watering and allowing the plants to extract the soil moisture.

Experiment I (1981)

Seeds were planted on 30 January 1981, emerged 6 days later, and
were thinned to two plants per pot 29 days after emergence. On 2 April,
all pots were fully watered and the pots designated to receive the
flooding treatment were placed in the water-filled containers. The
drought treatment received no additional nutrient solution until the


- 9-






- 10 -


termination of the experiment on 16 April. Nutrient solution was added
to the well-watered treatment as needed, generally every other day.
Measurements of gravimetric soil water content, leaf water potential,
nodule water potential, leaf diffusive resistance, root respiration, and
nitrogenase were collected 1, 4, 6, 8, 10, 12, and 14 days after the
imposition of the water stress treatments.

Experiment II (1982)

Seeds were planted 24 February 1982 and emerged 7 days later.
Water stress treatments arranged in a randomized complete block design
and identical to those described for Experiment I were initiated on 29
March, and soil water and plant physiological characters were determined
5, 9, 12, 13, 14, 17, and 25 days after imposing the water stress treat-
ments. After measurements were completed on day 14, the droughted plants
were re-watered with nutrient solution and flooded plants were removed
from the flooding containers and allowed to drain. Measurements on
days 17 and 20 represented recovery from the previously imposed water
stresses. The experiment was terminated on 28 April.

Experiment III (1982)

Experiment III was specifically designed to examine the relation-
ship between leaf water potential and nodule water potential as plants
progressively became more water stressed. Seeds were planted in 50
pots on 2 September 1982. Beginning 23 September water was withheld
from 30 pots. During the following 10-day period, leaf and nodule water
potentials were monitored as the plants dried. Several well-watered
plants were also sampled on each measurement date.

Measurements

All measurements of gravimetric soil water content and plant
physiological parameters were made only during periods of high light
intensity between 1100 and 1300 h EST.

Leaf (TL) and nodule (ynod) water potentials. Four, 1-cm leaf
discs were removed from uppermost, fully-expanded leaflets and placed
in a thermocouple psychrometer sample chamber. The chamber was then
immediately sealed to a Spanner-type thermocouple psychrometer (J. R.
D. Merrill Speciality Equip. Co. Model 84-13). After completing all
of the other measurements, the root system and surrounding soil were
carefully removed from the pots. Soil was separated from the root
system, nodules were selected and removed from the root system with
small forceps. The four or five detached nodules were quickly blotted
with a cloth towel to remove adhering soil particles and surface
moisture. The nodules were then quickly placed in a sample chamber
and attached to a thermocouple psychrometer unit. The thermocouple
psychrometer assemblies were then transported to the laboratory and
placed in a thermostatically controlled water bath at 30 C. After 4
h of vapor and temperature equilibration, the psychrometric output was
recorded using a strip chart recorder and a dewpoint microvoltmeter






- 11


(Wescor Model HR 33T) operating in the psychrometric mode. To
determine PL and Ynod, output was compared to calibration curves
which were constructed individually for each thermocouple psychro-
meter using NaCl solutions. Previous research had indicated that
an equilibration time of 4 h was adequate for both leaf and nodule
tissue.

Stomatal resistance. A steady-state diffusion porometer (Li-Cor
Model LI-1600) was used for measuring leaf diffusive resistance on
one of the uppermost, fully-expanded leaflets. Measurements of the
resistance of the abaxial and adaxial leaf surface were made and total
leaf resistance was calculated assuming the resistances act in parallel.

Gravimetric soil water. Soil samples were collected from each
pot by taking a soil core, including the entire depth of the soil,
from each pot. The soil sample was placed in a tin container, capped,
transported to the laboratory, and then weighed. The soil sample
was then dried overnight at 100 C before determining the dry weight
of the soil. Soil water is expressed as a percentage of the soil dry
weight.

Oxidation-reduction potentials. On several dates during the
experiment, oxidation reduction potentials of the water and saturated
soil were measured in the flooded treatments. Potentials were deter-
mined with a portable Orion meter and redox probe.

Nitrogenase activity. Intact root systems were excised and
gently separated from the soil. The entire root system and adhering
soil were quickly placed in a 75 ml serum vial. The vial was sealed
with a rubber serum stopper and 7.5 ml of air in the vial was replaced
with acetylene, resulting in a 10% acetylene atmosphere within the
vial. Vials were incubated at ambient laboratory temperatures.
During the incubation period, 0.5 ml gas samples were taken at 0, 30,
60, and 90 min. and injected into a gas chromatograph (Varian Model
940) fitted with a flame ionization detector. Rates of nitrogenase
activity were calculated from linear regression lines fitted to the
time sequence measurements. After the incubation period, the nodules
were detached from the root system, dried, and weighed. Specific
nodule nitrogenase activity was calculated by dividing nitrogenase
activity of each vial by the dry weight of the nodules.

Root respiration. Evolution of C02 from the root-nodule complex
was measured by withdrawing 0.3 ml of air from the incubation vials
and injecting it into an infrared C02 gas analyzer (Beckman Model 205)
(Clegg et al., 1978). Gas samples were analyzed for C02 every 10 min.
for a period of 90 min. Rates of C02 evolution were calculated from
the linear regression analysis and specific CO2 evolution expressed as
ug C02 evolved min-1 g dry wt root-1.

Gravimetric soil water content for the well-watered treatments
fluctuated with additions of nutrient solution but generally remained
between 4 and 10% in 1981 and 3 and 8% in 1982 (Figure 1). Although















1982
IS.E. of mean


O
0



0
CI)
O
-h-

E

0
CD


N
N >


i I I A


0 2 4 6


8 10


4


8 12


16 20 24


Days After Treatments Were Imposed


Figure 1. Gravimetric soil water contents of control and drought treatments imposed on soybeans
in 1981 and 1982. S. E. represents the average standard error of the mean.


control
adroughted




0
I
/
I

\ I

I
Nfl. ,
.,, a L 1J






- 13 -


on a couple of dates the soil water content of the well-watered plants
appeared somewhat low, no visual stress symptoms were observed and other
physiological processes were not affected by limiting soil moisture.
However, soil moisture contents in the droughted treatment declined
rapidly after the first 4 to 8 days after withholding water. As the
gravimetric soil water content decreased below about 1%, plants became
severely stressed and had difficulty extracting the remaining soil
moisture. After rewatering the plants on day 14 in 1982, gravimetric
soil water contents returned to values of the well-watered treatment.
The slightly lower soil water contents in 1982 as compared to those
observed in 1981 in both treatments are probably due to slight differ-
ences in particle size distributions of the soil. Although the soils
were quite similar, they were obtained from different sources in each
of the two years.

No differences were observed in leaf water potentials (vL) in
well-watered and flooded plants (Figure 2). In both 1981 and 1982,
the well-watered and flooded plants maintained TL between -0.8 and
-0.4 MPa, suggesting that the flooding treatment did not result in
tissue desiccation. Often flooding results in reduced water uptake
and water stress symptoms similar to those observed during drought
stress (Kramer, 1951). This response was not observed either visually
or with leaf water potential measurements. In both years, YL was
reduced as a result of limiting soil moisture after about 12 days of
water withholding. Between days 10 and 12, iL dropped rapidly, an
occurrence often observed in very sandy soils because of their low water
holding capacity. In 1981, severe desiccation of the plant leaves
occurred as indicated by PL only as low as -1.4 MPa. After rewatering
on day 14, TL of previously droughted plants recovered to values of
well-watered leaves. The leaf water potential data suggest no effect
by the flooding treatment, a rapid progression of desiccation in
plants subjected to drying soil, and complete recovery after rewatering.
The rapid decrease in TL was not observed until the gravimetric soil
water content dropped below about 1% (Figure 1).

Leaf diffusive resistance closely paralleled changes in TL (Figure
3). Throughout both experiments in 1981 and 1982, leaf diffusive resis-
tances for flooded plants were similar to those of the well-watered
plants. The droughted plants exhibited increases in leaf diffusive
resistances after 10 days in both years. Rather large decreases in
leaf diffusive resistance were observed as TL declined very slightly
(Figure 2). After rewatering, leaf diffusive resistance recovered to
control values.

Flooding had no effect on nodule water potentials (Figure 4).
However, Ynod was reduced with reductions in soil water content,
reaching water potentials as low as -1.6 to -1.8 in 1981 and 1982,
respectively. In 1981, there was an obvious trend for reduced Tnod
during the early days after withholding water (days 4, 6, and 8),
suggesting that Tnod may be slightly more sensitive to reductions in
soil water potential than the other measured parameters. In fact, if
data from both studies are examined it is apparent that Tnod declined
with any reduction in gravimetric soil water content, at least, within
the limits imposed in these experiments (Figure 5). As gravimetric





















2``~~0


S.E. of mean

o control
* flooded
a droughted


'.-,


0 2 4 6 8 10 12 14


-.4


-.8


-1.2


1982


...


-S kvf


0o



4--
C
0"




-4--
0
i___


\ /
\ d
\0 "
mean
D4
m-nlln


4 8 12


I I I I I


16 20 24


Days After Treatments Were Imposed

Midday leaf water potentials of well-watered, flooded, and droughted soybeans in
1981 and 1982. S. E. represents the average standard error of the mean.


-.4


-.8


-1.6- 1 ---- I
o control
flooded
-2.0 droughted


Figure 2.


.,e *


...














S1300

r)
1100

0
900
U)

S700
cr

a)
> 500
nU)
:3

-300

S4-
_-j o


1981


S.E. of mean

control
flooded
Sdroughted


.n

p'

/
/
/
/
/
/
/






/


___


.1982


1300


1100


900


700


500


300


100


IS.E.of


mean


o control
* flooded
o drou'ghted


0 2 4 6 8 10
Days After


12 14.
Treatments


I A I


4 8 12 16 20 24
Were Imposed


Figure 3. Midday leaf diffusive resistance of well-watered, flooded, and droughted soybeans
in 1981 and 1982. S. E. represents the average standard error of the mean.


0/
I
I
/
I


"


r
q
9


I I


A |


A A


J I I n














1981


S.E. of


- ---0--.. ---..


mean


0
CL





0
- -

C
-o





2:
O
"4--




0


-.2


-.6


-1.0


1.4


-1.8


1982
. .- .----


SI

i I
II


I
I
I
I,
ri
01


S.E. of mean.

o control
* flooded
odroughted


, i ,


I I I I I I I I I I I I 1


0 2 4 6


8 10


4 '8


16 20 24


Days After Treatments Were Imposed


Midday nodule water potentials of well-watered, flooded, and droughted soybeans
in 1981 and 1982. S. E...represents the average standard error of the mean.


1-L
-.2


-.6


-I.0


-1.4


-1.8


- o control
Flooded
a droughted


Figure 4.


A


. I A


UJ


i


I I







- 17 -


I 3 5 7 9 11


GRAVIMETRIC SOIL WATER


(%)


Figure 5. The relationship between gravimetric soil
and midday soybean nodule water potential
1982.


water content
in 1981 and


-1.0


0
. 6 *0
0

B N


o 1981
* 1982


13
13






- 18 -


soil water contents of about 1% were approached, further reductions in
soil water caused dramatic reductions in Ynod. In 1982, as with the
other parameters measured, Ynod returned to values equivalent to the
well-watered Wnod after rewatering on day 14 (Figure 4).

The relationship between YL and Ynod was evaluated in all three
experiments that were conducted. However, particular emphasis was
placed on this relationship in Experiments II and III. As indicated in
Figure 6, there was not a 1:1 relationship between TL and 'nod. At
the high TL (>-1.0 MPa), 1nod was generally equal or higher than L-.
However, as the leaf water potential decreased below about -1.0 to
-1.2 MPa, Ynod became considerably lower than TL. Most likely, as TL
decreased below -1.0 to -1.2 MPa, stomatal closure occurred (Figures
2 and 3) resulting in the inhibition of transpiration. Stomatal
closure would help to prevent further dehydration and would tend to
maintain TL at potentials near those where the stomata closure was
triggered. Conversely, since nodules lack stomata, a progressive
reduction of Tnod with decreasing soil water content would be expected
if delivery of water to nodules from root vascular tissue could not
resupply the water lost from the nodules to the soil. In the sandy
soil medium used in these experiments, the upper soil layers, where
most nodules were sampled, became extremely dry and warm and probably
represent a steep water potential gradient from nodule to soil. Other
research has shown that nodules lose water very rapidly when subjected
to desiccating environments (Sprent, 1971). It appears that with severe
water stress in sandy soil medium, the nodules may continue to dry to
water potentials below the potentials of leaves. However, it is not
entirely clear whether this drying is simply due to a very steep water
potential gradient between the nodule and soil coupled with the inability
of the nodule to decrease water loss, to increased resistance to water
flow from the root to the nodule, a combination of both, or other processes.
Nevertheless, it is possible that nodules may be subjected to severe
water deficits which may result in irreversible damage (Sprent, 1972),
whereas leaves are protected from severe desiccation by stomatal control
mechanisms.

Contrary to the other physiological parameters measured, flooding
seemed to affect nitrogenase activity differently (Figure 7). In both
years, nitrogenase was apparently reduced immediately after the flooding
treatment was imposed. The initial decrease in nitrogenase activity is
probably due to the depletion of oxygen in the root environment. The
lack of oxygen may reduce nitrogen fixation in one of two ways. Firstly,
it will decrease or eliminate oxidative phosphorylation, which is necessary
for ATP production, and this reduction in energy should cause concomitant
reductions in nitrogenase activity. The activity may only be depressed,
because some ATP may still be produced by some oxidative phosphorylation
which continues at the low oxygen concentrations. Metabolic activity
may shift to fermentation which is capable of less efficient ATP forma-
tion. Secondly, if roots and bacteroids shift their metabolic patterns
to fertmentation, less reduced carbon substrate will be available for
the production of both ATP and electrons required by nitrogen fixation.
Bisseling et al. (1980) reported that, in Pisum sativum inoculated with







- 19 -


-.2


-2.2 -1.8 -1.4 -1.0 .6

LEAF WATER POTENTIAL (MPa)


The relationship between midday soybean leaf water
potential and midday nodule potential for two
experiments in 1982. .The line drawn represents
a 1:1 relationship between the two potentials.


1.01


-1.4F


-2.2


* 0 0

0
00, .0


1:1 / oo


S- 0
0
0 *
0
0
0
/ 0


/ 0


0
0 O
0 02 (1982)

o o Experiment 2 (1982)
o o ExperimentS (1982)


Figure 6.


'"


-1.8















1981

/.J ""'A.



)\-
'*- --


o control
flooded
o droughted

SS.E. of mean

S I i I I I I I I I I I I i I*"


0 2 4 6 8 10


70%


60
50
40
30
20
10
0
O


12 14


1982


o control
* flooded
a droughted
I S.E. of mean


53--0




(I
-I
I
I
I

' /
'I
b0


i "I A I I I I I I I I I I I


8 12


Days After Treatments Were Imposed

Figure 7. Midday nitrogenase activity (acetylene reduction) of well-watered, flooded, and droughted soybeans
in 1981 and 1982. S.E. represents the average standard error of the mean. Rates of nitrogenase
activity in both years are expressed as umoles C2H4 h-l g nodule .


24
21


-D
"0
0
(I




o
ci





'-

E
ZL


16 '20


24


18


k~'






- 21


Rhizobium leguminosarum, waterlogging decreased nitrogenase activity.
They further reported that there was a decrease in the active nitro-
genase enzyme and suggested that the iron protein of the nitrogenase
complex was repressed under waterlogging conditions.

Nitrogenase activity of the flooded plants did, however, tend to
increase during the duration of the experiment until activities higher
than those for well-watered controls were observed by the termination
of both experiments (Figure 7). There was evidence of the initiation
and development of new nodul-es on the flooded plants. These new nodules
were formed at the soil surface on adventitious roots which were rapidly
initiated after the imposition of the flooding treatment. It is likely
that these young nodules increased in number and activity throughout the
experiment, resulting in the enhanced rates of nitrogen fixation observed
in Figure 7. Many plants are also capable of producing aerenchyma tissue
when flooded (Yoshida and Tadano, 1978). Observations of the roots of
the flooded soybeans suggest that the adventitious roots initiated after
the flood treatment was imposed were large and spongy in texture. In
addition, there were gross anatomical changes near the crown of the root
system. While the increase in nodulation may be the major reason that
flooded plants increased and maintained their nitrogenase activity, the
possibility of aerenchyma tissue may also play a role in the increased
activity.

Nitrogen fixation was reduced after 10 days of withholding water
in both 1981 and 1982 and activities declined to essentially zero as
stress became progressively more severe (Figure 7). After rewatering
on day 14 in 1982, nitrogenase activities of previously droughted plants
increased to rates higher than the well-watered controls. Such enhanced
rates after rewatering have also been observed by others (Sprent,1972)
and may result from utilization of an abundant supply of carbohydrates
which accumulated in the nodules during the stress period or from the
removal by transpiration of reduced nitrogen compounds that had pre-
viously inhibited nitrogenase activity during the drought.

Nitrogenase activity in well-watered and droughted plants was
closely related to both nodule water potential (Figure 8) and
respiratory activity of the root-nodule complex (Figure 9). Data
presented in Figure 8 demonstrate the extremely sensitive nature
of nitrogenase activity to the fnod. Even as 'nod declined from -0.2
to -0.4 MPa, nitrogenase activity was reduced. Even slight reductions
in both YL and Tnod can be expected to result in some inhibition of
nitrogenase activity. Plants in these experiments never reached a
point where the water stress caused terminal damage to the tissue.
Plants were rehydrated and quickly regained the activity of the well-
watered plants. In most instances, drought stressed plants were wilted
at midday after the tenth day of the experiment.

It is also interesting to note that nitrogenase activity declined
more than root respiration (Figures 7 and 9), similar to data presented
by Sprent (1971). Sprent also presents information to indicate that
water stress produces physical damage in the nodule. It is not fully






- 22 -


70
1982






< 3O

S> *


C 0


U 0
0
! (I .





-2.2 -1.8 -1.4 -1.0 -.6 -.2

NODULE WATER POTENTIAL (MPa)


Figure 8. The relationship between midday soybean nodule water
potential and-midday nitrogenase activity.














1981. .



.J --


o control
* flooded -
odroughted

S.E. of mean "-


4

-3

2

I


1982
control
*flooded
odroughted


----a


SS.E. ofmean


I I I I A I B p ~ I I I I a .I P *


0 2 4


OL


8 10


a a


8 12


a a 1 1 a


16 20


24


Days After Treatments Were Imposed


Midday root respiration of well-watered, flooded,
represents the average standard error of the mean


and droughted soybeans In 1981 and 1982. S. E.


r
0
0
a:





C)
Ec


Figure 9.


N
'ii


-


L" r r _


--


I~


6V


t -






- 24 -


understood whether this relatively larger decrease in nitrogenase
activity is because of some regulatory activity, the sensitivity of
the metabolic functions measured, actual physical damage in the
tissues during the stress period, or differences in the amount of
water that is available to the root system as compared to the nodule.


B. Aeschynomene and Desmodium, Experiments I, II, III

Experiment I

Twenty-eight accessions of Aeschynomene were rooted in sand and
transplanted to 15-cm black plastic pots containing a mixture of sand,
peat, and perlite (2:1:1). Plants were allowed to establish and grow
for 4 weeks before the pots were arranged in a completely random design
in two flooding tanks. For 1 week, water was maintained at 8 cm below
the soil surface. For the second week, water was maintained at 4 cm
below the soil surface. Visual observations and nitrogenase activities
indicated that the flooding treatment had not resulted in detrimental
effects on any of the accessions. In order to test the hypothesis
that soil moisture affects nitrogenase activity, 16 plants were removed
and allowed to dry for 20 days until the plants wilted, 16 plants were
maintained as well-watered controls, and a corresponding group of
plants remained flooded for an additional 20 days. After 20 days,
midday measurements of yL, leaf osmotic potential, leaf turgor potential,
Tnod, and nitrogenase activity were measured. The plants were then
harvested, and root and nodule dry weights were determined. Percent
nodulation was expressed as nodule dry weight as a percentage of total
root dry weight. Data reported are for a range of accessions with no
evaluation made of any differences which may be caused by different
accessions.

Leaf water potential, leaf turgor potential, leaf osmotic potential,
and nodule water potential were all reduced by the 20 days of withholding
water (Table 1). The flooding treatment did not affect the L', leaf
turgor potential, or leaf osmotic potential, however, there was a trend
for Ynod to remain slightly higher than that of the well-watered controls.

Nodule weight, percent nodulation, and nitrogenase activity also
declined in response to the drying treatment (Table 2). Root weight
was apparently stimulated by the drought period, a response that is
often observed during drought. The flooded plants exhibited no adverse
effects in response to the treatment imposed.

Experiment II

Four accessions of Aeschynomene and one of Desmodium heterocarpon
were grown in sand:peat:perlite mixtures in 15 cm pots to compare the
response of the two species to short-term water stresses. After 10
weeks of growth, pots were either flooded with water to surface level
or water was withheld. Three days after the treatments were imposed,
nitrogenase activity and components of leaf water potential were measured
at midday.






- 25 -


Table 1. Leaf water, osmotic, and turgor potentials of Aeschynomene
leaves and water potentials of nodules as affected by
flooding and soil water deficit.


Treat t Leaf Water Leaf Osmotic Leaf Turgor Nodule Water
Treatment
Potential Potential Potential Potentiala

-------------------------a-----------------

Droughted -1.15 0.9 -1.36 1.2 0.21 0.5 -1.07

Flooded -0.58 0.9 -1.21 0.6 0.63 0.8 -0.25

Control -0.72 0.3 -1.20 0.4 0.48 0.3 -0.45


a Values represent unreplicated observations.






- 26 -


Table 2. Effect of three water treatments on root and nodule
weight, percent nodulation, and nitrogenase activity
in nodules of Aeschynomene americana.



Percent Nitrogenase
Treatment Nodule Wt. Root Weight erceNodulationa Activityb
Nodulationa Activityb

----mg---- ----mg-----

Droughted 26.4 4.4c 336.4 + 4.9 6.8 + 0.3 9.05 + 1.01

Flooded 31.4 1.5 227.4 4.0 13.0 0.6 21.61 + 1.84

Control 32.9 6.7 200.5 2.8 13.5 + 0.9 18.58 + 1.94


a Nodule dry weight x 100/total root dry weight.


b umoles C2H4
c All values


h-1 / gram dry wt. nodule.
determined after 21 days of treatment.






- 27 -


Although Aeschynomene accessions subjected to flooding had leaf
water relations and nitrogenase activities which were similar to those
of the well-watered plants (Tables 3 and 4), flooding dramatically
reduced the nitrogenase activity of Desmodium plants. The flooding
effect also reduced leaf turgor potential to zero (Table 3). Visual
stress symptoms (yellowing and wilting) were also observed in flooded
Desmodium plants. Such symptoms were not observed with Aeschynomene
plants. Drought stress reduced the leaf water and turgor potentials
and nitrogenase activities of both Aeschynomene and Desmodium (Tables
3 and 4). Accession UF 186 appeared somewhat more resistant to the
drought.

Experiment III

Aeschynomene seedlings were subjected to varying concentrations
of 600 molecular weight polyethylene glycol (PEG). Root systems of the
small seedlings were placed into test tubes containing PEG solutions of
-1.0, -2.5, and -5.0 bars osmotic potential. Plants were then maintained
in a laboratory environment for a 7-day period. After 3 and 7 days of
treatment, nitrogenase activities were determined for plants grown in
each PEG concentration.

Nitrogenase activity was depressed in the PEG solutions after 3
days and declined even more after 7 days of stress (Figure 10). Even
at an osmotic potential of only -1 bar, nitrogenase activities were
reduced by 50 and 85% after 3 and 7 days of stress, respectively,
suggesting the extreme sensitivity of nitrogenase activity to drought
stress. Even though the plants subjected to the -1 bar osmotic stress
appeared turgid, nitrogenase activity was significantly depressed.


C. Cowpea and Bean

An experiment evaluating the response of cowpeas and beans to both
flooding and soil drying was conducted in 1981. The inoculated seeds
of both crops were planted 15 October 1981. Flooding and soil drying
treatments were imposed beginning 4 December and lasted for 17 days.
Data were collected for soil and plant parameters, as described in Part
A above, at midday 3, 5, 7, 10, 12, 14, and 17 days after the treatments
were imposed.

Unfortunately, very little difference in leaf water potential of
cowpeas was observed for any of the treatments (Figure 11). Considering
the variation inherent in the data, we conclude that leaf water potential
was not affected by the flooding or drought treatment during the time
period encompassed by this experiment. Apparently, the soil did not dry
sufficiently during the 17-day period to cause leaf water potential
reductions. If the experiment had continued for a longer period, we
would have expected the drought treatment to reduce leaf water potentials.

Despite the lack of a difference in leaf water potential, the drying
treatment did result in slightly higher leaf diffusive resistances on the






- 28 -


Table 3. Water stress effects on leaf water (TL) and turgor (,p)
potentials of Aeschynomene and Desmodium accessions.


Genus


viscosa (UF 369)

americana (UF 57)

americana (UF 186)

americana (UF 255)

heterocarpon


Control Flooded Droughted
'L 'p TL Tp 'L 'p
--------------------MPa-----------------

-.31 .25 -.17 .31 -.70 -.12a

-.27 .34 -.37 .33 -1.60 -.20

-.44 .30 -.29 .45 -1.35 .29

-.72 .37 -.30 .51 -1.20 -.01

-.34 .24 -.32 -.01 -.95 -.21


a Calculation of Yp yields some negative values since apoplastic
dilution of the cell sap after freezing causes a slight under-
estimation of 'p.






- 29 -


Table 4. Comparison of water stress effects on nitrogenase activity
in Aeschynomene and Desmodium.


Genus Control

A. villosa (UF 369) 2.58 + 0.67a
(100)b

A. americana (UF 57) 5.13 + 0.27
(100)

A. americana (UF 186) 1.88 0.63
(100)

A. americana (UF 255) 3.45 + 0.81
(100)

D. heterocarpon (UF 20) 6.49 + 0.89
(100)

a umoles ethylene formed h-1 gram dry
the mean.

b Percent of well-watered control.


Flooded Droughted

2.43 0.47 0.20 0.06
(95) (8)

5.65 0.64 0.11 0.03
(110) (2)

1.94 0.24 0.76 0.12
(103) (40)

3.97 0.66 0.09 0.03
(114) (3)

0.18 0.06 2.40 0.56
(3) (37)

weight standard error of


s







O
--"
0
0


h 3-
S ,L-^ 7 DAYS TREATMENT
S\o--o 3 DAYS TREATMENT
.'ERROR BARS ARE SE.M.
LI LJ 2 -'-

Z .. "'
0

O0 17
2 O -.
0 1 2 3 4 5
PEG-INDUCED OSMOTIC POTENTIAL BARS)
Figure 10. The effect of PEG imposed stresses for 3 and 7 days on nitrogenase activity of Aeschynomene americana.



















S.E.of mean


* CONTROL
o FLOODED
o DROUGHTED


I I I I I I


DAYS AFTER TREATMENTS WERE IMPOSED


Figure 11. Midday leaf water potentials of well-watered, flooded, and droughted cowpeas
in 1981. S. E. represents the average standard error of the mean.


-0.2


COWPEA


-0.4k


-0.6-


-0.8-


-1.0






- 32


last four days of the experiment (Figure 12). The flooding treatment,
however, did not influence leaf diffusive resistance.

Cowpea nodule water potential for the droughted plants decreased
markedly between day 14 and 17, dropping from -0.6 MPa on day 14 to
-1.9 MPa on day 17, indicating severe desiccation of nodule tissue by
the last day of the experiment (Figure 13). Generally, the flooding
treatment had little effect on the nodule water potentials, however,
there was a trend for the flooded nodules to maintain slightly higher
water potentials than those of the well-watered controls.

Nitrogenase activity of the cowpea plants showed a similar response
to that observed for diffusive resistance (Figure 14). The drought stress
inhibited nitrogenase activity on the last four days of the experiment,
despite no reduction in leaf water potential or nodule water potential
until day 17. Again, these observations suggest that nitrogenase activity
may be reduced in the very early stages of drought. The flooding treat-
ment appeared to have little effect on nitrogenase activity. Root respira-
tion of flooded plants was slightly inhibited initially after imposing
the treatment, but seemed to increase throughout the duration of the
experiment and tended to be higher than in the well-watered controls
by the termination of the experiment (Figure 15). The drought stress
inhibited root respiration, especially on the last two days of the
experiment. Root respiration, although slightly depressed, was not as
sensitive to water stress as was nitrogen fixation.

In summary, flooding had little effect on the physiological processes
measured in cowpeas. Nitrogen fixation was particularly sensitive to the
imposed drought and was affected earlier and to a greater degree than
stomatal resistance, leaf or nodule water potential, or root respiration.

Results from the experiment which evaluated the response of bean
plants to either flooding or drought are shown in Figures 16a-18. As
with cowpeas, apparently the drought stress was not imposed for a long
enough period of time to reduce leaf water potentials (Figure 16a).
However, flooding reduced leaf water potential by 0.2 to 0.3 MPa through-
out the experiment. It is probable that the flooding treatment reduced
uptake of water by roots, thus leading to slight reductions of water
potentials in the leaf tissue. Although the drought treatment had no
effect on leaf diffusive resistance, flooding resulted in partial stomatal
closure on all sampling dates (Figure 16b), probably in response to the
lower leaf water potentials and reduced water uptake by the roots.

Nodule water potential was unaffected by the flooding treatment and
reduced only on the last day of the experiment by the imposed drought
(Figure 16c). Nodules were also observed to form on adventitious roots
of beans during the flooding treatment. At the same time, older nodules
that were subjected to flooding became very mushy and dark in color
after about the tenth day of the treatment. Leaves of the flooded bean
plants also became quite yellow after about 10 days.

Nitrogenase activity was almost completely inhibited by imposing
the flooding treatment (Figure 17). Very low rates of activity were










COWPEA


3000,


*CONTROL
o FLOODED
a DROUGHTED



I
[] I
0I I

I V


E

U)


CO

IZ





LLU




_j
LjI

-J


Figure 12.


)4 8 12 16 20
DAYS AFTER TREATMENTS WERE IMPOSED

Midday leaf diffusive resistance of well-watered, flooded, and droughted
cowpeas in 1981. S..E. represents the average standard error of the mean.


20001


I000


/ S.E.of mean



2 I I 2 I I


600k


200k


a a











-0.2




-0.6 -

-J


Z -1.0
W
OL
I-



QC

Ld

0
z
O


a CONTROL
o FLOODED
o DROUGHTED


Figure 13.


I I I I I I I I I I
0 4 8 12 16 20
DAYS AFTER TREATMENTS WERE IMPOSED

Midday nodule. water potentials of well-watered, flooded, and droughted cowpeas
in 1981. S. E. represents the average standard error of the mean.


COWPEA


os.
/o


\ S.E. of mean


I
I
I
I

I'


O'DZ










COWPEA


S.Eof mean


51


' -0





zo






0
E
E>


\ -


4 8
DAYS AFTER
Midday nitrogenase activity
droughted cowpeas in 1981.


12 16 20
TREATMENTS WERE IMPOSED


(acetylene reduction) of well-watered, flooded, and
S. E. represents the average standard error of the mean.


*CONTROL
FLOODED
DDROUGHTED


Figure 14.


. I













COWPEA


.IO
100


z 8
0 8


rE





80
2O4
0



0 20
C31


S.E.of mean I


.---.-


*CONTROL
FLOODED
a DROUGHTED


Figure 15.


II I I I I I
4 8 12 16
DAYS AFTER TREATMENTS WERE IMPOSED
Midday root respiration of well-watered, flooded, and droughted
1981. S. E. represents the average standard error of the mean.


I I


cowpeas in










-0.2


-0.61


BEAN


0 y ,

\o. .oI1"0.


/ 0


-I.0-


* CONTROL
o FLOODED
o DROUGHTED


S.E.of mean I


SI i I i


Figure 16a.


SI I I I I


0 4 8 12 16 20
DAYS AFTER TREATMENTS WERE IMPOSED
Midday leaf water potentials of well-watered, flooded, and droughted beans in
1981. S. E. represents the average standard error of the mean.


1.8-









BEAN
3000
o FLOODED
DROUGHTED


2oo00


,o1


E

0
Oz


UJ
U
Q
U-

U
E3
LL



d
_1


0


0/
o
oV/A'


o

/
/
/
/



I
I
/
/
/


S.E.of mean


200-


I I I


i I i I


Figure 16b.


0 4 8 12. 16 20
DAYS AFTER TREATMENTS WERE IMPOSED
Midday nodule water potential of well-watered, flooded, and droughted beans
in 1981. S. E. represents the average standard error of the mean.


600-





















\ S.E. of mean
I


*CONTROL
o FLOODED
o DROUGHTED


9 3 I I I I a S


3 4 8 12
DAYS AFTER TREATMENTS


Figure 16c.


16 20
WERE IMPOSED


Midday leaf diffusive resistance of well-watered, flooded, and
beans in 1981. S. E. represents the average standard error of


-0.2


BEAN


Q.6


-1.41


droughted
the mean.


- 1


-I..0


-1.8














L.




-c-
:o










0
0
ti









0


*CONTROL
FLOODED
n DROUGHTED


o
N,,)~ C~
~cc .~ O9'


DAYS AFTER TREATMENTS WERE IMPOSED


Figure 17.


Midday nitrogenase activity (acetylene reduction) of well-watered, flooded, and
droughted beans in 1981. S. E. represents the average standard error of the mean.


\
\
\
b S. E. of mean
\
v
\
\ i


! !






- 41 -


observed, even after only 3 days exposure to flooding. The drought
stress also reduced nitrogen fixation on the last two measurement
dates, apparently in direct response to lowered nodule water potentials
(Figure 16c). By the last day of the experiment, nitrogenase activities
were essentially zero in both the flooded and droughted treatments
(Figure 17).

Root respiration was not reduced by the drought stress until the
last two measurement dates (Figure 18), after which it was reduced by
40 to 50%. Except for the measurements taken on day 12, root respira-
tion in the flooded plants was reduced below levels observed for the
control plants. As was the case for cowpeas, flooding reduced root
respiration, but not as severely as nitrogen fixation.

The data suggest that nitrogen fixation in bean was particularly
sensitive to flooding. In fact, even short-term flooding almost
completely inhibited nitrogen fixation as nodules became dark and
mushy. Leaf water potentials and leaf diffusive resistance also
reflected the detrimental effects of the flooding treatment. Toward
the end of the experimental period, drought stress reduced nodule water
potential, root respiration, and nitrogenase activity. These reductions
were observed before any stress symptoms were observed in leaf water
potentials or diffusive resistance.


D. Alfalfa

The effect of water stress, both drought and flooding, on nitro-
genase activity in alfalfa were studied in a greenhouse experiment
during the spring of 1982. Alfalfa seeds (cultivar Florida 77), inocu-
lated with a peat based Rhizobium culture, were planted on 24 February
1982. The seedlings emerged four days later, and treatments were imposed
77 days after emergence. The water stress treatments lasted for 10 days.
Measurements were made, starting at midday, on the first, third, fifth,
sixth, eighth, and tenth days after treatments were imposed. The anatomy
of the alfalfa plant, especially the small size of the leaflet and nodule,
made it impractical to obtain measurements on leaf diffusive resistance
and nodule water potential.

Leaf water potentials remained stable in the control plants during
the stress period, exhibiting very little variation from day to day
(Figure 19). Flooded plants were reasonably stable for the first six
days, then decreased, with the average water potential one MPa lower than
the control plants 10 days after the treatments were imposed. Drought
stressed plants exhibited a marked decrease on the fifth day of the
stress period. Rewatering of the plants restored the leaf water potential
to that of the control levels.

The depletion of soil water during the drying cycle is shown in
Figure 20. The percent water in the soil was reduced to below 1% within
five days after the stress was imposed. Rewatering rapidly restored the
soil moisture to 6%, slightly above the average level for the controls.











120 [


.JOO
..00
o






o60

E





c20
Z4


F


BEAN


S.E. of mean


-n
- -


F


I


* CONTROL
o DROUGHTED
o FLOODED


I I


I I


20


DAYS AFTER TREATMENTS WERE IMPOSED
Figure 18. Midday root respiration of well-watered, flooded, and droughted beans in 1981.
represents the average standard error of the mean.


S. E.










ALFALFA


-2h


S.E. of mean


*CONTROL
o FLOODED
o DROUGHTED


i I I I I I I --


I 2 3 4 5 6
DAYS AFTER TREATMENTS


Figure 19.


7 8 9 10
WERE IMPOSED


Midday leaf water potentials of well-watered, flooded, and droughted
alfalfa in 1981. S. E. represents the average standard error of the
mean.


-31


I I


i I I I I I I I I I I














0-'
8-
II
S.E. of mean

6-

U) .


\ / -\
4 %



r 2-


CONTROL \ /

%a
u-" -D
I I I I I 1 I I I
0 I 2 3 4 5 6 7 8 9 10
DAYS AFTER TREATMENTS WERE IMPOSED
Figure 20. Gravimetric soil water content of well-watered and droughted alfalfa in
1981. S. E. represents the average standard error of-the mean.






- 45


Nitrogenase activity for the control and drought stressed plants
parallels leaf water potentials (Figure 21). Nitrogenase activity is
reduced to less than one-third of the control plants by day eight of
the stress, however, rewatering restored the nitrogenase activity to
that of the control. Flooding reduced nitrogenase activity to less than
one-sixth of the control in only 3 days, and it remained depressed for
the duration of the stress period. This rapid and almost total loss of
nitrogenase activity suggests that some metabolic function in the
nodule is irreversibly destroyed by flooding.

The response of root respiration, which closely parallels nitro-
genase activity and leaf water potentials is presented in Figure 22.
Once again, the control plants remain relatively stable throughout the
10-day stress period, while the drought stressed plants show a reduction
in respiration on day six. Rewatering the drought stressed plants
restores respiration to control levels. As observed in the nitrogenase
activity, respiration is depressed by flooding to about 50% that of the
control plants.









ALFALFA


S.E. of mean


-r
5-
-0








0
z


-


\0

0 %
% D.


"' -


I 2 3 4 5 6 7 8 9 10
DAYS AFTER TREATMENTS WERE IMPOSED


Figure 21.


Midday nitrogenase activity (acetylene reduction) of well-watered, flooded, and
droughted alfalfa in 1981. S. E. represents the average standard error of the mean.


301


2400

2100

1900

1600

1300


*CONTROL
0 DROUGHTED
o FLOODED


/
/
/
/
/
/
.1
/


0\.

0~~0 ~LOZI~









ALFALFA


S.E. of mean


a





-
c:



Q^u


3000

2700

2400

2100-

1900

1600

1300

1000


000

I I a I I 1 I I I a
I 2 3 4 5 6 7 8 9 10
DAYS AFTER TREATMENTS WERE IMPOSED


Figure 22.


Midday root respiration of well-watered, flooded, and droughted alfalfa
S. E. represents the average standard error of the mean.


*CONTROL
a DROUGHTED
o FLOODED


' /

D
o iS
C//


700-

400-

200k

100-

0-


in 1981.














CHAPTER III. FIELD EXPERIMENT, SOYBEANS


An adjunct experiment was conducted to evaluate the effect of
water deficits on field-grown soybeans. Soybeans were grown in field
plots located at the Irrigation Research and Education Park on the
University of Florida campus. Although the field experiment was
designed to study the response of soybeans to six water management
treatments, for purposes of the nitrogen fixation study, only a well-
watered and stressed treatment were monitored during mid-reproductive
growth. The drying cycle imposed on the stressed treatment lasted
for 19 days and was relieved by irrigations before recovery rates were
determined. The well-irrigated treatment was irrigated when the soil
water potential at 15 cm reached -15 to -20 KPa. Measurements of
canopy carbon exchange rates, leaf water potential, leaf stomatal con-
ductance, percent nodule moisture, and nitrogenase activity were made
periodically between 1000 and 1500 hours during the drying cycle.

Leaf water potential, leaf conductance to water vapor, nodule
moisture, apparent canopy carbon exchange, and nitrogenase activity
declined as the duration of the stress period increased (Figures 23-
27). All physiological processes, except apparent canopy carbon
exchange rate returned to control levels upon adequate rewatering. The
incomplete recovery of apparent carbon exchange was attributed to
decreased leaf area index resulting from leaf senescence during the
stress period. Nitrogenase activity was reduced to almost 0 by day 16,
and remained very low until the soil profile was rewatered (Figure 26).
It is interesting to note that nitrogenase activity showed only slight
recovery on day 21. The soil profile was only partially filled by the
irrigation on day 20, and although the majority of the nodules were in
wet soil, the nitrogenase activity continued to be depressed. This
suggests that nodule water is derived from roots deeper in the soil and
that the nodules were apparently not capable of complete rehydration
from immediately surrounding soil moisture.

While nitrogenase activity showed a decreasing linear trend with
decreasing percent nodule moisture, the data in Figure 28 indicate
that there is a sharp end point once the nodules dry to a nodule
moisture of about 50 to 55%.

Although nitrogenase activity appeared to decline concurrent with
the other physiological processes, the magnitude of the depression near
the end of the drying cycle was greater for nitrogenase activity. When
canopy carbon exchange was reduced by about 50-60%, and when stomata
were only partially closed, nitrogenase activity dropped to essentially
zero.

The data support the hypothesis that decreases in nitrogenase
activity are primarily caused by nodule dehydration and tissue damage
and not directly related to short-term changes in photosynthesis.


- 48 -












IRRIGATED


-I.OH


DRY


1.4k


-1.8 -


DAYS OF


Figure 23.


DRYING CYCLE


Effect of a soil water drying cycle on soybean leaf water potential. Arrows
on the abscissa at days 20 and 22 indicate rewatering of the stressed plants.


T


20


24


-0.2


-0.6


1:




























Figure 24. Effect of a soil water drying cycle on soybean leaf conductance to water.
Arrows on the abscissa at days 20 and 22 indicate rewatering of the stressed
plants.

0n
I





.06


.05-


E .0 \
.4 IRRIGATED
LU \

.03
C) /

0 .02- -DRY


.01-


I I I I
U-



0 4 8 12 16 20 24
DAYS OF DRYING CYCLE







1.3
z


X
W
z
0
m 0.9
a:
<-177

n E 0.7
0 ca
Z E

H 0.5
z

a:
a. 0.3


\ T




DRY\

\ /


H


DAYS OF DRYING CYCLE


Figure 25.


Effect of a soil water drying cycle on soybean apparent canopy carbon exchange.
Arrows on the abscissa at days 20 and 22 indicate rewatering of the stressed plants.


/-


/
/
/
/
'9


\{/


', I


20


24


































Figure 26. Effect of a soil water drying cycle on soybean nodule nitrogenase activity. Arrows
on the abscissa at days 20 and 22 indicate rewatering of the stressed plants.



I
(n




35

I'
30 I'


-Io I
S' 25I

S1 IRRIGATED
20

z N\
0 a, \ ] -- / \
\ I
I

5. DR


SI I
0 4 8 12 16 20 24 28
DAYS OF DRYING CYCLE













80




Ut
70

o IRRIGATED /

w/
: 60





t 50 DRY 1

REWATER
V V



DAYS OF DRYING CYCLE

Figure 27. Effect of a soil water drying cycle on soybean nodule moisture
content.





























Figure 28. The relationship of percent nodule moisture to nitrogenase activity during a soil
water drying cycle.



01
07i







21-



18

*
*

>- 15



Z- *



wo9


S*






OO
0 *










80 70 60 50 40 30 20 10

PERCENT NODULE MOISTURE
3-





PERCENT NODULE MOISTURE










CHAPTER IV. SUMMARY


The current study revealed important characteristics about the
relationship of plant water status and metabolism to nitrogen fixa-
tion by legumes in sandy soils. Results of these experiments suggest
that plant and nodule water status are closely linked to nitrogen
fixation. However, often nitrogen fixation appeared more sensitive
to water stresses than some of the other measured physiological
parameters. In all cases, by the time visible stress was observed
or reductions in leaf water potentials or stomatal conductance
occurred, nitrogen fixation was also depressed.

This study has shown that drought stress decreases symbiotic
nitrogen fixation and root respiration in legumes. The changes
observed during the drought stress periods suggest that these may
be tissue damage, and these observations, coupled to the measured
physiological changes support the hypothesis that drought stress
directly affects the interactions between Rhizobium and the host
plant, as suggested by Sprent (1976).

Flooding provided a mixed response in the observed crops.
Soybeans, cowpeas, and Aeschynomene showed little adverse effects
during the time they were flooded, while alfalfa, Desmodium, and
bean exhibited sensitivity to the imposed stress. This sensitivity
was most readily noted by reduced root respiration and nitrogen
fixation.

Legume crops in Florida can be subjected to periods of water
stress, both drought and flooding, during their growth. In most
instances, these stresses result in lower leaf water potentials,
stomatal closure, decreased photosynthesis, reduced nitrogen fixa-
tion, and impaired respiratory functions. Any of the impairments,
if of sufficient magnitude, can result in reduced yields and
inefficient utilization of other management inputs. Physiological
changes which occur in response to water stresses should ultimately
help integrate efficient water management schemes into management
practices.


- 58 -












LITERATURE CITED


Albrecht, S. L., J. M. Bennett, and K. H. Quesenberry. 1981. Growth
and nitrogen fixation of Aeschynomene under water stressed conditions.
Plant and Soil 60:309-315.

Bisseling, T., W. van Stavern, and A. van Kammen. 1980. The effect of
waterlogging on the synthesis of the nitrogenase components in
bacteroids of Rhizobium lequminosarum in root nodules of Pisum
sativum. Biochem. and Biophy. Res. Comm. 93:687-693.

Bourget, S. J., and R. B. Carson. 1962. Effect of soil moisture stress
on yield, water-use efficiency and mineral composition of oats and
alfalfa grown at two fertility levels. Can. J. Soil Sci. 40:7-12.

Brolmann, J. B. 1978. Flood tolerance in Stylosanthes, a tropical
legume. Proc. Soil and Crop Sci. Soc. of Fla. 37:37-39.

Burris, R. G., and P. W. Wilson. 1957. Methods for measurement of
nitrogen fixation. In Methods and Enzymology. S. P. Colowick and
N. 0. Kaplan (eds.), 4:3-55. Academic Press, New York.

Clegg, M. D., C. Y. Sullivan, and J. D. Eastin. 1978. A sensitive
technique for the rapid measurement of carbon dioxide concentrations:
Plant Physiol. 62:924-926.

Dilworth, M. J. 1966. Acetylene reduction by nitrogen-fixing prepara-
tions from Clostridium pasteurianum. Biochem. Biophys. Acta
127:285-294.

Engin, M., and J. I. Sprent. 1973. Effects of water stress on growth
and nitrogen-fixing activity of Trifolium repens. New Phytol.
72:117-126.

Evans, H. J., and S. A. Russell. 1971. Physiological chemistry of
symbiotic nitrogen fixation by legumes. In Chemistry and Biochemistry
of Nitrogen Fixation, J. R. Postgate (ed.-, pp. 191-244. Plenum
Press, London.

Fishbeck, K., H. J. Evans, and L. I. Boersma. 1973. Measurement of
nitrogenase activity of intact legume symbionts in situ using the
acetylene reduction assay. Agron. J. 65:429-433.

Gallacher, A. G., and J. I. Sprent. 1978. The effect of different
water regimes on growth and nodule development of greenhouse-grown
Vicia faba. J. Exp. Bot. 29:413-423.

Huang, C-Y., J. S. Boyer, and L. N. Vanderhoef. 1975a. Acetylene
reduction (nitrogen fixation) and metabolic activities of soybean
having various leaf and nodule water potentials. Plant Physiol.
56:222-227.


- 59 -






- 60 -


Huang, C-Y., J. S. Boyer, and L.
acetylene reduction (nitrogen
having low water potentials.


N. Vanderhoef. 1975b. Limitation of
fixation) by photosynthesis in soybean
Plant Physiol. 56:228-232.


Kilmer, V. J., 0. B. L. Bennett, V. F. Stahly, and D. R. Timmons. 1960.
Yield and mineral composition of eight forage species grown at four
levels of soil moisture. Agron. J. 52:282-285.

Kramer, P. J. 1951. Causes of injury to plants resulting from flooding
of the soil. Plant Physiol. 26:722-736.


Lawn, R. J.,
soybeans.
Crop Sci.


and W. A. Brun. 1974. Symbiotic nitrogen fixation in
I. Effects of photosynthetic source-sink manipulations.
14:11-16.


Ludlow, M.
pasture
grasses


M., and G. L. Wilson.
plants. IV. Basis and
and legumes. Aust. J.


1972. Photosynthesis of tropical
consequences of differences between
Biol. Sci. 25:1133-1145.


Mack, A. R. 1973. Soil temperature and moisture conditions in relation
to the growth and quality of field peas. Can. J. Soil Sci. 53:59-72.

Mague, T. H., and R. H. Burris. 1972. Reduction of acetylene and
nitrogen by field grown soybeans. New Phytol. 71:275-286.

Mclvor, J. G. 1976. The effect of waterlogging on the growth of
Stylosanthes guyanensis. Tropical Grasslands 10:173-178.

Mederski, H. J., and D. L. Jeffers. 1973. Yield response of soybean
varieties grown at two moisture stress levels. Agron. J. 65:410-412.


Minchin, F. R., and J. S. Pate. 1975.
salt regime on nitrogen fixation in
of an optimum root environment. J.


Effect of water, aeration and
a nodulated legume definition
Exp. Bot. 26:60-69.


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

Pankhurst, C. E., and J. I. Sprent. 1975. Effects of water stress on
the respiratory and nitrogen-fixing activity of soybean root nodules.
J. Exp. Bot. 26:287-304.

Pate, J. S. 1958. Nodulation studies in legumes. II. The influence
of various environmental factors of symbiotic expression in the
vetch (Vicia sativa L.) and other legumes. Aust. J. Biol. Sci.
11:496-515.

Quesenberry, K. H., S. L. Albrecht, and J. M. Bennett. 1982. Nitrogen
fixation and forage characterization of Aeschynomene spp. in a
subtropical climate. In Biological Nitrogen Fixation Technology for
Tropical Agriculture, P. H. Graham and S. C. Harris (eds.). pp. 347-
354. NifTAL, Honolulu.









Salter, P. J., and D. H. Drew. 1965. Root growth as a factor in the
response of Pisum sativum L. to irrigation. Nature 206:1063-1064.

Schollhorn, R., and R. H. Burris. 1966. Studies of intermediates in
nitrogen fixation. Fed. Proc., Fed. Amer. Soc. Biol. 24:710.

Schwinghamer, E. A., H. J. Evans, and M. D. Dawson. 1970. Evaluation
of effectiveness in mutant strains of Rhizobium by acetylene reduc-
tion relative to other criteria of N2 fixation. Plant and Soil
33:192-212.

Sprent, J. I. 1969. Prolonged reduction of acetylene by detached
soybean nodules. Planta 88:372-375.

Sprent, J. I. 1971. The effects of water stress on nitrogen-fixing
root nodules. I. Effects on the physiology of detached soybean
nodules. New Phytol. 70:9-17.

Sprent, J. I. 1972. The effects of water stress on nitrogen-fixing
root nodules. II. Effects on the fine structure of detached soybean
nodules. New Phytol. 71:443-450.

Sprent, J. I. 1973. Growth and nitrogen fixation in Lupinus arboreus
as affected by shading and water supply. New Phytol. 72:1005-1022.

Sprent, J. I. 1976a. Nitrogen fixation by legumes subjected to water
and light stresses. In Symbiotic Nitrogen Fixation in Plants, P. S.
Nutman (ed,). pp. 405-520. Cambridge University Press, Cambridge,
U. K.

Sprent, J. I. 1976b. Water deficits and nitrogen-fixing root nodules.
In Water Deficits and Plant Growth, Vol. VI, Soil Water Management,
Plant Responses and Breeding for Drought Resistance. T. T. Kozlowski
(ed.). pp. 291-315. Academic Press, New York.

Yoshida, S., and T. Tadano. 1978. Adaptation of plants to submerged
soils. In Crop Tolerance to Suboptimal Land Conditions. pp. 233-254.
ASA Publication, Madison, Wisconsin.


- 61 -












PUBLICATIONS


Quesenberry, K. H., S. L. Albrecht, and J. M. Bennett. 1982.
Nitrogen fixation and forage characterization of Aeschynomene spp.
in a subtropical climate. In Biological Nitrogen Fixation Technology
for Tropical Agriculture, P. H. Graham and S. C. Harris (eds.).
pp. 347-354. NifTAL, Honolulu.

Albrecht, S. L., J. M. Bennett, and K. J. Boote. Relationship
of nitrogenase activity to plant water status in field grown soybeans.
Submitted to Field Crops Research.

Albrecht, S. L., and J. M. Bennett. Nitrogen fixation in forage
legumes in relation to plant water and metabolic status in drought
stressed and flooded conditions. In preparation.

Bennett, J. M., and S. L. Albrecht. Response to nitrogen fixa-
tion to soil moisture stress in beans and cowpeas. In preparation.

Bennett, J. M., and S. L. Albrecht. Soybean nitrogenase activity
as affected by plant water stress. Temporal changes and the relation
to plant and nodule water status. In preparation.


- 62 -




Full Text

PAGE 1

WATER IiRESOURCES researc center Publication No. 69 THE EFFECT OF EXCESS AND LIMITED SOIL MOISTURE ON NITROGEN FIXATION OF SEVERAL LEGUMINOUS CROPS By Jerry M. Bennett Agronomy Department University of Florida Gainesville and S. L. Albrecht USDA-ARS Gainesville UNIVERSITY OF FLORIDA

PAGE 2

TABLE OF CONTENTS ... 3 ABSTRACT .... 4 CHAPTER I. INTRODUCTION ..... 5 II. GREENHOUSE EXPERIMENTS ..... 9 A. SOYBEANS EXPERIMENTS I, II, AND II I .............. 9 B. AESCHYNOMENE AND DESMODIUM, EXPERIMENTS I, I I, AN D I I I ... 24 C COW PEA AN D BEAN .... 27 D. ALFALFA 41 Ill. FIELD EXPERIMENT, SOyBEANS ............................ 48 IV. SUMMARy ... 58 LITERATURE CITED 59 PUBLISHED PAPERS ... 62 -2 -

PAGE 3

ACKNOWLEDGEMENTS We are indebted to Carmen Harris for the many hours of technical assistance provided in this study. We also appreciate the assistance of Ann Newitt, Gary Moore, Mark Moser, and Andrea Givens. Thanks are also expressed to Dr. K. J. Boote for allowing us to utilize soybean field plots for an adjunct experiment and for providing data on carbon exchange rates. We also thank Dr. K. H. Quesenberry for his cooperation and for providing greenhouse space for the Aeschynomene and Desmodium experiments. Our appreciation is extended to Carolyn Meyer for final assembly and typing of this report. -3 -

PAGE 4

ABSTRACT Experimental greenhouse work was conducted to examine the effects of legume crop species and soil moisture content on leaf and nodule water status, stomatal resistance, root respiration, and nitrogen fixation. Data were collected on forage and grain crops that are important in Florida and increasing in importance in other semi-tropical and tropical climates. The effect of both flooding and drought on the capacity of Aeschynomene americana, Desmodium heteroCaryon, cowpea (Vigna unuicu1ata L.), alfalfa (Medicago sativa L. soybean (Glycine max L.), and bean (Phaseo1us vulgaris L.) to fix nitrogen was examined in a series of experiments. Evaluations of 28 accessions of Aeschynomene suggested a tolerance to flooding. The capacity td fix nitrogen in flooded plants remained about the same as the well-watered controls, and the number of nodules was found to increase on lateral roots which initiated after the flooding was imposed. In contrast, flooded Desmodium, bean, and alfalfa showed a marked decrease in nitrogenase activity, with a concomitant decline in plant water status and root respiration. Drought stressed Aeschynomene, Desmodium, and alfalfa plants showed a decrease in the abil ity to fi x nitrogen whi ch was associ ated with reductions in leaf water potentials. In drought stressed soybeans, large reductions in leaf and nodule water potent1a1s, stomatal conductance, and root respiration occurred as gravimetric soil water content decreased below about 1%. In general, nitrogenase activity declined along with other physiological processes, however, there was an indication that nitrogen fixation was more severely inhibited by drought. In an adjunct field study, nitrogen fixation was reduced by soil water deficits more than photosynthesis or stomatal conductance, suggesting that reductions in nodule water status had a direct effect on nitrogen fixation. Rewatering previously drought stressed plants restored control levels of the measured physiological parameters within 24 hours. Flooded soybean plants increased rates of nitrogen fixation during the flooding period, while other plant water stress indicators remained similar to those of well-watered plants. The increase in nitrogenase activity was probably due to young nodules which initiated on newly formed lateral roots. In cowpeas, flooding had little effect on.either water status or nitrogen fixation activity. Bean plants were sensitive to the flooding treatments and even though plant and nodule water potentials varied little from normally watered plants, nitrogenase activity was severely inhibited. While drought stress was detrimental to nitrogenase activity in all species examined, differential responses to flooding were observed. Desmodium, alfalfa, and bean were qu;,te susceptible to flooding while soybean and cowpea were moderately tolerant and Aeschynomene was very tolerant. -4 -

PAGE 5

CHAPTER I. INTRODUCTION Next to photosynthesis, the reduction of atmospheric nitrogen to ammonia is probably the most important reduction reaction on earth. The ability to reduce atmospheric nitrogen to ammonia is confined to certain prokaryotic organisms, including the genus Rhizobia, which. mqy live symbiotically with plants. Agriculturally, legumes are the most important of the symbiotic systems involving Rhizobia. Before 1966, nitrogen fixation was determined by either total nitrogen analysis or measuring 15N2 incorporation (Burris and Wilson, 1957). These methods are time consuming and expensive, however, the discovery that the nitrogenase enzyme complex can also reduce acetylene to ethylene (Dilworth, 1966; Schollhorn and Burris, 1966) provided the basis for a rapid, inexpensive assay, using gas chromatography to separate the two hydrocarbons. Soi 1 moi sture has long been known to be a factor that wi 11 i f1uence the growth of plants. Members of the Leguminosae family are generally intolerant of either an excess or a deficiency of water in their root environment, and there are numerous reports that water stress will affect both yield and quality of several legume crops (Kilmer et al., 1960; Bourget and Carson, 1962; Mack, 1973). The effect of soil moi sture on nitrogen fi xation has received re1 ati vely 1 itt1 e attenti on until the past decade, but there are several reports in the literature which suggest that symbiotic nitrogen fixation by legume root nodules is very sensitive to either desiccated or flooded soils. Nitrogen fixation by root nodules requires that the soil be able to supply optimum amounts of both water and certain gases, notably nitrogen and oxygen (Evans and Russell, 1971). The simultaneous need for the soil to be wet enough for adequate water supply to the root system and dry ,enough for gaseous exchange allows a very narrow optimum for nodule nitrogenase activity (Huang et al., 1975a). There have been examinations of water stress effectS-on-nitrogenase in a number of representative legumes including Trifolium repens (Engin and Sprent, 1973), Glycine max (Sprent, 1971, 1972; Huang et al., 1975a, 1975b), Vicia faba (Sprent, 1972), Phaseolus vulgaris (Sprent, 1976a), Lupinus arboreus (Sprent, 1973), and Aeschynomene americana (Albrecht et 1981). The pattern of response to drought stress is similar in most legume species. Nitrogenase activity falls to undetectable levels at about 40 percent of maximum fresh nodule weight (Sprent, 1972). Nodule activity can be restored to that of desiccated nodules by watering, full activity being resumed in one or more hours (Sprent, 1972). Plants with meriste matic nodules can recover from more damaging kinds of stress by regrowth of existing nodules, which can occur two to three days after watering (Engin and Sprent, 1973). On plants with spherical nodules severe stress causes nodule loss (Sprent, 1973). Recovery is slower and 5

PAGE 6

-6 -involves the formation of new nodules. Huang (1975b) found that inbibition of activity at low water potentials could be partially reversed by exposing the shoots to high concentrations of C02. In addition to reduci'ng the nitrogenase activity of existing nodules, water stress affects the growth of young nodules and the formation of new nodules. Sprent (1976a) reports that in young P. vulgaris plants, stressed to the point of wilting, nitrogenase activity was reduced by 90 percent. Nodule number and size were also depressed by this treatment. It is not known if this depression of nodulation is caused by fewer Rhizobia being present in the rhizosphere and available for nodulation, or if the stress affects the infection process. Water stress usually occurs gradually and progressively as moisture is lost from the soil. Sprent (1972) has shown that under these conditions the of the lower leaves of a legume is a good indication of suboptimal nitrogenase activity. However, under simulated wind conditions, if the soil remains moist, the shoots of both G. max and P. vulgaris will wilt, without a noticeable effect on nitrogenaseiactivity in the nodules (Gallacher and Sprent, 1978). In contrast to Huang and co-workers (1975a, b) who suggest that the inhibition of photosynthesis activity accounted for the inhibition of nitrogenase at low water potentials, Sprent (1976b) suggests that nodule stress occurs when the root systems cannot supply sufficient water to export materials from the nodules and replace the water lost from the nodule to the drying soil Sensitivity to excessive soil moisture stress has been reported for several legumes (Minchin and Pate, 1975; Minchin and Summerfield, 1976; Sprent, 1972). Mague and Burris (1972), Schwinghamer et El. (1970), and Sprent (1969) have reported that waterlogging depresses nitrogen fixation, largely as a result of oxygen deficiency. Sprent (1976a) has shown that nodule number, size, and water content are also affected by waterlogging. Minchin and Summerfield (1976) report that total dry weight of Vigna unguiculata nodules was reduced by 60 percent after only 8 days of waterlogging. Prolonged waterlogging did not affect the percentage nitrogen content of the various plant parts, but they found that the plant dry weight could be reduced by as much as 60 percent. There does appear to be some resistance to the effects of excessive soil moisture. Sprent (1976a) found in vulgaris that there was some variation among nodules formed with different strains of rhizobia in their ability to withstand waterlogging. There is evidence that some tropical forage legumes are flood tolerant (Bro1mann, 1978; McIvor, 1976; Quesenberry et a1., 1982), and A. americana cultivars have been shown to have good rates of nitrogen fixation activity in flooded conditions (Albrecht etEl., 1981). There can be an increase in nodule activity .as the soil dries from flooded conditions and this is usually explained by increased gas

PAGE 7

-7 -diffusion through the soil (Fishbecket al., 1973; Sprent, 1972). Although inhibitory desiccation, thatlbeyond what is required for maximum nodule activity has been explained in terms of reduced nodule respiration (Engin and Sprent, 1973; Sprent, 1972), the reduction in respiration was not as severe as that displayed by nitrogenase activity. Pankhurst and Sprent (1975) attempted to measure the water potential of nodules by the Shardakov dye method and by measuring the water potential of sand equilibrated with stressed nodulated root systems using a dewpointpsychrometer. They found that loss of up to about 25 percent in fresh weight (equivalent to -8 to -10 bars) results in reversible effects on nitrogen fixation, but beyond this, the effects became progressively more severe. The relationship between carbon fixation, nitrogen fixation, and water potential has been investigated to some extent. Minchin and Pate (1975) found the ratio of fixed nitrogen to C02 absorbed decreased as soil water suction was increased. They also found the effects of drought stress were considerably smaller than those for waterlogging. These observations are in agreement with the hypothesis that the nitrogen fixing process .is more sensitive than carbon fixation to stress, even though the nodules are closer to the normal water supply than the leaves. However, Huang et al. (1975a, b) describe a series of experiments using intact soybeans-Which suggest that the reduced supplies of photosynthate were responsible for the depression of nitrogenase activity under drought stress. Legumes are notorious for their extravagent use of water, as shown by ludlow and Wilson (1972) who found for the tropical grasses Pennisetum purpureum and Sorghum album a transpiration ratio (g H20 transpired/g C02 fixed) of just under 80 compared to 180 for the legumes wightii and Calopogonium mucunoides. There are many reports of moisture stress reducing the yields of 1 egumes, and the responses may vary among vari eti es (Mederski and Jeffers, 1973). This may be related to cessation of root growth at flowering (Salter and Drew, 1965). After flowering, and into the podfilling stage, nitrogen fixation decreases in many species (Pate, 1958) and may reflect competition for photosynthate between nodules and developing seeds (Lawn and Brun, 1974). This competition, linked to no new growth by the roots, and hence impaired water uptake would make the symbiotic plant especially sensitive to water stress at this time. Although numerous experiments have been conducted to elucidate some of the effects that soil moisture deficits can have on nitrogenase activity, relatively few experiments have been concerned with soil waterlogging. Very few researchers have included more than one crop species in their experiments for comparative evaluations. A large portion of the experiments reported in the literature concerning water stress effects on nitrogen fixation were conducted with crops, soil types, and environmental conditions which are not encountered in Florida. The purpose of this research was to utilize several crop

PAGE 8

-8 -and forage legumes used extensively in Florida, and to emphasize sandy soils, which are commonly found in Florida, to examine the effects of water stresses on nitrogen fixation as related to other physiological changes. In addition to the greenhouse work done for this project, an adjunct project was carried out in the field, utilizing field conditions.

PAGE 9

CHAPTER II. GREENHOUSE EXPERIMENTS A series of greenhbuse studies were conducted during 1981 and 1982 utilizing Aeschynomene americana, Desmodium heterocar on, alfalfa (Medicago sativa L.), bean (Phaseolus vulgaris L. ,cowpea Vigna unguiculata L.), and soybean (Glycine max L.J. Although several ex periments were conducted at various time intervals, procedural details were very similar for all. greenhouse experiments with the exception of crop species evaluated, durations of stress treatments, inoculants used, and other obvious variations. A complete description will only be given for the first soybean experiments. These details generally apply to all other greenhouse experiments as well. A. Soybeans, Experiments I, II, and III Three greenhouse experiments designed to examine the effects of water stresses on nitrogen fixation of soybeans in relation to other physiological processes were conducted in the spring of 1981 and 1982 and the fall of 1982. In all experiments, 'Cobb' soybeans were grown in 25 cm diameter black plastic pots containing a freely-draining washed, coarse sand. Before planting, seeds were coated with a commercial preparation of Rhizobium japonicum using gum arabic as an adhesive. Pots were arranged in a randomized complete block design. During the experimental periods, greenhouse temperatures generally ranged between a minimum of 12 C and a maximum of 35 C. On clear days, photosynthetically active radiation (PAR) at midday inside the green house was approximately 1500 uE m-2 s-l. After germination, the photo peri od was extended to prevent floral i nducti on by provi ding addit iona 1 light from 1800 to 2400 h with incandescent lamps. Plants were thinned to two per pot after seedling establishment. Except during the water stress treatment periods, plants were watered as needed with a nitrogen-free, half-strength Hoagland's nutrient solution. The soil was leached weekly with water to prevent buildup of salts. Flooding treatments were imposed by placing the treatment pot inside a dark plastiC container filled with water. The plastic containers were 27 cm diameter, which allowed complete submer sion of the smaller pot, but little room for mixing of water and aeration between the two container walls. The water level in the flooded pots was maintained approximately 5 cm above the soil surface. To retard algal growth, aluminum foil was secured over the top surface of the pot to exclude light. Drought cycles were imposed by simply withholding watering and allowing the plants to extract the soil moisture. Experiment I (1981) Seeds were planted on 30 January 1981, emerged 6 days later, and were thinned to two plants per pot 29 days after emergence. On 2 April, all pots were fully watered and the pots designated to receive the flooding treatment were placed in the water-filled containers. The drought treatment received no additional nutrient solution until the -9 -

PAGE 10

10 termination of the experiment on 16 April. Nutrient solution was added to the well-watered treatment as needed, generally every other day. Measurements of gravimetric soil water content, leaf water potential, nodule water potential, leaf diffusive resistance, root respiration, and nitrogenase were collected 1, 4, 6, 8, 10, 12, and 14 days after the imposition of the water stress treatments. Experiment II (1982) Seeds were planted 24 February 1982 and emerged 7 days later. Water stress treatments arranged in a randomized complete block design and identical to those described for Experiment I were initiated on 29 March, and soil water and plant physiological characters were determined 5, 9, 12, 13, 14, 17, and 25 days after imposing the water stress treatments. After measurements were completed on day 14, the droughted plants were re-watered with nutrient solution and flooded plants were removed from the flooding containers and allowed to drain. Measurements on days 17 and 20 represented recovery from the previously imposed water stresses. The experiment was termi nated on 28 April. Experiment III (1982) Experiment III was specifically designed to examine the relationship between leaf water potential and nodule water potential as plants progressively became more water stressed. Seeds were planted in 50 pots on 2 September 1982. Beginning 23 September water was withheld from 30 pots. During the following 10-day period, leaf and nodule water potentials were monitored as the plants dried. Several well-watered plants were also sampled on each measurement date. Measurements All measurements of gravimetric soil water content and plant physiological parameters were made only during periods of high light intensity between 1100 and 1300 h EST. Leaf and nodule water potentials. Four, l-cm leaf discs were removed from uppermost, fully-expanded leaflets and placed in a thermocouple psychrometer sample chamber. The chamber was then immediately sealed to a Spanner-type thermocouple psychrometer (J. R. D. Merrill Speciality Equip. Co. Model 84-13). After completing all of the other measurements, the root system and surrounding soil were carefully removed from the pots. Soil was separated from the root system, nodules were selected and removed from the root system with small forceps. The four or five detached nodules were quickly blotted with a cloth towel to remove adhering soil particles and surface moisture. The nodules were then quickly placed in a sample chamber and attached to a thermocouple psychrometer unit. The thermocouple psychrometer assemblies were then transported to the laboratory and placed in a thermostatically controlled water bath at 30 C. After 4 h of vapor and temperature equilibration, the psychrometric output was recorded using a strip chart recorder and a dewpoint microvoltmeter

PAGE 11

-11 -(Wescor Model HR 33T) operating in the psychrometric mode. To determine and output was compared to calibration curves which were constructed individually for each thermocouple psychro meter using NaCl solutions. Previous research had indicated that an equilibration time of 4 h was adequate for both leaf and nodule tissue. Stomatal resistance. A steady-state diffusion porometer (Li-Cor Model LI-1600) was used for measuring leaf diffusive resistance on one of the uppermost, fully-expanded leaflets. Measurements of the resistance of the abaxial and adaxial leaf surface were made and total leaf resistance was calculated assuming the resistances act in parallel. Gravimetric soil water. Soil samples were collected from each pot by taking a soil core, including the entire depth of the soil, from each pot. The soil sample was placed in a tin container, capped, transported to the laboratory, and then weighed. The soil sample was then dried overnight at 100 C before determining the dry weight of the soil. Soil water is expressed as a percentage of the soil dry weight. Oxidation-reduction-potentials. On several dates during the experiment, oxidation reduction potentials of the water and saturated soil were measured in the flooded treatments. Potentials were determined with a portable Orion meter and redox probe. Nitrogenase activity. Intact root systems were excised and gently separated from the soil. The entire root system and adhering soil were quickly placed in a 75 ml serum vial. The vial was sealed with a rubber serum stopper and 7.5 ml of air in the vial was replaced with acetylene, resulting in a 10% acetylene atmosphere within. the vial. Vials were incubated at ambient laboratory temperatures. During the 'incubation period, 0.5 ml gas samples were taken at 0, 30, 60, and 90 min. and injected into a gas chromatograph (Varian Model 940) fitted with a flame ionization detector. Rates of nitrogenase activity were calculated from linear regression lines fitted to the time sequence measurements. After the incubation period, the nodules were detached from the root system, dried, and Specific nodule nitrogenase activity was calculated by dividing nitrogenase activity of each vial by the dry weight of the nodules. Root respiration. Evolution. of C02 from the root-nodule complex was measured by withdrawing 0.3 ml of air from the incubation vials and injecting it into an infrared C02 gas analyzer (Beckman Model 205) (Clegg et al., 1978). Gas samples werQ analyzed for C02 every 10 min. for a period of 90 min. Rates of C02 evoluti.on were calculated from the linear regression analysis and specific C02 evolution expressed as ug C02 evolved min-l g dry wt root-l. Gravimetric soil water content for the well-watered treatments fluctuated with additions of nutrient solution but generally remained between 4 and 10% in 1981 and 3 and 8% in 1982 (Figure 1). Although

PAGE 12

-_ 1981 -L -fQ9 3 :=7 cB 'U --5 L -to \ \ \ \ \ \ \ \ \ \ \ '0, ..... ill E I S.E. of mean > o t5 l 0 contro I I -0 droughted ..... ... ..... 0 , '0 "', ... '0-----0---'II, ,1982 9 7 5, :3 I S.E. of mean o control odroughted 0, \ \ \ \ \ \ \ 'Q-o I I I I I I I I I P" "' ... .... '0o 2 4 6 8 .. 10 12 14 4 8 12 16 20 24 Days Af1er Treqtments Were Imposed Figure 1. Gravimetric soil water contents of control and drought'treatments imposed on soybeans in 1981 and 1982. S. E. represents the average standard error'of the mean. ..... N

PAGE 13

13 -on a couple of dates the soil water content of the well-watered plants appeared somewhat low, no visual stress symptoms were observed and other physiological processes were not affected by limiting soil moisture. However, soil moisture contents in the droughted treatment declined rapidly after the first 4 to 8 days after withholding water. As the gravimetric soil water content decreased below about 1%, plants became severely stressed and had difficulty extracting the remaining soil moisture. After rewatering the plants on day 14 in 1982, gravimetric soil water contents returned to values of the well-watered treatment. The slightly lower soil water contents in 1982 as compared to those observed in 1981 in both treatments are probably due to slight differences in particle size distributions of the soil. Although the soils were quite similar, they were obtained from different sources in each of the two years. r No differences were observed in leaf water potentials in well-watered and flooded p1 ants (Fi gure 2). In both 1981 and 1982, the well-watered and flooded p1 ants mai ntai ned between -0.8 and -0.4 MPa, suggesting that the flooding treatment did not result in tissue desiccation. Often flooding tesu1ts in reduced water uptake and water stress symptoms similar to those observed during drought stress (Kramer, 1951). This response was not observed either visually or with leaf water potential measurements. In both years, was reduced as a result of limiting soil moisture after about 12 days of water withholding. Between days 10 and 12, dropped rapidly, an occurrence often observed in very sandy soils because of their low water holding capacity. In 1981, severe desiccation of the plant leaves occurred as indicated by only as low as -1.4 MPa. After rewatering on day 14, of previously droughted p1 ants recovered to values of .. well-watered leaves. The leaf water potential data suggest no effect by the flooding treatment, a rapid progression of desiccation in plants subjected to drying soil, and complete recovery after rewatering. The rapid decrease in was not observed until the gravimetric soil water content dropped below about 1% (Figure 1). Leaf diffusive resistance closely paralleled changes in (Figure 3). Throughout both experiments in 1981 and 1982, leaf diffusive resistances for flooded plants were similar to those of the well-watered plants. The droughted plants exhibited increases in leaf diffusive resistances after 10 days in both years. Rather large decreases in leaf diffusive resistance were observed as declined very slightly (Figure 2). After rewatering, leaf diffusive resistance recovered to control values. Flooding had no effect on nodule water potentials (Figure 4). However, was reduced with reductions in soil water content, reaching water potentials as low as -1.6 to -1.8 in 1981 and 1982, respectively. In 1981, there was an obvious trend for reduced during the early days after withholding water (days 4, 6, and 8), suggesting that may be slightly more sensitive to reductions in soil water potenti al than the other measured parameters. In fact, if data from both studies are examined it is apparent that declined with any reduction in gravimetric soil water content, at least, witHin the limits imposed in these experiments (Figure 5). As gravimetric

PAGE 14

-.41 lAO I ::Ii ...................... '-,0 (L --' o +C ill o (L -.8 -1.2 ill +o .3 -2.0 /' -I s. E. of mean --0 \ \ \ \ \ \ \ \ \ \ \ \ 0 control *fboded \ '13 t 0 droughted \ ill 2.4 ... I I I "" ........ 0 o 2 4 6 8 10. 12 14 Days After Treatments o. 1982 -4 -.8 -1.2 ........ ........ .... ,q \ \ \ \ I \ 0 ... 0 \ I S.E. of o control *flooded _2.0l 0 droughted ---'-T,l I I I I I I I I I I 4 8 12 16 20 24' Were Imposed' Figure 2. Midday leaf water potentials of well-watered, flooded, and droughted soybeans in 1981 and 1982. S. E. the average standard error of the mean. I --' .",.

PAGE 15

ll300 1981 13001 .... 1982 .... '" .!!!, 1100 IP ill I U / 900 /' .1n I S.E. of mean /0 700 I 0::: 0 control OJ -if flooded 1100 900 /'1 p'" \ I \ I \ q \ q I \ 700" .' I S.E. of mean control If flooded > 500 0 droughted I ,/ ,,' I o droughted I c:;; '\ I \ I \ I 500 Lf' I '4-" A I \ 300 (:) 300 4o \ ,.....a \ \,p---100 I I I I I 100 L o 2 4 6 8 10 12 14., .. 4 8 12 16 I Days After Treatments Were Imposed Figure 3. Midday leaf diffusive resistance of flooded, and droughted soybeans in 1981 and 1982. S. E. represents the average standard error "of the mean. 20 n 24

PAGE 16

-o n 2 __ W_ I .--k __ .__ ,. .---.......... ........ o -.6 -t C OJ ........ 0-----0 ...... ...... 0-----.... -"0 & I S. E. of mean OJ -t-OJ :J -1.8 u o control *flooded o droughted '\ \ \ .\ \b -.2 1982 -.6 -1.0 ) "1.4 -:1.8 0 .... I I I ...... I '''''tJ---'Y I \ I \ I \ I \ I , \ I q, '0 __ .,.A.. ------0 I s. E. of mean o control flooded (j droughted o Z I I I I I I I' I I ..", I I I I '. I I I I I I o 2 4 6 8 10 12 '4' 8 12 16 20 24 Fi gure 4. Days After Treatments Were Imposed Midday nodule water potentials of flooded, and droughted soybeans in 1981 and 1982. S. average standard error 6f the mean. .. ..... 0'1

PAGE 17

-17 --.2 0 0 0 0 c 0 0., 0 a.. 00 0 co -.6 0 -.. -l-
PAGE 18

-18 -soil water contents of about 1% were approached, further reductions in soil water caused dramatic reductions in 'l'nod. In 1982, as with the other parameters measured, 'l'nod returned to values equivalent to the well-watered 'l'nod after rewatering on day 14 (Figure 4). The relationship between 'l'L and 'l'nod was evaluated in all three experiments that were conducted. However, particular emphasis was placed on this relationship in Experiments n and III. As indicated in Figure 6, there was not a 1 :1 relationship between 'l'L and 'l'nod. At the high 'l'L (>-1.0 MPa), 'l'nod generally equal or higher than 'l'L. However, as the leaf water potential decreased below about -1.0 to -1.2 MPa, 'l'n d became considerably lower than 'l'L. Most likely, as 'l'L decreased be90w -1.0 to -1.2 MPa, stomatal closure occurred (Figures 2 and 3) resulting in the inhibition of transpiration .. Stomatal closure would help to prevent further dehydration and would tend to maintain 'l'L at potentials near those where the stomata closure was triggered. Conversely, since nodules lack stomata, a progressive reduction of 'l'nod with decreasing soil water content would be expected if delivery of water to nodule.s from root vascular tissue could not resupply the water lost from the nodules to the soil. In the sandy soil medium used in these experiments, the upper soil layers, where most nodules were sampled, became extremely dry and warm and probably represent a steep water potential gradient from nodule to soil. Other research has shown that nodules lose water very rapidly when subjected to desiccating environments (Sprent, 1971). It appears that with severe water stress in sandy soil medium, the nodules may continue to dry to water potentials below the potentials of leaves. However, it is not entirely clear whether this drying is simply due to a very steep water potential gradient between the nodule and soil coupled with the inability of the nodule to decrease water loss, to increased resistance to water flow from the root to the nodule, a combination of both, or other processes. Nevertheless, it is possible that nodules may be subjected to severe water deficits which may result in irreversible damage (Sprent, 1972), whereas leaves are protected from severe desiccation by stomatal control mechanisms. Contrary to the other physiological parameters measured, flooding seemed to affect nitrogenase acti vi ty di fferent1y (Fi gure 7).' In both years, nitrogenase was apparently reduced immediately after the flooding treatment was imposed. The initial decrease in nitrogenase activity is probably due to the depletion of oxygen in the root environment. The lack of oxygen may reduce nitrogen fixation in one of two ways. Firstly, it will decrease or eliminate oxidative phosphorylation, which is necessary for ATP production, and this reduction in energy should cause concomitant reductions in nitrogenase activity. The activity may only be depressed, because some ATP may still be produced by some oxidative phosphorylation which continues at the low oxygen concentrations. Metabolic activity may shift to fermentation which is capable of less efficient ATP formation. Secondly, if roots and bacteroids shift their metabolic patterns to fertmentation, less reduced carbon substrate will be available for the production of both ATP and electrons required by nitrogen fixation. Bisse1ing (1980) reported that, in Pisum sativum inoculated with

PAGE 19

-.; .. 19 -.2 0 .. 0.. ..... 0 -.6 .. -l .. 0 0 0 0 0 Z 0 0 -2.2 0 -Experiment 2 (1982) c 0 Experiment 3 (1982) -2.2 -1.8 -1.4 -1.0 -.6 -.2 LEAF WATER POTENTIAL (MPo) Figure 6. The relationship between midday soybean leaf water potential and midday nodule potential for two experiments in 1982 The line drawn represents a 1:1 relationship between the two potentials.

PAGE 20

I Q) ,:J u 0 c Ol I UN (/) Q) 0 E :t.. 24 21 18 15 12 9 63 0 .. 1981 'k-._.--(:(_ .......... /" ,/ ... / I -' ---q -----0---\ o control. flooded o droughted I s. E. of mean \ '\ 0, 0----_-0 80 701' 60 40 30 20, 10 o control flooded o droughted I S.E of mean o ... .... ... .... ................ H ./.><0 / .r ....... /" 'TI 1982 ? --------0 .;fr-'_._._._.--/t ./ I t/I I I '-v-n .' \ \ \ I if 0 I I I I I ,. I I I o 2 4 6 8 10 12 14 r 4 8 12 16 '20 24 Fi 7. Days After. Treatments Were Imposed Midday nitrogenase activity (acetylene reduction) of well-watered, flooded, and droughted soybeans in 1981 and 1982. S.E. represents the average standard error of the mean. Rates of nitrogenase activity in both years are expressed as C2H4 h1 g nodu1e. N. o

PAGE 21

-21 -Rhizobium 1eguminosarum, waterlogging decreased nitrogenase activity. They further reported that there was a decrease in the active nitrogenase enzyme and suggested that the iron protein of the nitrogenase '. complex was repressed under waterlogging conditions. Nitrogenase activity of the'f100ded plants did, however, tend to increase during the duration of the experiment until activities higher than those for well-watered controls were observed by the termination of both experiments (Bigure 7). There was evidence of the initiation and development of new nodules on the flooded plants. These new nodules were formed at the soi 1 surface on adventi ti ous roots whi ch were rapi d1 y initiated after the imposition of the flooding treatment. It is likely that these young nodules increased in number and activity throughout the experiment, resulting in the enhanced rates of nitrogen fixation observed in Fi gure 7 .. Many p1 ants are a1 so capab1 e of produci ng aerenchyma ti ssue when flooded (Yoshida and Tadano, 1978). Observations of the roots of the flooded soybeans suggest that the adventitious roots initiated after the flood treatment was imposed were large and spongy in texture. In addition, there were gross anatomical changes near the crown of the root system. While the increase in nodulation may be the major reason that flooded plants increased and maintained their nitrogenase activity, the possibility of aerenchyma tissue may also playa role in the increased activity. Nitrogen fixation was reduced after 10 days of withholding water in both 1981 and 1982 and activities declined to essentially zero as stress became progressively more severe (Figure 7). After rewatering on day 14 in 1982, nitrogenase activities of previously droughted plants increased to rates higher than the well-watered controls. Such enhanced rates after rewatering have also been observed by others (Sprent,1972) and may resu1 t from uti1 i zation of an a,bundant supply of carbohydrates which accumulated in the nodules during the stress period or from the removal by transpiration of reduced nitrogen compounds that had pre viously inhibited nitrogenase activity during the drought. Nitrogenase activity in well-watered and droughted plants was closely related to both nodule water potential (Figure 8) and respiratory activity of the root-nodule complex (Figure 9). Data presented in Figure 8 demonstrate' the extremely sensitive nature of nitrogenase activity to the Even as declined from -0.2 to -0.4 MPa, nitrogenase activity was reduced. Even slight reductions in both and can be expected to result in some inhibition of ni trogenase activity. P1 ants in these experiments never reached a point where the water stress caused terminal damage to the Plants were rehydrated and quickly regained the activity of the we11-watered plants. In most instances, drought stressed plants were wilted at midday after the tenth day of the experiment. It is also interesting to note that nitrogenase activity declined more than root respiration (Figures 7 and 9), similar to data presented by Sprent (1971). Sprent also presents information to indicate that water stress produces physical damage in the nodule. It is not fully

PAGE 22

>r> IU
PAGE 23

L 0 0 0:: CJl (\,1-0 u (Jl E 1981 /f:f-._.-.ir. 1982 8 8 ..... / '. oconfrol o----..:.---=i ,.-c::'._k/ _--'-. 7 *flooded ;::.. --"-" odroughted .-.. tI ..... If ..... "'q ...... 6 1\. 6 \ ""If \ 5 \ 5 \ o control \ .... --:........ .J "1 '4 \ 4 -tr flooded", \ .......... \ X' odroughted 3 --' '0 \ *_.-. ---q' 3 \ I 0 .... b-d I S.E. of mean --2 I S.E. 9( mean 2 '"1J I Ol J 0 I I I i I I I '-P' I I j I I I o 2 4 6 8 10 12' 14 ,.' 4 8 12 16 20 24 Days After Treatments Were Imposed" figure 9. Midday root respiration of well-watered, flooded. and droughted soybeans in 1981 and 1982. S. E. represents the average standard error of mean. w

PAGE 24

-24-understood whether this relatively larger decrease in nitrogenase activity is because of some regulatory activity, the sensitivity of the metabolic functions measured, actual physical damage in the tis.sues during the stress period, or differences in the amount of water that is available to the root system as compared to the nodule. B. Aeschynomene and Desmodium, Experiments I, II, III Experiment I Twenty-eight accessions of Aeschynomene were rooted in sand and transplanted to 15-cm black pots containing a mixture of sand, peat, and perlite (2:1 :1). Plants were allowed to establish and grow for 4 weeks before the pots were arranged in a completely random design in two flooding tanks. For 1 week, water was maintained at 8 cm below the soil surface. For the second week, water was maintained at 4 cm below the soil surface. Visual observations and nitrogenase activities indicated that the flooding treatment had not resulted in detrimental effects on any of the accessions. In order to test the hypothesis that soil moisture affects nitrogenase activity, 16 plants were removed and allowed to dry for 20 days until the plants wilted, 16 plants were maintained as well-watered controls, and a corresponding group of plants remained flooded for an additional 20 days. After 20 days, midday measurements of leaf osmotic potential, leaf turgor potential, and nitrogenase activity were measured. The plants were then harvested, and root and nodul e dry wei ghts were determi hed. Percent nodulation was expressed as nodule dry weight as a percentage of total root dry weight. Data reported are for a range of accessions with no evaluation made of any differences which may be caused by different accessions. Leaf water potential, leaf turgor potential, leaf osmotic potential, and nodule water potential were all reduced by the 20 days of withholding water (Table 1). The flooding treatment did not affect the leaf turgor potential, or leaf osmotic potential, however, there was a trend for to remain slightly higher than that of the well-watered controls .. Nodule weight, percent nodulation, and nitrogenase activity also -declined in response to the drying treatment (Table 2). Root weight was apparently stimulated by the drought period, a response that is often observed during drought. The flooded plants exhibited no adverse effects in response to the treatment imposed. Experiment II Four accessions of Aeschynomene and one of Desmodium heterocarpon were grown in sand:peat:perlite mixtures in 15 cm pots to compare the response of the two species to short-term water stresses. After 10 weeks of growth, pots were either flooded with water to surface level or water was withheld. Three days after the treatments were imposed, nitrogenase activity and components of leaf water potential were measured at midday.

PAGE 25

-25 Table 1. Leaf water, osmotic, and turgor potentials of Aeschynomene leaves and water potentials of nodules as affected by flooding and soil water deficit. Treatment Leaf Water Potential Leaf Osmotic Leaf Turgor Nodule Water Potential Potential Potential a -------------------------MPa---------------------------Droughted -1.15 0.9 -1.36 1.2 Flooded -0.58 0.9 -1.21 0.6 Control -0.72 0.3 -1.20 0.4 0.21 0.5 0.63 0.8 0.48 0.3 a Values represent unreplicated observations. -1 .07 -0.25 -0.45

PAGE 26

-26 Table 2. Effect of three water treatments on root and nodul e weight, percent nodulation, and nitrogenase activity in nodules of Aeschynomene americana. Treatment Droughted Flooded Control Nodul e Wt. ----mg----26.4 4.4c 31.4 1.5 32.9 6.7 Root Wei ght ----mg-----336.4 4.9 227.4 4.0 200.5 2.8 Percent Nodulation a 6.8 0.3 l3.0.6 13.5 0.9 a Nodule dry weight x 100/total root dry weight. b umoles C2H4 h-l / gram dry wt. nodule. c All values determined after 21 days of treatment. Nitrogenase Acti vityb 9.05 1.01 21 .61 1 .84 18.58 1.94

PAGE 27

-27 Although Aeschynomene accessions subjected to flooding had leaf water relations and nitrogenase activities which were similar to those of the well-watered plants (Tables 3 and 4), flooding dramatically reduced the nitrogenase activity of Desmodium plants. The flooding effect also reduced leaf turgor potential to zero (Table 3). Visual stress symptoms (yellowing and wilting) were also observed in flooded Desmodium plants. Such symptoms were not observed with Aeschynomene plants. Drought stress reduced the leaf water and turgor potentials and nHrogenase acti vi ties of both AesGhynomene and Desmodi urn (Tabl es 3 and 4). Accession UF 186 appeared somewhat more resistant to the drought. Experiment II I Aeschynomene seedlings were subjected to varying concentrations of 600 molecular weight polyethylene glycol (PEG). Root systems of the small seedlings were placed into test tubes containing PEG solutions of -1.0, -2.5, and -5.0 bars osmotic potential. Plants were then maintained in a laboratory environment for a 7-day period. After 3 and 7 days of treatment, nitrogenase activities were determined for plants grown in each PEG concentration. Nitrogenase activity was depressed in the PEG solutions after 3 days and declined even more after 7 days of stress .(Figure 10). Even at an osmotic potential of only -1 bar, nitrogenase activities were reduced by 50 and 85% after 3 and 7 days of stress, respectively, suggesting the extreme sensitivity of nitrogenase to drought stress. Even though the plants subjected to the -1 bar osmotic stress appeared turgid, nitrogenase activity was signtficantly depressed. C. Cowpea and Bean An experiment evaluating the response of cowpeas and beans to both flooding and soil drying was conducted in 1981. The inoculated seeds of both crops were planted-15 October 1981. Flooding and soil drying treatments were imposed beginning 4 December and lasted for 17 days. Data were coll ected for soi 1 and pl ant parameters, as descri bed in Part A above, at midday 3, 5, 7, 10, 12, 14, and 17 days after the treatments were imposed. Unfortunately, very little difference in leaf water potential of cowpeas was observed for any of the treatments (Figure 11). Considering the variation inherent in the data, we conclude that leaf water potential was not affected by the flooding or drought treatment during the time peri od encompassed by thi s experi.ment. Apparently, the soil di d not dry sufficiently during the 17-day period to cause leaf water potential reductions. If the experiment had continued for a longer period, we would have expected the drought treatment to reduce leaf water potentials. Despite the lack of a difference in leaf water potential, the drying treatment did result in slightly higher leaf diffusive resistances on the

PAGE 28

-28 Table 3. Water stress effects on leaf water and turgor potentials of Aeschynomene and Desmodium accessions. Genus Control Flooded Droughted --------------------MPa---------------------A. viscosa (UF 369) -.31 .25 -.17 .31 -.70 .12a A. americana (UF 57) -.27 .34 -.37 .33 -1.60 -.20 A. americana (UF 186) -.44 .30 -.29 .45 -1 .35 .29 A. americana (UF 255) -.72 .37 -.30 .51 -1.20 -.01 D. heterocarpon -.34 .24 -.32 -.01 -.95 -.21 a Calculation of yields some negative values since apop1astic dilution of the cell sap after freezing causes a slight undereitimation of

PAGE 29

29 -Table 4. Comparison of water stress effects on nitrogenase activity in Aeschynomene and Desmodi urn". Genus Con 'tiro 1 Flooded Droughted A. vi 11 osa (UF 369) 2.58 0.67a 2.43 0.47 0.20 0.06 (100)b (95 ) (8) A. americana (UF 57) 5.13 0.27 5.6"5 0.64 0.11 0.03 (100) (11 0) (2) A. americana (UF 186) 1.88 0.63 1.94.24 0.76 0.12 (100) (103) (40) A. americana (UF 255) 3.45 0.81 3.97 0.66 0.09 0.03 (100) (114 ) ( 3) D. heterocarpon (UF 20) 6.49 0.89 0.18 0.06 2.40 0.56 (100) ( 3) ( 37) a umo1esethylene formed h-1 gram dry weight standard error of the mean. b Percent of well-watered control.

PAGE 30

o o a t----: 3 > 'U : ,(Y W '\.9' 2 (f), a, 'z a t--Z to ---1 o \ \ \ \ \ \ \ \ \t, ... .... .... .... ..... 6. Do 7 DAYS TREATMENT 0---0 3 DAYS TREATMENT ERROR BAR S ARE SE.M. ......... .... ........ -=so 0 1 ... 2 3 4 5 ':.. ,-----------,---2 PE G INDUCED OSMOTIC 'POTE,NTIAL EBARS) w a Figure 10. The effect o,f PEG imposed stresses for 3 and 7 days on nitrogenase activity of Aeschynomene americana.

PAGE 31

" -c 0-::E -0.4 ..J i= z -0.6 o 0-0:: W -0.8 3:. l.L. W ..J -1.0 COWPEA" 0" /,_ o 1 o \ 0 ., S.E. of mean ; \ 8 /' .. )14" '\ 0' 0' \; '. I >0 '. /,' 0" '0 ,0 I o CONTROL 0 o FLOODED o DROUGHTED o 4 8 12 16 20 DAYS AFTER TREATMENTS WERE IMPOSED Figure 11. Midday leaf water potentials of well-watered, flooded, and droughted cowpeas in 1981. s. E. represents the average sta.nda rd error of the mean. ..

PAGE 32

-32 -last four days of the experiment (Figure 12). The flooding treatment, however, did not influence leaf diffusive resistance. Cowpea nodule water potential for the droughted plants decreased markedly between day 14 and 17, dropping from -0.6 MPa on day 14 to -1.9 MPa on day 17, indicating severe desiccation of nodule tissue by the last day of the experiment (Figure 13). Generally, the flooding treatment had little effect on the nodule water potentials, however, there was a trend for the flooded nodules to maintain slightly higher water potentials than those of the well-watered controls. Nitrogenase activity of the cowpea plants showed a similar response to that obs.erved for diffusive resistance (Figure 14). The drought stress inhibited nitrogenase activity on the last four days of the experiment, despite no reduction in leaf water potential or nodule water potential until day 17. Again, these observations suggest that nitrogenase activity may be reduced in the very early stages of drought. The flooding treatment appeared to have little effect on nitrogenase activity. Root respiration of flooded plants was slightly inhibited initially after imposing the treatment, but seemed to increase throughout the duration of the experiment and tended to be higher than in the well-watered controls by the termination of the experiment (Figure 15). The drought stress inhibited root respiration, especially on the last two days of the experiment. Root respiration, although slightly depressed, was not as sensitive to water stress as was nitrogen fixation. In summary, flooding had little effect on the physiological processes measured in cowpeas. Nitrogen fixation was particularly sensitive to the imposed drought and was affected earlier and to a greater degree than stomatal resistance, leaf or nodule water potential, or root respiration. Results from the experiment which evaluated the response of bean plants to either flooding or drought are shown in Figures 16a-18. As with cowpeas, apparently the drought stress was not imposed for a long enough period of time to reduce leaf water potentials (Figure l6a). However, flooding reduced leaf water potential by 0.2 to 0.3 MPa through out the experiment. It is probable that the flooding treatment reducea uptake of water by roots, thus leading to slight reductions of water potentials in the leaf tissue. Although the drought treatment had no effect on leaf diffusive resistance, flooding resulted in partial stomatal closure on all sampling dates (Figure 16b), probably iri response to the lower leaf water potentials and reduced water uptake by the roots. Nodule water potential was unaffected by the flooding treatment and reduced only on the day of the experiment by the imposed drought (Figure 16c). Nodules were aJso observed to form on adventitious roots of beans during the flooding treatment. At the same time, older nodules that were subjected to flooding became very mushy and dark in color after about the tenth day of the treatment. Leaves of the flooded bean plants also became quite yellow after about 10 days. Nitrogenase activity was almost completely inhibited by imposing the flooding treatment (Figure 17). Very rates of activity were

PAGE 33

COWPEA 3000 I E *CONTROL tJ) o FLOODED W oDROUGHTED u Z i=! 0 C/) I I C/) I 0 I W I \ I 0:: I \ I I I \ I W \. I \ .,' > \ \ I b' p C/) ::l / La.. 600 La.. / Q -j LL 1 S.E.of mean
PAGE 34

COWPEA /00 -c --' I-Z -1.0 W b a.. l.'t: W -1.4 IW :5 -1.8 o o Z o ., / 0 o \/ o CONTROL o FLOODED oDROUGHTED \ \ \ \ \ SE.of meon I \ \ \ \ \ \ \ \ \ \ \ \ \ \ b I I I I I I o 4 8 12 16 20 DAYS AFTER TREATMENTS WERE IMPOSED Figure 13. Midqay nodule. water potentials of well-watered, flooded, and droughted cowpeas in 1981. S. E. represents, the average standard of the mean.

PAGE 35

I. +-Q) >-:;10 t--g -c I8 0"0 o COWPEA 'kCONTROL o FLOODED DROUGHTED SEof rrean 1 I I 4 8 12' 16 20 DAYS ,AFTER TREATMENlS WERE IMPOSED Figure Midday nitrogenase activity (acetylene reduction) of well-watered, flooded, and droughted cowpeas in 1981. S. E. represents the standard errdr of the mean. w U'1.

PAGE 36

+-28 oL. -E 80 a::L. _0' a."': en 60 wa::E 1--0 40 a::g Q) N 8'20 0' =l -o COWPEA ./0 /.0................. ,S. Eo of mean I p". 0"0 CONTROL o FLOODED DROUGHTED I I o 4 8 12 16 20 DAYS AFTER TREATMENTS WERE IMPOSED Figure 15. Midday root respiration of well-watered, flooded, and droughted cowpeas in 1981. S. E. represents the average standard error of the mean. W 0'1

PAGE 37

-I I-:z -/.O W b a. n: W -1.4 3: lJ.. L5 -1.8 -' BEAN .. ;-* CONTROL o FLOODED S.E. of mean I o DROUGHTED I o 4 8 12 16 20 DAYS AFTER TREATMENTS WERE IMPOSED Figure 16a. Midday leaf water potentials .of well-watered. floodeq, and droughted beans in 1981. s. E. average standa,rd error of the mean.

PAGE 38

I :::::I. E en w .U Z en w 0:: W > en ::J LL LL o LL w 3000 2000 .....J. 200 Bf:AN CONTROL FLOODED oDROUGHTED .f / I / /\ / .., /O, ...... j. /. 0 ". j' \ /. /0 i 0/0 \i if 5.E. of mean I o 4 8 12 16 20 DAYS AFTER TREATMENTS WERE IMPOSED Figure 16b. Midday nodule water potential of well-watered, flooded, and droughted beans in 1"981. S. L represents the average standard error of the mean. w ex>

PAGE 39

", 0 a. --1 -0.6 J-Z w b -1.0 a. 0:: W -1.4 W --I :::> 0 0 -1.8 Z BEAN ... ;-* CONTROL o FLOODED o DROUGHTED : I (" /0 \'\. \0' 'go \ \ \ \ \ S.E. meon I \ \ \ \ \ \ \ \ \ \ o I I I I I I o 4 : 8 12 16 20 DAYS AFTER TREATMENTS WERE IMPOSED Figure 16c. Midday leaf diffusive resistance of well-watered, flooded, and droughted in 1981 .. S. E,. represents the standard error of the mean. -..

PAGE 40

-BEAN *CONTROL o FLOODED 0 DROUGHTED ,.0, , \ \ \ \ \ \ \ \ \ b \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ S. E. of mean 0, .\ \ "-0 _._0--..... 0 'L ........... ,,_.---.__ u .--0 4 8 12 16 20 DAYS AFTER TREATMENTS WERE IMPOSED Figure 1'7. Midday nitrogenase activity (acetylene reduction) of well-watered, flooded, and droughted beans in 1981. s. represents the average standard error of the mean. I .a::. .0

PAGE 41

-41 observed, even after only 3 days exposure to flooding. The drought stress also reduced nitrogen fixation on the last two measurement dates, apparently in direct response to lowered nodule water potentials (Figure l6c). By the last day of the experiment, nitrogenase activities were essentially zero in both the flooded and droughted treatments (Figure 17). Root respiration was not reduced by the drought stress until the last two measurement dates (Figure 18), after which it was reduced by 40 to 50%. Except for the measurements taken on day 12, root respiration in the flooded plants was reduced below levels observed for the control plants. As was the case for cowpeas, flooding reduced root respiration, but not as severely as nitrogen fixation. The data suggest that nitrogen fixation in bean was particularly sensitive to flooding. In fact, even short-term flooding almost completely inhibited nitrogen fixation as nodules became dark and mushy. Leaf water potentials and leaf diffusive resistance also reflected the detrimental effects of the flooding treatment. Toward the end of the experimental period, drought stress reduced nodule water potential, root respiration, and nitrogenase activity. These reductions were observed before any stress symptoms were observed in leaf water potentials or diffusive resistance. D. Alfalfa The effect of water stress, both drought and flooding, on nitrogenase activity in alfalfa were studied in a experiment during the spring of 1982. Alfalfa seeds (cultivar Florida 77), inoculated with a peat based Rhizobium culture, were planted on 24 February 1982. The seedlings emerged four days later, and treatments were imposed 77 days after emergence. The water stress treatments lasted for 10 days. Measurements were made, starting at midday, on the first, third, fifth, sixth, eighth, and tenth days after treatments were imposed. The anatomy of the alfalfa plant, especially the small size of the leaflet and nodule, made it impractical to obtain measurements on leaf diffusive resistance and nodule water potential. Leaf water potentials remained stable in the control plants during the stress period, exhibiting very little variation from day to day (Figure 19). Flooded plants were reasonably stable for the first six days, then decreased, with the average water potential one MPa lower than the control plants 10 days after the treatments were imposed. Drought stressed plants exhibited a marked decrease on the fifth day of the stress period. Rewatering of the plants restored the leaf water potential to that of the control levels. The dep1 eti on 0 f soi 1 water-during the drying cyc1 ei s shown in Figure 20. The percent water in the soil was reduced to below 1% within five days after the stress was imposed. Rewatering rapidly restored the soil moisture to 6%, slightly above the average level for the controls.

PAGE 42

,I .' 120 I "i +-100 o e Z E Q e80 0:. c 0.. ._ ta' 60 o:-g og40 Cl) (\J 8 =l -o o BEAN o I I \ I /', I J\' ,'.... I \ \ .... I '''0 /. \ .0..... I \ \ I '. / \\ /" "0 \ \ PI" '\\ \ / \\ __ '\,' I ............ j' ,0 .......... CONTROL o DROUGHTED o FLOODED S.E. of mean I I I I I I 4 8 12 16 20 DAYS AFTER TREATMENTS WERE IMPOSED I I Figure 18. Midday root respiration of well-watered, flooded, and droughted beans in 1981. S. E. represents the avera'ge standard error of the mean.'

PAGE 43

. ALFA.LFA -c a. :::E -I ..J f-Z w -2 f-a a. 0:: -3 ..J.. _. 0, '* 8--::'-' 1;' __ '..... It, P ............ -0 '. 1" I I \ '..... I' '. I \ "'0 '0.......,' \. ......... I \ I '0 ',. I \ I \ I I \ I \ I of mean \ I '," __ ---0 --. 0--..... lJ---'" l.J.. W-4 ..J CONTROL o FLOODED o DROUGHTED I I I I I I I I I I I o 1 3 4 5 67 8 9 10 DAYS AFTER TREATMENTS 'WERE IMPOSED Figure 19. Midday leaf water potentials of well-watered, flQoded, and droughted alfalfa in 1981 S. E. represents the average standard error of the mean. ..

PAGE 44

0 8 -0:: w 3: --' 6 0 en u 0:: 4. w -ALFALFA I .. S.E. of mean , , '=\. CONTROL o DROUGHTED , d. 'D ... ... 0 ....... ....... I I I I I I I I I I I I I I I I o I 2 3 4 5 6 7 8 9 10 DAYS AFTER TREATMENTS WERE IMPOSED Figure 20. Gravimetric soil water content of well-watered and droughted alfalfa in 1981. S. E. the average standar:d error of :the mean .. . \

PAGE 45

45 -Nitrogenase activity for the control and drought stressed plants parallels leaf water potentials (Figure 21). Nitrogenase activity is reduced to less than one-third of the control plants by day eight of the stress, however, rewatering restored the nitrogenase activity to that of the control. Flooding reduced nitrogenase activity to less than one-sixth of the control in only 3 days, and it remained depressed for the duration of the stress period. This rapid and almost total loss of nitrogenase activity suggests that some metabolic function in the nodule is irreversibly destroyed by flooding. The response of root respiration, which closely parallels nitrogenase activity and leaf water potentials is presented in Figure 22. Once again, the control plants remain relatively stable throughout the la-day stress period, while the drought stressed plants show a reduction in respiration on day six. Rewatering the drought stressed plants restores respiration to control levels. As observed in the nitrogenase activity, respiration is depressed by flooding to about 50% that of the control plants.

PAGE 46

2700 'T >C 2400 1-.2 5...:.0. 2100 1900 "0 W 1600 CJ)t... 1300 1000 a::: enG) 700 1--100 o ALFALFA CONTROL o DROUGHTED FLOODED 0 .......... ........... I 5.E. of mean / ./ / / / "-0 ----0-'-'---<:>-'-,---<:> 'f I 'i ., I I I 2 3 4 5 6 7 8 9 10 DAYS AFTER TREATMENTS WERE IMPOSED < Figure 21. Midday nitrogenase activity (acetylene reduction) of well-watered, flooded, and droughted alfalfa in 1981. S. E. 'represents the "average standard error of the mean.

PAGE 47

3000 ALFALFA 2700 I 5.E. of mean ,. 2400 4->-5 2100 1900 "0 Q) 1600 ,WE 'en 1300 ; , 1000 / / 0::(1) 700 /' *CONTROL / -0 ZE 400 o DROUGHTED FLOODED 200 0, 100 0 -.......0-_--0---.--0 I 'i ,. I I I I I 2 3 4 5 6 7 8 9 10 DAYS AFTER TREATMENTS WERE IMPOSED Figure 22. Midday root respiration of well-watered, flooded, and droughted alfalfa in 1981. s. E. represents the average error of mean. .j:::> ...... I'

PAGE 48

CHAPTER III. FIELD EXPERIMENT, SOYBEANS An adjunct experiment was conducted to evaluate the effect of water deficits on field-grown soybeans. Soybeans were grown in field plots located at the Irrigation Research and Education Park on the University of Florida campus. Although the field experiment was designed to study the response of soybeans to six water management treatments, for purposes of the nitrogen fixation study, only a we11-watered and stressed treatment were monitored during growth. The drying cycle imposed on the stressed treatment lasted for 11.9 days and was relieved by irrigations before recovery rates were determined. The well-irrigated treatment was irrigated when the soil water potential at 15 cm reached -15 to -20 KPa. Measurements: of canopy carbon exchange rates, leaf water potential, leaf stomatal con ductance, percent nodule moisture, and nitrogenase activity were made periodically between 1000 and 1500 hours during the drying cycle. Leaf water potential, leaf to water vapor, nodule moisture, apparent canopy carbon exchange, and nitrogenase activity declined as the duration of the stress period increased (Figures2327). All physiological processes, except apparent canopy carbon exchange rate returned to control levels upon adequate rewatering. The incomplete recovery of apparent carbon exchange was attributed to decreased leaf area index resulting from leaf senescence during the stress period. Nitrogenase activity was reduced to almost 0 by day 16, and remained very low until the soil profile was rewatered (Figure 26). It is interesting to note that nitrogenase activity showed only slight recovery on day 21. The soil profile was only partially filled by the irrigation on day 20, and although the majority of the nodules were in wet soil, the nitrogenase activity continued to be depressed. This suggests that nodule water is derived from roots deeper in the soil and that the nodules were apparently not capable of complete rehydration from immediately surrounding soil moisture. While nitrogenase activity showed a.decreasing linea'r trend with decreasing percent nodule moisture, the data in Figure 28 indicate that there is a sharp end point once the nodules dry to a nodule moisture of about 50 to 55%. Although nitrogenase activity appeared to decline concurrent with the other physiologi cal processes, the magnitude of the depression near the end of the drying cycle was greater for nitrogenase activity. When canopy carbon exchange was reduced by about 50-60%, and when stomata were only partially closed, nitrogenase activity dropped to essentially zero. The data support the hypothesis that decreases in nitrogenase activity are primarily caused by nodule dehydration and tissue damage and not directly related to short-term changes in photosynthesis. -48 -

PAGE 49

-0.2 -c a. ---1 -0:6
PAGE 50

Figure 24. Effect of a soil water drying cycle on soybean leaf conductance to water. Arrows on the abscissa at days 20 and 22 indicate rewatering of the stressed plants.

PAGE 51

.05 ,.--., I en E ---.04 w u z u03 ::::> c z o u .02 1.L. w ...J .01 0 4 \' \ \ \ ... t'," \ ,I \ 'I ..... DRY" t I ......... -", ...... "" ""',,, ..... .",. -,. ..... .",. ..... "" 8 12 16 20' DAYS OF DRYING CYCLE 24 (J'1 --'

PAGE 52

w (!) Z :r: u x w z o CD a:: 1:3 1.1 0.9 IE 0.7 00' ZE u IZ W a:: a. 0.5 0.3 o 4 f\ [] \ I \ / ... \ /. \. / \' ',J/ .\ ,I 't \1 8 12 16 20 24 DAYS OF DRYING CYCLE Figure 25. Effect of a soil water drying cycle on soybean apparent canopy carbon exchange. Arrows on the abscissa at days 20 and 22 indicate rewatering of the stressed plants. 0"1 N

PAGE 53

Figure 26. Effect of a soil water drying cycle on soybean nodule nitrogenase activity. Arrows on the abscissa at days 20 and 22 indicate rewatering of the stressed plants.

PAGE 54

30 ....-.I >-.!! 25 I-:J "'tJ >0 -c 1-0' I -en . I I ., -I ., J \ I \ I \ -I \ I \ I I \. DR I \ \ I 8 12 16 20 DAYS OF DRYING CYCLE 24 28 U1 I

PAGE 55

W 0: 70 C/) o W -.J ::J 60 o o z r z w (J 0: 50 w 0.. IRRIGATED' ",. I I I I I I \ I \\ ,...)/ / r REWATER 40 __ J__ __ -L __ __ __ 12 14 16 19 21 23 DAYS OF DRYING CYCLE Figure 27. Effect of a soil water drying cycle on soybean nodule moisture content. U1 U1

PAGE 56

Figure 28. The relationship of percent nodule moisture to nitrogenase activity during a soil water drying cycle.

PAGE 57

3 o 70 60 50 40 30 80 PERCENT NODULE MOISTURE 20 10 I U1 -.....J I .; .... ,'

PAGE 58

CHAPTER IV. SUMMARY The current study revealed important characteristics about the relationship of plant water status and metabolism to nitrogen fixation by legumes in sandy soils. Results of these experiments suggest that plant and nodule water status are closely linked to nitrogen fixation. However3 often nitrogen fixation appeared more sensitive to water stresses than some of the other measured physiological parameters. In all cases, by the time visible stress was observed or reductions in leaf water potentials or stomatal conductance occurred, nitrogen fixation was also depressed. This study has shown that drought stress decreases symbiotic nitrogen fixation and root respiration in legumes. The changes observed during the drought stress periods suggest that these may be tissue damage, and these observations, coupled to the measured physiological changes support the hypothesis that drought stress directly affects the interactions between Rhizobium and the host plant, as suggested by Sprent (1976). Flooding provided a mixed response in the. observed crops. Soybeans, cowpeas3 and Aeschynomene showed little adverse effects during the time they were flooded, while alfalfa, Desmodium, and bean exhibited sensitivity to the imposed stress. This sensitivity was most readily noted by reduced root respiration and nitrogen fixation. Legume crops in Florida can be subjected to periods of water stress, both drought and flooding, during their growth. In most instances, these stresses result in lower leaf water potentials, stomatal closure, decreased photosynthesis, reduced nitrogen fixation, and impaired respiratory functions. Any of the impairments, if of sufficient magnitude, can result in reduced yields and inefficient utilization of other management inputs. Physiological changes which occur in response to water stresses should ultimately help integrate efficient water management schemes into management practices. 58 -

PAGE 59

LITERATURE CITED Albrecht, S. L., J. M. Bennett, and K. H. Quesenberry. 1981. Growth and nitrogen fixation of Aeschynomene under water stressed conditions. Plant and Soil 60:309-315. Bisse1ing, T., W. van Stavern, and A. van Kammen. 1980. The effect of waterlogging on the synthesis of the nitrogenase components in bacteroids of Rhizobium 1eguminosarum in root nodules of Pisum sativum. Biochem. and Biophy. Res. Comm. 93:687-693. Bourget, S. J., and R. B. Carson. 1962. Effect of soil moisture stress on yi e1 d,,1 water-use effi ci ency and mi nera 1 composition, of oats and alfalfa grown at two fertility levels. Can. J. Soil Sci. 40:7-12. Bro1mann,'J. B. 1978. Flood tolerance in Stylosanthes, a tropical legume. Proc. Soil and Crop Sci. Soc. of Fla. 37:37-39. Burris, R. G., and P. W. Wilson. 1957. Methods for measurement of nitrogen fixation. In Methods and Enzymology. S. P. Co10wick and N. O. Kap1 an (eds.) ,4: 3-55. Academi c Press, New York. Clegg, M. D., C. Y. Sullivan, and J. D. Eastin. 1978. A sensitive technique for the rapid measurement of carbon dioxide Plant Physiol. 62:924-926. Dilworth, M. J. 1966. Acetylene reduction by nitrogen-fixing prepara ,tions from Clostridium pasteurianum. Biochem. Biophys. Acta 127:285-294. Engin, M., and J. I. Sprent. 1973. Effects of water stress on growth and nitrogen-fixing activity of Trifolium repens. New Phytol. 72:117-126. Evans, H. J., and S. A. Russell. 1971. Physiological chemistry of symbiotic nitrogen fixation by legumes. In Chemistry and Biochemistry of Nitrogen Fixation, J. R. Postgate (ed."), pp. 191-244. Plenum Press, London. Fishbeck, K., H. J. Evans, and L. I. Boersma. 1973. Measurement of nitrogenase activity of intact legume symbionts in situ using the acetylene reduction assay. Agron. J. 65:4.29-433-.--Gallacher, A. G., and J. I. Sprent. 1978. The effect of different water regimes on growth and nodule development of greenhouse-grown Vicia faba. J. Exp. Bot. 29:413-423. Huang, C-Y., J. S. Boyer, and L. N. Vanderhoef. 1975a. Acetylene reduction (nitrogen fixation) and metabolic activities of soybean having various leaf and nodule water potentials. Plant 56:222-227. -59 -

PAGE 60

60 Hua'ng, C-Y., J. S. Boyer, and L. N. Vanderhoef. 1975b. Limitation of acetylene reduction (nitrogen fixation) by photosynthesis in soybean having low water potentials. Plant Physiol. 56:228-232. Ki.lmer, V. J., O. B. L. Bennett, V. F. Stahly, and D. R. Timmons. 1960. Yield and mineral composition of eight forage species grown at four levels of soil moisture. Agron. J. 52:282-285. Kramer, P. J. 1951. Causes of injury to plants resulting from flooding of the soil. Plant Physiol. 26:722-736. Lawn, R. J., and W. A. Brun. 1974. Symbiotic nitnogen fixation in soybeans. I. Effects of photosynthetic source-sink manipulations. Crop Sci. 14:11-16. Ludlow, M. M., and G. L. Wilson. 1972. Photosynthesis of tropical pasture plants. IV. Basis and consequences of differences between grasses and legumes. Aust. J. Biol. Sci. 25:1133-1145. Mack, A. R. 1973. Soil temperature and moisture conditions in relation to the growth .and quality of field peas. Can. J. Soil Sci. 53:59-72. Mague, T. H., and R. H. Burris. 1972. Reduction of acetylene and nitrogen by field grown soybeans. New Phytol. 71 :275-286. Mclvor,J. G. 1976. l"heeffect of waterlogging-on the growth of Stylosanthes guyanensis. Tropical Grasslands 10:173-178. Mederski, H. J., and D. L. Jeffers .. 1973. Yield response of soybean varieties grown at two moisture stress levels. Agron. J. 65:410-412. Minchin, F. R., and J. S. Pate. 1975. Effect of water, aeration and salt regime on nitrogen fixation in a nodulated legume -definition of an optimum root environment. J. Exp. Bot. 26:60-69. Minchin, F. R., and R. J. Summerfield. 1976. Symbiotic nitrogen fixation and vegetative growth of cowpea (Vigna unguiculata (L.) Walp.) in waterlogged conditions. Plant and Soil 45:113-127. Pankhurst, C. E., and J. I. Sprent. 1975. Effects of water stress on the respiratory and nitrogen-fixing activity of soybean root nodules. J. Exp. Bot. 26:287-304. Pate, J. S. 1958. Nodulation studies in legumes. II. The influence of various environmental factors of symbiotic expression in the vetch (Vicia sativa L.) and other legumes. Aust. J. Biol. Sci. 11 :496-515. Quesenberry, K. H., S. L. Albrecht, and J. M. Bennett. 1982. Nitrogen fixation and forage characterization of Aeschynomene spp. in a subtropical climate. In Biological Nitrogen Fixation Technology for Tropical Agriculture, H. Graham and S. C. Harris (eds.). pp. 347-354. NifTAL, Honolulu.

PAGE 61

Salter, P. J., and D. H. Drew. 1965. Root growth as a factor in the response of Pisum sativum L. to irrigation. Nature 206:1063-1064. Schollhorn, R., and R. H. Burris. 1966. Studies of intermediates in nitrogen fixation. Fed. Proc., Fed. Amer. Soc. Biol. 24:710. Schwinghamer, E. A., H. J. Evans, and M. D. Dawson. 1970. Evaluation of effectiveness in mutant strains of Rhizobium by acetylene reduction relative to other criteria of N2 fixation. Plant and Soil 33:192-212. Sprent, J. I. 1969. Prolonged reduction of acetylene by detached soybean nodules. Planta 88:372-375. Sprent, J. I. 1971. The effects of water stress on nitrogen-fixing root nodul es 1. Effects 00' the phys i 01 ogy 0 f detached soybean nodules; New Phytol. 70:9-17. Sprent, J. I. 1972. The effects of water stress on nitrogen-fixing root nodules. II. Effects on the fine structure of detached soybean nodules. New Phytol. 71 :443-450. Sprent, J. I. 1973. Growth and nitrogen fixation in Lupinus arboreus as affected by shading and water supply. New Phytol. 72:1005-1022. Sprent, J. 1. 1976a. Nitrogen fixation by legumes subjected to water and light stresses. In Symbiotic Nitrogen Fixation in Plants, P. S Nutman (ed.). pp. 405-520. Cambridge University Press, Cambridge, U. K. Sprent, J. I. 1976b. Water deficits and nitrogen-fixing root nodules. In Water Deficits and Plant Growth, Vol. VI, Soil Water Management, Plant Responses and Breeding for Drought Resistance. T. T. Kozlowski (ed.). pp. 291-315. Academic Press, New York. Yoshida, S., and T. Tadano. 1978. Adaptation of plants to submerged soils. In Crop Tolerance to Suboptimal Land Conditions. pp. 233-254. ASA Publication, Madison, Wisconsin. 61

PAGE 62

PUBLI CAT! ONS Quesenberry, K. H., S. L. Albrecht, and J. M. Bennett. 1982. Nitrogen fixation and forage characterization of Aeschynomene spp. in a subtropical climate. Biological Nitrogen Fixation Technology for Tropical Agriculture, P. H. Graham and S. C. Harris (eds.). pp. 347-354. NifTAL, Honolulu. Albrecht, S. L., J. M. Bennett, and K. J. Boote. Relationship of nitrogenase activity to plant water status in field grown soybeans. Submitted to Field Crops Research. Albrecht, S. L., and J. M. Bennett. Nitrogen fixation in forage legumes in relation to plant water and metabolic status in drought stressed and flooded conditions. In preparation. Bennett, J. M., and S. L. Albrecht. Response to nitrogen fixation to soil moisture stress in beans and cowpeas. In preparation. Bennett, J. M., and S. L. Albrecht. Soybean nitrogenase activity as affected by plant water stress. Temporal changes and the relation to plant and nodule water status. In preparation. -62 -