65- 12,187

JLEMON, E d g a r R o th w ell, 1 9 2 1 SOIL AERATIO N AND ITS CHARACTERIZATION.
M ich igan S tate U n iv e r s ity , P h .D ., 1952
A g r ic u ltu r e , s o i l s c ie n c e

U n iv e rs ity M icro film s, Inc., A n n A rb o r, M ic h ig a n

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w ith o u t perm ission.

SOIL AERATION AND ITS CHARACTERIZATION

By
E d g a r Rothwell L e m o n

A THESIS
S u b m i t t e d to the School of G r a d u a t e Studies

of M i c h i g a n

Sta te Co l l e g e of A g r i c u l t u r e and A p p l i e d Science
in pa r t i a l f u l f ill ment of the r e q u i r e m e n t s
for the degree of
D O C T O R OF P H I L O S O P H Y

D e p a r t m e n t of Soil Scien ce

1952

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

a-

BIOG-RAPHICAL SKETCH
The author was horn August 22, 1921 In Buffalo,
New York, the son of &retta and A. Bertram Lemon,

He

attended elementary and high school In Buffalo.
From 1940 to 1943 he attended Cornell University,
receiving the bachelor of science degree in agriculture
in 1943.
During World War II he served from 1943 to 1946 as
a deck officer in the United States Navy,
From 1946 to 194? the author and his wife Donna,
operated a dairy farm at Oakfield, New York.
He re-entered Cornell University in September, 194?
and was awarded the master of science degree in agriculture
in September, 1948.

During his graduate studies at Cornell

University he held a Research Assistantship, with major
interest in Soil Physics.
Since September, 1948 he has been a graduate student
at Michigan State College, holding a Teaching Assistantshlp
in Soil Science until September, 1950.

From September, 1950

to July, 1951 the author held a Graduate Council Fellowship.
During his graduate work at Michigan State College he
majored in the study of Soil Physics.
11

SO S'IX S'
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

SOIL AERATION AND ITS CHARACTERIZATION

By
Edgar Rothwell Lemon

AN ABSTRACT
Submitted to the School of Graduate Studies of Michigan
State College of Agriculture and Applied Science
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY

Department of Soil Science
Year

ApproVed

1952

j.. )?7.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

SOIL AERATION AND ITS CHARACTERIZATION
By
Edgar Rothwell Lemon
ABSTRACT
An investigation was made of the single physical pheno­
menon of soil aeration with special emphasis on the methods
used In the past to characterize soil aeration.

Preliminary

investigations of these older methods substantiated the
belief that some new methods were sorely needed to measure
quantitatively the aeration status In the root zone of the
soil.
The possibility of measuring oxygen diffusion potentiometrically based upon the principle of oxidation of the chromic
chromous system was investigated with little success.
Another method of measuring oxygen diffusion, based upon
the electrolysis of oxygen at the surface of a platinum mlcroelectrode with an applied voltage betvfeen the electrodes, gives
promise of furnishing a new, simple, rapid and inexpensive
method of determining the rate of oxygen supply to an
environment similar to that in the liquid film surrounding
an actively respiring plant root.
Evidence has been presented.indicating that the factors
controlling the diffusion rate in the gaseous phase of the
soil extend also into the liquid phase.

There is evidence

that at lower porosities moisture film thickness is of greater
importance in controlling oxygen supply to the root surfaces
than at higher porosities.
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

ACKNOVn^EDGMENT
Th.e author wishes to express his sincere appreciation
to Professor A. E. Erickson for directing his graduate work
at Michigan State College.

His willing assistance and

constant readiness to discuss questions that arose In
research have been of great value In the completion of the
doctorate problem.

Thanks are due to Professor L, M. Turk

for his kindly counsel as head of the author's special
Guidance Committee.
Probably no one person Is more responsible for the
author's completion of the graduate program than his wife,
Donna Lemon.

Her encouragement and devotion during the

author's Illness, while completing the doctorate work, has
been of Inestimable value.

111

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w ith o u t perm ission.

TABLE OF CONTENTS
I.
II.

I N T RO DUC TIO N

L I T E R A T U R E REVI EW

A.
B.
C*

D,
III.

The Soil A t m o s p h è r e ...............
Processes of G-aseous Interchange
Operating in the S o i l ............
Methods of Characterizing Soil Aeration
The composition of the soil atmosphere
Methods based on physical-chemical
properties of the soil . . . .
Methods based on gaseous interchange
A bioelectrical method ............
Summary of Findings in the Literature

2
5
10
11
12
14
17
1?

E X P E R I M E N T A L ...........................

19

A.

19
19
22
34

B.

C.

D.

IV.

.

A Preliminary Greenhouse Experiment .
Experimental methods and design . ,
Discussion of results ............
Summary of r e s u l t s ...............
A Potentiometric Method of Measuring
Oxygen Diffusion in the Soil . ..
.
Theoretical considerations
. . .
Experimental methods and discussion
of r e s u l t s .....................
Summary of r e s u l t s ...............
An Amperometrie Method of Measuring
Oxygen Diffusion in the Soil . . .
Theoretical consideration . . . .
Discussion of laboratory experimental
methods and r e s u l t s ............
A Second Greenhouse Experiment . . .
Experimental methods and design , .
Discussion of results ............
Summary of results . . . . . .

LOOKING AHEAD

36
36
38
57
58
59
66
72
72
83
96

........................

100

LITERATURE CITED ........................

102

A P P E N D I X .....................

107

1.

Tabulated results of measurements
made during a preliminary green­
house experiment, with Analyses
of V a r i a n c e ..................... 107

R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.

TABLE OF CONTENTS (Continued)
APPENDIX
2.

Tabulated results of measurements
made during a second greenhouse
experiment, with Analyses of
V a r i a n c e ........................... II3

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

I.

INTR ODU CTI ON

P l an t growth. Is dependent upo n the rates at w h i c h
the environment can supply nutrients and energy to the
organism a n d remove certain byproducts.

In the soil these

factors are controlled largely b y four physical phenomena:
temperature, moi s t u r e content, a era tio n and resistance to
root penetration.

I n a practical

sense,

it has l o n g been

k n o w n that these factors have to be integrated in the
p r oper w a y to give m aximum plant growth.
however,

in a fundamental way,

L i t t l e is known,

of how these factors o p e r ­

ate integrally or independently.

Thi s is due p r i m a r i l y

to the complex and dynamic properties of the
This inves tig ati on was

soil.

started to study the single

p h e n o m e n o n of soil ae ration with special

emphasis on the

methods used in the past to characterize soil aeration.
A

new m e t h o d is p r o p o s e d to measure the rate at w h i c h

oxygen is supplied to the root environment in the s o i l .

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission.

2.

II.

LITERATURE REVIEW

Rather complete reviews of the recent literature on
the relation of aeration and plant growth have been given
by Russell

(4l) and Taylor (4?).

Suffice to say, soil

aeration Is very Important to plant growth.

This review

will be limited to the mechanistic aspect of soil aeration
and the methods used In the past to characterize soil
aeration*
A*

The Soil Atmosphere

The composition of the soil atmosphere Is the
result of two major processes acting simultaneously;
1.

Oxygen Is used and carbon dioxide given off by the

biological activity In the soil,

2.

At the same time,

diffusion and other processes of gaseous Interchange are
constantly replenishing the soil atmosphere from the
atmosphere above the soil.

As a net result, the composi­

tion of the soil air at any one time Is determined by the
difference In the velocity of these two processes.
The first of these processes, the absorption of
oxygen and the evolution of carbon dioxide as a result
of biological respiration In the soil has been ably
Investigated by Russell and Appleyard (40).

They found

that the carbon dioxide content and the number of bacteria

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

3.
In the Boll fluctuated In the same way seasonally, with
two maxima, one In late spring, and another In early fall,
with minima In winter and summer.

The fluctuations are

due to such seasonal factors as temperature and rainfall.
The amplitude of these fluctuations Is Influenced by the
type and quantity of organic matter, cultivation and species
of plants growing on the soil.

Boynton and Reuther (5 )

also found that the oxygen and carbon dioxide varied sea­
sonally to a marked extent.

Most of the earlier workers

have established the following characteristics of the soil
atmosphere:
1.

The sum of the carbon dioxide and oxygen Is

very near that of the atmosphere.
2.

The carbon dioxide content Increases with depth,

while the oxygen content decreases.
3.

The carbon dioxide content depends upon organic

matter content and porosity.
4.

Rainfall and temperature exhibit a marked Influ­

ence on the gases In the soil air.
5.

The quantity of these gases vary markedly with

the season.
Russell and Appleyard (4o) found that this free air
filling the pore space was not the only air In the soil.
Upon evacuation of all the free air from the pore spaces
they observed a gas slowly being evolved from the soil

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

4.
for several days.

Its characteristic feature was a very

low concentration of oxygen and a very high proportion of
carbon dioxide.

The volume of obtainable gas depended on

the amount of moisture present In the soli. Indicating
that the gas was partially dissolved In the soil solution
although part probably was adsorbed on the soil colloids.
On this matter the present writer has stated (25):
“There Is little doubt that the free soil air
and that dissolved or adsorbed air are at
equilibrium. Also, It Is probably true that
plants grow In closer association with the
dissolved or adsorbed air than with the free
soil air. In that case, a slight change of
the oxygen and carbon dioxide concentration
of the free air due to Inhibited aeration
may deplete the oxygen concentration in the
dissolved phase enough to become critical.
Too little attention has been paid to the
oxygen and carbon dioxide concentrations as
a function of distance from the root surface,
especially close to the root surface.“
This possibility of two atmospheres In equilibrium
In the soil permits the operation of aerobic and anaerobic
systems along side each other in the soil, according to
Russell and Appleyard (40).

When the aerobic system Is

dominant, nutrient balance Is maintained, but when the
anaerobic system Is dominant, continued growth of certain
organisms becomes Inhibited and those that survive are
forced to use the various forms of combined oxygen present
or reduce certain multlvalent elements.

Many of the

reduced substances found In the soil are supposed to be

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

5.
toxic to plants In low concentrations (6) such as, phenols,
complex aldehydes, nitrites, hydrogen sulfide, sulfites.
Iron and manganese.

So far little is known about the toxi­

city of these compounds as they exist in the soil.
In addition to toxic compounds formed under reducing
conditions, some elements, nitrogen and phosphorus in
particular, are thought to be made less available for
plant growth,

Sturgis (^5) states that low oxygen concen­

tration in the soil slows up the decomposition processes
of bacteria which liberate
matter.

nitrogen from the soil organic

The accumulated reduced substances cause the

soluble phosphates to be changed to less available forms
particularly in alkaline soils.
Poor aeration also increases the availability of some
elements that are necessary for plant growth.
B*

Processes of Gaseous Interchange in the Soil

- I n the above section, consideration has been given
to processes that reduce the oxygen content of the soil and
increase the carbon dioxide content.

Some indirect effects

of gaseous composition and biological activity were dis­
cussed briefly.
Of equal importance in determining the composition of
the soil atmosphere are those processes that tend to main­
tain the oxygen and carbon dioxide at the levels of

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

6,
concentration found in the atmosphere above the soil.
The interchange of gases between the soil air and the
atmosphere is the result of two distinct mechanisms.

The

first is viscous or mass flow through the pores of the soil
which is a function of the pressure and volume relation­
ships.

The second is gaseous diffusion.
Total pressure differences may result from changes in

temperature, barometric pressure changes, rainfall and wind
action.
The effect of temperature according to Romell, as
quoted by Keen (20), is of little significance in as much
as the volume of soil gases changes as the absolute temp­
erature and wide temperature fluctuations are confined to
shallow depths.
Both Romell (38,39) and Buckingham (8) have made
calculations indicating that changes in barometric pressure
can be discounted as important factors influencing the flow
of air into or out of the soil.
The uneven surface and presence of vegetation on the
soil greatly decreases the velocity of wind over the soil
surface.

This prevents wind action from being an effec­

tive agent in causing transfer of air between the soil and
the atmosphere.
Rainfall or Irrigation displaces an equivalent volume
of soil air drawing air behind It and releasing dissolved

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

7.
oxygen.

However, the debrease In effective pore space due

to increased moisture and the intermittent nature of rain­
fall in the temperate region discounts this factor as an
Important one in soil air renewal.
The process most important in causing an interchange
of gases between the soil air and the atmosphere above the
soil is that of diffusion (8,20,32,37*38,39).

This process

does not depend upon viscous flow, but only upon the diff­
erence in partial pressure of the gas or gases in question.
Concentration gradients or differences in partial pressures
always exist in normal soils, thus diffusion is a continuous
process.

As stated before, these concentration gradients

are due to the using up of oxygen and the evolution of
carbon dioxide in the respiration processes of biological
materials usually present in the soil.
Hannen (l4) was one of the earlier workers to study
diffusion as such in soil.

He artificially packed soil

into ten-inch cylinders below which he placed a chamber of
carbon dioxide.

After a ten hour period of diffusion he

analysed the gas in the chamber.

Blake*s translation (l)

of his conclusions are quoted below:
1.

“The diffusion of carbon dioxide from the soil,
at constant temperature, depends principally
on the sum of the cross section of the pores.
Therefore, the larger the total pore space, the
larger the absolute amount of diffused gas and
vice versa.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

8.
2.

"Every decrease in pore space such as that
effected by a more or less high moisture con­
tent results in a decrease in the quantity of
gas diffusing. Therefore, the finer grained
the soil, the more compact the soil components,
the greater will be the water capacity of the
soil, and the lesser the amount of carbon
dioxide that is given up to the atmosphere
by diffusion and vice versa.

3.

"The diffused carbon dioxide decreases directly
as the depth of the soil column, but the de­
crease is not proportional to the height of the
column, but is relatively less (as the height
of the soil column increases).

4.

"The soil types which become saturated by atmos­
pheric water, and into which rain infiltrates
very slowly, the diffusion of carbon dioxide is
more or less considerably reduced."

Buckingham (8 ) was one of the first American workers
to evaluate the Importance of diffusion processes in rela­
tion to soil aeration.

He measured diffusion rates by

running streams of air and of carbon dioxide across oppo­
site ends of a one-inch column of soil and noting the
resulting dilution of the two.

Buckingham concluded that ^

the rate of diffusion was proportional to the square of the
free pore space; being dependent upon texture, structure
and moisture content only so far as these factors affect
the free pore space.
Smith and Brown (42) attempted carbon dioxide diffu­
sion measurements in the field but concluded that accurate
determinations of the rate of diffusion could not be made
because of complications arising from the production of

R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w ith o u t perm ission.

9.
carbon dioxide in the soil Itself.

With alr-drled soil

packed into a cylinder these authors found diffusion of
carbon dioxide through the alr-drled soli to be a linear
function of porosity between the range of 36 to 65 percent
porosity.
Hagan (I3) In California and Penman (32) In England
each independently started to use carbon disulfide vapors
In the study of diffusion through soils.

Both Investiga­

tors used disturbed samples packed in soil columns of
varying porosity.

Hagan controlled porosity by adding

moisture and found that diffusion was a linear function
of porosity but approached zero not at zero porosity, but
In the porosity range of 26 to 29 percent.

Penman found

that diffusion was a linear function of porosity* between
15 and 60 percent.

A backward extrapolation of his curve

brings It to the origin.

Penman stated, however, that In

these lower ranges of porosity the relation becomes less
reliable as the pore space In the soil Is decreased by
wetting.

Small air pockets are formed that are isolated

and Ineffective as diffusion channels, but still contribute
to the pore space.

This could account for zero diffusion

at porosities greater than zero.
Penman (32) has suggested that Buckingham didn't
obtain a linear relationship of diffusion to porosity
because his data were taken before a steady state had been
reached.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w ith o u t perm ission.

10.
Very recently a renewed interest in soil aeration
and the diffusion processes has been fostered by Haney (3^)
and Taylor (^6 ) working under M, B. Russell at Cornell
University and by Blake (2) working under J. B. Page at
Ohio State University.

Raney and Taylor using oxygen and

Blake using carbon disulfide have all substantiated the
previous findings that diffusion is a linear function of
porosity in the soil, at least for porosities between 20
and 50 percent.

Below 20 percent porosity Raney (35) found

a curvilinear relationship when sand was wetted.

This

finding substantiated the predictions of Penman (32).
Taylor (46) and Blake (l) have shown that the elope of the
linear relationship of diffusion to porosity differs among
soils.

Blake (l) has suggested that these differences in

slope are due to differences In soil structure.
C.

Methods of Characterizing Soil Aeration

Past methods used in the characterization of soil
aeration can be divided into four general groups:

1, The

determination of oxygen and/or carbon dioxide in samples
of gas extracted from the soil;

2.

Methods based on

physioal-chemloal properties thought to be dependent upon
the aeration status of the soil;
gaseous Interchange;

3.

Methods based on

4. A bioelectrical method.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

11.
The composition of the soil atmosphere. The measure­
ment s of carbon dioxide and oxygen In the atmosphere were
the first methods of characterizing soil aeration.

The

determinations of carbon dioxide and oxygen Were made as
èarly as 1852 by Bousslngault and Lewy (4),

Since that

time many Investigators have determined the composition of
gases aspired from the soil.

Russell and Appleyard (40)

In 1915 made extensive studies throughout the year of the
fluctuations of oxygen and carbon dioxide in the soil
atmosphere.

Their work and, more recently, the studies of

Boynton and Reuther (5 ) have well established the charac­
teristics of the soil air.
The main objection to this method of evaluating the
aeration status Is that the source of the gas sample la not
known.

The gas drawn from the soil may be taken from many

pores In a small volume of soil or may be drawn from a few
pores extending Into a large volume of soil.

This measure­

ment then, gives no Idea of the amount of oxygen available
for plant growth in a given volume of soil.
Leather (24) overcame this objection by removing a
known volume of soil and determining the composition of
the entrapped gases.

Karsten (19) removed a known volume

of soil from the ground and placed It In a previously de­
gassed solution.

He then determined the oxygen content by

a polarographlc method using a dropping mercury electrode.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

12.
Leather (24), Rueeell and Appleyard (4o) and Lundegardh (27) all have recognized the Importance of the
dissolved gases In the soil solution, but no one has made
any measurements of these gases.

Karsten did measure the

oxygen In the liquid phase along with the oxygen In the
gaseous phase, however.

g

Methods based on physical chemical properties of
the soil. A number of methods of evaluating the aeration
status based on physical chemical properties of the soil
have been used.

Vine, Thompson and Hardy (49,50) and

Hardy (15) placed bright strips of iron In the soil.
After a period of time the quantity of rust which had
formed was used as an Index of aeration.

Hoffer (l6) has

used the quantity of ferrous iron present In the soil as
an index of aeration.

Both of these methods are strongly

dependent upon moisture and temperature.

The reduction of

ferric iron to ferrous is dependent upon biological acti­
vity.

Since aeration is only one of the factors affecting

biological activity. It Is questionable that a measure of
a biochemical end product Is a good criterion of aeration.
Oxldatlon-reductlon potentials (redox potentials)
have been used In situ and on soil samples to characterize
soil aeration (6,31,53,43).

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

13

.

Peach, et. al. (31) have summarized the results of
these investigations thus:
"The oxidation-reduction potential is a measure of
the immediate intensity of reduction, and is con­
trolled largely by the dissolved oxygen in the soil
solution. Upon depletion of this dissolved oxygen,
the micro-organisms will utilize the combined forms
of oxygen in oxidizing the organic matter and, at
the same time, reducing various inorganic compounds
such as ferric oxide, manganese oxide, and nitrates
which are present in the soil. The intensity
factor of reduction is governed by the supply of
free oxygen while the quantity of iron, manganese,
and other compounds reduced and brought into solu­
tion will be primarily determined by the amount of
readily oxidizable organic matter in the soil,"
Direct redox potential measurements give a measure
of the intensity factor.

An idea of the capacity factor

has been obtained from potentiometric titrations of soil
samples with an oxidizing agent such as potassium perman­
ganate (43).

Taylor (4?) concluded from the literature

that, in general, quite drastic conditions must prevail
before present day methods of measuring redox potentials
can detect reducing conditions.

Plants may suffer from

lack of oxygen before these conditions prevail in the
soil.
The use of such physical chemical methods as reviewed <5^
above have failed to relate closely or consistently to
plant response.

This may be due to a lack of sensitivity

in present day methods, their static nature and their
dependency upon so many factors other than oxygen supply
alone.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

14
Méthode based on praseous Interohartge.

.

The rate of

loss of carbon dioxide from the soil has been suggested as
a means of evaluating soil aeration (17,27,29,37,38).

It

has been assumed that the diffusion of carbon dioxide out
of the soil has an Inverse relation to the amount of oxy­
gen diffusing into the soil.

This assumption may not be

sound, however, since anaerobic respiration may account for
considerable carbon dioxide produced In the soil.

Field

measurements are of questionable value since the volume of
soil Involved Is unknown.

Laboratory measurements with a

known volume of soil are not valid either since the radical
change In environment greatly affects the carbon dioxide
production.
Klrkham (21) has developed a method of measuring the
permeability of soil to viscous flow of air both In situ
and In the laboratory,

Blake (l) has pointed out that many

early studies In Europe have dealt with the characterization
of soils on the basis of their air permeeblllty.

Romell (36 ),

however, has stated that when one seeks to use permeability
of a soil to air under pressure as a measure of aeration,
one does so with the hypothesis that mass flow plays the
principal role In soil.air renewal and that diffusion Is
I
secondary.

Since air permeability varies with the method

by which It Is determined, It differs In Its nature from
diffusion and does not take Into account the effects of

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

15.
biological activity.

It is evident that these measurements

are of limited value in making inferences regarding the
aeration status of the soil,
Blake and Page (2 ) have adapted Penman's (32) method
of measuring carbon disulfide vapor diffusion to field use.
A porous cup with a known amount of liquid carbon disulfide
was placed in a hole bored into the soil.

The hole was

plugged and at the end of three hours the cup was removed
and weighed.

The loss of carbon disulfide was used to

evaluate the rate of diffusion of the gas through the soil.
The solubility of carbon disulfide in the soil water
is appreciable.

This may cause an error in determining

true diffusion.

The additional fact that the effect of

biological activity is not accounted for in this measure­
ment limits the value of the method for evsJ-uating soil
aeration.
A method that will measure the rate at which oxygen
will diffuse into the soil and at the same time will inte­
grate the reducing effects of biological respiration would
seem to offer the most promise of successfully evaluating
soil aeration conditions. Cannon and Free (9 ) have stated
that the rate of supply of oxygeh to the roots is of
fundamental importance in plant growth.

Hutchins (18) in

1926 recognized the importance of these principles.

He

devised a method of measuring the oxygen-supplying power

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

16.
to an absorber burled in the soil.

Presumably only a few

measurements were made because of the bulky apparatus and
the difficulty of making measurements.
It was not until 19^9» however, that these principles
were again applied to the characterization of soil aeration,
Taylor (46) then developed the theory and necessary mathe­
matical calculations to enable Raney (34) to successfully
make field measurements with an oxygen deficient absorber
much simpler In operation than Hutchins'

(18).

A closed metal cylinder was burled In the soil at the
desired depth.

This cylinder which had been flushed free

of oxygen with nitrogen gas was then opened at the bottom
to the soil atmosphere and diffusion allowed to proceed
for thirty minutes.

At the end of this period, the ports

In the bottom of the cylinder were again closed, the gases
Inside thoroughly mixed and then analyzed for the amount of
oxygen present.

The relative amount of oxygen that had

diffused Into the cylinder was then taken as a measure of
the oxygen-supplying power at that particular point In the
soil.

Although this method Is based on sound principles,

there are a few limitations.

The root environment Is dis­

turbed considerably when the probe (diffusion cylinder) is
Inserted Into the soil, the method Is costly and slow and
the equipment heavy and cumbersome.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission.

17.
A "bioelectrlcal method.

It Is believed that the

oharacterlzatlon of the conditions at the Interface
between the root surface and the soil system offers the
greatest' possibility of ascertaining the Influence of
soil aeration on plant growth.

Because of this, measure­

ment of an Inter-dependent system In which both the soil
and plant are constituent parts seems to offer great
possibilities.

The electrical potential developed between

the root and Its growth medium is such a system (3 »30,^7 >^8).
It has been shown that these bioelectric potentials respond
to changes In the aeration status of the growth medium.
However, careful studies made by Taylor (4?) have indicated
that bioelectric potentials are not solely dependent upon
aeration, but are strongly Influenced by environmental and
genetic factors that control cell respiration.

Since

aeration is only one of the factors Influencing cell
respiration, the utility of such measurements to characterize
soil aeration Is at present not in sight,
D.
1.

Summary of Findings in the Literature

The amount of oxygen found In soil gases Is the

resultant of two sets of forces, namely, those biological
respiration processes that tend to reduce the oxygen
content and those processes that tend to maintain the oxy­
gen at the level found in the atmospheric air.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

18.
2.

There le little or no Information concerning

the composition, amount, or Influence of gases in the
liquid phase of the soil,
3.

The physical process most Important In main-

Î’

talnlng Interchange of gases between the soil and the
atmosphere la diffusion.
4.

Diffusion Is a continuous process dependent

upon a concentration gradient and has a linear relation­
ship to water free pore space In the soil, at least, above
20 percent free pore space,
5.

The composition of a sample of aspirated gas from

the soil may be a poor estimate of aeration conditions
close to the. Immediate root environment,
6.

Most physical-chemical methods of characterizing

soil aeration are dependent upon biological activity.
Since aeration Is only one of the factors Influencing this
activity. It Is questionable whether these methods are
valid

ones.
7.

Methods of measuring the oxygen diffusion rate

to a point In the soil give promise of truly characterizing
aeration.

Present day methods, however, have Inherent

objectionable features,
8.

The use of the bioelectric potential developed

between the root cells and the soil as a-measure of aeration
has too many limitations to be of value at present.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

19.

III.
A.

EXPERIMENTAL

Preliminary Q-reenhouse Experiment

Experimental methods and design.

A greenhouse

experiment was initiated In the spring of 1949 to test
three different methods of characterizing soil aeration
and to investigate the possible deleterious effects of
ferrous and manganous ions on plant growth under reduced
aeration in the soil.
A well-aggregated Brookston clay loam soil was taken
from a field that had grown alfalfa for the past two years.
The soil was air dried, and then screened into three ranges
of aggregate sizes:
<3.2 mm.

12.8 to 6.4 mm., 6.4.to 3,2 mm., and

An 8-8-8 commercial fertilizer was mixed with

each aggregate separation at the rate of 10 tone per acre.
Ferrous sulfate and manganous sulfate were mixed with the
appropriate aggregate ranges at the rate of 10 tons per
acre.

The soil aggregates were then uniformly packed into

thirty-six tile pots 18 inches deep, with an Inside diameter
of 6.5 inches, that were painted inside and out with black
tygon plastic paint.
The experiment consisted of four replications of
three levels of aeration as determined by aggregate size
with three treatments; a check, an added increment of iron

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w ith o u t perm ission.

20.
and an added Increment of manganese.
Lincoln soybean seeds were planted in each pot.
Later the plants were thinned to four per pot.

Soybeans

were chosen because of their sensitivity to an excess of
ferrous or manganous ions.
The pots were irrigated every day to their approximate
original weight of thirty percent moisture.

At this mois­

ture percentage* the moisture equivalent point, the air
free porosities for the three aggregate ranges were:
16.7 percent for the <3*2 mm. aggregates; 31.4 percent for
the 6.4 to 3.2 mm. aggregates; and 34.4 percent for the
12.8 to 6.4 mm. aggregates.

A glass fiber mulch was placed

over the soil of each pot to help reduce slaking of the
aggregates during the addition of water.
Three methods of characterizing the aeration condi­
tions in the pots were used:

1.

The measurement of the

oxygen and carbon dioxide content in a sample of soil
atmosphere drawn from the 15-inch depth.

2.

The deter­

mination of the redox potential at the 15-inch depth.
3.

The measurement of the bioelectric potential developed

between the plant root and the soil.
In order to sample the soil atmosphere at the 15-inoh
depth, 3 mm. glass tubing sampling wells were permanently
installed in each pot by first making a hole in the soil
with a wire of slightly less diameter than the glass tubing.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

21,
and then gently inserting the glass tubing.

Each well had

a valve on top to prevent air from the atmosphere from
entering the well.

The oxygen and carbon dioxide content

was determined on a 100 cc. sample of soil air with an
Orsat analyser (ll).
Platinum electrodes were prepared according to the
method of Quispel (33) and inserted into the pots in the
same manner as the glass sampling well.

Redox potentials

developed at the platinum electrodes were measured
against a saturated calomel electrode in contact with the
soil moisture films at the soil surface.

A Beckman Model

G vacuum tube battery operated potentiometer was employed
to determine the potentials developed.
Bioelectric potentials were measured according to
the method of Taylor*(4?).

Two saturated calomel electrodes

were used to measure the potential.

Contact with the soil

was made with a saturated KCl bridge and contact with the
plant was made with a tap water saturated cotton string
wrapped around the plant Just below the cotyledons and
connected to the electrode.

Care was taken that no water

ran down the stem to short out the potential.
The potentials developed were measured with the same
instrument used to make the redox potential determinations,
*By personal communication July, 19^8.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission.

22.
To demonstrate the manner in which the bioelectric
potential Is Influenced by the aeration status around the
root zone, a tomato plant in a 6-lnoh clay pot filled with
a coarse sandy soil was grown to a height of 10 Inches and
then brought Into the laboratory for treatment.

The pot

and root system were submerged in water for a period of
time and then allowed to drain.

This cycle was repeated

and potentlometrlc measurements were made periodically
throughout the demonstration.

Figure 1 , taken from data

in Table 1 , deplete the manner In which the bioelectric
potential varies with the aeration status of the root
medium.

It was anticipated that the soybeans In the green­

house experiment would show the same general trend of
potentials as a function of aeration as the tomato did.
At the completion of the greenhouse experiment, the
plants were harvested and fresh weights determined.
Discussion of results.

The manganese-treated plants

died soon after germination or only grew e.t a very slow
rate, exhibiting extreme manganese toxicity symptoms.
There appeared to be no correlation between the levels of
aeration and the degree of toxicity.

Due to the poor

stand of plants on the manganese-treated pots, these data
were discarded.
Aeration measurements and final harvest weights from
the check and Iron treatments are reported In Appendix 1

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

23-

TABLE 1
A DE340NSTBATI0N OF THE EFFECT OF THE AERATION
STATUS ON THE BIOELECTRIC POTENTIAL OF A
TOMATO PLANT ROOT SYSTEM

T r e a t m e n t

D a y

H o u r

P o t e n t i a l
( m

0

F l o o d e d

11

II

It

II

II

II

w

II
II

D r a i n e d
It

It

II

II

II

II

II

1

II

It

II

2

II

It

II

II

II
F lo o d e d
u

5

H

II

«

II

II

6

II

7

II

w

II

II

II

8

It

II

II
II
II
D r a i n e d
II
w
II
II
II

9
10
11
II
II

12
II

13

II

II

14

II

II

11
11
12
01
02
03
03
04
05
11
04
11
02
03
10
10
10
01
04
10
10
04
10
11
05
09
11
01
05
10
04
11
05
09
03

i l l i v o l t e )

3 0

A .M .

40
00
10
00
00

A .M .

4 5

P . M .

0 5

P .M .

-

20
00
00

P .M .

4.

A .M .

4 .3 2

P .M .

4-40

3 0

A .M .

3 0

P . M .

N .
P .M .
P .M .
P .M .

4-10
4-10
4. 1
4- 6
-20
4- 5
0
5
9

1 5

P .M .

*•35
*-35
4-20

00

A .M .

4 .5 0

1 5

A .

4 5

A .M .

m

3 0

P .M .

00
00

P .M .
A .M .

3 0

.

4.3 0

4-28
4-48

A . M .

4-15
4-25
4. 8

3 0

P .M .

4-20

3 0

A .M .

4-10

4 5

A .M .

1 5

P .M .

4-15
5

3 0

A .M .

0

3 0

A .M .

1 5

P .M .

- 1 5
4 .3 0

1 5

P .M .

1 5

A .M .

4-28
4-38

4 5

P .M .

4 .1 0

3 0

A .M .

4 .4 5

44
10

P .M .
A . M .

4 5

P .M .

-20
4-85
4-35

.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

24.

Figure 1,

The effect of the aeration status on the
Bioelectric potential of a tomato plant
root system

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission.

VO

oj

O

00

VO

CM

o
o

00

o

VO

o

c

CM

O

SÎHOATITTH

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

o

CM

00
>
»
a
t
o

25.
with analyses of variance.

Mean values of aeration

measurements appear in Table 2.

Statistical methods show

no significant differences between treatments and levels
of aeration as they affected fresh weights of the soybean
plants.

However, analyses of variance on the methods of

characterizing soil aeration show very significant diff­
erences between the three levels of aeration and also
between the weekly measurements made over the five-week
test period.

This effect of time of measurement can be

attributed to poor control over temperature and moisture
conditions of the soil.

Undoubtedly the soil In the pots

became progressively drier between each watering because
of the more rapid uptake of moisture by the plants as they
became larger.

The last measurements were taken on a day

when the temperature was near 100° P.

This temperature

effect overshadowed the moisture factor In moèt of the
fifth-week measurements, particularly on the fine aggregate
soli.
Inspection of Figure 2 reveals that the carbon
dioxide content of the soil air steadily decreased In the
finer aggregated soil for the first four weeks, but rose
sharply at the fifth week.

The two larger aggregate ranges

showed a slight Increase In carbon dioxide content for the
first four weeks and rose more sharply the fifth week.

It

Is understandable that the fluctuations In carbon dioxide

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

26,
TABLE 2
MEAN VALUES OF MEASUREMENTS MADE
BY THREE METHODS OF CHARACTERIZING- SOIL AERATION
IN A PRELMINARY G-RSENHOUSE E}3ERIMENT

(A)

Redox Potentials

Aggregate Size

(Positive millivolts)
Weeks
2
4
3

1

5

^ 3 . 2 ram.

434

427

43 1

445

415

6.4 - 3*2 mm.

453

472

478

475

486

12.8 - 6.4 mm.

433

445

45 8

46 0

46o

(B)

Bioelectric Potentials

(Negative millivolts)

<3.2 mm.

62.3

80.5

S3.5

9 1 .0

72.5

6.4 — 3.2 rara.

33.0

62.0

69.0

66.8

7 1 .1

12.8 - 6.4 mm.

40.5

5 1 .2

67.0

60.6

48.8

(C)

Carbon Dioxide

(Percentege)

<3.2 rara.

4.0

3.3

2.9

2.3

5.5

6,4 - 3,2 ram.

1.3

1.4

1.55

1.56

2.1

1.2

1.2 ■

1.3

2. 2

12.8 - 6.4 rara.

.96

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w ith o u t perm ission.

27.

Figure 2 .

Percentage of carbondioxide in the soil
atmosphere at the 15-inch depth over the five
week test period

R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.

12.8 - 6.4 mm. AGOREGATGS
6.4 - 3,2 mm. AGGREGATES
ram. AGGREGATES
<3.2

0>
•c)
•ri

K
O,
ti

•ri

c
o

u
cd

o

c
0)

O
u
<v
A,

0

_l

±

2

3

WEEKS AFTER PLANTING

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

28content were greater on the finer aggregates because the
processes of diffusion were arrested where the porosities
were less.

Evidently the drying out process mentioned

above had little effect on the two larger aggregate ranges,
for the carbon dioxide content changed very little over
the earlier portion of the experiment.

The effect of the

high temperature was reflected, however. In a rise in the
fifth-week measurements on the larger aggregates as well
as the finer ones.
The same general trend of aeration conditions was
exhibited by the redox measurements shown in Figure 3 .
The potentials gradually rose during the first four weeks
and dropped sharply in the finer aggregates in the fifth
week.

The assumed temperature effect in the fifth week

didn't Influence the redox potentials in the two larger
aggregate ranges.

Perhaps in the larger aggregate ranges

the diffusion processes maintained the oxygen concentration
more nearly constant so that the temperature effect was
minimized.

'

'

It should be pointed out that the relative positions
of the redox potentials of the two larger sized aggregates
were the reverse of what was to be expected.

The larger

aggregate» exhibited a lower redox potential indicative of
poorer aeration than that found in the intermediate sized
aggregates.

This may have been due to a moisture relation-

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

29.

Figure 3 .

Redox potentials at a soil depth of 15 inches
over the five week test period

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

520

X = 12.8

500

O =

6.4

• = < 3.2

6. ij' mm. AGGREGATES
3.2 mm. AGGREGATES
m m . AGGREGATES

en
"o 46o
-M
r-t
r4
U-Uo -

^20

400

WEEKS AFTER PLANTING

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

30.ship.

One can theorize that in the larger aggregated soil

the larger sized pores allowed a more rapid infiltration
of water to the lower levels.

The aggregates near the top

of the*pot may have not then been thoroughly wetted.

In

this case there would be an increase in the water content
in the smaller pores of the larger aggregated soil and yet
allow the large ones to remain open.

In such a case, the

percentage of carbon dioxide could be lower in the large
pores of the larger aggregated soil and the redox potential
could be lower in contrast to that in the intermediate
aggregate range.
The curves in Figure 4 demonstrate the trends of the
bioelectric potentials produced between the plant root and
the soil systems.

The relative positions of the curves

are as expected in that the poorer the aeration status the
lower the potential.

Although the two larger aggregate

curves are close, they are significantly different.

The

<

largest aggregate range has the highest potential.

This

would be predicted by the carbon dioxide measurements, but
would be in contradiction to the redox measurements. The
general trends of the bioelectric potentials over the fiveweek period were exactly opposite to those which would be
expected from the trends of the other aeration measurements.
However, the temperature factor did evidently Influence the
potentials at the fifth-week measurements because of the

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

31.

Figure 4.

Bioelectric potentials between the soybean
root and soil systems over the five week test
period

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

12.8
6.4
f 3 .2

-6.4

-3.2

-50

-6o

-80

-90

-100
WEEKS AFTER PLANTING

R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.

ram. AaOREŒATSS
mm. ACK3-REGATE0
mm. AGGREGATES

32.
marked change In the slopes of the large and small aggre­
gate curves.

Why the intermediate aggregate soil did not

exhibit this effect is not clear.
It is of interest to point out that with all three
methods of measuring the aeration status of the soil the
curves for the finer aggregate range are strikingly alike
in general shape for the five-week test period.

These

measurements on the finer aggregates are probably more
valid because the porosity of this aggregate rsnge is
nearer to that found under natural field conditions.

Also

at this lower porosity, as pointed out before, aeration
conditions fluctuate over a wider range due to the diff­
erences in moisture and temperature over the test period.
This is indicated by the shapes of the three curves for
the finer aggregates in Figures 2, 3, and 4.
Evidently the factors that influenced the carbon
dioxide content and the redox potentials operated in the
same general way.

The bioelectric potentials, however,

reflected an inverse relationship.

Taylor (47) has found

that as the concentration of a nutrient solution increases
the bioelectric potential decreases.

Since a heavy appli­

cation of fertilizer was made to the soil prior to planting
the beans, it is possible that as the pots were drier over
a longer period of time as the test period advanced, the
effect of the increased concentration of salts in the soil

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

33.
solution overshadowed the effect of the Increased availa­
bility of the oxygen and drove the potentials down instead
of up for the first five weeks.

Increased temperature and

sunlight at the fifth-week measurement evidently increased
the respiration rate despite the decreased availability of
oxygen.
The Influence of carbon dioxide concentration per se
on the bioelectric potentials has had little study.

The

similarity of the two curves suggests a relationship.
Since, however, the bioelectric potentials are a result of
respiration, the mass action effect should have decreased
the respiration rate and thus lowered the potential.

It is

not likely then that the carbon dioxide concentration per
&e was the dominant factor operating to influence the
potentials.

The available oxygen to the roots bore an

inverse relationship to the carbon dioxide content in the
soil air.

These results, appearing in Appendix 1 , were not

graphed, but in general the inverse relationship held for
all measurements.

According to these results, the bioelec­

tric potentials fluctuated in a manner opposite to those
reported by other investigators.

They (3,30,^7) have found

that increased oxygen concentration Increased the bioelectric
potential.

One would conclude then that oxygen was not the

critical factor per se operating to influence the fluctua­
tion of the bioelectric potentials over the five-week period.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

3^ .

Again, it is pointed out, that the general levels of the
potentials for the three different aggregate ranges were
directly related to the aeration status of the soil.
Summary of results.

1.

There was no significant

difference between levels of aeration and plant growth.
The manganese treatments were too drastic to be of any
value and the iron treatments exhibited no significant
differences from the untreated checks in total fresh
weights.
2.

The percentage of carbon dioxide in the soil air

extracted from the larger pores correlates well with the
aeration levels determined by aggregate size.

The fluctua­

tions between weekly measurements were accentuated on the
finer aggregates and can be attributed to poor control
over moisture and temperature.
3.

The redox measurements showed the same general

trend of aeration conditions over the five-week test
period, but the relative positions of the potentials for
the two larger aggregates are reversed from what would be
expected for both the carbon dioxide content determinations
and the bioelectric measurements.

This may be due to a

higher moisture content in the small pores of the larger
aggregate range.

This argument would substantiate the

belief that the sampling of the gas from the larger pores
of a soil does not necessarily reflect the degree of
aeration in the smaller pores.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w ith o u t perm ission.

35.
4.

The relative levels of the bioelectric potentials

developed between the plant root system and the soil for
the different soil aggregate ranges correlates significantly
with the porosity of the aggregates.

However, the trends

over the five-week test period are in direct opposition to
what was expected based on levels of aeration per se.
These findings give added proof to the beliefs of Taylor (4?)
that this method is responsive to the many complicated
factors that affect respiration, of which oxygen supply is
one.

This experiment demonstrated that over the weekly

measurements some environmental factor other than oxygen
was influencing respiration and thus the bioelectric
potential.
Since the relative positions of the bioelectric
potentials correlate well with the porosities of the three
aggregate sizes and the carbon dioxide content of the
larger pores in the soil, the relative positions of the
redox measurements for the two larger aggregate ranges
are anomalous, despite the reasons given above.

The root

system of the plant is in closer association with the finer
soil pores than the larger pores.

If this is the case

then the bioelectric potentials are measuring the aeration
status in the finer pores just as do the redox potentials.
The question then arises as to what caused the higher re­
dox potentials in the lower porosity medium.

The need for

further investigation was indicated by these results.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w ith o u t perm ission.

36.
5.

Although such environmental factors as moisture

and temperature were inadequately controlled, the evalu­
ation of the methods used to characterize aeration
conditions in the soil substantiated the belief that some
new methods were sorely needed to measure quantitatively
the aeration status in the root zone of the soil,
B. A Potentiometric Method of Measuring Oxygen Diffusion
Theoretical considerations.

Buckingham (8 ) and

Penman (32) have shown that diffusion accounts for most
of the gaseous interchange occurring in the soil.

This

suggests that some method of measuring oxygen diffusion
to a point in the soil might be of value.

Hutchins (l8)

and Raney (3^) have developed methods based upon measuring
the rate at which oxygen diffuses to an oxygen deficient
absorber buried in the soil.

Their methods, however,

involve the use of costly and bulky equipment and the
measurements are time-consuming.

Consequently, an investi­

gation was undertaken in an attempt to devise a simple,
inexpensive and rapid means of measuring oxygen diffusion
in the soil.
Since the potentials developed at the surface of a
metal reflect the state of oxidation of the materials in
a solution wetting the metal, it was reasoned that perhaps
this principle could be used in the development of a

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

37.
suitable method for measuring oxygen diffusion in soil.
If one could find a reversible oxidation-reduction system
that is easily oxidized by free oxygen, it should be
possibl'e to measure the time rate of change of potential
which is a function of the time rate of change in the
oxidation status of the solution wetting the metal electrode,
this, in turn, being dependent upon the rate at which oxygen
diffuses to the solution wetting the electrode.

In other

words, the measured time rate of change in potential
developed at the metal electrode would be a function of
oxygen diffusing to the solution surface wetting the metal.
The chromic-chroraous system can be used for this purpose.
In their study of the chroraic-chromous potential
Forbes and Richter (12) encountered considerable difficulty
because of the unnoticed presence of a small pinhole in a
rubber connection of their apparatus which permitted inward
diffusion of oxygen.

As a result of this, they state, “We

believe that a more sensitive test for the presence of
oxygen in a gas otherwise incapable of reacting with the
substances in the cell could scarcely be devised."
The chemical equation for this reaction according to
Stone and Skavinski (44) is:

Eq. 1

4 Cr^^ » 4

. Og

4

=

2HgO f 4

and this is related to the familiar Nernst equation
according to Forbes and Richter (12) as:
Eq. 2

E = Eo + RT/F

In

Cr^3/Cr^2

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

38.
Referred to normal hydrogen electrode as zero* with
correction for Junction potentials, using a
Eq. *3

E =

Hg

electrode,

-0.400 4- 0.065 log Or^^/ Or^^

The experimental value O .065 as against the theoretical
0.058 may be due to complex formation and/or repression of
ionization.
From Eq. 3 when
Eq. 4

kE
At

-

t = time: .

Zik).0400 4.0 .06') log
A t

Cr^^/Gr*'^)

and the quantity Or^^/Or^^ is dependent upon the oxygen
diffusion rate to the system, according to Eq. 1 .

This

theory was tested experimentally.
Experimental methods and discussion of results. The
first phase of this investigation involved the study of
the chroraous-chromic system under controlled laboratory
conditions in order to learn the characteristics of the
?
system.
Chromic sulfate was prepared according to the
»
direction of Llngane and Pecsok (26) to give a concentration,
of 0.1 N chromous sulfate in 0.1 N sulfuric acid after the
oxidized solution had been reduced.

The apparatus used

for the reduction process and the metering of the solution
is shown schematically in Figure 5»

The three-way stopcock

and the stopcocks above and below the reagent reservoir
were adjusted to allow gas pressure from a tank of nitrogen

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

39.

Figure 5 .

Schematic diagram of charging system used in
the potentiometrio studies.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

Zinc reducer
column
2 ml. microburette
Potentiometer

Calomel
r '1

Reagent ^
Reservoir

Soil Probe
\\

3 way stopcock
------- f

Exhaust

Nitrogen supply
Scrubber

Charging cell

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w ith o u t perm ission.

40.to force the chromic solution up through the zinc reducer
column.

The reduced chroraous solution that flowed from

the top of the column passed over into the micro-burette
reservoir.

This reservoir, the micro-burette, the

connecting tubing and the charging cell (or other apparatus
to be mentioned later) were previously swept free of oxygen
with oxygen free nitrogen.

This was accomplished by turning

the three-way stopcock so that tank nitrogen pa.seed through
the alkaline pyrogallol scrubber and the humidifier and
then through the rest of the system.

By allowing a slow

stream of gas to flow through the system to the right of
the three-way stopcock it was possible to meter out the
desired amount of chroraous solution from the micro-burette
into the charging cell or other apparatus as desired.
Lingane and Pecsok (26) have suggested that the
chroraous solution can be kept for a considerable period
of time above mossy zinc without deteriorating.

However,

it was found that the period of solution stability was
limited by the gradual reduction of hydrogen ions by the
zinc.

This finally raised the pH to such a value that

hydrolytic precipitation of the chroraous ions occurred.
For this reason, the solution was only reduced as needed
by the simple method described above.
For the first Investigations the charging cell was
replaced by a closed chamber equipped with a mercury seal

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

41.
stirrer similar to that suggested by Maolnnes (28) for
potentiometrio titration.

The potentials were measured

with the vacuum tube potentiometer.
A potentiometrio titration of ohromous solution
was made with 1.5 percent hydrogeii peroxide, using a plati­
num electrode against a saturated calomel electrode.
curve, taken from Table 3, Is shown In Figure 6.

The

It

Indicates a simple reaction with no complex formation.
The reaction Is probably:
Eq. 5

4Cr*^ *

+ ZHgOg

=

ifHgO *.

4Cr'*'^

Another series of titrations was made with the
chromous solution by slowly drawing through the chamber
at a uniform rate two different concentrations of gaseous
oxygen.

In this Investigation an amalgamated copper

electrode was used Instead of platinum.

The resulting

curves are developed from Table 4 In Figure 7.

These

curves were obtained numerous times with excellent agree­
ment, provided the solution was stirred at the same uniform
rate and the gaseous mixtures were drawn through the titra­
tion chamber at the same uniform rate.
It will be noted that the time rate of change In
potential was dependent upon the concentration of oxygen
In the gaseous mixture drai-m through the chamber, the lower
the concentration, the slower the time rate of change In
potential.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

42

TABLE 3
POTENTIOMETRIO TITRATION OF 83 MILLILITERS OF
0.1 N. CrSOj^ IN 0.1 N. HpSOiL WITH 1.5^ HpO,
USING A PLATINUM ELECTRODE AGAINST
A SATURATED CALOMEL HALF CELL

Milliliters of H^Og

Potential (millivolts)

0

-400

12.0

-380

24.0

-365

24.2

- 40

24.6

4-200

25.0

1-300

30.0

4.390

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

» .

,

^3 *

Figure 6.

Potentiometrio titration curve of ohromous
sulfate with hydrogen peroxide using a
platinum electrode against a saturated
calomel electrode

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

-800

-6oo

CD

1-200

0

10

20

30

iK)

M l l llllter p of 1 .5%

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

The general shape of the curves Is of particular
Interest.

Two reactions appear to be taking place.

There

«

is a slight break in the curve near the beginning of the
titration with a more distinct one occurring at about
-320 millivolts.

The 10 percent oxygen curve would show

this same break, if it were completely plotted,
Laitinen and Kolthoff (23) state that the reduction
product of oxygen on a platinum surface is hydrogen peroxide.
Kolthoff and Lingane (22) have found that the reduction of
oxygen at a dropping mercury electrode is a two-step
reaction,
Eq. 6

Og

Eq, 7

H 2O2

4. 2H‘*‘^

2e“

4- 2H*^ + 2e"

=
=

'^2 ^ 2

(sicid solution)
(acid solution)

Perhaps a two-step reaction similar to that suggested
above by equations 6 and 7 was taking place.

Since a single,

simple reaction appeared to take place when hydrogen per­
oxide was used in the titration and there appeared to be
two reactions occurring when free oxygen was used, one might
conclude that hydrogen peroxide was an intermediate product
of the reduction of oxygen.

To the present author's know­

ledge, no similar reports have been made in the literature
concerning the oxidation of the chroraous ion by free
oxygen.

These findings warrant future investigation.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

^5 .

TABLE 4
POTENTIOMETRIO TITRATION OF 20 MILLILITERS OF
'0.1 N. CrSOh IN 0.1 N. HpSOn WITH FREE OXYGEN
IN TWO DIFFERENT CONCENTRATIONS USING
AN AMALGAMATED COPPER ELECTRODE AGAINST A
SATURATED CALOMEL ELECTRODE.
THE GAS MIXTURES
WERE DRAWN AT A UNIFORM RATE OVER THE STIRRED SOLUTIONS

21^ Oxygen

10^ Oxygen

Time
(min.)

Potential
(millivolts)

Time
(min. )

Potential
(millivolts)

1
3
45

-520
-518
-500
-492
-479
—468
-458
-448
-435
—424
-412
-400
-392
-378

0
5
10
15
20
25
30
35
40
45
50
55
60
65
70

-630
—625
-615
-600
-570
“550
“525
—50C
-475
-400
-430
—4l0
-395
-370
-210

6
7

8
9
10
11
12
13
14
15
17
18
19
20
21
22
23
28
33
58

-332
-260
-210
-198
-192
-I83
-180
-150

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

46

Figure 7.

Potentiometrio time rate of change titration
curves of ohromous sulfate with 10^ and 20%
oxygen using an amalgamated copper electrode
against a saturated calomel electrode

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

.

-6oor
10% o
-500

21% 0

300

-200

-100
10

20

ko

Time in mlnutee

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

The titration studies, with oxygen over the stirred
chromous solutions, were sufficiently encouraging to Justify
the Investigation of oxygen diffusion to a quiet solution
of reduced chromium.
The apparatus used in these investigations Is
diagrammed in Figure 8; it consists of a four-inch
cuhlcally shaped chamber made of plexiglass plastic.
One side of the cube was removable to permit access to the
inside of the diffusion chamber.

This was easily sealed

with petroleum when the apparatus was in operation.

At

the top of the chamber a hole was drilled to permit
diffusion.

Riding on top of this surface was a greased

slide that operated between two firmly attached guides.
This slide contained an opening the same size and shape
as the hole into the diffusion chamber.

The sample holder,

a piece of cylindrical brass tubing with a diameter of
3 inches and a height of 2 inches, fit snugly Into a recess
in the slide.

By smearing petroleum around this Joint an

air-tight seal was maintained.

To keep the sample rigid,

a heavy screen was placed In the recess of the slide to
act as a support for a finer gauze screen that was placed
in the bottom of the sample holder to contain the sample
of fine material.
The oxygen sensitive half cell consisted of a glass
vial with an inside diameter of

$/Q

Inches almost filled

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

-700

-6oo
10 %

-500
ai

21%

O

-200

-100
0

10

20

30

Time in mlnutee

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

47.
The titration studies, with oxygen over the stirred
ohromous solutions, were sufficiently encouraging to Justify
the investigation of oxygen diffusion to a quiet solution
of reduced chromium.
The apparatus used in these investigations is
diagrammed in Figure 8; it consists of a four-inch
cuhicelly shaped chamber made of plexiglass plastic.
One side of the cube was removable to permit access to the
inside of the diffusion chamber.

This was easily sealed

with petroleum when the apparatus was in operation.

At

the top of the chamber a hole was drilled to permit
diffusion.

Riding on top of this surface was a greased

slide that operated between two firmly attached guides.
This slide contained an opening the same size and shape
as the hole into the diffusion chamber.

The sample holder,

a piece of cylindrical brass tubing with a diameter of
3 inches and a height of 2 inches, fit snugly into a recess
in the slide.

By smearing petroleum around this Joint an

air-tight seal was maintained.

To keep the sample rigid,

a heavy screen was placed in the recess of the slide to
act as a support for a finer gauze screen that was placed
in the bottom of the sample holder to contain the sample
of fine material.
The oxygen sensitive half cell consisted of a glass
vial with an inside diameter of 5/8 inches almost filled

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

48.

Figure 8.

Schematic diagram of diffusion apparatus

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

■D
O
Q.
C

g
Q.
■D
CD

C/)
C/)

3" braBP cylinder
■8D

To Micro-burette

r
!:
I;

3.
3"
CD

■D
O
Q.
CD

C

a
O
3
"O
o

CD

]KC1 Bridge to calomel
Nitrogen
intake

3- to potentiometer

Q.

1:

Exhaust

■D
CD

I^ ^ hole rubber stopper

Gr30^ flolutlo^

C/)
CO

film

I L_,

Hg electrode»

Cu wire

49.
with pure mercury.

Enough ohromous solution was placed

on the pool of mercury to thoroughly wet the metal (about

0.25 milliliters).

Contact with the mercury pool was made

with a copper wire, and a KOI bridge connected the chromous
solution film with the saturated calomel half cell.
The method, In operation, consisted of pushing the
slide to the position indicated in the diagram and then
sweeping the chamber free of oxygen by a stream of oxygenfree nitrogen.

The chromous solution was then metered out

from the micro-burette and the half cell allowed to come to
equilibrium*

When equilibrium was obtained, the slide

holding the sample holder was moved until the opening in it
coincided with the opening in the diffusion chamber and
diffusion allowed to take place through the brass cylinder.
The time rate of change in potential was then determined.
The curves plotted in Figure 9, taken from data in
Table

5» for the empty sample holder and the holder filled

with 0,25 to 0.1 mm. sand show marked differences in oxy­
gen diffusion rate as indicated by the differences in the
time rate of change of the potential.
Many similar results were obtained, but agreement
among replications of like treatments was not good.
Unexplained inherent errors began to appear at this phase
of the problem.

Despite these difficulties, work was

started on the manufacture of a soil probe using this
principle to measure oxygen diffusion.

It was hoped that

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

50.

TABLE 5
POTENTIOMETRIO TITRATION OF 0.25 MILLILITERS OF
0.1 N. OrSOi+ IN 0.1 N. H^SOk BY OXYGEN DIFFUSION
USING A QUIET SOLUTION OVER A COPPER-MERCURY POOL
ELECTRODE AGAINST A SATURATED CALOMEL ELECTRODE

Diffusion Cylinder Empty

Diffusion Cylinder Filled
with 0.1 - 0.25 mm. sand

Time
(min.)

Potential
(millivolts)

Time
(min.)

0
5
10
15
19
20
21
22
23
24
25
26
27

-620
-615
-605
-585
-520
“505
-465
-421
“325
-280
-262
-242
“235

0
5
15
20
25.
28
29
30
32
33
34
35
38
40
41
43
44
45
46
47
49

Potential
(millivolts)

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

-645
-645
-620
—605
“585
-555
“535
-520
-500
-490
-480
-470
“435
-405
-380
“350
“315
-295
“275
“255
-220

51.

Figure 9.

Representative potentiometrio time rate of
change titration curves obtained in diffusion
studies with the diffusion cylinder empty and
filled with sand

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

-800

-700

-6oo

Sand filled
cylinder

Empty
cylinder

-300

-200
0

10

20

Time in minutes

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

52.
continued study of the system would reveal ways of elimi­
nating the serious errors.

Considerahle time and effort

was spent with no avail, devising many different mechanical
variations in an attempt to make a probe sufficiently
accurate to be of practical use.

Figure 10 represents

schematically one such design tested.

The other designs

were similar in principle with slight variations that are
to be discussed later.
The probe consisted of two pieces of aluminum tubing,
one fitting snugly inside the other.

The inside tube was

closed at one end by a pointed plastic tip and contained
a calomel cell at the upper end.

The lower portion of the

inside tube contained the diffusion chamber and the oxygen
sensitive cell.

This chamber was sealed above with a two-

hole rubber stopper and below with a rubber gasket firmly
meeting the outside tubing.

The interface between the two

tubes was coated with stopcock grease to prevent downward
diffusion of oxygen between the tubes and to act as a
lubricant.

In operation the diffusion chamber was effec­

tively sealed when the outside tube was down on the rubber
gasket.

By slipping the outer cylinder up, the ports into

the diffusion chamber were exposed and diffusion allowed
to take place.
In order to charge the probe, it was inserted through
a snugly fitting hole in the rubber stopper sealing the

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

53.

Figure 10.

Schematic diagram of potentiometrio diffusion
prohe

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

Calomel cell

G-round glapB etopcook

Potentiometer

KCl bridge

Scale 1:1
12“ In length

$000“ OD (outer sleeve)
,
ID (outer sleeve)

Platinum wire
Rubber peal

.^375" OD (inner tube)
.3675" ID (inner tube)
Tvo hole rubber stopper
,^100“ OD (inner tube)
.3750” (2) dla. of diffusion
ports
GrCl^ solution film
Hg electrode

Plastic tip

R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.

5^.
charging cell as pictured in Figure 5-

The diffusion

chamber was opened and swept free of oxygen with oxygenfree nitrogen, then the chromous solution was metered out
from the micro-burette into the diffusion chamber in a
sufficient quantity to entirely wet the mercury (or platinum
as discussed below).

In order to prevent creeping,

contamination, and to secure a plane surface, the inside of
the Inner tube, the fine capillary KCl bridge and the
platinum wire were all coated with paraffin.

Only the very

bottom of the platinum wire was free to make contact with
the mercury because of the paraffin insulation.

The diffu­

sion chamber was closed by lowering the outside cylinder to
the rubber seal.

The whole probe was then removed from the

charging cell and was ready for operation.
In operation the probe was to be inserted into the
soil, the outside tube raised to allow oxygen diffusion
into the diffusion chamber through the ports and the change
in potential measured by the same potentiometer that was
used in the previous studies.

Successful tests in free air

were not obtained, so that the instrument was not tried in
the soil.
For some time various probe designs employed platinTira
electrodes Instead of mercury.

Platinum tends to come to

equilibrium more quickly than mercury because the catalytic

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

55.
reduction of oxygen on a platinum surface occurs more
rapidly than on mercury.

However, hydrogen tends to be

set free at a platinum surface because of the low overvoltag-e for hydrogen and Is not set free at a mercury
surface because of the high overvoltage.

The liberation

of gaseous hydrogen could seriously cause accountable
errors In the measurements.

It was not proven In these

experiments, however, that this was the case.

Nonetheless

attempts were made to obtain satisfactory results by
employing mercury electrodes using either copper wire or
platinum wire contacts.

These electrodes gave no better

results than the platinum electrodes.
Diffusion potentials may have developed In the
necessarily long KCl bridge system.

In order to minimize

this possibility a small calomel cell was made In the
plastic tip of the probe.

Reducing the bridge length

appeared to have little Influence In reducing the radical
behe.vlor of the electrode.

It would appear that this error

was Insignificant In that It did not seem to affect the
titrations when the solutions were stirred, even though
the KCl bridges were quite long.
Different concentrations of chromous sulfate In sul­
furic acid were tried, but here again no light was shed
on the causes of the errors Involved,

Solutions of 0.1 N

chromous chloride In 0,1 N hydrochloric acid were also

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission.

.56.
used with little success.
Purity of reagents Is of extreme Importance, since
any foreign metals either In the electrode or the solution
tend fo reduce hydrogen.

This possible error was minimized

by extreme care In the preparation of the system.

However,

the author feels that perhaps contamination of some sort
may account for the failure of this system to work
satisfactorily.
Perhaps the failure can be attributed to the erratic
diffusion of the oxygen In the solution phase and erratlo
reduction of oxygen at the electrode surface.

If this was

the case, only a statistical evaluation of a large number
of measurements would truly differentiate between various
rates of oxygen diffusion to the solution surface.

Since

errors In the method only appeared when the solution was
not stirred, this might be the difficulty.
Some radical change In design, such that the indi­
cator solution could be stirred, should be tried in the
future.
Another good possibility for the development of a
method of measuring soil aeration appeared during these
Investigations.

Both chromous solutions and solutions of

sodium anthraqulnone-B-sulfonate exhibit distinctive color
changes when going from the reduced state to the oxidized
state.

Sodium anthraqulnone-B-sulfonate was tested

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

57.
potentlometrlcally and displayed the same radical behavior
as the chromous solutions did when not stirred,

Colori-

metrically this organic compound might be more suitable
because of its distinct color change from red to colorless
upon oxidation (7).

The development of some colorimetric

method of measuring oxygen diffusion in the soil warrants
further investigation.
Summary of results.

1.

Stirred solutions of reduced

chromium wetting a mercury electrode exhibit a time rate
of change in potential dependent upon the partial pressure
of oxygen above the solution surface as predicted by theory.
2,

The nature of the titration curve Indicates a two-

step reaction in the reduction of oxygen.

The intermediate

product may be hydrogen peroxide,
3,

Unstirred solutions give results too erratic to

be of practical value.

This may be due to the erratic

diffusion process in the liquid phase itself and/or the
erratic activation of oxygen at the electrode surface,
4,

A colorimetric method of measuring oxygen diffusion

may offer possibilities for future investigation.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

58.

C.

An Amperometrlc Method of Measuring Oxygen Diffusion
It is believed by Russell (4l) that the evaluation of

conditions at the interface between the root surface and the
soil system presents the greatest possibility of ascertaining
the influence of soil aeration on plant growth.

This inter­

face is a cell-wall liquid-phase boundary with a gas-water
boundary delimiting the liquid phase on the opposite side.
Therefore, factors that operate to influence the movement
of oxygen through the liquid phase, in. addition to the oxy­
gen diffusion rate to the gas-water boundary through the
gaseous phase, are important in the ultimate supply of
oxygen reaching the respiring cells in the plant root.
In the water-fllm cell-wall portion of the oxygen
chain, movement is probably by diffusion only.

Under normal

conditions, the direction of oxygen diffusion is down a
concentration gradient from the soil atmosphere into and
through the water films of the soil to the cell walls of
the root.

This concentration gradient may at times, if not

always, be very marked (9).

This suggests that some method

of measuring oxygen diffusion, down a considerable concen­
tration gradient, through the liquid phase to a reducing
surface similar to that of a plant root would be of
fundamental importance.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

59.
Theoretical oonalderationa.

Under certain conditions

the current resulting from the electrolytic reduction of
oxygen at a platinum electrode, within limits, is governed
solely by the rate at which oxygen diffuses to the elec­
trode surface from the surrounding solution (22).
The manner in which the electrode current varies with
respect to the applied e.m.f. is shown in Figure 13 developed
from data in Table 6.

It will be noted that as the e.m.f.

increased in the electrolysis cell, in this case platinum
vs. saturated calomel, only an exceedingly small current,
"the residual current," flowed through the cell until the
decomposition potential of oxygen was reached at about minus
0.2 volts.

At this point continuous reduction began.

As

the voltage was increased after the decomposition potential
was reached, the current did not increase indefinitely, but
gradually approached a limiting value and finally became
constant and independent of further increases in the applied
potential, as the "plateaus" of Figure 13 indicate.

This

state of affairs held until the e.m.f. was raised above
0.85 volts, when the current again began to rise.

This

second rise was due to a second reaction, the reduction of
ionic hydrogen to molecular hydrogen.
The "limiting currents" in the "plateau" region are
caused by a virtually complete state of concentration polari­
zation at the platinum surface, as pointed out by Kolthoff

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

6o.
and Lingane (22).

As a result of the decomposition of

oxygen, the concentration of this reducible material is
depleted to a concentration practically equal to zero close
to the surface of the platinum electrode, and this loss is
compensated by diffusion of a fresh supply of oxygen from
the body of the solution.

With an excess of some indiff­

erent salt present in the solution, the limiting current is
determined almost entirely by the rate of diffusion and is
called the "diffusion current."
Kolthoff and Lingane^ (22) have successfully applied
Pick*s classical diffusion theory to the electrolysis
phenomena occurring at various types of electrodes.

In this

study it is assumed that the concentration of oxygen at the
electrode surface is zero.

The diffusion of oxygen to the

electrode is a maximum only when this is true.
The diffusion process in the vicinity of a platinum
microelectrode depends upon the geometric characteristics
of the electrode and the diffusion field surrounding it.
The diffusion at platinum wire microelectrodes, such as
used in these studies, is unsymmetrioal and not subject to
exact mathematical interpretation.

However, an approxima­

tion can be made by assuming linear diffusion to a flat
electrode.

Under these conditions, the following equations

will then apply;

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

6l*

Eq. 8

If
^

Eq. 9

C^_t

»

n F A

IX

= o,t)
.y

n F A D

\,ax/x=o,t

= ° ^

from Eq. 8

and differentiating

Eq. 10

-

i^

=

Eq. 9

when

x « o,

1/2

nP C A (D/7Tt) ^

i.j.

=

current in amperes at time (t) in seconds

n

- number of electrons used per molecule of oxygen
electrolyzed.

F

=

the Faraday, 96,500 coulombs

A •

= Area of electrode in square centimeters

^x=o,t = Flux at the electrode surface (X distancefrom the
electrode surface equals zero) at time t or the
number of moles of oxygen diffusing to the electrode
at time t.
D

=

Diffusion coefficient of oxygen in square centime­
ters per second.

C

- Initial uniform concentration of oxygen

^x,t

= Concentration of oxygen in moles per cc. at
distance
face

t

x

centimeters from the electrode sur­

seconds after beginning diffusion (by

closing circuit).
y

a

= a variable of integration

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

62.
According to equation 8, it Is possible to calculate
the number of moles of oxygen diffusing to the electrode
per second.

If the electrode area is known, amperes of

current flow can be used to calculate the oxygen diffusion
rate in grams per square centimeter per minute at a given
time after the circuit has been closed.
In Figure 11 are shown curves plotted from equation 9,
indicating how far the concentration gradient should extend
at various times after the beginning of electrolysis*

Each

curve thus represents the fraction of the initial concen­
tration

as a function of distance from the electrode

at a given instant.

The concentration gradient at the

electrode surface is very large, for very small values of
time, but gradually decreases to zero as time approaches
infinity.

Figure 17 developed from data obtained in a

second greenhouse experiment, to be considered later,
demonstrates this decrease in current with time as the
diffusion gradient boundary extends further and further from
the electrode.
Equation 10 predicts that the current at a given
instant should be proportional to the concentration of oxy­
gen in solution.

Figure l4, taken from data in Table 6,

shows the linear relationship between current flow and
oxygen concentration in a soil suspension, at given time
periods after electrolysis begins.

This linearity indicates

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

63.
that substances other than oxygen had not been reduced at
this potential (0.8 volts).
There is some uncertainty as to the nature of the
chemical reaction occurring at the platinum electrode.
Laitinen and Kolthoff (23) state that hydrogen peroxide is
the reduction end product of oxygen, according to equation 6,
Two arguments support this conclusion.

In the first place,

they found hydrogen peroxide in the solution near a large
platinum cathode.

However, Davies and Brink (10) have

pointed out that, unfortunately, no mention was made of the
potential at which the electrolysis was carried out.

The

second argument given is that with the assumption of a two
electron reaction (instead of a double reaction involving
four electrons, according to equations 6 and 7) the calcu­
lated diffusion coefficient agrees with the available values
given in the literature,

Davies and Brink obtained data

using platinum microelectrodes that indicated a four elec­
tron reaction, although in one instance an electrode showed
a two electron reaction.

They explain this by postulating

that, in some cases, hydrogen peroxide is absorbed or
combined chemically with some active agent.

If the hydrogen

peroxide is not so "inactivated" it will be reduced and the
apparent four electron reaction will take place.
It is believed that the reaction involves four
electrons.

Data obtained in potentiometric studies with

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

61^.

Figure 11.

Ciirrent-distance curve at various times after
beginning linear diffusion. Calculated from
Equation 9, assuming D = 4. _$6 x 10~5 cm. 2/gec.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

o w
*
x r\

E-i
M

O
O

o

o

o
rH

oo

VO

o

o

o

o

o

K o
o

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

Q

65.
chromous solutions, discussed previously, suf^gest that the
titration of chromous sulfate with hydrogen peroxide, using
a platinum electrode, was a simple reaction.

However,

titration of the chromous sulfate with oxygen using an
amalgamated copper electrode suggested a double reaction.
Although one may not be Justified in drawing conclusions
involving two different types of electrodes, nevertheless
hydrogen peroxide was activated at the platinum surface in
the presence of chromous ions and a more complex reaction
occurred at the amalgamated electrode when oxygen was used.
Kolthoff and Lingane (22) have shown that there is a
two-step reaction in the reduction of oxygen at the dropping
mercury electrode.

The second polarographic wave obtained

coincided with that obtained with an air free solution of
hydrogen peroxide.

This double reaction is in agreement

with the double reaction obtained in this study with the
amalgamated electrode.
Another argument that supports the four-electron
theory was brought out by the microelectrode measurements of
oxygen diffusion obtained in a greenhouse experiment to be
discussed later.

The oxygen diffusion rates, calculated

from these measurements, are in better agreement with the
data obtained by Hutchins (18) if the four-electron reaction
is considered to be correct (if n = 4 in equation 8),

If

the two-electron reaction is assumed to be correct (if n = 2

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

66.
In equation 8) the discrepancies between Hutchins' results
and those obtained In the present study are twice as great.
Discussion of laboratory experimental methods and
results.

No evidence was given by Karsten (19) In his study

of oxygen In soil suspensions using the dropping mercury
electrode that substances other than oxygen In the soil are
reduced at the voltages used for the reduction of oxygen.
He used six different soils that varied widely In texture
and organic matter.
Despite Karsten*s failure to report any difficulties
arising from the reduction of substances other than oxygen
In the soil, it was considered advisable to study this
possible difficulty using a Brookston clay loam soil.
A one to one soil In water suspension was placed In
a closed container equipped with a mercury seal stirrer
such as used In the potentiometric studies mentioned above.
A platinum microelectrode h mm. long made of 25 gauge wire
was placed well Into the suspension and contact between a
large saturated calomel cell and the suspension was made
with a KOI bridge.

The container had an Inlet tube below

the surface of the suspension and an outlet tube above the
surface.

In this way gas of a known partial pressure of

oxygen could be bubbled through the suspension and maintained
at a steady flow above the suspension.

The laboratory

Instrument diagrammed In Figure 12 maintained the desired

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

67.

Figure 12.

Schematic diagrams of the field and laboratory
Instruments used In the amperometrlc studies.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

Laboratory Inatrumetot
+ L-

vW\VWV\WV'WV\’

\
<V>
\WMA/WV\AVr

y

Field Instrument
j.“_

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

68.
voltage between tbe calomel cell connected to the positive
lead and the platinum electrode connected to the negative.
lead and also measured the current flow through the circuit.
►

A one and one-half volt dry cell was used as a source
of power.

The upper 1200-ohm slide wire resistor was

adjusted so that the potential across the second (lower)
86-ohm slide wire resistor was one volt as Indicated by the
voltmeter

V

when the Upper left circuit key was closed.

It was then possible to select any voltage between 0 and 1.0
volt by positioning the second slide wire resistor.

By

closing the lower circuit key this selected voltage was
maintained across the electrodes and the current flowing
through the circuit could be computed from the galvanometer
readings and the resistance In the two resistor shunt.
galvanometer,

The

0-, used was a Leeds and Northrup wall type of

320 megohm sensitivity.

The upper shunt resistor was a

0 to 9*999 ohm decade type and the lower one was a 0 to
999.9 ohm decade type.
The procedure for the soil suspension studies was to
bubble gas of a known oxygen concentration through the
stirred soil suspension for five minutes.

The gas flow and

the stirrer were stopped for one minute prior to making the
microelectrode measurements In order to allow the suspension
to become still.

The circuit was closed and current readings

made at 10, 20, 30 and 6o seconds at the desired voltage*

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w ith o u t perm ission.

69.
The cycle was repeated at 0 .1 -volt intervals from 0 , 2 volte
to 1.0 volt for three different oxygen ooncontratlone.
Figure 13, developed from data In Table 6, exhibits the
manner in which the current flow depended upon the
potential between the electrodes and the oxygen pressure in
the soil suspension 20 seconds after the circuit was closed.
A plot of the current flow vs. the oxygen pressure at 10
and 20 seconds after closing the circuit at a potential of
0.8 volte reveals a linear relationship between the current
flow and oxygen tension in the soil suspension.

This is

shown in Figure 1 4 and Table 7 .
This excellent linear relationship indicates that no
substance other than oxygen was being reduced at this voltage.
This particular voltage was chosen because here the slopes
of the current voltage curves were the least.

Any e.m.f.

between 0.0 and 0.2 volts could be used, although current
and oxygen tension are proportional, within the experimental
error, only in the “plateau” region.

Also the effect of

variation in applied voltage is least in the “plateau” region.
For these reasons, the voltege is always adjusted to some
constant value on the “plateau”.
In view of the fact that no substance in the Brookston
clay loam suspension interfered in the measurement of oxygen
tension with the platinum microelectrode, it was believed
that if the electrode is completely wet by the moisture in

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

70.

TABLE 6
THE .INFLUENCE OF VARIOUS PARTIAL PRESSURES OF OXYG-EN
IN A 1 : 1 BROOKSTON CLAY LOAM SOIL-WATER SUSPENSION
ON THE CURRENT FLOW AT VARIOUS VOLTAG-ES 20 SECONDS
AFTER DIFFUSION BEGINS USING A PLATINUM MICRO­
ELECTRODE AGAINST A SATURATED CALOMEL ELECTRODE
Volts

Current in Microamperes
193 mm.
3 . 7 mm.*
742 mm.

0 .2

.0 2 0

.450

0.100

0 .3

.025

1 .10 0

2.250

0 .4

.0 3 0

1.650

4.500

0 .5

.050

1.900

6.250

0.6

.100

2.150

7.700

0 .7

.300

2.300

8. 500

0.8

.650

2.450

9.250

0 .9

1.500

4.300

9.250

1.0

4.250

7.000

12.100

^Partial pressure of oxygen in mm. of mercury

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

71

Figure 13.

Current-voltage curves for three different
concentrations of oxygen above a Brookston
clay loam soil suspension

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

o

rH

CD

O

• vO

O

00

\o

ca

âdjadm'eoao'tj^

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission.

o

72.
the Boil, the microeleotrode could be used to measure
oxygen diffusion rates In the soil.

If this measurement

could be made successfully, it would be possible to determltio the supply of oxygen to the environment similar to
that of an actively respiring plant root in the soil.
This procedure was tested ea^erimentally under
carefully controlled greenhouse conditions,
D.

A Second Greenhouse Experiment

Experimental methods and design.

Information gained

from the preliminary greenhouse experiment raised many
questions for further study.

In particular the testing

of methods for measuring aerôtion under more carefully
controlled moisture and temperature conditions was suggested.
To this end, a similar experiment was initiated to test
two new methods of characterizing soil aerat ion *along with
two of the older methods and also to investigate further
the possible toxic effects of the ferrous and manganous
ions on plant growth under reducing conditions.
A well-aggregated Brookston clay loam soil was taken
from a field that had grown alfalfa for two years.

This

soil was similar to that used in the earlier experiment,
as it was taken from the same area.

The soil was air-dried,

and screened into three finer aggregate size ranges than
used previously; 3*2 to 1.6 mm., 1.6 to 0.8 mm., and <0.8 mm.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

73.

TABLE 7
THE INFLUENCE OF VARIOUS PARTIAL PRESSURES OF OXYG-EN
IN A 1 : 1 BROOKSTON CLAY LOAM SOIL-WATER SUSPENSION
ON THE CURRENT FLOW AT 0.8 VOLT 10 AND 20 SECONDS
AFTER DIFFUSION BEGINS USING A PLATINUM MICROELECTRODE AGAINST A SATURATED CALOMEL ELECTRODE

Partial Pressure of
Oxygen in mm. of Mercury

Microamperes
10 sec

20 sec.

3 .7 mm.

0.93

0. 65

152

mm.

3.30

2. 45

742

mm.

13.05

9.25

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

74

Figure 1 4 .

The relation between oxygen tension and
current flow at the platinum micro-electrode*
at 0.8 volts potential 10 and 20 seconds
after diffusion begins

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

14
10 SEC
12

10
20 SEC

m

0)

u
03

I
u
Ü

0

200

600

800

Oxygen Concentration
ram. of mercury

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission.

75.
A

16-8 commercial fertilizer was mixed with each aggregate

separation at the rate of 10 tons per acre.

Ferrous sulfate

was mixed with the appropriate aggregate ranges at the rate
of 16 tons per acre as previously, but the manganese sulfate
was reduced in application to 1 ton per acre.

The same tile

pots were used as in the previous experiment.

This time,

however, two 9/l6-lnch ports were drilled through each pot,
one to accomodate the constant moisture tension syphon
system, and the other to facilitate the measurement of gaseous
oxygen diffusion by a method similar to Raney's (3 4 ), without
disturbing the soil surface.

The outside of the pots were

painted with aluminum paint in order to better control the
soil temperature during sunny days.
A layer of sand 2 inches deep was put into each pot
and the soil aggregates were uniformly packed into the pots
to give a depth of 40 cm. above the constant water table.
The pots were set up with a constant water level syphon
system to maintain a tension of Uo cm, at the soil surface.
Moisture tension studies were made, in the laboratory, on
core samples of the three aggregate sizes to enable the
calculation of air-filled pore space in the pots at various
heights above the water table.
This experiment consisted, as in the previous greenhouse
experiment, of four replications of three levels of aeration

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

76.
(as determined by aggregate size ranges) with three treat­
ments; a check, an added Increment of iron and an added
Increment of manganese.
After the soil had been flooded from the bottom and
allowed to come to equilibrium with the constant water table,
uniform 6-Inch tomato plants, var. Michigan Forcing, were
selected and planted one per pot with as little disturbance
of the soil as possible.

Tomato plants were chosen because

of their high sensitivity to a deficiency of oxygen In the
root medium.
Greenhouse temperature was maintained at approximately
75 ° Fahrenheit.

Supplemental artificial light was supplied

for the plants as needed.
Four methods of characterizing the aeration In the
pots were used:

1.

The measurement of oxygen content In a

sample of soil atmosphere drawn from the 8-lnch depth;
2.

The determination of the redox potential at the 8-lnch

depth;

3-

The measurement of oxygen diffusion In the gaseous

phase at the 8-lnch depth, employing a small diffusion chamber
similar In principle to Raney's;

4.

The measurement of oxy­

gen diffusion through the liquid phase with the platlnùm
microelectrode at four depths, 2 , 4 , 6, and 8 Inches.

Plant

growth rates were determined by periodic height measurements.
Redox potentials were measured as before and the soil
atmosphere was sampled by aspiring approximately 50 cc. of

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

77.
gas from the eight-inch depth through one of the drilled
ports in the side of the pot.

The gas was drawn directly

through a Beckman Model D Oxygen Analyser for analysis.
Between measurements the ports were closed with rubber
stoppers.
The method used in the measurement of oxygen diffu­
sion in the gaseous phase was similar to that described by
Raney (3^).

However, it was necessary to use a much smaller

diffusion chamber.

A schematic drawing of a diffusion cham­

ber probe used is shown in Figure I5.
The probe consisted of two pieces of aluminum tubing,
one fitting snugly inside the other in the same manner as
that described in Figure 1 0 ,

The closing device at the

bottom operated in the same way except that a threaded nut
above a collar was necessary to force the bottom of the out­
side tubing firmly against the rubber seal when the probe
was closed.
In operation, the closed probe was connected to the
charging system as shown schematically in Figure I6,

The

three-way stopcocks A and B were positioned to allow tank
nitrogen to pass through at stopcock A, circulate clockwise
through the entire system and exhaust through stopcock B.
In this way, all the oxygen, except that contaminating the
incoming nitrogen, was swept out of the system.

The probe

was then disconnected from the charging system and inserted

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

78.
Into the side ports of the pots filled with soil.

In order

to prevent the disturbance and compaction of the soil by the
insertion of the probe, a piece of glass tubing of the same
diameter as the probe was placed in the port of each pot
while the dry soil was packed into the pots.

After the pots

were flooded, the glass tubes were removed, leaving wells
for the insertion of the probe.
Once the probe was inserted into a well through the
side port of the pot, the threaded nut was unscrewed from
the inner tube so that collar and nut slid off allowing the
outside tube to be raised, exposing the many small ports at
the other end of the tube to the soil atmosphere in the
sampling well.

The small space between the walls of the

port and the walls of the probe were sealed with putty to
prevent inward diffusion of oxygen from the atmosphere.
Diffusion was allowed to take place for ten minutes.

During

this period oxygen diffused into the probe from the soil and
nitrogen diffused out.

At the end of this diffusion period,

the outside tube was lowered once again, along with the
collar and nut, and the probe was sealed by screwing the nut
down.
system.

The probe was then removed and connected to the charging
This time the stopcocks A and B were positioned so

that the system was a closed one with no inlet or outlet.
The gas in the whole system was circulated by depressing
and releasing the squeeze-bulb several times.

The expansion

chamber which was made of a simple toy balloon prevented any

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

79.
appreciable pressure building up in the system during the
depressing of the squeeze-bulb.

The directional flow valves

permitted the gas to be circulated in a clockwise direction
only. . When the gas in the system was thoroughly mixed the
percentage oxygen was determined by the Beckman Model D
Oxygen Analyser.

By simple calculation it was possible to

determine the diffusion rate, D, in the soil and compare it
with the diffusion rate, D^, in free Air.

The value D q had

f

been previously determined by opening the nitrogen-filled
probe to free air for ten minutes in the same manner as
described above for diffusion in the soil.

A correction

factor of 0.016 had to be subtracted from both D and Dq
values to account for the 0.7 percent oxygen in the tank
nitrogen used.

Outside of this correction factor the method

of calculation was that used by Raney (3 ^)*
Of particular interest here is the platinum microelectrode.

Pyrex glass tubing, 4 mm. in diameter, was

drawn to a taper, then a short piece of 25 gauge platinum
wire was sealed into the glass allowing 4 ram. of exposed
wire to protrude from the sealed.end,

A mercury-copper wire

contact was made in the usual way with the platinum wire
inside the tubing.

This electrode was simply forced into

the soil to the desired depth.

The large calomel cell

ground-glass Joint was slightly but firmly pressed into the
surface of the soil to make good contact with the moisture

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

80.

Figure 15.

Schematic diagram of the modified Raney
diffusion probe used during the second
greenhouse experiment

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission.

Exhaust

-Intake

3 /8 “ Hex nut

Two hole rubber stopper

Scale 1 : 1

Collar

1 2 “ In length

5000“ OD (outer sleeve)
ID (outer sleeve)
^375” OD (inner tube)
3675” ID (inner tube)

075“ (^9 ) dia. of diffusion
^10“ OD (inner tube)
Por-ts

Rubber seal

Plastic tip

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

81.

Figure 1 6 .

Schematic diagram of soil probe diffusion
chamber connected to the charging and analyser
system

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

Beckman Oxygen Analyser
Model D

Silica gel water absorber

Squeeze
bulb

3 way stopcock A

Directional Flow Valve

3 way stopcock B
Directional Flow Valve .
I-- 1
__________

_______

Expansion chamber
Soil Probe
Diffusion Chamber

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

_

82.
films of the soil.

The lead from the calomel cell was

connected to the positive pole and the lead from the microelectrode to the negative pole of a simple, inexpensive
instrument that maintained a voltage of 0,8 volts across
the electrodes and measured the current flow when the
circuit was closed.

A schematic diagram of the field instru­

ment used in this experiment is pictured in Figure 12.

Power

was supplied by a 1 .5-volt dry cell and the desired voltage
was determined by the position of the slide on the 2000-ohm
slide wire resistor.

Current flow was measured directly in

microamperes on a Model 50, 0 to 50 microampere D C micro­
ammeter made by Meters Inc., Indianapolis, Indiana.
In operation both keys on the Field Instrument were
closed simultaneously and diffusion allowed to proceed for
five minutes before a reading was taken.

It was felt that

only after this period of time was the diffusion rate to
the microelectrode representative of the oxygen supplying
power of the soil near the electrode.

Assuming linear

diffusion, an approximate calculation can be made from
equation 9» graphed in Figure 11, of the distance from the
electrode surface to the “diffusion gradient boundary" at
various times after the circuit is closed.
five minutes, this boundary Is between

k

At the end of

and 5 Qini. from

the electrode surface providing no gas liquid interfaces
have been reached.

With a maximum aggregate size of 3.2 mm.

)
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

83.
It seemed that a five-minute diffusion period was suffi­
cient.

In practice it was found that the change in diffu­

sion rate was very slow after five minutes and became
almost steady within a three-minute period.

This is

demonstrated in Figure 17, developed from data in Table 8.
Harvest weights of the tomato plants were made at
the end of the five-week test period.
Discussion of results.

The curves plotted in Figure 18*

for the three aggregate separations, show the oxygen diffusion
rates as found by the platinum microeleotrode at the indicated
depths.

Each point is the mean of one measurement obtained

in each of nine replications.

These microelectrode data

are treated on an aggregate range basis irrespective of
treatment.

This would give a total of twelve replications.

Three of the replications of aggregate range were used in
preliminary studies with the microelectrode and are not
reported here.
The similarity of shape of the three curves is striking.
Since the porosity changes very little (Figure 19) between
the surface and the eight-inch depth for any one soil, a
change in oxygen concentration must be the explanation.
Diffusion theory predicts that the shape of the curves must
be due to the greater rate of change in the concentration
*The remaining figures and accompanying tables are developed
from the complete data tabulated in Appendix 2 with
Analyses of Variance,

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

84

.

TABLE 8
THE COMPARISON OF CURRENT
(OXYGEN DIFFUSION RATE TO A PLATINUM MICRO-ELECTRODE)
WITH TIME AFTER DIFFUSION BEGINS AT THE 2" DEPTH IN
THREE DIFFERENT SIZED AGGREGATE RANGES OF A BROOKSTON CLAY LOAM
SOIL. VALUES OBTAINED DURING A SECOND GREENHOUSE EXPERIMENT

Ti m e
(s e c , )

Microamperes

3.2 - 1.6 mm.

1.6 - 0.8 mm.

<0. 8 mm.

10

14.0

14.0

1 3 .0

20

13.4

1 3 .2

1 2 .5

30

1 3 .0

1 2 .5

1 1 .5

60

1 2 .2

1 1 .3

9.9

120

11.0

10.2

7.9

180

10.2

9 .5

6.9

300

9 .5

S.5

5.8

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

85.

Figure 17.

Current-time curves obtained in three different
sized aggregates of a Brookston clay loam soil
by the platinum micro-eleotrode

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

o
o

03

o

CM
CO

Q
%

(H

O

CM

VO

OO

iH

O

o

§
H
W

O
O

CM

C3

B9a9doii9oaoTH

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

86.

gradient near the surface and decreases with depth,
approaching a constant rate of change at the four—inch
depth.
-Figure 19 shows the relationship between diffusion
rate and porosity as a function of depth for the three
aggregate ranges.

These curves were partially developed

from the data presented in Figure 18,
Penman (32) and more recently Blake and Page (2),
Taylor (46) and Raney (35) have shown that diffusion is
a linear function of porosity in the normal ranges of
porosities found in the soil.

Raney (35) found that below

22 percent porosity there was a curvilinear relationship,
possibly due to the closing of the connecting channels
between the pores.

The dashed curves in Figure 19 relate

porosity to diffusion rate at the indicated depths*
results are in agreement with Raney's data.

These

Above a poro­

sity of about 20 percent the curves approach linearity
while below 20 percent there is a rapid drop in the
diffusion rate with decreasing porosity.

This relationship

was observed at all soil depths.
The great drop in diffusion rate in the surface two
inches is indicated by the greater distance between the zeroinch dashed curve and the two-inch dashed curve in compari­
son to the distance between the lower two-inch increments.
It is noticed also that the slope of these dashed curves

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

87

.

TABLE 9
THE RELATION BETWEEN SOIL DEPTH AND OXYO-EN DIFFUSION RATE
AS MEASURED BY A PLATINUM MICRO-ELECTRODE
AFTER FIVE MINUTES DIFFUSION

Depth
(In. )

gm. X 10 ®/cra.2/min.
0.8 - 1.6 mm.
< 0,8 mm.

1.6 — 3*2 mm.

2

33. B

84.3

91

4

20.2

73.8

73.8

6

6.2

55.5

59.8

8

2.2

43.8

51. B

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

88

Figure 18.

The relation between oxygen diffusion rate
and depth of soil in three different sized
aggregates measured by the platinum mioroeleotrode after five minutes diffusion

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

# » < 0.8

1 ,6"mm.- ACKÎRE»ATES
0.8 tara. ACKJREaATES
ram. AOOREOATES

120

100

s
C B
O O•
■H

80

<0 O'

3

to

%-*CO
■H I

O O 60
c
«> K
to
>
* 01
K E
O

q)

bC

20

0

0

2

4

6

Depth of Soil in Inches

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

8

10

89.
(when nearly linear) decreases with depth.

Diffusion

theory predicts this phenomenon since with decreasing depth
oxygen concentration decreases.

Thus Increased porosity

has a decreasing effect on diffusion rate and at aero concen­
tration of oxygen the slope of the line would be zero.
The solid curves in Figure 19 Indicate the diffusion
rate of oxygen In the individual aggregate ranges as a
function of porosity.

Again it Is shown that the diffusion

rate falls rapidly with little change in porosity until the
water table Is approached.

This may be due to the lowering

of oxygen concentration, with decreasing depth, due to
biological activity.
Figure 20 summarizes graphically the mean results
obtained at the eight-inch depth by the four different
methods used to characterize aeration.

These mean values

were calculated for all treatments and temperatures
compounded.

Results of analysis of variance showed no

significant dlfferenOes In temperature in the Raney diffu­
sion method or the redox potentials.

Temperature did, however,

have a significant effect on the percent oxygen In the soil
atmosphere.

All microeleotrode measurements were made at

75° F.
By visual inspection of the data, differences of high
significance are apparent between the aggregate ranges for
the percent oxygen In the soil atmosphere and the data

R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w ith o u t perm ission.

90.

TABLE 10
THE RELATION BETWEEN POROSITY AND OXYG-EN DIFFUSION RATE
AS AFFECTED BY SOIL DEPTH AND AGG-REG-ATE SIZE
Depth
(in. )

Porosity

gm. x 10 ®/om.^/min,

1.6 - 3.2 mm.
0

.2
4
6
8
10

.322
.320
.317
..315
.305
.297

128*
92
74
6l
50
39*

0.8 - 1.6 ram.
0

2
4
6
8
10

.280

120 *

.270
.262
.250
.24o
.224

83
67
54
44
33*

<0.8 ram..
0
2
4
6

8
10

.050
.043
.038
.035

.028

■

70*
34
21
10

3

.024

*Extropolated values

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

91.

Figure 19.

The relation between oxygen diffusion rate
and depth of soil in three different sized
aggregates as a function of porosity measured
by the platinum micro-electrode after five
minutes diffusion

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

^

1 4-0

X

o

e

0

3.2 - 1.6 mm. AGGREGATES
1,6 - 0.8 mm. AGGREGATES
ram. AGGREGATES /
<0.8
/

t

120

100
«

c
M *H
M .0

ë BO
O
M

ta

O*

(B

b GO

S s
g k 6o
M œ

g
I
c5
^0

20

Aggregate size
Depth In soil

ram.
Inohee
J

10

.20

30

.40

FRACTION OF TOTAL VOLUME
OCCUPIED BY AIR

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

50

92
obtained by the Raney diffusion method.

.

Treated statisti­

cally for analysis of variance the platinum mlcroeleotrode
data were highly significant between the aggregate ranges
but for the redox potentials there was no significant diff­
erence between the two larger aggregate ranges.
It Is of Interest to compare the results of the two
different methods used In measuring oxygen diffusion.

For

the two larger aggregate ranges the differences In rates of
diffusion through the gaseous phase (D/D^) were more accen­
tuated than the differences In the diffusion rates through
the liquid phase as determined by the microelectrode.

This

relation may be due to the fact that the data for the D/Dq
were obtained at different temperatures some of which were
higher than

F.

while the micro electrode measurements

were all taken at 75° F.

A rise In temperature would tend

to accentuate the differences In the diffusion rates between
the aggregate ranges.

In the consideration of the finer

aggregate range, however, diffusion through the liquid phase
was relatively much slower than through the gaseous phase
despite the effect of the temperature differences.

This may

Indicate that the thickness of the moisture film Is of more
Importance In controlling the oxygen supply to the root
surfaces, at lower porosities, than In the higher porosity
ranges.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

93.

TABLE 11
MEAN VALUES OF ALL MEASUREMENTS MADE BY
FOUR.DIFFERENT METHODS OF CHARACTERIZING- SOIL AERATION CONDITIONS
IN THREE DIFFERENT SIZED AG-G-REOATE RANGES AT THE 8» DEPTH OF
BROOKSTON CLAY LOAM SOIL DURING A SECOND GREENHOUSE EXPERIMENT

(A)

Mlcro-electrode

Oxygen Diffuelon'

<0.8 mm.

(B)

(C)

2.2

0,8 - 1.6 mm.

43.8

1.6 - 3.2 mm.

51.8

Raney Diffusion Probe
<0.8 mm.

(D/Do)
.154

0.8 - 1.6 mm.

.395

1.6 - 3,2 mm.

.651

Redox Potentials
<0.8 mm.

(gm. x 10" /cm.“/mln.)

(Millivolts)
-421

0.8 - 1.6 mm.

1-469

1.6 - 3.2 ram.

4-497

(D) Percent Oxygen In Soil Atmosphere

0.8 - 1.6 mm.

3.5%
17.7%

1.6 - 3.2 mm.

IS.7%

<0. 8 ram.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w ith o u t perm ission.

9^.

Figure 20.

Mean values of all measurement8 made at the
8" soil depth by four different methods of
characterizing aeration during the second
greenhouse experiment

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

80

.8

Oxygen Diffusion

Oxygen Diffusion
Raney's Method

.7

70

.6

6o

.5

50

gm. X 10“®/ora^/mln.

^0
.3

30

.2

20

.1

10

1
0

<

0.8

1.6

-

0.8

3.2

-

1.6

<

AOORmATE RANGE (mm.)

%

0.8

1.6

-

0.8

3.2

-

1.6

AGGREGATE RANGE (mm.)

Oxygen in Soil Air

Redox Potential
(rallllvoltp)

f6oo
%

f400

15
(0.

4-200
0

10

-200

-400
-6oo

<

0.8

1.6

-

0.8

3.2

-

AGGREGATE RANGE (mm.)

1.6

<

0.8

1.6

-

0.8

3.2

-

AGGREGATE RANGE (mm.)

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

1.6

95.
The comparison of the static measurements, the per­
cent oxygen in the soil air and the redox measurements,
give added evidence that at the lower porosities the oxygen
supply- is considerably reduced in the liquid phase on a
relative basis to that in the gaseous phase.

Reducing

conditions were quite pronounced in the liquid phase of
the finer aggregates as indicated by the very low redox
potential.

The fall of oxygen, however, in the soil air was

not as pronounced on a relative basis.
While there is a significant difference between the
oxygen content in the soil air for the two larger aggregates,
there is no difference between the redox potentials obtained
in the two larger aggregates.

Evidently, as stated before,

the reduction of aeration conditions have to be quite
drastic to reflect significant differences in redox potentials.
Since the reduction of some compounds that may be toxic in
the reduced state is strongly dependent upon the redox status
of the soil, the evidence presented could be used to reason
that, for some plants at least, the lack of oxygen per se
may become a critical factor to plant growth sooner than the
effects of toxic reduced compounds.

The response in plant

growth in this experiment cannot throw any light on this
problem, however, because there was no significant difference
between the plants grown on the two larger aggregates, or
between plants grown on the treated soils.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission.

96.
The effect of aeration on plant growth in the three
different aggregates is illustrated in Figure 20.

The rate

of plant growth was slower and final yields were lower when
grown on the finer aggregates than when grown on the two
larger aggregates,
Taylor (4?) has suggested that an approximate value

of

D/Dq

=

.Ill

for the oxygen diffusion rate in the

gaseous phase may be a critical point where plants begin
to suffer from lack of aeration.

This experiment suggests

a value between .3 and ,4 for tomatoes at 75° F, if the
eight-inch depth can be taken as a proper sampling level.
As far as the oxygen diffusion rate in the liquid phase is
— ft

concerned, values between 3 0 and 40 grams times 10”

per

square centimeter per minute at the eight-inch depth may
be critical for tomato plants at 75° F.

Summary of results.

1.

Measurements made in the

soil, in situ, based on the principle of the platinum
microelectrode, give promise of furnishing a new, simple,
rapid and inexpensive method of determining the rate of
oxygen supply to an environment similar to that in the
liquid film surrounding an actively respiring plant root.
2,

Evidence has been present indicating that the

factors controlling the diffusion rate in the gaseous phase
of the soil extend into the liquid phase as well*

The

additional factors that are unique in their effects on

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

97 .

TABLE 12
AVERAG-E WEEKLY HEIGHT OP TOl^ATG PLANTS
GROWN ON THREE DIFFERENT SIZED AGGREGATE RANGES OF
A BROOKSTON CLAY LOAM SOIL IN A SECOND GREENHOUSE EXPERIMENT
FINAL AVERAGE HARVEST WEIGHTS ARE ALSO CITED

Weeks
3.2

- 1.6

Height in Centimeters
1 . 6 - 0 . 8 ram.
ram.
< 0 . 8 mm.

1

10.6

9.4

2

15.6

14.1

11.25

3

28.8

27* 2

20.4

4

47.6

45.0

34.0

5

62.8

6 1 .8

41.25

Harvest
Weights

225*6 gm.

216.6

8.8

gm.

56.0

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

gm.

98

Figure 21.

.

Average growth curves and final green weights
of tomato plants grown on three different
sized soil aggregates during the second green­
house experiment

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission.

300

70

60

200

z
M
50

&
M
EZ
w

100

3.
AOGRKÜATE SIZE IN MM

Ü

z
n
30
s
cb
d)
B
w

20

10
•

0

0.8

<

X

X

X

2

3

^

WEEKS FROM PLANTING DATE

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

mm. AGGREGATE
ram. AGGREGATE
ram. AGGREGATE:

99.
oxygen diffusion in the liquid phase of the soil are little
understood at present.

Future research into this matter is

needed.
3.

There is evidence that below 20 percent airfree

pore space the straight line relationship between porosity
and diffusion is not maintained,
4.

The thickness of the moisture films may be of

more importance in controlling the oxygen supply to the
plant roots at lower porosities, than in the higher poro­
sity ranges.
5.

The comparison of all four methods used to

characterize aeration shows the same general trends depending
on the size of the soil aggregates.

However, quite drastic

changes in aeration conditions have to be attained before
significant differences are reflected in redox potentials.
6.

Some value considerably greater than

D/D^ = .111

was a critical point where the tomato plants in this experi­
ment began to suffer from a lack of oxygen. A value between
O
30 and 4o grams times 10* per square centimeter per minute
may have been the critical point at the eight-inch depth
for the oxygen diffusion rate in the liquid phase.
7.

The effects of added increments of iron and

manganese on plant growth were statistically insignificant.

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

100.

IV.

LOOKING AHEAD

-The Influence of physical conditions on p l an t g r o w t h
can be d i v i d e d Into four In tera cti ng attributes:
moistur e content,
potential

a e r a t i o n a n d penetrability.

temperature,

Of these the

concept of soil moisture ha s be e n of great value

In the s t u d y of soll-raolsture p l a n t - g r o w t h relationships.
So far no sound concept of equal u t i l i t y has b e e n found to
q u a n t i t a t i v e l y express
aeration.

soil temperature,

pe net ra b i l i t y or

It has b e e n the purpose of this study to devis e

a q u a n t i t a t i v e m e t h o d of m e a s u r i n g a e r a t i o n In order to
es tab l i s h a sound basis for the deve lop men t of a hypothesis
of soil aeration.

Once a fundamental

concept of practical

u t i l i t y has b e e n established for each of the physical p h e n o ­
mena,

an u n d e r s t a n d i n g of their Inte gra ted effect on plant

g r o w t h can be accomplished.

Not until then can m a n

c o m p le tel y control the soil environment so that plant
yields closely a p p r o a c h their p r o to pla smi c limits.
I n the present study a new m e t h o d of m e a s u r i n g the
rate of oxyg en supply to the p l ant root has b e e n proposed.
The u t ili ty of the m e t h o d was

s u c ces sfu lly t es ted for Its

a b i l i t y to correlate plant g r o w t h a n d a era t i o n conditi ons
where aeration

conditions w e r e k n o w n to be a d e q u a t e In one

case a n d l i m i t i n g In another.

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w ith o u t perm ission.

101.
The way now appears to he open for further research
to separate the direct effects of oxygen deficiency per se
from the effects of compaction,

and the effects of toxic

materials p r o d u c e d in poorly aerated soils.

Also the

effects of aeration on many physiological processes can
he more adequately elucidated.

The uptake of ions and

other problems of plant nutrition,

along w i t h respiration

studies offer fertile areas for further study.

Soil

aeration may indirectly influence plant growth through its
effect on disease incidence.

Future investigation along

these lines is indicated.
As more informstion is gathered on the subject of
soil aeration,

further applications will be forthcoming.

R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.

102 ,

LITERATURE CITED

1-

Blake, Or. R. The aeration of soil a with special
emphasie on the diffusion process in soil air
renewal. Doctor's Dissertation, Ohio State,
University (1949).

2. Blake, G-. R. and Page, J. B. , Direct measurement of
gaseous diffusion in soils. Proc. Soil Scl. 3oc.
Amer. I3 : 37-42 (1948).
3. Blinks, L. R., Darsie, M. L. and Skow, R. K.
Bioelectric potentials in Halicyatis, VII The
effects of low oxygen tension. Jour. Q-en. Phys.
22: 225-279 (1938).
4.

Bousslngault and Lewy. Memoir© sur la composition de
l'air confine dans la terre vegetale. Ann, de
Chemie et Physique 37: 5-50 (I853).

5. Boynton, D . and Reuther, W. Seasonal variation of
oxygen and carhon dioxide in three different
orchard soils during 1938 and its possible
significance. Proc. Amer. Soc. Hort. Scl. 36:
1-6 (1939).
6. Bradfield, R., Batjer, L.P., and Oskamp, J. Soils in
relation to fruit growing in New York IV The
significance of the oxidation-reduction potential
in evaluating soils for orchard purposes. New York
Agr. Exo. 3 t a . Bui. 592 (1934).
7.

Brady, L. J ., Determination of small amounts of oxygen
in gases. Ind. Eng. Chem.. Anal. Ed. 20: 1033
(1948).

8. Buckingham, R. Contributions to our knowledge of the
aeration of soils. U.S.D.A. Bur, of Soils Bui. 25
(1904).
■ '
9. Cannon, W, A. and Free, E. E. Physiological features
of roots, with special reference to the relation
of roots to aeration of the soil. Carnegie Inst.
Wash. Publ. 368 (1925).

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

103.
10.

Davies, P . W. and Brink, P. Microelectrodes for
measuring local oxygen tension in animal tissues.
Rev. Scl. Inst. 13: 52^-533 (1942).

11.

Dennis, L. M.
New York

12.

Forbes, G. 3. and Richter, H. W. The measurement of
oxidation potentials at mercury electrodes. II.
The chromic chroraous potential. Jour. Amer.
Chem. 3oc. 39: 114o (1917).

13.

Hagan, R. M.
solids.

14.

Hannen, F. Untersuchungen ueber den Einfluss der
physikalischen Beschs.ffenheit des Bodens auf die
Diffusion der Kohlensaeure. Forsch. Geb. Agrikult. Phys. 15 : 6-26 (1892).

15.

Hardy, F. Studies on aeration of cacao soils in
Trinidad. Part III Gaseous diffusion in certain
cacao soil types in Trinidad. Tropical Agr. 20:
13-24 (1943).

16.

Hoffer, G . N. Fertilized corn plants require wellventilated soils. Reprint A-1-45. Am. Potash
Inst. Inc., 1155 Sixteenth St., N.W. Washington
6 , D.C. Reprint from Better Crops with Plant
Food.

17 .

Humfield, H. A method for measuring carbon dioxide
evolution from soil. Soil 3ci. 30: 1-11 (1930),

18.

Hutchins, L. M. Studies on the oxygen-supplying power
of the soil together with quantitative observa­
tions on the oxygen-supplying power requisite
for seed germination. Plant Physiol. 1 : 95-150
(1926). .

19.

Karsten, K. S. Root activity and the oxygen require­
ments in relation to soil fertility. Amer. Jour,
Bot. 26: 855-860 (1939).

20.

Keen, B. A. The physical properties of the soil.
Chapter X The soil atmosphere. Longmans, Green,
and Co., New York (1931)-

G-as A ^ l y s i s . The MacMillan Company,
(1913) .

Movement of carbon disulfide vapor in
Hilgardla l4: 83-118 (1941).

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

104

.

21.

Kirkham, D.
Field method for determination of air
permeability of soil in its undisturbed state.
Proc. Soil Soi. S o c . A m e r . 11: 93-99
(1946).

22.

Kolthoff, I. M. and Lingane, J. J.
Polarography.
Interscience Publishers,. Inc., N evr Y o r k
(1946 ) .

23.

Laitenen, H. A. and Kolthoff, I. M,
Voltammetry w i t h
stationary microelectrodes of platinum wire.
Jour. Phys. C h e m . 4 5 : I 0 6 I
(1941).

24.

Leather, J. W.
Soil gases.
Memoirs Dent. Agr.
(Chem. Series)
4: 83-13^ (1913)•

23.

Lemon, E. R.
The effect on plant growth of manganous
and ferrous ions ag related to soil aeration.
Master's Dissertation, Cornell Un i v e r s i t y
(1948).

26.

Lingane, J. J. and Pecsok, R. L.
Preparation of stan­
da r d chroraous sulfate or chroraous chloride
solutions of determined concentration.
Ind. E n g .
Chem.. Anal. E d . 2 0 : 423
(1948).

27.

Lundegardh, H,
C a rbon dioxide evolution of soil and
crop growth.
Soil S c i . 23: 414-433
(1927).

28.

Maclnnes, Duncan A.
The principles of electr o c h e m i s t r y ,
R e i n h o l d Publishing Corporation, New Y o r k
1939.

29.

Marsh, F. W.
A laboratory apparatus for the measure­
ment of carbon dioxide evolved from soil.
Soil
Sci. 2 3 : 2 3 3 -2 6 1
(1 9 2 8 ).

30.

Osterhout, W. J. V.

plant cells.

India

Electrical phenomena in large

Physiol. Rev. I6 : 216-237

(1936).

31.

Peech, M . a n d Boynton, D.
Susceptibility of various
New Yo r k orchard soils to reduction upon w a t e r ­
logging.
New Y o r k Agr. Exp. Sta. B u i . 6 6 7 :
1-20
(1 9 3 7 T:

32.

Penman,

H. L.

Q-as and vapour movement in the soil.

I The diffusion of vapours through porous solids.
Jour. Agr. Scl. 30: 438-462 (l94o) II The
diffusion of carbon dioxide through porous solids.
Jour. Agr. Soi. 30: 370-381 (1940).

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

105

.

33-

Quispel, A. Measurement of the oxidation reduction
potentials of normal and inundated soils.
Soil Sci. 63: 265-275 (1947).

34.

Raney, W. A. Field measurement of oxygen diffusion
through soil. Soil Sci. Soc. Amer. Proc. l4:
61-65 (1949).

35.

Raney, W, A. Oxygen diffusion as a method of
characterizing soil aeration. Doctor's
Dissertation, Cornell University (1950).

36. Rom ell, L. G-. As quoted by Handb. der Bodenlehre;
Julius Springer, Berlin. VT : 315. 1930.
37. Rom ell, L . G-. Mechanism
5 (n.e.): 374-384
38. Rom ell, L. G-.
factor.

of soil aeration.
(1935).

Ann. Agon.

The aeration of soil as an ecological
Int. Inst, of Ag. Bui. 1257 (1922).

39. Rom ell, L. G-. Soil aeration. Int. Rev, of the Sci.
and Practice Agr. N.S. 1: 285-300 (1923).
40.

Russell, E.J. and Appleyard, A. The atmosphere of the
soil; its composition and causes of variation.
Jour. Agr. Sci. ?: 1-48 (1915).

41. Russell, M. B. Soil aeration.and plant growth.
graph, Agronomy Department, Cornell Univ.

Mono­
(1950).

42.

Smith, F. B. and Brown, P. E. The diffusion of carbon
dioxide through soils. Soil Sci. 1^: 413-423
(1933).

H-3.

Starkey, R. L. and Wight, K. M, Anaerobic corrosion of
iron in soil. American G-as Assoc. 420 Lexington
Ave. , New York 17, New York (1945).

44.

Stone, H. W. and Skavineki, E., R. Quantitative absorp­
tion of oxygen; Critical factors in the applica­
tion of acid chrômous solutions. Ind. Eng. Chem.,
• Anal. Ed.. 1?: 495 (1945).

45.

Sturgis, M, B. Changes in the oxidation-reduction
equilibrium in soils as related to the physical
properties of the soil and the growth of rice.
La. Agr. Exp. Sta. Bui. 271: 1-37 (1936).

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

106.

46.

Taylor* S. A. Oxygen diffusion in porous media as
a measure of soil aeration. Soil Sci. Soc.
Amer. Proc. l4: 55-^1 (1949).

4?,.

Taylor* S. A. Soil air-plant growth relations with
emphasis on means of characterizing soil
aeration. Doctor's Dissertation* Cornell Univ.
(1949).

48. Tenderloo, H. J. C .* Vervolde, G-. J, and VoorsouiJ*
A. J. Z. Electrochemical behavior of ion exchanging
substances. Potential measurements of plant roots.
Rec. Trav. Chim. 69: 97-104 (1944).
49. Vine* H.* Thompson* H. A., and Hardy* F.
Studies on
aeration of cacao soils in Trinidad. Part I.
Tropical Agr. 19: 175-181 (1942).
50. Vine* H.* Thompson* H. A.* and Hardy, F.
Soil air
composition in certain cacao soil types in
Trinidad. Part II. Tropical Agr. 19: 215-223
(1942).

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

107

APPENDIX

1

Tabulated results of measurements made during
a preliminary greenhouse experiment, with Analyses
of Variance

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

CD

■D
O
Q.
C

g
Q.

APPENDIX 1

TABLE 1

HARVEST WEIGHTS OP SOYBEAN PLANTS

■o
CD

(/)
o'
=5

Aggregate Range
8

5
cS'
z
i
=3
CD

C
p.

1

2

3

4

1

Check

165

255

230

190

192

Iron

180

190

l40

122

170

ReplJ.cation
■p
0}
e
p
s
k

12.8 - 6.4 mm.

6.4 - 3.2 mm.

<3.2 mm.

3

4

1

198

195

210

202

150

188

170

135

2

2

3

4

195

200

210

170

109

181

CD

■o

I
c

a
O
3
■o
O

Source

&

Total

o

c
■o
CD

(
/)
œ
o'
3

ANALYSIS OF VARIANCE
Degrees of Freedom

Sum of Squares

Mean Square

23

25418

1105

Aggregate Range

2

420

210

Treatments

1

11971

11971

Interaction

2

715

357

18

12312

684

Error

Ratio

.307
1.75

.522

H
O
O)

CD

"D

O
Q.

APPENDIX 1

C

8
Q.

TABLE 2

PERCENTAGE OF CARBON DIOXIDE IN SOIL AIR
■D
CD

CO
o'
3

C/)

"8O
ci'

Replication
Week

3

4

1

2

3

4

1

2

3

4

o
Xi

O

3.5
2.8
3.8
2,8
11.0

5.0
4.8
4.0
2.5
12.5

7.0
3.0
2.4
2.5
5.5

1.0
1.0
2.0
1.0
0.5

1.0
1.0
1.6
1.8
2.2

1.8
1.0
1.3
1.5
2.2

1.2
1.0
1.8
2.0
3.2

1.2
2.0
0.8
1.2
2.2

1.0
1.4
0.8
1.2
1.5

0.5
0.8
1.8
1.5
2.2

1.0
1.8
1.0
1.2
3.0

4.0
1.6
3.0
2.2
5.5

3.0
5.5
4.8
2.0
2.5

3.5
3.8
1.8
1.5
2.0

3.0
1.4
1.4
1.0
2.1

1.4
2.2
1.2
1.8
2.5

1.8
2.0
1.2
1.6
2.2

1.0

c

1
2
3

1.0
1.0
1.6
1.0

1.0
0.6
1.0
1.5
1.8

1.0
0.8
1.4
1.8
2.2

1.0
1.0
0.8
0.8

1.0
1.0
1.9
1.5
3.0

3"
CD

O
Q.

2

12 . 8 - 6 .4 mm.

3.0
3.8
2.6
4.2
3.2

3.
CD

1

6.4 - 3. 2 mm.

1
2
3
4'
5

<D

"D

<3.2 mm.

Aggregate Range

g
hi

H

C

k
5

a
O
■D
O

2.2
1.2
1.8
2.5

1.8

1.8

3

CD

Q.

ANALYSIS OF VARIANCE
Source

Degrees of Freedom

Total
■D
CD

(/)
C/)

119

Sum of Souares

Mean Square

Aggregate Range

2

319.95
124.76

Weeks

4

37.76

9.44

Interaction

8

20.24

2.53

105

137.19

1.31

Error

**Slgnlfleant greater than

Ratio

62.38

47.618**
’ 7. 206**
1.9
»
<

110,
o o
pvooo M>vo

vr\vf\o o O

0
0
o o o
,

O O vc\o O
CM vr»o O O t n

o

"H
P
cd
oü

VO
1
oo
*

C\J
rH

CM

M

O O '■oo o
P OvO P P
vorovr\vr\it

o o o o o

o o o o »o

O O VAO O

S > P 5 S -S -

O O V^O o

CQ
S
ê

g
s
G
CVJ

d

ry

o o in o o
cO(T\vr\vn

1

w

i

M
>
M

E4

VOo vr\ VT\o
g

g

g

CM

o o o o o
to 03 O ov o
^ ^ vo^ VO

O VO o o o o o
p
vnvnvfy g g g g g

O

H

S

xr\xovoiO o
0 \ O P CM O
^ u^v^xr\vr\

O

00 P

VOO

as ov.? o o
^ ^-3- VOVO

S
g

vnvr^o O vr^

o o o o
VO V O ( S .

o

VO
-et--M
"

_:t

B

CM

■

0

o vr\ o O
(T'lVOO

O VOO o o
C
OC O VOVOO
^

^

-d -

P4
g
1

CM
•
cn

vf\vr\0O VT\vn
CM
1 }-

V

-j

.z)" ^

O VOOVO O
rH

CM C ^ V O CM O
^
-'Ù-

o o o p
VOVOv o v p ^

o
^

^

j -

v o p o o o
C O - = t '- j ' t o CM
_d"
^

ol

03

«tH
0

0
o
OJ «H
P
p
cd
0Î
bO o
Qi p
k
H
bc A
bO ®
cc;
<

CM c n ^

v r\

k
4-,
O

m
0)
o

O
o
pVO
CO

o
VO
c*VO

o
cco
VO
rH

o
o
CM
CM
«H

WL
rH
es CvJ --t CO
O
rH
r4
rH

o

03

Qj
g
%
cu
ü
<

<0
n

0

C

+J

bd

o

R eproduced w ith perm ission o f the copyright owner.

O
CO
VO
CM
03
rH

Q)

(O
UOJI

o

CO
rH
(H

P-l

<D

3 îO 9 q 0

o

03
Ov
P

0)

a

0)

o

§

rH CMCO-z^ VO

5:

O

(O VO
VO CO
rH CO
I— 1

1

Q)
H

rH

to
ÏH
q

§

VO
rH

[V_

VO
«
rH

CQ

ë
M

CQ

CM
CM

0)
k

3

cr»

X
M

M
O

CO 0 \ 0

VO

^-3-

VO

X

g

•

M
w

o

133
Ü
W
0)

1.

<X>

M

S’

•P

a
0)
p
cd
bO
o

O
*H
P
O
o
S
to
u
Î4
to
o
M
a
i
•p bo <u p
Fr
D 0
O bo <
S
V
v
M W
IH <

F urther reproduction prohibited w itho ut perm ission.

0
«5
O
P

•H
F
bo
P
03
*
*

■o

I
§
Û.

APPENDIX 1
BIOELECTRIC POTENTIALS

TABLE 4
(NEG-ATIVE MILLIVOLTS)

"O
CD

Aggregate Range
Replication
Week

8
■D

Ai

o

C3.
û

(D

A

Ü

6.4 - 3 .2 mm.

< 3.2 mm.

C/)
C/)

12.8'- 6.4 mm.

1

2

3

4

1

2

3

4

1

2

3

4

1
2
3
4
5

70
85
92
115
90

55
78
85
80
75

70
85
90
90
75

45
85
85
90
60

38
70
80
85
80

42
78
82
70
70

45
70
65
60
60

15
35
75
60
70

75
64
90
100
60

50
70
55
65
20

30
55
52
85
80

85
90
90
70
40

1
2

60
85
65

40
80
65
85
90

80
81
95
90
60

77
70
92

35
50
80
50
90

15
55
50
60
70

20
67
85
70
50

55
70
85

20
10

22
30
84
.40
50

42
25
45
60
70

20
68
75
60
70

CD

3.
3"
CD

C
O

CD

"O
O
Q.
C
a
O
3
"O
O

U

H

3
4
5

100
110

80

50

80

45
25

80

0

ANALYSIS OF VARIANCE
Mean Square

Source

CD

Ratio

t.
1—k

3
"
0
C

"O
CD

3

119

54305

Aggregate Range

2

12643

Weeks

4

Interaction

Total
.

6321.5

40.42**

2392

598

3.82**

8

22844

2855

105

16426

3

(/)'
C/Î
o'
3

Error

^^Significant greater than

156.4

18.25
H

H
H

- j-

m

oo

ON
iH

On
rH

o

CM

O

a \

C3N
iH

VfN
•
ON
rH

CTn
rH

C7N
rH

O

00

O

UN

CM

rH

ON
iH

ON
1— i

ON
rH

ON
rH

CT\
rH

iH

00

CO

OO

00

UN

0C>
H

Cn .
iH

ON
rH

00
H

CTn
iH

ON
rH

O

O

UN

ON
iH

O
iH

On
rH

CNiH

q
CM

OO

VTN

O

O

CM

o<— 1

ON
H

C7N
H

OO
rH

CTN
rH

VT\

CM

CM

00

h J-

OO
rH

ON
rH

OrH

00
rH

ON
iH

H

m

O

iH

i

On
1— 1

iH

O
CM

ON
rH

vr\

NO

M-\

M-\

cn
rH

NTN
1— 1

00
iH

!>.
rH

00

m

CM

I

O
to
g

ON
1— 1

V)

I

On
H

CM
O
CM

CTN
iH

CM
rH

O
CM

ë
2a

UN

I

I

112,

CM

o\
rH

00
H

cü
<

ON
iH

CM

cn

(n

ON
rH

GO
H

N

•H

UN

»

O

cû

m

CO

VÛ
I

O

o

H

oo

00

o.
M

NO
iH

OiH

ON
rH

CM

O

NO
H

On
rH

00

00
*
CO
rH

NO
iH

00
1— 1

CM

CM

CN.
iH

00
rH

tv
H

O-

00
•
NO
H

UN
rH

CM
rH

H

N

V
00
rH

NO
rH

H

CM

(D

g"
S
«D

d
o
<H
-P
(d
o
A

fS

cn

-ri-

VN

C»N

UN

M

0>
a>

3[oaqO

u o a i

R eproduced w ild perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.

113.

APPENDIX

2

Tabulated results of measurements made during
a second greenhouse experiment, with Analyses of
Variance

R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.

CD

■D
O
Q .

C

APPENDIX 2 TABLE 1

g
Q .

GROWTH RATES OF TOMATO PLANTS

"O

(Centimeters)

CD

C/)
C/)

Aggregate Range
Replication

1.6'- 3.2 mm.

0.8 - 1.6 mm.

<0.8 mm.
1

2

3

4

1

2

3

4

-

10
14
30
51
69

11
17
33
54
79

9
13
25
38
52

—
—
-

8
11
18
39
57

11
16
32
51
65

12
19
33
53
66

-

10
15
27
42
60

9
12
24
41
57

_
—
—
—
-

11
16
32
49
66

8
14
28
41
48

10
14
27
45
66

—
-

-

—
—
—
-

7
10
18
36
50

9
14
22
36
46

—
—
—
-

8
14
29
51
65

9
14
28
43
58

12
18
31
49
67

—
—
—
-

11
16
28
52
69

13
18
33
56
76

12
17
29
43
53

—
—
-

1

2

3

id
w

1
2
3
4
5

9
12
20
28
34

11
17
28
40
46

10
9
—
—
-

•p
c
o
sG
•po
k
oH
u
E4

1
2
3
4
5

6
6
»
—
-

11
14
23
32
39

6
7
—

1
2
3
4
5

10
11
_
—
—

8
—
9
—
-

4

CD

Week

■8D
CQ'

si

3.

3"
CD
CD

■D
O
Q .

C

a
O
3
"O
O
CD
Q .

(1
gO
% «0
G (D

aG

-

ANALYSIS OF VARIANCE
FOR PLANTS GROWN ON THE TWO LARGER AGGREGATE RANGES

■D
CD

C/)
C/)

—

Source

Deerrees of Freedom

Total

Sum of Souares

Mean Souare

17

7124

Aggregate Range

1

84

84

Treatments

2

880

440

Interaction

2

155

77

12

6005

500

Error

Ratio

.168
,88
.154

H
fr

(D

"O
O
Q.

APPENDIX 2

C

g
Q.

TABLE 2

PERCENTAG-E OXÏGEN AT 8“ DEPTH
■D
CD

O)
o"
3

Aggregate Range

3

Pemp. 86°F.
73°
86°

C/)

Renlloatlon

CD

■8D
(3û"‘
i
3

M0

<D
0

0.8 - 1.6 mm.
1

2

3

1,6 - 3.2 mm.

4

1

2

4

3

16.3
17.4

16.8
18.0

17.2
18.6

16.8
18.3

18.4
19.3

18.2
18.9

18.0
18.8

18.2
19.2

16.8

18.4

18.8

18.8

19.5

19.4

19.2

19.7

17.8
18.5

16.4
17.2

18.2
18.5

17.5
18.5

17.9
18.8

18.4
19.5

18.2
18.9

18.3
19.2

18.7

17.0

18.2

18.8

19.3

18.8

19.1

19.3

16.7
17.2

16.8
17.8

16.8
18.2

16.0
17.8

16 :
18.5

18.0
18.8

17.3
19.0

18.2
19.0

19.4

18.8

18.7

18.0

19.3

18.8

19.4

19.4

CD

86°
3.

G

3"
CD
CD

■D
O
Q.

llo

H

C

a
O
■D
O
3

CD

Q.

86 °

I
à Q)
6003
G
Q)
q) C
S

86 °

75°
740

ANALYSIS OF VARIANCE
"D
CD

Sum of Squares

(/)
C/)

Total
Treatment
Temperature
Interaction
Error
f

Significant beyoncL

71
2
2
4

‘

63
.

1% level

57.10
0.51
21.71
1.85
32.03

Mean Square
.8042
.2550
10.8550
.4625
.5084

Ratio
-

.501
21.351**
. .909

H
H

116 .

(M

en

vn

4 4

4

4 4

CM vn
cnat
vn\n

vn
CM
VO

o o
CD adad VT\

4 4

I

Vû

CM

03
VO

5
en
H

â
g

00

H
I

+3

03

O
>

O

rH

en

4

4 4

ad-^

vnvn
CO ov
-d- ad-

4 4

4

4 4

vOrH

xno
ad* \f\

o
VO
vn

O-rH

4 4

4

4 4

CM vn
VO rH
vnvv

O
o
VO

rH Cv-

4

o o

enad-

vno

4 4

o
at
at
4

vn

COvp
C3\jp
ad- vn

vn
vn
VO

Cv-

vn
4

O

^ 2

4 4

vnvn
enov
ad-ad-

vn
vn
vn

o vn
vnco

4

VO

O
OV
vn

4 4

4

O
adcn

vnvn
00 CM
ad- vn

CM
1— I
vn

4

4

00 vn
O-VQ
cnar
4 4

4 4

vno
t>- cn
ad- vn

rH

O
C7V
cn

O
o
tn
4

o vn
cncM
ad- ad-

4

o o
oo rH
enai4 4

o
o
vn
4

o
CM
vn

tnw

4 4

O

o o

00
ad- adad-ad-

4 4

vno
rH VO
vnvn
4 4

4

O
CM
CM

CO vn
c^gv
^ ad-

O .
ov
vn

ê

rH rH O

O O C

ql
pi

en O
^
o

O O cso M -\rH

rH 03 VO O-vû Cv-VO OC
I>—:à- vn
P-VO c*-

§

4

VO

O-cnrH

\T \0 CV
rH CMC*-

l>—=i-

Q)
k

O
vn
vn
CM
Cvj
VO

^

v
H
-P

4

vno
CM ovnvn
4 4

O

rH

cMcnvc

O

<D
S

o

oo e n

rH CX3-^
rHVO

^ 0 0 vp o - ^
e * 'i^ O iiH CMO-

4

00

CM

4

CVJ

4

4

vnuv
ooadcnvT'.

CM
advn

4 4

4

4

4 4
rH O

O
£Svn
4

4 4

4 4

o p
oadvnvn
4 4

4

0>
u

Dû

O
VO
vn
4

g

s
<p
0

CMVO CMH crv Cv-OOVO

o-ovfH cncno-ovrH
O VO Cv-CM vO vO rH C*^
V O 0 - 0 0 CM 0 0 -d" O - ^

C
Den o

CM

rH rH

r4M3 OV

1

M
Q

03

oco
1 1

1

vnvn
CM CM
ad-ad1 1

O CM

vn

CM vn

ad1

2 ’
^

1 1

1 1

vn
CM
vn
t

CM

vno
CM o
CM ad
1 I

o
o
cn
1

at at
at at
t 1

vn
at
at
1

CM O
CM Q
ad-^
1 1

vn

OO vn

p

1 1

at
1

vn
3

(U

S
<n
GO

CM

vn
CM
ad1

o vn
1 t

O
vn
m
1

1 1

vn
CM
en
1

e
'§
o>
<D
U

O

i—
4

CM

1 1

ad1

CM CM

o
p

1 I

0)

-P

k
o o
o c o
VO cnvo vnaj-

0 0 cv-oo t v Cv-

§
hH
■p
cd

S •ort

<D

j!

1

<P
O
m

rH CM CM^ tn rH cncn
VO

o-

<0

m
P,
n

G)

S

1

00 vn

&

O C O o o
VO cnvo vn^ar
00

t> -c o c v ,o -

<D

o c o O Q
VO cnvo vnat
oo O-OO

•
A
8
03
EH

03

p

-H

o

cd

o

c: p

Q)

8S8U

uoai

(H

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

ü
k

§

03

tg a co
o

c
o

u
3

0) -H p

p p Ha

p

Cd oJ O
bO O

S3 g

(d

u 03 P
p, 03
■ a s
p 03 6 p s m 'Pi'S

o SH 03 a

E-«

EH M

w

117
c^vncnv^o
^ cn-^t -it ^
I
VO

cn

1—1

00 rHVO-^ o
CM
^

rH
^

cnz)^ ^

i il
S
P4

%

1

00 cn-zt c n o
• • • • •
m z t zj- zt

C
OS
<D
S

Cf\00 o o o

cvm-=^ Ov-:t
. . . . .
cncM cncM CM

O V O m-zt CM
# # $ # *
.zt CM cncncn

VO
rH
CM

rH

=

s

C»VV0 rH 00 CO
cncn-d cncn

O CJ'CnrH o
. . . . .
cn
cncM cncncn

o

EH

S

k
«0
u|
CQ

cncM cncncn

CM

X
M
O
g

rHVO cnrHOO
# • • • «
zf cn-ztzzt cn

CMOO-zt m C M
• # ♦ * •
-d- CM cncncn

CO
1

1

cn

cnmcMO-rH
-d" cn-cf- cn-d"

00 rH c n ^ CM
. . . . .
CM cncn cncn

CQ

M

a
a

cn

CM

1—1
0)
g
&

c
o

0>

•H

at
<D

"S

bo O

coco cnrH cn
. . . . .
CM CM cncnc^\

mi>- cnrH Ov
cncM cncncM

oo Ov cnvo cn
CM CM cncncn

m o o cv. cMCJv

cv-co m v o CM

cncM cncncM

cncM cn cncn

cn-^ -=i VO 0 “ vo-j- mcjvo• é • • •
rH rH rH rH rH 1—If—IrH rH rH

CO m r H o o
• • • • •
rH rH CM CM CM

p..CO 00 VO cv.mcn.zt CV.
• • • « •
1—11—1rH 1—11— 1 1—1rH rH <—1rH

rH-zt O O cn

m o o cn-^t cn
. . . . .
cn CM c*’'»cn cn

cncno ov^
. . . . .
rH rH CM rH rH

- ic

CM

r4

CM

en

w

i
g

0

CO
l>-

CM

VO

CM

œ
w

U
g

(W
0
1

cn VO
cn m
(V- rH
orH

m

CM
O
OV

OO
OO

-zt

O
rH

VO
VO
rH

-j-

00

CQ

CQ

M

CQ

1t

i

t

oc

O m o CO rH
cncM cncM CM

CM r-i CM CM rH
H CM VO OO O
. . . . .
CM rH rH rH CM

Ê
Ov

O O-

CM

VO

CQ
Q)

<ü

VO VO cnco [V. 00 zt- cnoo CVO V O O-Ov O
# e * # #
# # # # *
CM rH rH iH CM •i-H rH I—1rH rH 1—1rH rH rH i—1
• 4 • • •
k k k k k
k PcJ k b k
k k k k k
O OO O O
O O C3 o O
O O O C3 O
VO cnvo m z )
VO cnvo m-z} VO cnvo m v o
CO 0-00 o-1>- oo Cv-OO Cv-Cv- GO Cv-OO Cv-OO
4z>
C3
0)

é
d)
EH
aïoôiîo

U

1

I—I

<0

CO

o
V

tN- VO

CM rH cnCM OO
VO C - O O rH ^ C^VrH o c o
« ♦ # » •
• • « • •
• • • • •
z
cnz^z± cn t- cn-zt .zt cn
cncn-^-^i*

CO

S

CM

tV
O-

•H

(D

1
CM
cn

o
4J

O-rHVO O C O
CM-:j^ c n H o o
• • • • *
# # * # #
-ct cn-=t
cn .zt cn-zt .zt cn

.

uoai

a:VU0mq.-Baaj,

8sau
-•eSu-ew

o
Ü
U
CS

o
CQ

rH
at
4Z>
o
EH

R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.

B

ctS

Q)
u

EH

0>
fH
0
+5
at

o
vH
+3
o

U

at

(U
k

a
o
Eh

k
ou
c
M

u
°
M

118.
O O V\v\ O VAO O

W^OVOiH

00 0 O \ O ^

\f\r4

# » * *

o\

M

0 - 0 \ ^

O

O

*
*

o
•
fH
4^

CO

O O ON

vi^ o

vr\-d-

OJ

i r\ 0~\

0 \ 0 \ 0 \ CA

O

v rv rA
• •

rH

•

»

vn
CM

a

OO C*- O WA
♦ • • •
C\! oo ÇVJ o

*

* cn
rH oo VO
ov CM

Q>
O
• •

f-tcovovo

fH

O

O -C A O
• • ' • •

O

u

v n 'JA v n
• • • •

§l
o

o vnvnrH ^ rH O O

fH

JdCM
CM
CM
ii-

CO

a
(0
<D

Jh
<D

CM w v n o
«

\o

m

Î

a

I

•

«

^ £>-vo VO
rH

A

o
•H

«

o oo

vr\

•

•

CM O
•

•

ONC^tVO-

vn vn CMCO
♦

•

•

CM O CM

»

•

•

C ^rH rH fH

I

O CO O - C M O O O

O cn o o

rH 00 rH CO

rH '-f'» rH fH

M
o

0)
k

00

o VOvo vn

o oo -d" ^

mcMoo

m

Vf

« « . »

xnc^-^ vn
o CO O-VO

vnco o o
0 \ CM-:t ^

vnn- vn cm
m o o o

O ^O O N

vncM l>-C n

O-OO O rH

OVO (TV CM
rH

vnOTv iH O

1

I

03

•

» • • •

00
crv

c m o o o

cO
• • • •

o
00
1—1 m
o

%

<

E-i

s;

3

*h

cm h

•

CD o - vn vn

O

^

vn

n- o

iH

o

CO oo vn oo
CM o
o
vo vn
vo
cn
00 CM

ë CO
CQ
>4

B

«

<M

•

•

•

rH c n o (TV
fH rH rH

•

•

•

•

I

%

0>

ê

§
H
S

o vn CMvfA

CO CM o oo

vnvn VO vo

CMoo VO vn

OVCV-Z^ CM

vn CMrH O

H

9-f
o
CQ
0)
<D

CM

+*

O
j
o

A

(S "mm z:C

-

I

CM^ v o oo

<u
o
u
G
o
CO

<D
■p

oJ

&

Q)
H'

CM^ VO 00

;mi q :o .
- 9'T ‘rnm 9"0>
9't
"835 sy""9^W3iB093^

I
<D

rH
0)
P
O
EH

O
P
0)
bO
(D
k
to
to
H*î

«)

G
o
•H

S

s 3 s s
CM VO CO

ov

o
fH

hL
d>
O

g j
a
4-3

m

o

•ri

Vi
vH

-p
0

JG
p
A
o
A

1

c

fH

I

&

■H
CQ