r, 4.33:?
v. :P .34. I
a» 5.

V .1. .
1:. ..

 

3.3 ‘ , . , . , , V . .firiafi.

. A , ‘ . , ‘ ....:.

 

gs l 8 097 5

 

Ilmill]ililflljliimgl;l um"
Michigan State
University

 

This is to certify that the

dissertation entitled
Response of Robinia Pseudo Acacia L.
To Varied Soil Matric Potential

presented by
Omar Essaady

has been accepted towards fulfillment
of the requirements for

_PI!_'D'— degree in M

 

M5u;.,... Mr .;...A ' r, ”\r -' r .-. p 042771

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

”DATE DUE .. DATE DUE DATE DUE

 

 

 

my 1 a: w

flair:

 

 

 

 

 

 

 

 

 

 

 

 

fie

 

 

 

 

 

 

 

 

 

 

 

 

 

 

II

 

 

 

 

 

 

 

j

MSU Is An Affirmative Action/Equal Opportunity lndltution

 

 

 

RESPONSE OF ROBINIA PSEUDOACACIA L.
TO VARIED SOIL MATRIC POTENTIAL

BY

OMAR RAMADAN ESSAADY

A DISSERTATION

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

DOCTOR OF PHILOSOPHY OF SCIENCE

Department of Forestry

1989

@0309é

ABSTRACT
RESPONSE OF ROBINIA PSEUDOACACIA L.
TO VARIED SOIL MATRIC POTENTIAL

BY

OMAR R. ESSAADY

Due to its fast growth and nitrogen fixation
capability, black locust has been under intensive
investigation as a potential forest tree for a variety of
uses. The goal of this study was to determine how
different regimes of moisture availability influence whole
tree growth and carbon allocation in black locust.
Seedlings were grown in the greenhouse from seeds during
1987 in 15 x 15 x 55 cm containers filled with 20 kg air
dried field soil-Mason builder sand mix (1:2 by volume).
Treatments began on May 13, 1987. A randomized block
design of 64 seedlings and 4 treatments each was
established. Dehydration to the minimum designated soil
matric potential and then rehydration was implemented.
Treatments were control (-.20 bars), moderately stressed
(-.50 bars) and stressed (-.90 and -2.0 bars). Matric
potential was monitored by tensiometers and gypsum blocks.
Seedling water potential was measured on July 1, and July
17. Half of each block was harvested, while the other half

was rewatered after a period of stress. Harvesting began

Omar R. Essaady

when seedlings were 5 months old and proceeded at various
intervals on a block by block basis. Water deficit
proportionally and adversly affected all tree components
with the exception of nodule dry weight. Root/shoot, and
fine root/leaf area ratios increased as stress increased.
Diameter, wood volume and wood density, were also
influenced by stress. Stress resulted in a decrease of
40%, 72% and 74% of the total dry weight, while the
decrease in height was 15%, 25% and 29% for treatments
-.50, -.90 and -2.0 bars, respectively. Stress increased
N% (percent nitrogen) in leaves and decreased it in fine
roots. P% (percent phosphorus) markedly increased
following stress in almost all tree components. There was
a perfect correlation between dry weight and N and P
content. N% was not correlated to the dry weight, while P%
was negatively correlated to dry weight. N and P were
highly correlated. The N/P ratio was negatively correlated
with treatments. Control treatments (-.20 bars) maintained
a N/P ratio of 17.0 throughout the growth period. However,
stress markedly reduced the N/P ratio. After stress was
interrupted the N/P ratio increased rapidly. The N/P ratio
appeared to be a potential indicator of plant water status.
Total leaf area was highly plastic in response to moisture
stress. In general, black locust exhibited considerable

morphological plasticity in response to drought.

ii

dedicated

to my parents

ACKNOWLEDGMENTS

The auther would like to express sincere thanks and
appreciation to Dr. Kurt Pregitzer for his patience,
support and guidance during the course of my Ph.D. program.

Sincere thanks and appreciation are also extended to
Dr. James Hanover for allowing me using the Tree Research
Center facilities, to Dr.'s Donald Dickmann and James Hart
for providing me with their labratory facilities and to Dr.
Peter Murphy for his continuous advice and assistance.
Also, I thank them for their suggestions and critical
review of the manuscript and for serving on my graduate
commmittee.

Finally, I am deeply grateful to Dr. Phu Nguyen and
Andrew Burton for helping me with the chemical analyses of
the experiment and for their advice and assistance whenever

needed.

iii

TABLE OF CONTENTS

Page

LIST OF TABLESOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0.00.0000... iv

LIST OF FIGURES ........ 0....OO...OOOOOOOOOOOOOOOOOOOOO Vi

INTRODUCTION ......... O...0.0.0....OOOOOOOOOOOOOOOOOOOO 1

HypotheseSOOOOOOOOOOOOIOOOOOOO0.0.0...0.0.00.0... 4
Overall Hypothesis............................ 4
Specific Hypotheses..... .................. .... 5

Objectives ........... O OOOOOOOOOOOOOOOOO OOOOOOOOOO 5

MATERIALS ANDMETHODS.OOOOOOOOOOOOOOOOOOO0.0.0.0000... 6

Soil Mix......................................... 6
Experimental Design.............. ........ . ....... 7
Harvesting and Sampling.. ........................ 12
Nutrient Analysis ............................... 14

StatiStiCal Analyses O O O O O O O O O O O O O O ..... O O O O O O O O O O 17
RESULTS ........... OOOOOOOOOOOOOOOOOOOOOOO0.00.00.00.00 18
Water Relations.................................. 18

The Effect of Stress on Individual Tree Components 24

iv

Wood Volume and Density..........................
Height and Diameter Growth.......................
Interrupted Stress...............................
The Effect of Stress on Nitrogen and Phosphorus

Accumulation.....................................
Interrupted Stress Treatments....................

$011 and Plant NitrogenOOOOOOOOOOOOOOOOOOOOOOOOOO
DISCUSSION 0000000000 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO.

Water Relations..................................
Effects of Water Stress on Seedling Growth.......
Root/shoot Ratio.................................
Nodule Weight....................................
Height and Diameter Growth............ ...... .....
Stress Interruption..............................
Nitrogen and Phosphorus Allocation in Relation to
Water Deficit....................................
Nitrogen-Phosphorus-Weight Relationships.........

Nitrogen/Phosphorus Ratio...... ..... ..........

Interruption of Stress........ ....... . ...... ..

SMARY AND CONCLUSIONSOOOOOO000......OOOOOOOOOOOOOOO.

BIBLIOGRAPHY ......... .0... ....... OOOOOOOOOOOOOOOOOOOOO

32

32

35

38

44

47

50

50

55

58

59

6O

61

63

66

67

68

70

73

LIST OF TABLES

Table Page

1. Quality control analysis of nitrogen and
phosphorus tissue concentration................. 16

2. Litter recovered throughout the experiment
(leaflets and compound leaves).................. 21

3. Average pressure chamber readings and soil matric
potential (in bars), on two dates, arranged by
treatments...................................... 22

4. Individual tree components arranged by treatments.
Data for stressed treatments................... 25

5. Ratios of stressed (ST), and stress-interrupted (SI)
treatments to the control...................... 27

6. Correlation coefficients (r) of selected variables
for the stressed treatments..................... 31

7. Wood density and volume in relation to height and
diameter, arranged by treatments................ 33

8. Individual tree components arranged by treatments.
Data for interrupted stress treatments.......... 36

vi

9.

10.

11.

12.

13.

14.

Nitrogen concentration of individual tree components

and nitrogen contents per tree, arranged by
treatment. Data for stressed treatments........ 39

Phosphorus concentration of individual tree components
arranged by treatments. Data for stressed
treatments ........... .......................... 40

Percent of nitrogen and phosphorus, and nitrogen/
phosphorus ratio, for the root, shoot, and for whole
seedlings, (stressed treatments)............... 43

Nitrogen concentration of individual tree
components, and nitrogen content per tree, for
interrupted stress treatments.................. 45

Phosphorus concentration of individual tree
components for interrupted stress treatments... 46

Nitrogen acumulation and soil phosphorus concentration
after harvest (arranged by treatment). Data for
stressed treatments............................. 48

vii

LIST OF FIGURES

FIGURE - Page

1.

The experimental design layout. Harvesting dates and
the number of blocks are shown outside the diagram.
Treatments, (-.20), (-.50), (-.90), and (-2.0) bars,
are represented by: A, B, C, and D, respectively.
Each treatment per block represents four boxes of
four seedlings. The first row of each block was
harvested as stressed treatment (ST), while the
stress was interrupted (SI) in the second row... 8

Soil moisture characteristic curve for the soil
used in the experiment. Numbers indicate the
standard error of the mean...................... 19

Soil matric potential and predawn and midday pressure
chamber readings. Data were taken on July 2 and July
17 and designated as either 1 or 2 in the number
following treatment letter designation. Observation
treatment (0) was similar to the control (A). Each
point on the curve represents an average of four
readings........................................ 23

Fine root/leaf area and root/shoot ratios for stressed
treatments. Numbers indicate standard error. Data
represent average of four blocks................ 28

Total dry matter for the whole seedling, for the
root and the shoot, arranged by treatment. Numbers
indicate the standard error. Data represent average
of four blocks.................................. 29

viii

6. Accumulated height growth for all treatments. The
arrow indicates the begining of moisture stress. 34

7. Nitrogen to phophorus ratio for different tree
components related to soil matric potential..... 42

ix

INTRODUCTION

The ever-increasing demand for forest products is
being met by introducing fast-growing tree species both in
reforestation and afforestation plans. World-wide, black
locust (392131; pseudgacacia) is the most widely planted
species after Eucalyptus.

Forest trees are often planted on poor sites because
better lands are devoted to agricultural crops. Generally,
these sites lack adequate moisture and nutrients. Success
or failure of any plantation program depends mainly upon
selection of proper genotypes and the amount and the
availability of essential resources, especially water and
nitrogen.

Experiments designed to control soil moisture vary in
complexity and importance (Kramer 1969 and Slavik 1979).
Exposing trees to a given soil matric potential and then
resoaking them is easier to control, more reliable, and it
is an experimental system that mimics nature (Kramer 1969).
Once seedlings are planted in the field they rarely grow
under constant soil matric potential. The dehydration
rehydration cycle is a dominant characteristic of most

newly planted sites.

The goal of this study was to determine how different
regimes of moisture availability influence whole-seedling
growth and carbon allocation in black locust.

Black locust was chosen for study because of its
ability to fix nitrogen and grow on poor soils (Boring
and Swank 1984, Boring et a1. 1984, Reinsvold and Pope
1987). Soil moisture availability is critically important
during black locust growth and it appeared to be an ideal
species to study.

Studies dealing with the effect of water stress on
black locust are limited. Brandle et al. (1973, 1977)
studied the effect of water stress on protein content of
this species. Shulte and Marshall (1983) studied the
effect of various solute potential concentrations on plant
water potential and stomatal conductance. The major focus
of the present study deals with the cumulative effects of
various moisture treatments on whole-tree carbon, nitrogen
and phosphorus concentration. The nitrogen fixation rate
of black locust has been studied by Boring and Swank
(1984), Boring et al. (1984) and Zimmerman et al. (1982).
Studies dealt with the effect of nitrogen fertilization on
rate of fixation are conflicting (Reinsvold and Pope 1987,
Roberts et al. 1983, Zimmerman et a1 1982). Reinsvold and
POpe (1987) concluded that a minimum amount of nitrogen is
required for nodule development.

Boring and Swank (1984) related the fast growth rate

of black locust to its high nitrogen fixation capacity as
well as to the large leaf area index. In soybean Sprent
and Gallacher (1978) found that water stress did not affect
nodule dry weight. Nitrogen fixation was related to the
biomass of the active nodules. Nodule activity was in
return related to the leaf area, because moisture
availability influences the supply of photosynthates. This
study also examines how water stress influences nodule dry
weight.

Most carbon allocation studies ignore the fine roots
which constitute a substantial and obviously critical part
of plants. Fine root separation is laborious, requires
fast elutriation and it is costly. Also, fine roots are
sensitive to environmental changes. Since fine roots are
responsible for most seedling water and nutrient
absorption, seedling growth is often highly correlated with
the proportion of fine roots. Compared with more
traditional root/shoot ratios, fine root/leaf area ratios
should be a more reliable indicator of the sensitivity of
seedlings to environmental changes. Root/shoot ratio
reflects carbon allocation through growth stages over a
longer period of time.

Knowledge of nitrogen and phosphorus translocation in
relation to water stress is limited and no studies have
been done on black locust. Nutrient translocation is very

sensitive to environmental changes and especially to the

water status of a plant. Both nitrogen and phosphorus
content are related to the metabolic activity of plants.
Their translocation is more immediately sensitive to stress
than are morphological changes. Also, the nitrogen to
phosphorus ratio could reflect the relative level of stress
that seedlings experience (Knight 1986).

In this study, fine root/leaf area ratio, root/shoot
ratio, nitrogen and phosphorus allocation and nitrogen to
phosphorus ratio were examined in relation to variable soil
matric potential through dehydration-rehydration cycles.

The correlations between soil matric potential,
seedling age and xylem pressure were also investigated to
determine the maximum stress that seedlings can tolerate,
and to determine the seedling day/night water potential at
which seedlings exhibited optimum growth. Stress
interruption was an important element considered in this
study. It is important to examine how seedlings recover
from moisture stress in terms of growth and whole tree

carbon allocation.
Hypotheses
Overall Hypothesis

The allocation of nitrogen and phosphorus to tree
components (leaves, shoots, roots of various sizes) is not

affected by moisture stress.

Specific Hypotheses

1. As moisture stress increases, the ratio of fine roots to
leaf area remains constant. Black locust may grow slower
in response to drought, but the relative proportion of
leaves to roots remains constant.

2. The relative allocation of total nitrogen and phosphorus
to various tree components remains constant regardless of
moisture availability.

3. Severely stressed trees recover from stress (after

watering) at the same rate as those mildly stressed.

Objectives

The objectives of this study were:

1. To investigate the growth and carbon allocation of black
locust under a range of soil moisture conditions.

2. To determine the level of stress at which black locust
seedlings would not be capable of faster recovery.

3. To compare fine root/leaf area and root/shoot ratios and
their correlation to water stress.

4. To examine nitrogen and phosphorus content, their
concentration, correlation to seedling dry weight, and the
allocation of nitrogen and phosphorus to various tree
components when trees are growing under stressful

conditions.

MATERIALS AND METHODS

Soil Mix

Sandy loam soil was collected from the upper 25 cm of
a plowed area where field crops had previously grown. The
chosen lot was on an elevated site, well-drained and devoid
of organic matter and herbicide residue. Soil was sieved
through a number 6 mesh screen and air-dried for 6 weeks on
the greenhouse floor. A cement mixer was used to mix the
field soil with air-dried and sieved Mason builders sand.
Random samples were drawn from each mix for moisture
characterization and nutrient analysis.

Paper containers (15 x 15 x 55 cm) were used to raise
the seedlings. Four paper containers were placed in a
plastic milkcase (30 x 30 x 30 cm) as a support. The upper
parts of the containers were encased in a plastic bag and
wood frames for reinforcement. Containers were arranged in
such a way that the distance between seedlings was 15 cm.
The containers were carefully filled with 20 kilograms of
the soil mix ( field sandy loam soil and Mason builders
sand, 1: 2 by volume).- The media were allowed to settle
using biweekly irrigation for about four months. Soil pH

was lowered from 7.9 to 6.7 by leaching and adding dilute

sulfuric acid in the irrigation water before sowing.
Experimental Design

The experiment was conducted at the Tree Research
Center operated by the Department of Forestry, Michigan
State University. A randomized block design was laid out
(Figure 1). There were four blocks, four treatments per
block, and sixteen trees per treatment. Two additional
cases were placed at the end of each block for observation
and additional measurements.

After preliminary experimentation on various methods
of inducing soil moisture stress, four treatments were
selected for the study:

Itgatmgnt A: well-watered treatment. Soil matric
potential did not fall below (-O.22) bars. In this
treatment care was taken to maintain well-aereated soil
media to avoid suppression of nodule activity.

Itgatmggt p: moderate stress treatment simulating the
natural environment. Soil matric potential in this case
did not fall below (-0.55) bars.

fittggfigg treatmggts Qt and Q: Soil matric potential
averaged (-O.94) and(-2.0) bars, respectively.

The dehydration-rehydration method was implemented in
this experiment. When the average soil moisture potential
reached the minimum allowable negative matric potential,

seedlings were irrigated until water drained through the

.3ou ccooom ecu ca AHmv coumsuuoucfl mo: muouuu

may mafia: .Asmv acoauoouu commuuum mm cuumo>ucn mc3 xooHn come no 30“

umuwm use .mmsfiapuou usou uo mmxon usou musumuumou xooda Hum usuaucouu

comm .>H0>Huommmou .9 use 0 .m .c an concomuunon one much Ao.~IV use

.om.lv .Aom.v ..c~.IV mucoaucous .acucowu on» ukuuso csonm mum mxoo~n
mo muses: ecu one mouse ocwuuu>uo= .uso>au cmfimoc Heucuafiummxo one "H ousvwm

«:3 «zap «:3 Np:—
sxcp '2: etc puxo

#1 l . .3
+

 

 

 

 

 

 

 

 

 

yi'li .irlllull'l

 

 

 

l

 

 

 

 

 

er
EF—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

>_ .62.. z. goo... __ 3.3... Zoo...

bottom of each container.

Black locust Robinia pseuggagagia (Kellogg Forest seed
source number 0450) seeds were scarified in concentrated
sulfuric acid for 50 minutes, rinsed and sown on March 31
1987. After two weeks, seedlings were thinned to one per
container. Some seedlings were grouped by number of leaves
and height and transplanted to minimize the variability in
the seedling size between blocks.

Seedlings were grown under a 24 hour daylength regime
with supplemental fluorescent light. Average day-night
temperature and relative humidity were, 22.2 C°, 15.6 C°,
and 62%, respectively. Irrigation and fertilization were
supplied as needed.

Three commercial fertilizers were used in the
experiment, NPK (20-20-20), Miracle-Grow with NPK contents
of 15%, 30%, and 15%, and ammonium nitrate (34-0-0).
Before sowing, a volume of 500 ml, 140 mg/liter NPK was
applied to each container. On April 8, 1987, a volume of
500 ml 204 mg/liter ammonium nitrate was applied to each
container. Similarly, on May 5 all treatments were
fertilized until saturation with ammonium nitrate at a
concentartion of 200 mg/liter. From June 26 to July 17,
Miracle-Grow fertilizer was applied at a rate of 200
mg/liter to each treatment at the time of irrigation. On
October 31, 300 mg/liter of NPK (20-20-20) were added to

interrupted stress treatments to examine the effect of

10

fertilization on nodulation. The actual amounts of
nitrogen and phosphorus were calculated according to the
amount of water retained in the containers at saturation by
referring to the soil moisture characteristic curve. The
amount of the nutrients available to the seedlings depended
upon the volume of the water added to each treatment during
irrigation.

Low soil moisture treatments were monitored by
tensiometers (models 2900 series E and 2100 series F,
SoilTest, Santa Barbara, California). For stressed
treatments, water content was measured by using calibrated
readings of gypsum blocks buried at 25 cm, as well as by
measuring soil moisture content through periodic soil
sampling. At least two gypsum blocks were installed in
every replication per treatment. Tensiometers were also
used for treatments C and D to observe the decline in the
soil moisture potential, and to prepare for gypsum block
measurements.

Soil moisture characteristic curves were constructed
from saturation to 1.0 bar by using PVC pipe sections 8 x
7.5 cm in height and diameter, respectively. Replicates
were run on one bar pressure plates in low pressure
containers. At tensions greater than one bar, a higher
range moisture extractor was used on five and ten bar
pressure plates. The PVC rings in the latter were 3.5 x

7.5 and 3 x 5 cm in height and diameter. Gypsum blocks

11

were calibrated by monitoring weights of four one-gallon
containers with two gypsum blocks buried at various depths
in each, along-side a tensiometer for comparison. In
addition, cylinders of 10 x 7.5 cm with one or two gypsum
blocks were used for calibration of soil matric potential
from one bar up to saturation. Rings of 5 x 7.5 cm with
one gypsum block in each were used for soil matric
potential more negative than one bar. .

The calibrated available moisture readings, determined
in the labratory were used to irrigate the stressed
treatments in the greenhouse. When soil matric potential
dropped below -.50 bar in the stressed treatments, usually
two readings per day were taken to insure timely
irrigation. The average readings of all gypsum blocks for
a given stressed treatment were periodically recorded until
they reached the irrigation point. When the soil matric
potential dropped below -.60 bars, the decrease in gypsum
block readings became more rapid requiring more frequent
readings. Soil matric potential of at least 5 containers
within each block were recorded for the control (-.20
bars) and the moderately stressed (-.50 bars) treatments.
Those treatments were irrigated when the designated average
soil matric potential was obtained.

A PMS pressure chamber (Boyer 1969) was used to
determine the water status of the trees at predawn and

midday. However, some difficulties were experienced during

12

the determination of xylem pressure. Either the rachii
were very thin or the pressure was high enough to push the
leaf out of the chamber. Eventually, it was possible to
measure one leaf per treatment per block at predawn, and
another at midday from a different side of each block.
Measured leaves were cut at 50 cm height, and they were
always cut as close to the stem as possible and immediately
placed in the chamber. Xylem pressure was measured twice,
on July 1 and July 17, 1987. The containers at the end of
each block were used for comparison, and the average soil

moisture potential was recorded.
Harvest and Sampling

Trees were well-watered until May 13 when the
treatments began. Height and diameter were measured from
the time treatments were initiated until harvest, at
various intervals. Also, during the stress period litter
was collected from each treatment, oven-dried and separated
into leaflets and rachii. Harvesting began when treatments
showed a significant difference in height.

Harvesting proceeded from the first block to the
fourth and occurred on the following dates: August 21,
September 18, October 4, and October 7, 1987. Eight trees
per treatment per block (of sixteen total) were harvested
on each date. The remaining eight trees were left for

further study of interrupted stress. Trees were well

13

watered and fertilized to determine how black locust
responded to interrupted stress (Figure 1).

At the time of the harvest leaflets of each tree were
stripped off the rachis and placed in the refrigerator,
while the stem was cut into sections of 10-15 cm and placed
in the cooler. All nodules and roots more than one mm were
hand-picked from the soil for each tree, while emptying the
contents of the container into a bucket. During the
harvest, three soil samples were collected at 5 cm, 25 cm
and at the bottom of the container. Samples were collected
for soil moisture content determination as well as for
nutrient analysis. An equal volume was collected from each
depth. Twenty-five percent of the total soil was removed
from the bucket and placed in plastic bags and immediately
elutriated to remove fine roots (roots <1 mm in diameter).
The washed fine roots were immediately placed in the
freezer. As time permitted a whole container was washed
from each treatment to determine the validity of the
sampling technique, which showed consistency with the
sampling procedure.

Leaf area was measured using a Li-Cor LI-3000 area
meter. At least two trees per treatment were measured. In
the case of stressed treatments with fewer leaves, leaf
area for the entire tree was determined. The weight of all
leaves from each tree was measured. Total leaf area (one

surface) and specific leaf weight (gram/dmz) were

14

determined for each tree. Leaf areas of nonsampled trees
were estimated from sampled trees of the same treatment and
the same block using the specific leaf weight of the
sampled leaves and the total dry weight of all trees.

Large roots (>1 mm), fine roots, and nodules were washed
and oven-dried. All tree parts were collected and treated

separately.
Nutrient Analysis

Tree components were ground separately, except the
nodules and the fine roots of treatments 8, C, and D.
Those samples were composited by treatment and block and
combined on a dry weight basis. Stems, leaves, and main
roots from individual trees were composited on an equal
weight basis and ground. A composite sample from each tree
component was prepared for each treatment per block and
three replicate samples from each treatment-block
combination were used for nitrogen and phosphorus
determination. Some parts of randomly selected trees were
ground and sampled separately to estimate the variability
as a result of compositing.

. Soil samples from each container were combined by
depth on an equal volume basis (air-dried). One sample was
drawn for nutrient analysis. In addition, the original
field soil and the mix before sowing were analyzed to

determine the change in nutrient status of the soil over

15

the course of the experiment. Total nitrogen and
phosphorus were determined by the Kjeldahl method using
Technicon II instruments. Two blank samples and two NBS
(National Bureau of Standards) samples (pine needles
containing 1.2% nitrogen and 0.12% phosphorus) were used
for quality assurance. Concentration standard of 9%
nitrogen was used to determine the total nitrogen for the
nodules and the leaves of the stressed treatments because
they contained more nitrogen the other seedling components.
Any run that overestimated or underestimated the NBS
standard was repeated (Table 1).

The stress-interrupted experiment was harvested on
November 12, 1987 for the first two blocks, and on December
12 for the third and the fourth blocks. Harvesting and
sampling procedures described earlier were implemented for
the second stage of the experiment. However, only vigorous
trees were harvested because some trees exhibited poor
shoot growth, and there was some mortality. Poor growth
was likely due to the change in the greenhouse growth
conditions, such as temperature and photoperiod, as winter
began.

To demonstrate the effect of moisture stress on the
growth of the stem, three trees per treatment per block
were randomly selected to determine wood volume as well as
the wood density. Tree sections were oven-dried and the

volume was calculated using the frustrum of cone fermula.

16

.eueauese ecu sw uees meassee uo senses ecu musemeumeu
uees msOuue>ueeco mo uecsss ecu use scuueu>eu uueuseue ecu eueOuusu Aso use A.u.eo «a

Aevs use .eueaaese ecu sw

.»~H.o mu esHe> usuocseocn Ammz. ueuuuuueu ecu “»~.H mu esae> seoouuus

mm: ueuuuuueo ece .eusueoous euexaese eseeuu udddmmm ecu ousu ueueuomuoosu Ammzo
euueuseum no seeusm HesOuuez uo eeueOuHseu deue>ee uo eoeue>e ecu useeeuseu euaseeme

 

 

ma
. moo. boo. wad. hmo.
om m no u «o ooo. odd. hHo. mnm.a meaauos
«a
hHo. omo. oma. oo.
mod 0 we u «n nHo. baa. Nod. o-.H uuoo.H eswu
h
moo. HHo. oao. omo.
mm v vn ooo. NHH. mvo. ohH.H muoou sfiefi
m
woo. woo. ono. mmo.
me m an ooo. had. uno. Nod.a Beum
NN
oHo. nao. odd. mno.
Nod Ha won wn Heo. mafia. who. mNN.H meeH
e s
moo. moo. «no. mho. ««.u.m
ov m wn boo. had. ono. ~vm.a seeE mucoeu
.eo : menu enmecaum geese mm: seeds em:
mo uecsss .osoo z esuocseocs seoouuus Heuueues useas

 

c .esOuueuuseosou eseeuu esuocseocs use seoouuus no eueaaese Houusoo auwaeso .H eases

17
Statistical Analyses

Stressed treatments had no missing values, unlike the
stress interrupted treatments. The analysis of variance
for a randomized block design with eight subsamples or less
(depending on the missing values or the variable to be
measured) was run on all tree components, as well as on
nitrogen, phosphorus, and nitrogen/phosphorus ratio.
However, N and P were not analyzed for all trees, rather
samples were composited by treatments in each block.
Analysis of variance for tree volume as well as the height
and diameter was run with three subsamples.

Duncan's multiple range test was used for the
comparison of the treatment means at the 5% probability
level. In the Results, a probability level of 5% indicates
significance, while the notation "highly significant"
indicates a probability of 1% or less.

Parameters such as standard deviation and standard
error of the mean for a given treatment were calculated for
all blocks regardless of block variability (harvesting
time). Pearson's correlation coefficients (r) between

variables were also calculated.

RESULTS

Water Relations

There were no differences among the bulk densities of
the containers which could have resulted in large
variabilitiy between or within treatments (data are not
shown).' Since the same weight of air-dried soil was added
to each container, water holding capacity of the containers
varied only slightly. Field capacity was estimated as 7.1%
water content, and the saturation point was 17.3%. The
growing medium was designed to estimate how fast the
species would respond to rapidly changing soil moisture,
and the experimental system worked well (Figure 2).

Treatments were imposed when the number of compound
leaves averaged 6 per tree and the average number of
leaflets totaled 36. After 7 weeks the number of leaves
was, 20, 18, 15, and 15, arranged according to treatments,
from nonstressed to severly stressed. However, no leaf
area data were collected at 7 weeks. A moderate stress of
-.40 bars had been exerted on all trees to enhance root
development before the experiment was divided into
treatments.

Leaf rolling and orientation were related to

18

19

 

 

 

-6 _
-5 P
2?
a
.o
V -4 "'
.4
e
0'4
.¢ -
5m
4.)
o
s
-2 '-
-1 -
O
0 5 10 15 . 20

moieture content (percent) oven-dry eoil

Figure 2: Soil moisture characteristic curve
for the soil used in the experiment.
Numbers indicate the standard error of
the mean.

20

treatments, but no data were collected. Leaflet rolling
was more pronounced on lower and larger leaflets. It was
almost a constant symptom in treatment B (-.50 bar). Acute
leaflet orientation was more pronounced in both stressed
treatments (C and D) and mainly on smaller leaflets.

Angles decreased as stress increased.

The most obvious response to moisture stress was a
decrease in the leaf area, i.e. leaf shedding. Table 2
shows the litter collected by treatment and block, and the
estimated leaf area. Nonstessed and moderately stessed
treatments supported larger leaflets: up to 20 cm2. These
treatments also produced larger compound leaves and larger
rachii. Size of the compound leaf, leaflet area, and rachis
length seemed to be highly correlated, but no data were
collected. Stressed treatments shed three times as much
leaf weight as moderately stressed treatments (Table 2). A
decrease in soil matric potential significantly decreased
total leaf area in the stressed treatments (C and D), while
the moderately stressed treatment (B) did not differ from
the control (-.20 bars).

Pressure chamber readings for two periods, July 1 and
July 17 (Table 3) are summarized in Figure 3. Each
observation was an average of four readings, where one
reading per treatment per block was recorded. Soil matric

potential, predawn, and midday xylem pressure were highly

correlated (p <0.01). Since irrigation schedules varied

21

Table 2. Litter recovered throughout the experiment (leaflets
and compound leaves).*

 

treament ** block number weight L.A

(bars) 1 2 3 4

 

. .............. grams..... ..... .............. dm2
-.20 23.16 16.91 43.81 14.81 1.54 a 2.81 a
-.50 13.79 45.02 27.42 35.55 1.90 a 3.71 a
-.90 82.25 82.91 58.90 85.21 4.83 b 13.00 b
-2.0 72.16 34.35 58.85 78.73 3.81 b 8.93 b
mean 47.84 44.80 47.25 53.58
*Treatment differences were highly significant (p <.01). Treatments

followed by the same letter were not significantly different.

Data under the blocks indicate the total litter weight for each
treatment (16 trees), while data for the weight and the leaf area
were averages for individual seedlings.

** Treatments represent the minimum soil matric potential before

rehydration.

22

.Heuuseuos Ouuuea Huom u

.m.£.m ee

.. Ho.o v a . unmouuuemum use: Hue one .om.o can om.o

seeauec usuhue> museuuuuueoo scuueaeuuoo uec scuueouuuu uo eeuoao ee Hues we aeuuseuos
Ouuuea Huoe use .Aheuuua use szeueuso mucuuem cuoc how emsuueeu Hecaeco eusueeum
.xooHc Hes eso .eosuueeu usou ueeea ue no eoeue>e se eu euceu ecu su osuueeu coem

.Oom.nm mes ha >Hsh so eunueuesseu Essuxes ecB

.N hush so .>He>uu0esueu .amo use

.OOe.>H .Oo~.~n euea auuuussc e>uue~eu use eusueuesseu asswsus use assuxea ucous\>eoe

 

 

 

~N.I n.mHI u.ol ho.l o.~nl o.vml o.NI
oo.HI o.nnl o.onl on.| v.nal o.w| om.l
om.| o.o~| ~.oal oH.| o.nHI o.o| om.l
nH.I o.mHu m.vn HH.| ~.na| o.hl oN.I
.m.:.m >euuua sseueus ««.m.=.m >euuus szeueum usesueeua
.0....0.0000000000000IOOOIOOOOIOOOOOOOOmHMQOOIOOOOOOOOOICOOoOOCOOO
OOOOOOOOOOONIH >H=n0000000000.000 0.0.0.000..." adgHOIOOOOODOOOO
e.euseaueeuu hc ueoseuue .eeueu ozu
so Aeuec suo Heuuseuom Ouuues auoe use mosuueeu necfieco euseueun emeue>< .n eases

23

.emsuueeu

usou no eoeue>e se euseeeuaeu e>uso ecu so usuos

coeu .Acv Houusoo ecu ou uenuluu ee: coo uselueeuu
sOuue>ueecO .sOuuesoqeeu ueuueu usesueeuu usuaoHAOu
uncles ecu su « so A Hecuue ee ueuesoueeu use ha
mash use m hush so sexeu one: eueo .eosuueeu hecIeco

essences heuuul use saeueua use neuuseuon Ouuuel Huom an enemas

 

 

.eaceuoa xeuEE I l .3833 Eseuea I Race—ca =8 II.
Econ—gum:
No 5 mm No 5 5 2 5 NO N<
Tifiilllillullill « . q a . _ 3 #

 

 

 

 

 

 

(seq) rexiueiod

24

from one treatment to another, the values in Figure 3 were
arranged according to the predawn xylem pressure as well as
according to the soil matric potential. The figure shows
that treatments with higher soil matric potential (more
recently irrigated) exhibited a predawn xylem pressure
around -7 bars. Their xylem pressure had almost doubled by
midday. On the other hand, trees under higher stress did
not show much variability between midday and predawn
readings. Observation trees at the end of each block were
treated as Treatment A (-.20 bar), their xylem pressure
data were plotted in Figure 3, but are not shown in Table

3.
The Effect of Stress on Individual Tree Components

Soil moisture stress had a pronounced effect on the
growth of all tree components, with the exception of nodule
dry weight (Table 4). Nonstressed and moderately stressed
treatments (A and B) produced larger nodules than the other
two treatments (C and D). In general, data in Table 4 show
that, out of 15 selected variables, 14 showed no difference
between the two highly stressed treatments (C and D). In
other words, soil matric potential around -1.0 bar had the
same effect as -2.0 bars. However, there was about a 1%
difference in soil moisture content between the treatments
(C and D). Treatments A and B ( -.20 and -.50 bars)

exhibited many significant differences. Specific leaf

25

 

 

Table 4. Individual tree components arranged by treatments. Data
for stressed treatments.
Treatment (bars)
A B C D
Component
-.20 -.50 -.90 -2.0
total dry weight (gm) 109.87 a 66.44 b 30.71 28.99 c ..
shoot dry weight (gm) 67.0 a 39.78 b 17.10 15.57 c
leaf area (dmz) 50.18 a 31.69 b 14.69 11.18 c
total leaf (gm) 26.88 a 16.20 b 5.57 4.77 c
leaflets only (gm) 22.44 a 13.16 b 4.41 3.82 c
stem (9D) 40.11 a 23.83 b 11.53 10.81 c
root dry weight (gm) 42.88 a 26.66 b 13.61 13.42 c
main root (> 1 mm) (gm) 23.74 a 14.61 b 5.25 5.51 c
fine root (< 1 mm) (gm) 18.72 a 11.63 b 8.10 6.78 c
nodules (gm) .420 .410 .310 .296
specific leaf weight* .450 a .425 a .310 b .345 b
root/shoot ratio *** .622 a .660 a .815 .857 b
fine root/leaf area *** .381 a .399 a .821 .957 b
height (cm) 191.86 a 162.67 b 143.52 c 135.61 c
diameter (cm) 1.10 a .86 b .65 c .60 c

 

*Specific leaf weight in gm/dmz.

** Treatment means were highly significant (p < .01). Means followed

by the same letter were not significantly different.

*** Root/shoot and fine root/leaf area ratios were significant at 5%.

26

weight, root/shoot ratio, and fine root/leaf area ratio
were not affected by moderate stress. Specific leaf weight
decreased as moisture stress became more severe. Stressed
treatments produced thinner leaves, and as time progressed
smaller leaves were typical in the stressed treatments.

In Table 5 data are presented as a ratio of the
control, i.e. the growth of trees in Treatment A represent
100%. Root/shoot ratio increased by 40%, and the fine
root/leaf area ratio almost trippled under the higher
stress treatments (Figure 4). Moisture availability
clearly affected carbon partitioning between root and shoot
(Figure 5). While the nonstressed treatment (A) had 39% of
its total dry weight in the roots, the stressed treatments,
(B, C, and D) accumulated, 40%, 44%, and 46%, respectively,
in their roots. Dry weight of fine roots increased more
than the larger roots as stress increased. Dry weights of
fine roots as percent of total weight were, 17%, 17.5%,
26.4% and 23.4% for treatments A to D, respectively. Stem
weight was not affected as much as leaves and fine roots.
Also, treatment B had a slightly larger root biomass than
the control on a percentage basis, while dry weight of the
main root declined in the other treatments. In general,
moderately stressed treatments grew about 60% as fast as
the controls, while highly stressed treatments grew only
half of that.

Rachii were separated from the leaflets because they

27

Table 5.
treatments to the control.

Ratios of stressed (ST), and stress-interrupted (SI)

 

Treatment (bars)

 

ST SI ST SI ST SI
Component

-0.5 -0.5 -0.9 -0.9 -2.0 -2.0
total weight ' 0.60 0.74 0.28 0.35 0.26 0.32
root weight 0.62 0.87 0.32 0.43 0.31 0.37
roots >1mm 0.62 0.91 0.22 0.43 0.23 0.33
roots <1mm 0.62 0.79 0.43 0.42 0.36 0.46
nodule weight 1.00 1.00 0.74 0.94 0.71 0.84
shoot weight 0.59 0.61 0.26 0.28 0.23 0.28
stem weight 0.59 0.62 0.29 0.29 0.27 0.31
leaflet weight 0.59 0.60 0.20 0.26 0.17 0.25
C. leaf weight* 0.60 0.59 0.21 0.26 0.18 0.23
leaf area 0.63 0.57 0.29 0.37 0.22 0.26
SLW 0.94 1.06 0.69 0.70 0.77 0.95
R/S ratio 1.06 1.44 1.31 1.94 1.39 1.39
FR/LA ratio 1.05 2.49 2.65 1.24 2.51 1.85
height 0.85 0.84 0.75 0.65 0.71 0.70
diameter 0.78 0.90 0.59 0.65 0.55 0.75

 

Each observation was represented by 32 trees for the stressed

treatments, while the number of trees for stress-interrupted

treatments varied.

* Compound leaf.

28

.exooHc noon «0 emeue>e ecu useeeuseu
eueo .uouue uueuseue eueo«us« euecasz .eusefiueeuu
ueeeeuue you nauueu uooce\uoou use eeue meeH\uoou esum “v eusous

 

 

 

 

 

 

«005302 D 38 .358. oz: I
was 5958..
ON. 8.- 8.- ON.
o
1 N6
L to 1
8 «o H.
to
1 so 0
8. 8.
. ad
3. N—
a.
L —

 

up.

29

.meOHQ
seen no emeue>e se useeeuaeu eueo .uouue uueuseue ecu

eueOuusu euecssz .useaueeuu uc ueoseuue .uooce ecu use
uoou ecu uou .osuaueee eHocs ecu you ueuuea auu Heuoa um enemas

296; .8. D .56; .85 a 2963 .32 l

 

 

 

 

1688658:
ON. 8.- OW..- ONR
. o
no no
~8—— e.—— low «W
e. e. n. I
«— od 9
he
1 cm
W...
. as
2.. L 8 m
1 8..
en ..o~u

30

constituted a large portion of the compound leaf weight.
Leaves of stressed treatments had 20% of their total weight
in the rachis, while the control had 18%. Total compound
leaf dry weight, leaflet weight, and leaf area were
severely affected by stress. The moderately stressed
treatment (B) lost 37% of its leaf area compared to the
control, while stressed treatments (C and D) lost 71% and
78% of their total leaf area, respectively.

Correlations between tree components were highly
significant (Table 6). Stem diameter was highly correlated
to stem dry weight (0.93), shoot dry weight (0.92), total
dry weight (0.91), root dry weight (0.86), and to leaf
weight (0.84). Stem diameter was also highly correlated to
leaf area (0.79) as well as to height (0.80). Similarly,
fine root weight was highly correlated to most other tree
components, except for nodule dry weight. Fine root weight
was highly correlated to leaf area (0.70). Leaf area, on
the other hand, was highly correlated to total dry weight
(0.87), with closer correlation to the shoot dry weight
(0.91). Treatments had the most pronounced effect on the
leaves. The correlation between treatment and leaf weight
was -.74. Correlation between treatments and leaf area was

-073.

31

Table 6. Correlation coefficients (r) of selected variables for
the stressed treatments.*

 

 

variable (r) d.f.
Fine roots and total dry weight 0.87 126
Fine roots and leaf area 0.70 126
Diameter and leaf area 0.79 126
Height and leaf area 0.71 126
Diameter and shoot dry weight 0.92 126
Height and diameter 0.80 126
Total root and total shoot 0.91 126
Diameter and total dry weight 0.91 ' 126
Dry weight and leaf area 0.87 126
Dry weight and height 0.82 126
Treatment and leaf area -0.73 126
Treatment and leaf weight -0.74 126
Stem weight and total weight 0.98 126
Diameter and height 0.74 46
Diameter and dry weight 0.90 46
Diameter and wood volume 0.87 46
Diameter and wood density 0.56 46
Height and dry weight 0.85 46
Height and wood volume 0.84 46
Height and wood density ** 0.32 46
Dry weight and wood volume 0.95 46
Dry weight and wood density 0.53 46
Wood volume and wood density 0.46 46

 

* All data were highly significant (p <.01%).
** Significant at ( p - .07).

32
Wood Volume and Density

Three trees per treatment from each block were sampled
for wood volume and wood density (Table 7). Wood volume
was affected by moderate stress (—.50 bars), but not wood
density. Severe stress affected both variables.
Correlation coefficients (r) among variables in Table 7 are
shown at the bottom of Table 6. Diameter exhibited a
higher correlation than height to wood density and volume.
Dry weight was closely correlated to wood volume (0.95)
rather than to wood density (0.53). However, both wood
volume and density were significantly correlated (0.46, p<

0.01).
Height and Diameter Growth

A decrease in tree height and stem caliper were the
most obvious visual symptoms of a water deficit. Leaf
biomass, on the other hand, varied quite dynamically
according to the time of irrigation. On the average, black
locust grew in height more than one centimeter per day.
Height and diameter were measured seven times from the time
of stress initiation to the harvest. Figure 6 shows the
cumulative height measurements. Moisture stress reduced
height growth after three weeks of stress, and the
treatments became distinguishable after 6 weeks, while
diameter decrease was apparent as early as 5 weeks ( data

not shown). Severly stressed treatments (C and D), were

33

Table 7. Wood density and volume in relation to height and
diameter, arranged by treatments.*

 

treatment diameter height dry weight volume density

 

(bars) (cm) (cm) (9m) (cm3) (cm3)
-.20 1.0 a 181.0 a 37.1 a 48.8 a 0.71 a
-.so .88 b 162.0 b 24.4 b 33.7 b 0.71 a
-.90 .66 c 139.0 s 11.6 c 17.6 c 0.65 b
-2.0 .63 c 137.0 c 11.9 c 18.4 c 0.63 b

 

* Data based on three randomly selected trees per treatment per
block. Total number of observations were 48 trees.

There was no significant difference between treatment means‘
followed by the same letter (p < .01).

34

.eeeuue eusueuos no ususuoec ecu eeueeuusu rouse

no eu30um

 

ece .euselueeuu Hue How cutouo ucUuec ueuedslseoc
o Eocene: II: 0 E8882. I
m “sesame: ...... . < Eesaeeh l l
5.82 we:
um um op mu m. o . o m o
o
om
. u.
9
r.
.\ 8. m
..... \ m
\ \ \ omu
\ ooN

 

35

not clearly separated. Treatment C was irrigated 10 times
during the stress period, while treatment D had 8
irrigations. At the fifteenth week, treatment D had
obtained more height than treatment C. Because treatment D
was irrigated while treatment C was under decreasing soil
matric potential.

Maximum growth in height for black locust occurred two
months after the sowing date, and then started to decline.
The average growth during the first two months was in the
vicinity of 0.50 cm per day: while during the fast growing
stage, it averaged more than 2 cm per day and then declined
steadily. Although treatments B, C, and D accumulated 60%,
28%, and 26%, respectively of the control dry weight, their
heights were, 85%, 75%, and 71% of the control. Moisture
stress did not affect growth in height as much as dry

weight.
Interrupted Stress

After stress was interrupted, there was a noticeable
increase in total dry weight and in growth of the roots,
resulting in an increase in the root to shoot ratio (Table
8). The moderately stressed treatment recovered faster
than the two severly stressed treatments, especially in
terms of height and stem dry weight. In general, the
moderately stressed treatment showed no significant

difference from the control with regard to dry weight of

36

Table 8.

for interrupted stress treatments.*

Individual tree components arranged by treatment.

 

Treatment (bars)

 

A B C D

Component

-.20 -.50 -.90 -2.0
total dry weight (gm) 141.9 a** 104.4 b 43.9 46.4
shoot dry weight (gm) 73.4 a 44.9 b 15.9 20.6
166: area (dmz) 39.4 a 22.6 b 14.9 10.2
total leaf (gm) 24.5 a 14.6 b 6.3 5.7
leaflets (gm) 20.1 a 11.9 b 4.6 4.9
stem (9m) 48.9 a 30.3 b 9.6 14.9
root dry weight (gm) 68.4 a 59.5 b 28.0 25.8
main root (> 1 mm)(gm) 45.1 a 40.8 a 17.1 14.8
fine root (< 1 mm)(gm) 22.4 a 17.7 b 10.0 10.2
nodules (gm) .96 .95 .90 .84
snw (gm/dmz) .51 a .54 a .37 b .47 a
root/shoot ratio .93 1.33 1.80 1.25
fine root/leaf area .63 1.57 .83 1.17
height (cm) 215.4 a 180.6 b 139.0 c 150.7 c
diameter (cm) 1.02 a .92 b .66 c .76 c

 

* Treatment means followed by the same letter were not
significantly different (p < .01).

37

large roots, specific leaf weight, root/shoot ratio, and
fine root/leaf area ratio. Only large roots of moderately
stressed treatments accumulated as much dry weight as the
large roots of the control. As with stressed treatments,
interrupted-stress treatments showed no difference in
nodule dry weight. Overall, the higher the stress that
seedlings experienced, the less responsive they were to the
alleviation of stress.

Specific leaf weight increased in all treatments
following the alleviation of stress indicating production
of thicker leaves. Leaf area, again was significantly
correlated with aboveground tree components (data not
shown). There was a significant correlation between leaf
area and shoot dry weight (0.86), leaf area and diameter
(0.87), and between leaf area and total dry weight (0.89).
In conclusion, when the stress was interrupted the
moderately stressed treatment grew more than 75% of their
control, and the severly stressed treatments (C and D) grew
more than 40%. Comparing Table 5 with Table 8, data
indicate that stress interruption resulted in a 14% gain in
total dry weight in the moderately stressed treatment,
while recovery was half of that in severely stressed

treatments.

38

The Effect of Stress on Nitrogen and Phosphorus

Accumulation.

Table 9 summarizes the nitrogen concentration in each
tree component. Stressed treatments accumulated more
nitrogen in the leaves and less in the fine roots, and they
did not differ from the control in the other tree
components. Nodules had the highest percent of nitrogen
followed by the leaves and then the main roots. Stem and
rachis did not differ in nitrogen concentration between
treatments. Although both stem and rachis accumulated the
least percent of nitrogen, on the average, the stem
exhibited a nitrogen concentration 0.3% higher than the
rachis (averaged over all treatments).

Phosphorus allocation was sensitive to stress (Table
10). A water deficit resulted in the accumulation of more
phosphorus in the stressed treatments. The higher the
stress, the more phosphorus accumulated. Large roots
accumulated more than the nodules, and the nodules had
slightly higher accumulation than leaves. For severly
stressed treatments, leaves and nodules had the same
percent of phosphorus accumulation and both were still
lower than the main roots. Stem and rachis had the least
percent accumulation of this element, with almost equal
percentages in both. Above ground phosphorus concentration

did not differ when comparing treatments C and D, while it

39

.Aao. v as presences

.cOuuew>ec ousoseum ecu mucemeumeu A.c.mv ««

.eeamaem mo uenac: ecu mucemeumeu .cv e

hauQMOHuwcowe uoc ewe: ueuuea ease ecu an ceaoaaou ecsea ucefiueeua

 

 

 

ow. ma. wo. mo. mo. om.
ma NH ha ma m rm
0 me. n~.v o ~m.a oe.~ cm.a cm.a em.m o.~|
on. me. no. we. me. on.
ma «a ma Ha . o he
0 mo. ¢Q.¢ n mm.H No.m Hw.d mm.H nm.n m.l
ma. NH. mo. we. No. me.
Ha NH ma NH m me
n mm.a ew.m s mo.~ ~o.~ mm.a. oa.a oa.n m.|
NH. mo. co. co. OH. NN. «4.0.m
Ha ma ha NH m an e c
e om.~ mm.¢ e mm.H wm.~ em.a hd.a vo.n ceefi o~.I
amuv eeeeeeeeeeeeeeeeeeeeeeeeeUCQOHereeeeeeeeeeeeeeeeeeeeeee
eeuu\z easooc uoou ecwu uoou cues Beue mucosa need uceaumeuu
.mucefiueeuu cemeeuue uou sumo .uceaueeuu an ceoceuue.eeuu Men
uceucoo cevouuuc use .mucecoeaoo eeuu Hescu>uccu uo cauueuuceocou semouuuz .m eacma

40

.c0wueu>eo cuecceum ecu mucemeumeu A.c.mv ««

.meamaem no genes: ecu mucemeumeu Ace «

.ae>eH an ecu us uceueuuuo haucMOMuucoum ewes mceea uceauseuu wucosm
.Aao. V dc uceueuuuo >Huse0uuucwum uoc euea ueuuea ease ecu an oesouHOH essea usesumeua

 

 

 

oo.o NH0.0 meo.o Ho.o No.0 vo.o
0H NH 5H ma m hm
M «N.o on nH.o U Nn.o 0 hH.o n mH.o 0N.o o.NI
no.6 Ho.o nHo.o Ho.o No.o mo.o
ma NH ma HA m h¢
m mN.O a NH.O O hN.o O hd.o n hd.o nN.O m.l
vo.o HHo.o HH0.0 moo.o Ho.o No.0
Ha NH ma NH m Nfi
m ON.o O vH.o n NN.o Q ¢H.o nm mH.o mH.o m.l
Nc.o Ho.o Hc.o noo.o Ho.o No.0 es fl.m
HH ma 5H NH m mm e c
m ON.O QM NH.o m VH.o m mo.o m mo.o wd.o Emma ON.I
.COOOCCCOOOOOO0......000000000OOOCCHCQOHmQOOOOOOOO0.00.00.00.0000
euccoc uoou ecwu uoou cues aeum mucosa used uceaueeuu
.muceaueeuu cemmeuum now sumo .uceaueeuu

ac oeocenus .mucecomaoo eeuu Hesow>wocu uo scuueuuceocoe msuocmmocm .oH eacea

41

differed when comparing the roots.

Data in Table 9 were divided by the data in Table 10
to generate Figure 7. This figure represents the N/P
(nitrogen/phosphorus) ratio for each tree component related
to soil matric potential. The N/P ratio decreased markedly
as water stress increased. The highest ratio was
maintained in nodules, followed by the leaves, while the
lowest ratio was in the rachis followed by the stem. Fine
roots had a higher N/P ratio in the stressed treatments
compared to the large roots, while this trend was reversed
in the control. The ratio was not very sensitive in the
three stressed treatments in the fine roots and the
nodules, while they were clearly separated in the main
roots and the leaves.

Nitrogen contents of tree components were calculated
for each tree to determine N allocation. N was added, for
each tree, for the whole root system, the shoot, and the
whole tree, and calculated for those parts on a percent
basis. The same procedure was followed for P and the N/P
ratio. These data are shown in Table 11. Nitrogen
concentration was decreased by stress in the root and did
not change in the shoot and in the total seedling.
Treatment 8 accumulated more nitrogen in the root system
than any other treatment. Phosphorus was increased by
stress in all parts: P% was increased markedly by stress

in total seedlings. All treatments had higher P% in their

42

.Heuuseuon euuuee Huoe ou oeusueu musecoeaoo
eeuu uceueuuwc now cause usuocnocn ou cemouuuz up encoum

.62 D 6....-. a an... i am.-. I

8.2.8 .8. 2... .8. use. so.» s62 .8.

 

 

 

 

 

 

 

 

 

 

ores: d/u

 

 

43

.xmo. «0 He>eH auuaucecoua
e ue useueuuuu >Huseewuuseum uos ewe: ueuueu esem ecu ac ue30-0u msees usesueeua

 

 

 

o h.oH 0 m.oa u e.oa 0 o~.o 0 o~.o o H~.o ha.~ ~H.~ 0e o~.~ o.~|
0 >.HH 0c e.HH o e.HH c ea.o 0 ea.o c ea.o HH.~ va.~ 0 uo.~ om.n
c h.na c s.ma h.nH c na.o c ma.o c eH.o u~.~ oH.~ c e¢.~ om.l
e 0.5H e 0.5d o.ha e NH.o e HH.o e ma.o OH.N mm.H e Hm.~ 0N.I
Heuou uoocm uoos Heuou uoocm uoou Heuou uoocm uoou muec
cuueu e\z m useeuem z useeueu useeueeuu
.Amuseaueeuu ueeeeuumv emsuaueem eaoca new use .uoocm .uoou ecu
now euueu msuocamoca\sewouuus use .msuocnmoca use semouuus no useouem .HH eacea

44

roots than their shoots. The N/P ratio markedly decreased
by stress and was similar for the root and the shoot.
Stress resulted in decrease in the N/P ratio of 19%, 31%
and 37% for in treatments 8, C, and D respectively, in

comparison to the control.
Interrupted Stress Treatments

When the stress was interrupted, treatments did not
show a difference in nitrogen concentration (Table 12).
Nodules maintained the highest percent of nitrogen followed
by the large roots and then the leaves. All treatments
accumulated more nitrogen in various parts, in comparison
to Table 9 with the exception of the leaves and the rachii.
Treatments showed no difference in phosphorus accumulation
with the exception of the nodules and the stem (Table 13).
In general, the control and the moderately stressed
treatments accumulated more phosphorus in the stem and in
the underground parts when the stress was interrupted in
comparison to stressed treatments. This element was
accumulated in all treatments in the fine roots as well as
in the nodules and the stem. Stress release resulted in a
decrease in phosphorus concentration in the leaf and the
rachis. All stressed treatments did not differ in
phosphorus accumulation in the stem. In the nodules, the
higher the previous stress the more phosphorus accumulated,

once the stress was alleviated.

45

.AHo. v av

useueuuHu >Huse0HuHsmHm uos eue3 ueuueH eaem ecu ha ue3oHHou mseea useEueeuB e

.musesoeaoo eeuu HHe
sH musesueeuu seeauec eoseueuuwu useOHuHsoHe 0s uezocm umeu emseu eHQHuHss m.seossc

.soHueH>eu uueuseum ecu wusemeumeu A.u.mv «e

.mumaHese ecu sH uees meHQsem mo necsss ecu musemeueeu .sv «

 

 

cu. MH. mo. mo. mo. m.
NH aH w m N N
U N0.H um.v ¢O.N mm.N mN.N OH.H om.N o.Nl
«H. «H. mo. OH. Ho.o ON.
9 OH 0 o v NH
O mm.H 5N.v mm.H Nm.N Hm.N ¢O.H mm.N ¢.I
mm. OH. VH. NH. Ho.o NN.
NH 0H 5 w v HN
n uh.N ho.v mH.N mH.m Nn.N mm. om.N m.l
Nm. HH. Ho. OH. Ho.o mm. es .6.m
OH HH 0 h u 0H s c
u oo.n mm.¢ mm.N Hv.m mm.H vo.H hh.N Coma ON.I
.‘EMHoOO OOOOOOOOOCOOOOOOOOOO0.0“:mOHmQCCOOOOOCOOOOOCCC-OCOOOOO
eeeuu\z eHsuos uoou esHu .uoou sHea aeum choeu ueeH useaueeuu

 

.museaueeuu meeuum ueumsuueusw new eeuu Hen
useusoo semouuHs .use musesomsoo eeuu HesuH>HusH Ho sowueuuseesoo seuouqu .NH eHceB

46

.soHueH>eu uueuseum ecu musemeuneu A.u.mv «4

.mo.

.meHmaem mo Hecass ecu musemeumeu Ase «

u Q

ue useoHuHsva eue3 museaueeuu eHsuos HHcs .Ho. n & ue useoeuHsaHm eue3 museaueeuu

 

 

 

seum .useueuuHu >Huse0HuHscHu uos ewes ueuueH eaem ecu ue3oHHou mseea museaueeua
No.o Ho.o Ho.o Ho.o Ho.o No.o
NH NH m m N h
0 5N.o ¢H.o 5N.o NH.o ¢H.o HN.o o.NI
Ho.o Ho.o No.0 Ho.o Ho.o mo.o
0 OH u o w HH
on mN.o MH.o 0N.c hH.o NH.o hH.o m.l
~o.o No.6 no.o Ho.o do.o no.6
NH uH h u v ON
ce HN.o mH.o mN.o uH.o NH.o NH.o m.l
No.o Ho.o Ho.o Ho.o Ho.c uo.o «e .u.m
mH HH u b e uH e s
e mH.o mH.o mH.o HH.o HH.o mH.o sees ON.I
.0...OOOOOOOOOOOOOOOOOOOOOOOOO0.0ucmOHmQOOOOOOOOOOOOOOOOOOOOOOO
eHsuos uoou esHu uoou sues aeum choeu ueeH usesueeuu
. .museaueeuu mmeuum
ueumsuueusu Mom musesomaoo eeuu HesuH>HusH no soHueuuseosoo msuocnmocm .nH eHceB

47
Soil and Plant Nitrogen

Basic field soil analysis indicated that the soil
contained 493 mg/kg nitrogen, based on 7 replicate
analyses. Soil mix analysis just before seed sowing showed
that the nitrogen content was 361 mg/kg soil, and 177 mg/kg
phosphorus, based on 16 composited samples. Since 20 kg of
air-dried soil was placed in each container, the soil mix
contained 7.22 gm of nitrogen at planting and 3.54 gm
phosphorus on a container basis. Table 14 represents the
nitrogen budget of the experiment. The amount of nitrogen
at seed sowing was known, as were the amounts of nitrogen
added as fertilizer and contained in the soil and plant
tissue following harvest. There was no difference in the
nitrogen concentration in the containers following harvest.
Also, the amounts of nitrogen added as fertilizer were
essentially the same when comparing treatments. Nitrogen
contents on a container basis after the harvest should
reflect the basic soil analysis, the amount of nitrogen
added as fertilizer as well as the input from fixation and
root turnover. Leaching was not under control and the
nitrogen fixation rate was not calculated. However, data
in Table 14 indicate that there was surplus nitrogen in the
soil in the control and in the stressed treatments C and D,
but not in treatment 8. Also nitrogen in the seedlings was
not included in the "difference" column calculations.

Nitrogen in the seedlings and nitrogen accumulated in the

48

. .AHH use a eeHceev aeumam uoou
ecu sH seoouuHs euos uec use .HHom ecu s« z eueHsesuee uos uHu Amuec om.lv usefiueeua «a

.m oc\oe osH use z.ox\ua nmv uesHeusoo HHoe uHeHh .msoHue>uemco no uecass usemeuneu
memecuseuem sH eueo .eeuu ecu mo museusoo sevouuHs msHe HHoe ecu sH ueueHsasoue
ussoae ecu eH ueueHsssooe seoouuuz .aHe>Huoemmeu .uesHeusoo ceee sH museaeHe cuoc now
means vm.n can -.s .e.« .m 6x\65 spa use .2 seem sexes Hen me: weer» uecu eeueoseru
@sHuseHn encuec mHm>Hese useHuusz .ume>uec ueuue sevouuus HHoe mssHs osHuseHm

ue wuseusoo HHoe ecu msHe soHueNHHHuueu sH ueuue semouuus musemeueeu semouuHs

uo reoseueuuucr .uesHeusoo Hem maeum use mc\cs sH eue museseHe cuoc you mussose ecee

 

 

cane cum. cams

ouH mm.o mNu.o mo~.o + Hmv.o mm.h van a o.NI

cenc cane Acne

qu hm.o emu.o OHn.o + Hhv.o 09.0 oou 0H om.l

cane sens can. _

HmH um.H mNm.H . «emoH.o I mu¢.o Nm.b won mH om.l

sens sens lune .

vVH mm.N Hen.N eNo.o + umv.o 05.5 mwn mN ON.I

mx\oa ueueHsesooe ............uesHeus00\Eeuo............. ox\ma .vHuuH Auecv

ueuue m seuouuus eeuu\z eoseueuuHu ueuue z ueuue z menses uauu

 

e.euseeueeuu ueeeeuum sou euec .ceuseaueeuu Nu ueoseuuev
ume>uec ueuue soHueuuseosoo msuocmeocn HHon use soHueHsssoee sevouuwz .eH eHceB

49

soil were included in Table 14 as the amount of nitrogen
accumulated. These amounts did not include the amount
added as fertilizer, or nitrogen absorbed from the soil
after sowing. Phosphorus absorption, on the other hand
increased by increasing growth, indicating more of this
element was absorbed by the control than by the moderately
stressed treatment. Phosphorus uptake by the severely
stressed treatments was minimum, which seemed to be related

to growth activity.

DISCUSSION

Water Relations

The effects of a water deficit on plant growth have
been extensively reviewed by Kramer (1969), Hsiao (1973),
Simpson (1981), and Turner and kramer (1980). The response
to moisture stress varies within and among species. The
duration and severity of moisture stress are two important
elements to be considered for a given species.

Temporary leaf wilting and change in leaf angle
orientation (active parahelionastic movement) are common
characteristics among legumes as a temporary morphological
adaptation to reduce radiation load on water stressed
leaves. The two phenomena were clearly observed in black
locust. Both were temporary on the control as well as on
other treatments depending upon the minimum allowed soil
matric potential. However, not all leaves within the same
tree exhibited these responses. They were mainly on lower
leaves or those in direct exposure to sunlight.

Leaf rolling became a constant morphological response
when the soil matric potential dropped below —0.40 bars,
especially on the lower half of the seedlings. In general,

leaf rolling, yellowing, senescence, and rachis senescence

50

51

were the gradual responses to a decrease in soil matric
potential.

Rotation of two opposite leaflets around their rachis
also reflected the water stress that seedlings were
experiencing. It was more common and explicit on smaller
leaflets while leaf rolling was more common on large
leaflets. Black locust leaflets rotate 90 degrees and the
angle increases between two opposite leaflets until a
maximum at 180 degrees. Leaflets of stressed treatments
did not completely close at night. Also, there was not
much change in the angle between day and night in the
severly stressed treatments. The larger the angle between
two leaflets the greater the water stress the seedlings
were experiencing. It seemed that there was a strong
correlation between leaf angle, seedling water status and
soil matric potential. However, no data were collected.

The other responses to water stress were defoliation
and production of smaller new leaves with altered anatomy.
When water stress was initiated leaf shedding was common
among all stressed treatments and leaf area decreased as
stress increased. Leaf expansion was significantly reduced
resulting in smaller thinner leaves, mainly on the lower
part of the seedlings. However, the upper leaves were
thicker and smaller in comparison to the control.

Water stress develops in plants when the balance

between water absorption and transpiration is not

52

maintained. Leaf rolling in black locust represents the
threshold of such an imbalance and is a useful visual
symptom in this species.

Plant water status or xylem pressure measurements are
correlated with seedling growth. Kramer (1987) summarized
the effect of water stress as either a direct effect on
cell enlargement resulting in a general decrease in leaf
area and plant size, or indirectly influencing stomatal
regulation and photosynthesis. Both effects were exhibited
in black locust resulting in a difference between
treatments in both growth parameters and pressure chamber
measuremnents. Although no direct measurements were taken
on stomatal activity in this study, pressure chamber
readings did reflect stomatal closure since negative xylem
pressure has to be neutralized by the pressure exerted by
the chamber through the stomata.

The pressure chamber technique for measuring water
potential is precise and reliable if properly used (Roberts
1987, Spomer 1985). Water absorption and translocation can
be affected by many factors as discussed by Kozlowski
(1987). In this study, equilibrium between soil matric
potential and plant water potential was not obtained and
the gap increased as soil matric potential decreased.
Height at which the leaf was cut and placed in the chamber,
transpiration at night (especially where the temperature

was above 15.6?C and resistance of water movement from the

53

soil to the root could have resulted in nonequilibrium
between soil matric potential and seedling water potential
in predawn measurements. In addition, the water column in
the xylem could have broken in response to stress resulting
in an increase in water flow resistance.

Hydraulic conductivity of the soil mix was very low
which could have been a factor in the slow recovery of the
stressed treatments at night. Studying black locust,
Schulte and Marshall (1983) demonstrated that mild moisture
stress resulted in lower xylem pressure potential readings
compared with severe stress and that the difference between
day and night readings decreased as stress became less
severe. They reported that duirnal differences in xylem
pressure potential mainly reflected changes in stomatal
conductance. Rawlins et a1 (1968) demonstrated that soil
moisture did not affect transpiration until it dropped to
-6 to -8 bars. After that transpiration decreased linearly
with decreasing soil matric potential down to -37 bars,
when the plant water potential was below -50 bars,
transpiration was zero.

Hinckley (1973) showed that wilting of black locust
leaves persisted in the dark after the plant water
potential fell below -23 bars, concluding that the after
effect of such stress on stomatal recovery was more
pronounced in the dark than in the light. Lower leaf

conductance upon rewatering after prolonged stress was also

54

reported by Schulte and Marshall (1983). Brandle et a1.
(1973) concluded that wilting was a means of controlling
excessive transpiration immediately following stress.
Xylem pressure decreased, on the average, from -7 bars
predawn to -15 bars midday at a soil matric potential of
-.22 bars. When the soil matric potential was -.87 bars,
the predawn xylem pressure potential decreased from -24 to
-32 bars at midday on July 2. At a soil matric potential
of -1.0 bars it decreased from -30 to -33 bars. However,
the gap between predawn and midday observations was
slightly lower (less negative) in stressed treatments.
Predawn plant water potential almost doubled during midday
in well-irrigated treatments (Figure 3). Since average
predawn and midday readings of the control as well as for
well-irrigated treatments were -7 and -15 bars
respectively, it can be concluded that such readings were
'normal' xylem pressure potential for black locust at that
age because the growth was not affected. 0n the other
hand, -30 bars was apparently the maximum xylem potential
that black locust could tolerate, which was approximately
double the midday xylem pressure potential of the control.
The soil mix used in the experiment was a sandy soil
which could be rapidly depleted of soil moisture. There
was no significant variation in soil moisture content at
-2.0 versus -6.0 bars. Because of the physical properties

of the soil (Figure 2), it can be concluded that -2.0 bars

55

had the same effect on black locust as a more negative soil
matric potential would have had.

Results from experiments on single species at a given
age can not be generalized (Brandle et al. 1973), and
plants with high sensitivity to a small decrease in water
potential may still maintain strong drought resistance
(Jarvis 1963). Certainly, the greenhouse environment, the
height at which leaves were cut for water potential
measurements, and other factors influenced results of this
experiment. Nevertheless, it appears that morphological
plasticity is one way black locust survives severe moisture

stress.
Effects of Water Stress on Seedling Growth

A water deficit markedly affected dry matter
accumulation of black locust through the effect on leaf
area and photosynthetic capacity, as in many other species
(Havranek and Benecke 1978). Sands and Rutter (1959)
demonstrated that a decrease in the soil water suction from
-0.1 to -0.5 bars decreased seedling dry weight by 15%,
while -1.5 bars reduced the weight more than 30%.

Comparing stressed treatments in Table 5, it can be
concluded that the shoot was more affected by stress than
the root system, and that the leaves were more affected by
water deficit than the fine roots. This trend indicates

that leaves probably ceased growth before fine roots in

56

response to moisture stress. Root growth in drying soil
depends upon the diameter of the roots. Fine roots (< 1
mm) increased in biomass in response to stress more than
large roots (> 1 mm). Black locust responded to stress by
producing a relatively greater biomass of small roots.
Bowen (1984) concluded that soil moisture availability has
a major effect on fine roots. Larson and Patashev (1973)
also found that shoots of oak seedlings decreased their
growth before roots. Sharp and Davis (1979) argue that
moisture stress can lead to salt accumulation in the roots
which enables them to grow after the leaves cease growth.
The close correlation between treatments and leaf area
indicates that a decrease in leaf area was probably a
mechanism to limit plant water loss and to help reduce
water stress. Stress resulted in a decrease in leaf
expansion and the final leaflets were smaller in Treatments
C and D. Water stress probably influenced cell division
and enlargement because fewer leaves were produced when
soil matric potential started to decrease. Also, the
compound leaves, leaflets, and leaf thickness were arrested
during stress, where moderate stress treatments expanded
their leaflets upon rehydration. In general, as soon as
stress was interrupted older leaves actually expanded if
not permanently wilted. Height growth was less affected by
stress (in a relative terms), probably because the apex is

a strong sink and remained so throughout the stress period.

57

Changes in the leaf characterization due to water
stress were detected by measuring specific leaf weight as a
function of leaf area and dry weight. Specific leaf weight
decreased as stress increased, especially in the lower
portion of the crown. Although the control and the
moderately stresssed treatments were significantly
different in leaf area, production of relatively smaller
leaflets in the latter did not affect the specific leaf
weight. This indicates that moderate stress arrested leaf
production without altering leaf characterization, and that
normal cell activity resumed as stress was released.
However, specific leaf weight is a function of other
factors besides water deficit, such as shade versus direct
sunlight leaves.

Dry matter production per unit leaf area was 2.2,

2.1, 2.1 and 2.6 gm/dmzfor the treatments, A, B, C and D,
respectively and exhibited no distinct relationship to
treatments. Leaf area and dry matter were highly
correlated. It appeared that a reduction in growth due to
lack of moisture was primarily related to a reduction in
leaf area and not a difference in the rate of
photosynthesis per unit leaf area. However, it should be
added that leaf size and area are not always a good
indicator of photosynthesic rate (Bhagsari and Brown 1986,

Hesketh et al. 1981).

58

Root/Shoot Ratio

Leaf area was slightly more affected by water stress
than the standing crop of fine roots, resulting in a
positive relationship between the fine root/leaf area ratio
and water stress. Root/shoot ratio for a given species
fluctuates during the growing season and it is influenced
by stage of growth as well as by environmental factors.
Kummerow (1980) argues that the relevent factor is not the
root/shoot ratio but the ratio of absorbing root surface to
the photosynthetically active leaf area of the plant.

Evans and Klett (1984) maintain that such a ratio
(root/shoot ratio) may not effectively predict water stress
susceptibility, suggesting the use of the ratio of leaves
to new roots.

Root/shoot and fine root/leaf area ratios reflect the
balance between absorption and transpiration and an
increase in these ratios could be viewed as a positive
adaptive mechanism. However, the latter is more
susceptible to the environmental changes in-shorter period
of time than the former. Black locust definitely decreased
its shoot growth more than roots in response to stress. At
moderate stress, root/shoot ratio was not changed in spite
of the significant change in the root and the shoot dry
weights. With increasing stress, seedlings produced
relatively more fine roots compared to large roots. Black

locust accumulated 61% of its total dry weight in the

59

shoot, of which 37% was accumulated in the stem. These two
percents were slightly altered by stress; the major effect
of water stress was on the leaves. The same seedlings
accumulated 39% of their total weight in the roots, of
which 22% were in the large roots. The general trend of
these data indicates that a water deficit resulted in rapid
defoliation, a decrease in large root biomass and an

increase in fine root biomass.
Nodule Weight

Nitrogen fixation or nodule activity depends upon the
supply of photosynthate (Dawson and Gardner 1979, Huang et
al. 1975, Sprent 1984). Environmental factors could affect
the nodules indirectly through photosynthate limitation.
Water supply could also limit nodule activity directly
regardless of photosynthetic supply (Sprent 1984). Boring
and Swank (1984) concluded that drought resulted in
reduction of nodule size. Sprent and Gallacher (1978)
concluded that water supply increases nodule size without
an effect on nodule weight. In general, nodule fresh
weight can be affected by water stress. In this study,
there was no difference in nodule dry weight among
treatments. However, there was a distinct difference in
the size of the nodules between treatments. In other
words, nodule fresh weight was different between

treatments, as well as nodule number, although no data were

60

collected for the latter. Control treatments (—.20 bars)
produced larger and fewer nodules than the stressed
treatments. Water stress affected the noudule's outer
cells more than the inner cells. Upon irrigation, new
nodules were developed. Continuous dehydration and
rehydration cycles resulted in production of smaller

nodules.

Height and Diameter Growth

As a fast growing species, black locust rivals hybrid
poplar in growth rate. The seed source involved in this
research exhibited substantial height growth. Boring and
Swank (1984) reported 8 meters of height growth in 3 years
for black locust. In this experiment, seedling growth rate
was 0.5 cm per day in the first two months after sowing,
increased up to 2 cm per day in the next two months, and
then started to decline.

Water stress started to affect height growth three
weeks after treatment initiation and treatments showed
distinctly different trends after six weeks. On the other
hand, caliper differences between treatments were
distinguishable after only five weeks of stress. Sharp and
Davis (1979) related the decrease in stressed plant weight
to the decrease in leaf area. Aiyelaagbe et al. (1986)
attributed a decrease in the diameter and internode

elongation to the decrease in cell elongation and

61

enlargement. In this study, height and diameter were less
affected by moisture stress relative to leaf area, but
there was a significant difference between treatments. The
height growth of stressed treatments were, 85%, 75% and

71% for the treatments 8, C and D compared to Treatment A.
Dry weights were 60%, 28% and 26% of treatment A,
respectively indicating that height growth was not as
sensitive, in a relative way, to fluctuating moisture
stress as weight gain.

Although only 48 trees were used for the statistical
analysis of wood volume and density (Table 7), the
statistical conclusions were consistent with the other
data, except for wood density. Moderate water stress did

not affect wood density, but severe stress did.
Stress Interruption

When the stress was released, leaf production occurred
mostly on the upper part of the stem. Preliminary studies
on black locust indicated that the optimum temperature for
growth is 20°C (McNiel 1975). A decrease in the greenhouse
temperature as fall progressed cannot be ruled out as a
factor in hindering the growth after the stress was
interrupted. However, the growth of the control seedlings
did not exhibit an appreciable decline.

Interruption of the dehydration-rehydration cycles

resulted in a general increase in the growth of all tree

62

components across all treatments. Production of thicker
leaves was common among all treatments. The burst of
growth following irrigation occurred mostly below ground.
Stressed treatments accumulated more than 75% of their
increased weight as a result of stress release in the root
system, and more than 60% of this weight was accumulated in
the large roots. This resulted in an increase in the total
root/shoot ratio and an increase in the fine root/leaf area
ratio. The increase in the latter was mainly attributed to
the slight increase in the leaf area compared to the larger
increase in the fine roots. Root/shoot ratio increased
50%, while fine root/leaf area ratio increased 65% compared
to the control. These trends indicate that root/shoot
ratios in black locust increase with age, and that the
species had 40% of its total weight in the root system by
the end of the experiment. The increase in the root weight
gain may be related to the fact that stress was interrupted
during the fall. Source-sink relationships vary with
season and the roots may have been a stronger sink in the
fall.

Although stressed treatments had enough time to
recover after stress release, they did not grow as would be
expected. This was primarily due to the decline in the
greenhouse temperature as well as to the after-effect of
the stress, especially in the Treatments C and D. Hsiao

(1973) maintained that the after-effect of water deficit is

63

proportional to the maximum stress reached before watering
and not to its persistance. Simpson (1981) concluded that
a tree's root growth can resume upon stress release even
after the shoot growth ceases late in the season. Pallas
et al. (1967) related the slow recovery of plants after
stress to root damage. The fact that trees responded to
the release of stress by growing primarily belowground was
probably due to the time of the year (Dickmann et al.

1989).

Nitrogen and Phosphorus Allocation in Relation to Water

Deficit

Nutrients are translocated in plants via the
transpiration stream. Depending on the stage of growth,
transpiration is more sensitive to water deficit than is
photosynthesis. Water stress results in redistribution of
nutrients within the plant, such as tranSlocation of
nutrients from wilting leaves. Also, plants respond to
stress by producing various compounds to decrease its
effect on growth and metabolism. Plants decrease their
nutrient absorptive capacity due to the decrease of
diffusion of these nutrients to the roots because of the
decrease in the soil moisture contents, as well as the
decrease in the absorbing root surface.

Phosphorus is the element most sensitive to soil

moisture content (Glass 1989). Its diffusion to the root

64

zone decreases substantially with the decrease of soil
matric potential, hence its absorption.

Total free amino acids increase with stress in the
leaves and the most pronounced increase is in the form of
proline which has been linked to water deficit resistance
(Clarke and Durkey 1981). However, Aspinall and Paleg
(1981) argue that proline accumulation is a response to any
stress, and is not necessarily an indicator of water
deficit. Hasegawa et al. (1984) reported an increase in
the concentration of various compounds such as sucrose,
nitrates and free amino acids such as proline, in stressed
tomato. Also, Brandle et al. (1973, 1977) and Hinckley
(1973) demonstrated that water stress increased polysome
concentration in black locust. So, an increase in nitrogen
in stressed leaves of black locust could be the result of
accumulation of these compounds in order for the species to
alleviate and adjust to a water deficit.

When nutrient concentrations reach a given point in
tissues, an increase in absorption becomes a function of
growth rate. Glass (1989) argues that the difference in
uptake among varieties are not the cause but rather the
result of differences in growth rate. Water stress
probably resulted in retranslocation of nitrogen and
phosphorus from wilting leaves down to the root system,
where roots are less impaired by stress and probably

function as a sink. Also, the assumption that water stress

65

cycles resulted in more nutrient absorption than
translocation to the shoot can not be ruled out. However,
translocation or immobilization varies between nitrogen and
phosphorus, and was controlled by water deficit as will be
demontrated.

Comparing nitrogen concentration in all tree
components, only leaves and fine roots showed a significant
difference. Moderate water stress did not alter nitrogen
concentration. Nitrogen concentration was higher in the
stressed treatments, perhaps because of the production of
compounds related to water stress as discussed above. In
contrast, the fine roots of stressed treatments (Treatments
C and D) had a lower nitrogen concentration. Perhaps this
could be due to fine root mortality and the retranslocation
of nitrogen to other seedlings parts.

In Treatments A, B, C, and D, black locust accumulated
44%, 45%, 46%, and 48% of its total nitrogen, and 44%, 45%,
45%, and 47% of its total phosphorus in the root system,
respectively. Thus on a content basis, moisture stress had
no real impact on partitioning of nitrogen to roots.

Phosphorus is associated with metabolic activity as a
source of energy. Phosphorus concentration substantially
increaSed as stress increased in all treatments and within
each tree component. That could indicate a general
decrease in growth rate since percent phosphorus was

negatively correlated to growth, especially in the shoot.

66

Since seedlings had experienced dehydration-rehydration
cycles, more phosphorus was absorbed than utilized due to
the decline in growth during the drying cycles. Phosphorus
was more sensitive to water stress than nitrogen: its
concentration showed a significant difference between all
treatments. Taking all treatments into consideration,
stress increased phosphorus by 30% in the root, 60% in the
shoot, and 30% in the total seedling, compared to the

control.
Nitrogen-Phosphorus-Weight Relationships

Since nitrogen accumulation is a function of growth
(Agren 1985, Ingestad 1982), there was an almost perfect
correlation between nitrogen content and dry weight. The
correlation between dry weight and phosphorus content was
0.95. The correlations between phosphorus concentration
and dry weight were, -.34, -.85, and -.70 for the root,
shoot and the total seedling. On the other hand, there was
no consistent correlation between percent nitrogen and dry
weight. The lack of correlation in the latter was related
to the large variation in percent nitrogen between tree
components.

Since both nitrogen and phosphorus accumulation depend
upon growth rate, the content of both elements was
negatively correlated to the treatments with more

significance in nitrogen than in phosphorus. On the other

67

hand, percent phosphorus was negatively correlated to the
treatments, especially in the shoot, while percent nitrogen
was not correlated to water stress. However, N/P ratio was

highly and negatively correlated to the treatments.
Nitrogen/Phosphorus Ratio

The N/P ratio is an indication of the amount of
nitrogen assimilated relative to the amount phosphorus
absorbed. Knight (1986) demonstrated that Acacia
melaggxylgn seedlings had a shoot/root ratio of 1.2 and N/P
ratio of 11.7, and both ratios decreased as moisture stress
became more pronounced. An N/P ratio of 17.3 for black
locust was calculated using Boring and Swank's (1984) data.
The control treatment (A) had a N/P ratio of 17.0 and the
ratio decreased as stress became more severe mainly due to
the increase in phosphorus. The control of interrupted
stress treatments maintained an N/P ratio of 17.0. In
other words, black locust maintained this ratio for the
whole experimental period.

The N/P ratios in Figure 7, from high to low for tree
components, were: nodules> leaves> large root> fine roots>
stem> rachii. The decreasing ratio with increasing stress
indicates that a high N/P ratio was associated with more
metabolic activity. Nodules were the major source of
nitrogen and also accumulated substantial amounts of

phosphorus due to their high energy (ATP) requirements. In

68

comparing well-watered with stressed treatments, if we
assume that N/P is a measure of metabolic activity it can
be concluded that nodule activity was affected by
decreasing soil matric potential but not as much as the
other seedling parts. In other words, nodules were the
plant part least affected by stress. There was no
significant difference in nodule dry weight among
treatments, neither in nitrogen or phosphorus -
concentration. Nodules of stressed treatments were thicker
and smaller indicating that lack of water affected the
outer part of the nodules. Figure 7 showed that there was
a higher N/P ratio in the fine roots than in the large
roots in the stressed treatments. Fine roots showed a
significant decrease in nitrogen concentration due to the

stress but a slight increase in phosphorus.
Interruption of Stress

Stress was interrupted as individual blocks were
harvested. For the first block, the seedlings were allowed
to grow without stress for 2.5 months, while the stress was
released for only two months in block four. Nitrogen
concentration decreased in leaves and rachii but increased
in the stem and the underground parts. The trend held in
all treatments. It was probably due to the onset of
shorter days and related to the retranslocation sequence

typical of dormancy. Comparing the control and both

69

stressed and interrupted stress treatments, it became
apparent that large roots were a storage site for nitrogen.
The main root apparently acted as a reservoir, especially
in the fall as the N/P ratio increased with time of
harvest. Roots of many tree species have been documented
to serve as storage sites for nutrients (Mutah 1972).

The N/P ratio for stress-interrupted treatments
increased after normal watering resumed. The greater the
previous stress, the more the increase in the N/P ratio.
This ratio showed a significant difference between
treatments. The control ratio did not change over the two
month growing period. Thus, it appears that the ratio of
N/P is responsive to metabolic changes induced by moisture

stress and its alleviation.

SUMMARY AND CONCLUSIONS

1. Water stress affected all tree components with the
exception of nodule dry weight. Treatments -.20 bars, -.50
bars, -.90 bars and -2.0 bars accumulated 61%, 60%, 56% and
54% of their total dry weight in the shoot, respectively.
The percentage of total dry weight allocated to fine roots
was, 17%, 17.5% and 25%, for the control, the moderately
stressed and the severly stressed treatments, respectively.
In general, stress resulted in a reduction of 40%, 72% and
74% of the total dry weight when comparing treatments -.50
bars, -.90 bars and -2.0 bars. As stress increased there
was a major reduction in leaf area. Height and diameter
were least affected by moisture stress. Height decreased
by 15%, 25% and 29% for treatments -.50 bars, -.90 bars and
-2.0 bars, compared to the control.

2. Interruption of stress resulted in more root than shoot
growth. Such growth was proportional to the after-effect
of stress experienced by seedlings. The more stress the
seedlings were exposed to, the less recovery they
exhibited.

3. Nitrogen concentration increased in the leaves and

decreased in the fine roots when trees were stressed. The

70

71

highest percent nitrogen was in the nodules followed by the
leaves, then the large roots.

4. There was an almost perfect correlation between nitrogen
and phosphorus accumulated and total dry weight. Nitrogen
concentration was not correlated to the dry weight, while
P% was significantly correlated to the shoot dry weight,
(-.85), and to the total seedling dry weight (-.70). Both
nitrogen and phosphorus were highly correlated (0.97). N%
was not correlated to the treatments in contrast to P%.
However, their ratio (N/P) was highly and negatively
correlated to the treatments.

5. N/P ratio decreased markedly as soil matric potential
decreased. The highest ratio was in the nodules followed
by the leaves, while stems and rachii maintained a lower
N/P ratio. Fine roots had a higher ratio than large roots,
but this trend was reversed in the control. Black locust
control treatments maintained a ratio of 17.0 throughout
the experiment.

6. Although height and diameter recovered after
rehydration cycles, wood volume and density were
permanently affected by stress.

7. _Black locust exhibited exponential growth after two
months of sowing date and grew almost 2 cm/day.

8. Black locust exhibited considerable morphological
plasticity in response to moisture stress. Shedding of

leaves and a relative increase in fine root production

72

occurred following stress.

 

BI BLI OGRAPHY

BIBLIOGRAPHY

Aiyelaagbe, I.O., M.O. Fawusi and O. Babalola. 1986.
Growth, development, and yield of pawpaw, gaziga
pgp§y§(L.) 'homstead solution' in response to soil
moisture stress. Plant Physiol. 93:427-435

Agren, G.I. 1985. Theory for the growth of plants
derived from the nitrogen productivity concept.
Physiol. Plant. 64:17-28.

Aspinall, D. and L.G. Paleg. 1981. Proline accumulation:
physiological aspects. In : L.G. Pelag and D.
Aspinall, eds. The Physiology and Biochemistry of
Drought Resistance in Plants. Academic Press N.Y. pp
206-242.

Bhagsari, A.S. and R.H. Brown. 1986. Leaf photosynthesis
and its correlation with leaf area. Crop Sci. 26:
127-132.

Boring, L.R., and W.T. Swank. 1984. The role of black

locust Egbinig pseudgggagig (L.) in forest succession.
J. Ecoology 72:749-766.

Boring, L.R. C.D. Monk and W.T. Swank. 1984. Symbiotic
Nitrogen fixation in regenerating black locust 3921313
pgeggggagia (L.) stands. Forest Sci. 30(2):528-537.

Bowen,G.D. 1984. Tree roots and use of soil nutrients.;n:
G. Bowen and E. Nambiar eds. Nutrients of Plantation
Forest. Academic Press, London. pp 147-154.

Brandle, J.R., P.A. Schnare T.M. Hinkley anf G.N. Brown.
1973. Changes of polysomes of black locust seedlings
during dehydration-rehydration cycles. Physiol.
Plant. 29:406-409.

73

74

Brandle, J.R., T. Hinckley and G. Brown. 1977. The
effects of dehydration-rehydration cycles on protein
synthesis of black locust seedlings. Physiol. Plant.

40:1-5.

Clarke, J.M. and R.C. Durkey. 1981. The responses of
plants to drought stress. In : G.M. Simpson, ed.
Effects of Water Stress on Plants. Praeger N.Y. pp
89-140.

Dawson, J.O. and J.L. Gordon. 1979. Nitrogen fixation in
relation to photosynthesis in A; glutingga. Bot. Gaz.
140:570-575.

Dickmann, D. I., K. S. Pregitzer, and P. V. Nguyen. 1989.
Net assimilation and photosynthate allocation of
ngglgg clones grown under short-rotation intensive
culture. Annual Progress Report, Department of Energy

Subcontract No. 86x-95903c. 80 p.

Evans, P.S. and J.E. Klett. 1984. The effects of dormant
pruning treatments on leaf, shoot and root production
from bare-rooted Malus garggntii . J. Arbor.
lO(1l):298-301.

Glass, A.D. 1989. Plant nutrition: An introduction to
current concepts. Jones and Barlett, Boston.
pp 103-122, 203-230.

Hasegawa, P.M., R.A. Bressan, S. Handa and A. Handa. 1984.
Cellular Mechanisms of tolerance to water stress.
Hort. Sci. 19(3):371-376.

Havranek, W. and V. Benecke. 1978. The influence of soil
moisture on the water potential,transpiration, and
photosynthesis on conifer seedlings. Plant and Soil
44:91-103.

Hesketh, J.D., W.L. Ogren, M.E., Hageman and D.P.Peters.
1981. Correlations among leaf COZ-exchange rates,
areas and enzyme activities among soybean cultivars.
Photosynth. Res. 2:21-30.

75

Hinkley, T. 1973. Responses of black locust and tomato
plants after water stress. Hort. Sci. 8(5):405-407.

Hsiao, T.C. 1973. Plant responses to water stress. Ann.
Rev. Plant physiol. 24:519-570.

Huang, C., J.S. Boyer and L.N. Vanderhoef. 1975.
Limitation of acetylene reduction (nitrogen fixation)
by photosynthesis in soybean having low water
potential. Plant Physiol. 56:228-232.

Ingestad, T. 1982. Relative addition rate, and external
concentration; driving variables used in plant
nutrition research. Plant, Cell and Environment 5:
443-453.

Jarvis, M.S. 1963. A comparison between the water
relations of species with contrasting types of
geographical distribution in the British Isles. In :
F.H. Whitehead and A.J. Rutter eds. The Water
Relations of Plants. Black Well, Oxford. pp 289-312.

Knight, P.J. 1986. Phosphorus and sulfur requirements of

blackwood Acacia malanczylcn seedlings.
Communications In Soil Science And Plant Analysis.

17(10):1121-1144.

Kozlowski, T.T. 1987. Soil moisture and absorption of
water by tree roots. J. Arbor. 13(2):39-46.

Kramer, P.J. 1969. Water relationships, a modern synthesis.
McGraw Hill N.Y. p 482.

. 1987. The role of water stress in tree
growth. J. Arbor. 13(2):33-38.

 

Kummerow, J. 1980. Adaptation of roots in water-stressed
native vegetation. In : S.C. Turner and P.J. Kramer
eds. Adaptation of Plants to Water and High
Temperature Stress. Wiley Interscience Publications
John Wiley and Sons N.Y. pp 57-73.

76

Larson, M. and I. Patashev. 1973. Effects of osmotic
water stress and gibberellic acid on initial growth of
oak seedlings. Can. J. For. Res. 3:75-82.

McNiel, R.E. 1975. Factors influencing nitrogen fixation
by several woody shrubs and trees. Ph.D.
dissertation. Department of Horticulture,Purdue
University. 135 p.

Mutoh, N. 1972. Further studies on the phospggrus economy
of the larch tree seedling by the use of . Jpn.
J. Botany. 20:339-362.

Pallas, J.E., B.E. Michel and D.G. Harris. 1967.
Photosynthesis, transpiration, leaf temperature, and
stomatal conductivity of cotton plants under varying
water potential. Plant Physiol. 42:76-88.

Rawlins, S.L., W.R. Gardner and F.N. Dalton. 1968.13
alga measurement of soil and plant leaf water
potential. Soil Sci. Am. Proc. 32:468-470.

Reinsvold, R.J. and P.E. Pope. 1987. Combined effect of
soil nitrogen and phosphorus on nodulation and growth

of 8551515 555555555515 (L-)- Can- J- For- Ree-
17:964-969.

Roberts, B.R. 1987. Methods for measuring water status
and reducing transpirational water loss in trees.
Hort. Sci. 13(2):56-61.

Roberts, D.R., R.W. Zimmerman J.W. Stringer and 8.8.
Carpenter. 1983. The effect of combined nitrogen on
growth, nodulation and nitrogen fixation of black
locust seedlings. Can. J. For. Res. 13:1252-1254.

Rueggs, J.J. and A.M. Alston. 1978. Seasonal and diurnal
variation of nitrogenase activity (acetylene
reduction) in Barrel Medic Magicacc
(Gaertn.) grown in pots. Aust. J. Agr. Res. 29:
951-962.

77

Sands, K. and A. Rutter. 1959. Studies in the growth
young plants of Einaa §¥1¥5§£Ii§5115 The relationship
of growth to soil moisture tension. Annals of Botany
23(90):269-283.

Schulte, P. and P. Marshall. 1983. Growth and water
relations of black locust and pine seedlings exposed
to controlled water stress. Can. J. For. Res. 13:
334-338.

Sharp, R. and W. Davies. 1979. Solute regulation and
growth by root and shoots of water-stressed
maize. Planta. 147:43-49.

Simpson, G.M. 1981. Effects of water stress on plants.
Praeger, N.Y. 324 p.

Slavik, B. 1979. Methods of studying plant-water
relations. Spring-Verlag, New York. pp 157-218.

Spomer, L.A. 1985. Techniques for measuring plant water.
Hortscience. 20(6):1021-1028.

Sprent, J. 1984. Nitrogen fixation. In : M. Wilkins
ed. Advanced Plant Physiology. Pitman Bath England.
pp 249-273.

. and A.E. Gallacher. 1978. Anaerobiosis in
soybean root nodules under water stress. Soil. Biol.
Biochem. 8:317-320

 

Turner, N.C. and P.J. Kramer. 1980. Adaptation of plants
to water and high temperature stress. John Wiley and
Sons. 482 p.

Taylor, B.K. 1967. Storage and mobilization of nitrogen
in fruit trees: a review. J. Aust. Inst. Agr. Sci.
33(1):23-29.

78

Zimmerman, R.W., D.R. Roberts and J.W. Carpenter. 1982.
Nitrogen fertilization effects on black locust
seedlings examined by acetylene reduction assay. In
B.A. Thielges ed. Proceeding of Seventh North

American Forest Biology Workshop, Lexington, KY. pp
341-349.

"I11111411411141.1111T