OVERDUE FINES ARE 25¢ PER DAY .
PER 1m

Return to book drop to remove
this checkout from your record.

 

 

 

 

 

 

 

 

 

 

Copyright by
JOHN WILLIAM HARRY DACEY
1979

GAS CIRCULATION THROUGH THE YELLOW WATERLILY

BY

John William Harry Dacey

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Zoology

1979

ABSTRACT
GAS CIRCULATION THROUGH THE YELLOW WATERLILY
By

John W.H. Dacey

The gases in the lacunar system of aquatic vascular plants
are important to the metabolism of these plants. Current models hold
that individual gases diffuse within these gas spaces according to
gradients in their partial pressures. These gradients are generally
thought to be the product of gas metabolism by the plant —-the varying
uptake and release of CO2 and 02 in a static gas phase. In this study

of Euphar luteum, attention was focused on the interaction between the

 

biological processes of the plant and the physical exchange processes
between the plant and its environment.

The internal gas spaces in the yellow waterlily act as a
pressurized flowbthrough system. Air enters the youngest emergent
leaves against a small gradient in total pressure. This pressurization
phenomenon is the result of a purely physical process of molecular
effusion, driven by gradients in temperature and water vapor between
the atmosphere and the lacunae of the youngest emergent leaves. As
the leaves mature and grow in size, they become more porous and lose
their capacity to maintain the positive pressures. Since the lacunae

of these leaves are continuous with the lacunae of the younger leaves,

John W.H. Dacey

the older leaves act to vent the elevated pressures of the younger
leaves.

The pressures generated in the young emergent leaves are
small (less than 0.002 atm), but they drive a fluid flow of gas down
their petioles to the rhizome, and from the rhizome up the petioles of
the older leaves to the atmosphere. Using gas tracer methods, it
was found that up to 60 m1/min of air (21% 02) was transported down
a single petiole to the rhizome buried in anaerobic sediment. The
efflux from the rhizome to the atmosphere carried gas rich in C02
and CH4. Much of the 002 was fixed by plant photosynthesis, while

the CH4 passes through the leaves to the atmosphere at an ecologically

significant rate.

I, having built a house, reject

The
And
One
One
The

The
Its
And

feud of eye and intellect,

find in my experience proof
pleasure runs from root to roof,
thrust along a streamline arches
sudden star, the budding larches.

force that makes the winter grow
feathered hexagons of snow,
drives the bee to match at home

Their calculated honeycomb,
Is abacus and rose combined.
An icy sweetness fills my mind,

A sense that under thing and wing
Lies, taut yet living, coiled, the spring.

Jacob Bronowski
"The Abacus and the Rose"
Science and Human Values

ii

ACKNOWLEDGMENTS

I would like to extend my sincere appreciation to Don Hall,
for bringing me to M.S.U. and encouraging my interdisciplinary study,
and to Mike Klug, for his enthusiastic commitment to me and my work.

I would further like to recognize and thank:

Karen Hogg Dacey, for everything...,

All those students of science I have known or read whose
inspiration was contagious,

My parents, whose spirit imbues my search for understanding.

iii

TABLE OF CONTENTS

LIST OF TABLES. . . . . . .

LIST OF FIGURES . . . . .
INTRODUCTION . . . . . .
CHAPTER 1. THE LITTORAL HABITAT . .

CHAPTER 2.

CHAPTER 3.

The Atmosphere. .
The Sediment . .

The Water Column .

The Lacunar Interphase.

LIFE BETWEEN PHASES . .
Duck Lake . . .

Nuphar luteum .

The Waterlily as Experimental

Plant Gas . . .

Sediment Gas . .

Chromatographic Analysis

Spring Emergence of the Waterlily

Leaf Efflux of Methane.

Fluid Flow. . .

GAS CIRCULATION . .

An Apparent Contradiction

A Flowahrough System .

iv

Plant

Page
vi

vii

11
13
15
15
18
21
22
23
30
31
38
44
55
55

65

CHAPTER 3 (cont'd)

SUMMARY . .

APPENDICES

REFERENCES CITED

Pressures and Flows . . .
A Pump. . . . . . .
"Seerosenphgnomen". . .

Mechanism . . . . . .

Physiological Significance. .

Nitrogen .

Conclusions

Appendix A
Appendix B
Appendix C
Appendix D
Appendix E

Gas Solubility .

The Lacunar Interphase

Fluid Flows . .

Thermal Transpiration.

Hygrometric Pressure .

Page

71
76
85
89
92
100
100

102

103
107
115
124
136

144

10.

11.

12.

13.

14.

LIST OF TABLES

Late winter plant and sediment gases. . .

Gas compositions in late winter Nymphaea. . . .
Pressurization of a leafless rhizome. . . . . .
Rate of CH4 efflux as a function of leaf age. . . .
Methane increases in the petioles of a torn leaf. . .
Methane in submerged and emergent leaves. . . . .
Methane concentrations in newly-emerged and older leaves.
Gas compositions throughout a shoot apex, August 19-21, 1978
02 and C02 in newly-emerged and older leaves (midafternoon)
Tracer efflux from a shoot apex . . . . . .
Ethane transport throughout a shoot apex. . . .
Comparison of ambient and lacunar gas compositions . .

Gas compositions in efflux petioles during daytime and
nighttime . . . . . . . . . . .

Page
32
33
37
43
44
55
56
57
62
65
66

81

95

Gas composition gradients along an efflux petiole (8.7 ml/min) 97

vi

LIST OF FIGURES

Figure
1. A generalized distribution of the principal gases
the plant environment. . . . . . .
2. Duck Lake . . . . . . . . . .
3. Growth habit of the yellow waterlily during early
4. Remote port sampling system . . . . . .
5. Home-made vacutainer. . . . . . .
6. Spring leaf emergence . . . . . .
7. Methane distribution throughout the plant during
summer growth . . . . . . . .
8. Bag on emergent leaf. . . . . . . .
9. Tracer movement through the petiole . . . .
10. Tracer elution curve. . . . . . . .
11. Flow as a function of applied pressure . . .
12. Modified interphase . . . . . . . .
13. Inverted beaker at t = 0. . . . . . .
l4. Inverted beaker at t. . . . . . .
15. Micromanometer . . . . . . . . .
16. Flow rate as a function of the mid-rib pressure in
influx leaf . . . . . . . . .
17. Mid-rib pressure as a function of light intensity
18. Lacunar pressure as a function of leaf temperature

vii

Page

10
17
20
25
28

36

40
42
47
49
53
64
70
73

75

78
80

83

Figure

19.

20.

Bl.

BZ.

C1.

C2.

D1.

D2.

D3.

E1.

E2.

The pressurized gas flOWBthrough system .

pH and P002 profile in Duck Lake sediments

Sediment-water exchanges.
The lacunar interphase . .
Venturi meter . .

Venturi flow. . . . .

Thermal transpiration (steady state).

Porous partition. . . .
Thermal transpiration (flow).
Hygrometric pressure. . .

Hygrometric pressure (flow) .

viii

Page
94
99

109
112
120
122
129
131
133
139

142

INTRODUCTION

It is generally accepted that aquatic vascular plants have
evolved from terrestrial ancestors. Even though these higher plants
originally evolved from earlier equatic forms, no more than 1% of
extant angiosperms are aquatic (Sculthorpe, 1967). The original
invasion of land by plants was accompanied by a number of adaptations
to the new environment. Some of these adaptations, most notably the
reproductive strategies that evolved for life on land, are not easily
lost on re-invasion of the aquatic habitat. There are, however, other
adaptations that must take place in order to support life in this "new"
environment.

In general, the transition from the terrestrial environment
to the aquatic has meant a move from a relatively dry, aerated habitat
to a wet, gas-poor one. As a result, aquatic plants have tended to
lose their adaptations against water loss (stomata are commonly absent
_ or nonfunctional; cuticle is reduced). 0n the other hand, they have
developed adaptations to counter the shortage of metabolic gases in
their environment. Aquatic plants are permeated by an extensive sys-
tem of internal gas spaces. These adaptations are characteristic of
obligate aquatic plants. Similar changes in morphology can be revers-

ibly induced within the lifetime of an amphibious plant by moving it

from one habitat to the other.

Considerable attention has been focussed on the developmental
aspects of the growth of lacunae in aquatic plants. Ecologically, the
dual embryonic origins of these gas spaces (lysigenous and schizogenous,
cf., Sifton, 1945 and 1957) suggest that the various plants have con-
verged in morphology in order to achieve an effect fundamental to
metabolism in the aquatic habitat.

In spite of the widely reputed role of gases in the metabolism
of these plants, there is very little understood about the dynamics of
the gases in these internal spaces or lacunae. It is generally accepted
that the primary role of the lacunae is the delivery of oxygen to
respiring plant parts buried in the anaerobic sediment (Hutchinson,
1975). Such an oxygen transport system may also be involved in detox~
ifying the highly reducing rhizosphere (Armstrong, 1975). Some recent
work (WiumrAnderson, 1971; Sondergaard and Sand-Jensen, 1979) has
demonstrated that these gas spaces may also serve to transport carbon
dioxide from the sediments to the photosynthesizing leaves of certain
plants.

Only in a very few cases have researchers studied the ex-
changes of gases between the plant and its environment. It has generally
been believed that gases diffuse through the stomata of emergent leaves
in accordance with models of simple molecular diffusion. It is also
considered that a similar, but slower, diffusive exchange occurs
between gases in the submerged organs of plants and the gases dissolved
in the surrounding water. Oxygen has received the most attention in
this respect, particularly the loss of oxygen from roots to the environ—
ment (Armstrong, 1964, 1978). Except for a hypothetical biochemical

pathway for the active transport of oxygen in plant tissue (Mitsui,

1964), oxygen transport in the plant is treated as a problem of
gaseous diffusion in a passive gas phase. This phase is, however,
made up of other gases which have their own dynamics within the
lacunae. The gas mixture is a fluid, and as a result the behavior
of a single gas cannot be.a priori separated from the dynamics of
the gas mixture as a whole. Furthermore, the behavior of this gas
mixture can only be understood within a framework that takes into
consideration of physical nature of the external environment. Such
an investigation is fundamental to the understanding of the dynamics
of gases in the lacunae.

The effects of a number of physical environmental parameters
on plant growth have been studied. Temperature, light intensity and
quality, turbulence, and sediment characteristics influence the
metabolism of aquatic plants (Gessner, 1952; Sculthorpe, 1967).
Although it has obvious importance in aquatic systems, pressure is
a physical parameter that has received surprisingly little attention.
Lab studies have shown that very small hydrostatic pressures (less
than 1 atm) dramatically impair the growth and reproduction of several
aquatic plants (Gessner, 1952, 1961). These observations have led to
the widespread generalization that vascular plants tend not to grow at
depths greater than 10 m (2 atm pressure). This generalization is
contradicted by the fact that such plants are found at considerably
greater depth in certain environments (Gessner, 1961), but no theo-
retical explanation has been forthcoming.

It is unlikely that the observed effects of hydrostatic
pressure on plant growth are a direct result of pressure per se, since

terrestrial plants can withstand gas pressures of over 100 atm.without

apparent injury (Scholander et.al., 1966). Furthermore, the effects
of hydrostatic pressure are absent in the lower aquatic plants (algae,
mosses, and liverworts) which lack lacunae (Hutchinson, 1975).

Clearly, hydrostatic pressure exerts a direct effect on gas
relations by changing partial pressures and therefore gas solubilities.
These changing relations are bound to have effects on the dynamics of
the gases in the lacunae. These dynamics are the product of a number
of processes: metabolic gas-consuming and -producing processes within
the plant; diffusive and fluid dynamic processes in the internal and
external gas phases.

A comprehensive understanding of the gas dynamics in the
lacunae requires the concepts of biology ang_physics. It is the pur-
pose of this dissertation to develop a physical framework incorporating
the dynamics of the environmental gases with those of the lacunar gases.

In Chapter 1, a physical perspective of the littoral habitat
is developed, with special attention to the emergent growth habit.
Several general (and testable) predictions are made about the possible
behavior of gases in the lacunae of these plants.

In Chapter 2, the considerations of the first chapter are

applied in a study of some aspects of the gas dynamics in Nuphar luteum

 

ssp. macrophylla. The emphasis is placed on the gases dissolved in the

 

interstitial water of the sediment, and the interaction between these
gases and those in the buried parts of the plant. During the study,
it became clear that there was a pattern of discrepancy between the
observations and the initial predictions.

In Chapter 3, the further investigation of these discrepancies

required a study of the physical exchange processes between the plant

and the atmosphere. By means of a purely physical process, the
pressures of the gases in the plant are increased over barometric,
thereby generating a gas flowbthrough system of great benefit to the

plant.

CHAPTER 1

THE LITTORAL HABITAT

From a purely physical perspective, the littoral habitat can
be broken up into three distinct phases. A phase is generally defined
as a region of uniformity in a system, separated from other homogeneous
parts of the system by boundary surfaces. Since the focus of this
research is on gases, obviously the atmosphere is one of the phases
under consideration. The gases of the atmosphere constantly exchange
with the gases dissolved in the surface of the second major phase -
the water column. Strictly speaking, the third phase, the sediment,
is made up of many phases - the interstitial water, the solid phases
presented by the various types of particulate matter comprising the
sediment, and the gas phase presented by bubbles trapped in the
interstices of the sediment. For the purposes of this study, the
sediment can be treated as a single phase, recognizing that it is not
truly homogeneous. In order to understand the value of partitioning
the environment in this way, it is necessary to further describe the
physical characteristics of the three major phases. Primary consider-
ation will be given to the gases of the atmosphere and the sediment
phases, since the exchanges between these phases and the plant dominate

the gas dynamics within the plant.

The Atmosphere

 

The gas composition of the atmospheric phase is N2 (78%) and
02 (21%), and its total pressure is 1.0 atm at sea level. Thus, the
partial pressures of N2 and 02 are 0.78 atm and 0.21 atm, respectively.
Three other gases are present in significant quantities -Ar (0.9%),
C02 (0.03-0.04%), and H20 (1-4%). water vapor is an important com-
ponent of the atmosphere even though it is rarely considered as a
component (most listings are for dry air). The amount of water vapor
in the atmosphere depends very much on local air temperatures and the
availability of liquid water for vaporization (by evaporation or
transpiration). The changes in the phase equilibria of water are
responsible for much of the earth's weather pattern, and as will be

discussed later in this thesis, they also play a fundamental role in

the gas dynamics in the yellow waterlily.

The Sediment

 

The sediment is extremely important in the growth of rooted
plants (Bristow and Whitcomb, 1971; McKay and Barsdate, 1970; Twilley
et.al., 1977). It is clear that these plants derive most of their
mineral nutrition from this source. It is also becoming apparent that
gases dissolved in the interstitial water (i.e., carbon dioxide) may
be important in the metabolism of a number of plants (Wium-Andersen,
1971; Sondergaard and Sand-Jensen, 1979).

Most of the volume of the sediment is water; the volume
percentages are variable depending on the nature of the sediment matrix.
These depend on the geological characteristics of the surrounding

terrain, and on the extent and character of biological processes

occurring in the vicinity.

The matrix presented by the particulate matter of the sediment
allows the accumulation of dissolved gases by restricting mixing with
the overlying water. It also has the secondary effect of trapping
bubbles of gas in the sediment interstices. The continuous production
of gases in the sediment inevitably leads to ebullition into the water
column. This process sometimes transports significant quantities of
gas to the atmosphere (Strayer and Tiedje, 1978; Robertson, 1978), and
may also act to accelerate the transport of dissolved compounds from
the interstitial water into the water column by turbulent mixing
(Ohle, 1958).

The compositions of individual bubbles lodged in the sediment
are determined, of course, by the partial pressures of the gases dis-
solved in the neighboring water. These pressures are in turn deter-
mined by kinetic properties, such as: the nature of the sedimenting
particulate matter, the metabolism of sediment organisms, diffusive
exchange within the sediment and across the sediment-water interface,
and groundwater flow processes.

The vertical distribution of gases in sediments has been the
subject of a number of studies (Kaplan, 1974; Reeburgh, 1969). Typical-
ly, the partial pressures of atmospheric gases decrease with increasing
depth in the sediments. Oxygen usually disappears within millimeters
of the sediment surface; nitrogen and argon generally decline much more
slowly, and never really disappear. Conversely, carbon dioxide and
methane concentrations increase sharply below the surface of the
sediment. The distribution of gases in the littoral habitat has been

generalized in Figure l.

Figure 1.-- A generalized distribution of the principal gases in the
plant environment.
Below the sediment-water interface the equilibrium
partial pressures of N2 and 02 fall dramatically. In
contrast, the partial pressures of CH4 and C02 increase
sharply below the interface. The presence of bubbles
trapped in the sediment interstices demonstrates that
the sum of the partial pressures of the dissolved gases
equals the hydrostatic pressure at that depth.

DEPIH (cm)

10

EQUILIBRIUM PARTIAL PRESSURE (atm)

 

 

 

 

O .5 1.0
o 1 .
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
:
50‘ 02 :NZ
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
: \Mafer
100 5
Sediment

150‘

 

 

a
'~'~. —".—
x. ,o‘
a
‘ I
I
\.

\
3
I” \
I
’ \
I’ \.
'1 I '\
. I .
’1 I. 2 PARTIAL Passsuass
. I

 

11

The Water Column

 

In the water column, there are also a number of processes
that significantly influence the composition of the dissolved gases.
The dominant factor is the continual exchange of gases with the
atmosphere, and as a result the partial pressures of the gases in the
water approach their partial pressures in the atmosphere (i.e., N2,
0.78 atm; 02, 0.21 atm; Ar, 0.01 atm).

The absolute solubility of gases is an inverse function of
water temperature, so the amounts of gases dissolved per unit volume
of water tend to decrease with increasing termperature while the
equilibrium partial pressures remain constant.1 (Appendix A)

There are also other processes that tend to move the water
away from this equilibrium. Photosynthesis by plants dependent on

dissolved carbon dioxide (either as C02, H 003, or H003) tends to

2

reduce the partial pressure of CO below atmospheric (0.03 atm).

2

 

1The choice of dimension in the discussion of dissolved gas
data must reflect the purpose behind the data analysis. The convention
of using mass/vol seems to be a poor choice in many applications. This
approach presumably follows the convention commonly applied to the
solution of solids in liquids. With solids, the notion of mass has
conceptual value; this is not true for gases. It is much easier to
think of gases in terms of pressures, volumes, or moles; it's virtu-
ally impossible to think of gases in terms of weights. It's tempting
to draw an analogy with the differences between the Imperial system
and the metric system of mensuration. Conceptualization can often be
hampered by adherence to traditional methods (cf., Appendix A).

Consider, for example, dissolved oxygen profiles for a lake.
These are universally presented in mg OZZLH 0. As discussed in
Appendix A, these values are of limited significance without tempera-
ture data, and even with such data it is difficult to make much more
than relative comments about the dynamics of the gases. However, if
the same information were presented in atm 0 (the equilibrium partial
pressure of O in water), it becomes immediagely clear how far the gas
composition OI the volume of water in question has diverged from the
equilibrium composition it seeks at lake turnover.

12

Photosynthesis in the water column also acts to increase the partial

pressure of O in this phase. Conversely, respiration tends to decrease

2

the partial pressure of O2 and increase that of C02.

Since C02 is more soluble than 02, each molecule of CO2
contributes relatively less to its partial pressure in the aqueous
phase. As a result, aerobic respiration near or at the sediment
surface may tend to decrease the sum of the partial pressures of the
dissolved gases. This hypothetical outcome is represented in Figure 1
by the slight depletion in the sum of partial pressures just above the
sediment surface.

In hardwater lakes, the partial pressure of CO2 in the sur-
face waters often exceeds atmospheric. This phenomenon results from
geochemical kinetic processes involving the degassing of a large
carbonate pool derived from groundwater and sediment respiration
processes.

The pressure relations of the gases in the water column are
very important to submersed vascular plants, but appear to be less
significant in the growth of emergent plants. This investigation of
the dynamics of gases in the yellow waterlily suggests that during the
summer growth period, the gases dissolved in the water column are of
little significance to the plant. These gases do appear to be impor-
tant, however, during those times of the year when the entire plant is
submerged (roughly November - April). Then the plants put out
morphologically distinct leaves that appear to act as efficient

exchangers of gas between the lacunae and the water column (Arber,

1920).

13

The Lacunar Interphase

 

Given the very simple principles underlying this discussion,
it is possible to make some very interesting and significant pre-
dictions about the interaction of the plant with its environment
(vis-a-vis gases). Since the plant under study is a waterlily, the
predictions outlined here will be framed in their terms.

The waterlily spends nearly half of its life without surface
leaves, from November through April in Michigan. It may or may not
have submersed leaves during that time, but even if it does, these
leaves are few and only project a short distance out of the sediment.
The roots and the rhizome form the vast majority of the plant bulk
(dry wt., volume, etc.), and these parts are for the most part buried
in the sediment. The low rates of plant metabolism during the winter
months (due to low light intensity and low temperatures) imply that
the plant is unlikely to be carrying out processes that greatly alter
the composition of the gases in the lacunae of the rhizome. The gas
spaces of the lacunae of the roots and rhizome can be likened to a
large, continuous bubble held in the sediment by a membrane of plant
tissue that is permeable to gas diffusion. The cumulative effect of
these diffusion processes means that the gas inside the lacunae should
ever-increasingly resemble the gas in the bubbles trapped in the
sediment interstices. That means that the volume percentages of
methane and carbon dioxide should be relatively high, and those for
atmospheric gases should be low. Furthermore, this convergence in gas
composition also means that the total pressure of the gases in the
lacunae should approach the pressure in the external bubbles (1 atm +

hydrostatic pressure). As will be discussed in the following chapter,

14

these predictions are upheld by observations made in the course of

this study. This research has lead to the first published discussion
of pressure equilibration between a plant and its anaerobic environment
(Dacey and Klug, 1979).

During the summer growth period, the plant extends above
the lake surface, and allows rapid diffusive exchange with the atmos-
phere. Considering the fact that the bulk of plant biomass is still
buried in the sediment, significant rates of exchange between the
plant and the sediment water should be expected. In a sense, the gas
spaces of the lacunae represent an extension of the atmosphere into
the sediment. In Appendix B, the possibilities for movement of gas
to and from the sediment water are discussed from a purely physical
perspective. In that appendix, the term "interphase" is borrowed
from mineralogy to emphasize the role of the lacunar gases as an
"interfacial phase". The continuity of this interphase suggests a
possible net flux of gas in one direction or the other, depending on
a number of physical and biological factors. Such a net flux could
have great significance in plant metabolism.

In the next chapter, the idea that the plant may act as a
source or sink for atmospheric gases is investigated. It is in this
study that a pattern of discrepance becomes apparent, leading to the
development of hypotheses fundamentally altering modern conceptions

of the factors regulating the distribution of gases in these plants.

CHAPTER 2

LIFE BETWEEN PHASES

As described by Sculthorpe (1967), there is typically a
zonation of plant growth habits along the shores of lakes: Emergent
plants in the shallow water, floating-leaved in deeper water, and
submersed plants in still deeper water. Although there are exceptions
in any littoral zone, it does point out a relationship between the
plants and their environment that is relevant to this study. It
suggests that the waterlilies, as floating-leaved plants, should
represent the deepest extension of atmospheric gases (at roughly
atmospheric pressure) into the submerged sediment. Although most
submersed plants do actually reach the surface when they flower (a
vestige of their terrestrial origin), they cannot be said to grow
with their leaves in the atmosphere. This is important, because the
effects predicted in Chapter 1 should be most easily observed in

plants extending from deep sediments to the surface.

Duck Lake (Kalamazoo Co., Michigan: 42°24'30"N, 85°22'57"W)

The plants used in this study grew along the perimeter of
Duck Lake, from the shoreline out to a depth of roughly 2 m. Duck
Lake is a kettle lake (mean depth 2 m; maximum depth 4 m), with no

surface inflows or outflows (Figure 2). The total area of the lake

15

16

Figure 2.--Duck Lake
The dashed line represents the outer extent of Nuphar
and Nymphaea growth in the lake. The map is adapted
from Miller (1972).

 

17

DUCK LAKE
sec. 5 nsasw.
KALAMAZOO COUNTY. MICHIGAN
23 o “25"“30 75 no
0 I!) 300

FEE?
CONTUUIS»NIIETEIS

     

18

is 12.6 hectares; roughly half of that is covered by a dense stand of

waterlilies (mainly Nuphar luteum, and some Nymphaea odorata). These

 

 

plants grow along most of the lake perimeter, while most of the central

area of the lake is occupied by a dense population of Ceratophyllum

 

demersum, coontail. Some physical and chemical parameters of the

central region of the lake for 1968—1969 are presented by Miller (1972).

,Nuphar luteum

 

The yellow waterlily perennates in lakes by means of a
horizontal creeping rhizome at or just below the surface of the sedi-
ment. The rhizome may be up to 10 cm in diameter and several meters
long, accounting for roughly 80% of the plant's biomass during summer
growth (Twilley et.al., 1977).

The rhizomes branch extensively, as shown in Figure 3, and
each growing shoot apex bears a rosette of leaves (and flowersl) that
rise up through the water column on long petioles. In Duck Lake, the
leaves commonly float on the lake surface for variable lengths of time,
and then rise up above the lake surface. This further elongation of
the petiole does not appear to influence the gas exchange properties
of the leaves, so throughout this discussion, leaves are called
"emergent" once they have reached the lake surface. This result is
presumably due to the fact that the stomata of the waterlilies are

found only on the upper surface of the leaves (Sculthorpe, 1967).

 

1It is an unusual feature of many waterlilies, including
Nuphar, that their solitary flowers are neither axillary or terminal;
they are borne on leaf sites in the parastichies (phyllotactic helices).
(cf., Weidlich, 1976)

19

.nuaoum wcfiuam haumo wcwuoo maaauoumS soaamh onu mo pans: :uaouOII.m ouawfim

20

\

E8.

., fistvflww

.. .p .

‘anofi .

I g .M
x' .

IQ

   

 

.
-I.

.
s

21

As the branching rhizome grows forward through the sediment,
adventitious roots grow down into the sediment. Some of these roots
are contractile, and act to pull the buoyant rhizome down into the
sediment. These roots appear to grow as far as a meter into the
sediment.

Much of the rhizome is leafless, and bears only the leaf
scars from previous years' growth and, of course the adventitious
roots (cf., Figure 3). The branching rhizome appears to be the domi-
nate means of plant reproduction within a lake. As the rhizome
elongates at its apices, it decomposes at an indeterminate rate in
its older parts. The eventual decay of branching points leads to

the vegetative reproduction of an individual plant into two.

The Waterlily as Experimental Plant

 

In order to understand the dynamics of the gas phase within
a plant, it is necessary to get analyzable samples from discrete regions
of the plant. Much of the published data on these dynamics has depended
on the extraction of gases from isolated plant parts under reduced
pressure. The large volumes of gas within the tissue of the yellow
waterlily make them ideal for gas sampling. Samples can be withdrawn
by syringe, and can subsequently be injected directly into a gas
chromatograph for compositional analysis. Furthermore, this sampling
can be done in gitg, avoiding the serious disturbances to plant inte-
grity and metabolism that other methods cause.

The actual percentage of plant volume occupied by gas in the

waterlily approaches 60% in the petioles, and lies between 20% and 40%

22

in the roots and rhizome.1 Not surprisingly, the large volumes occupied
by gases are a reflection of the role of gases in plant metabolism. As
has already been mentioned, the dynamics of gases have been studied in

a number of plants, but one of the most prominent is a study by Laing

(1940, 1940b)on “Nuphar advenum" in lakes in the vicinity of Ann Arbor,

 

Michigan.

The classification of waterlilies is confused, particularly
in the genus Nuphar. Under the classification system of E.0. Beal
(1956), the plant described by Laing would be called a subspecies of
_N. luteum. Under Beal's classification system the dominant yellow
waterlily in Duck Lake would probably be called N. luteum ssp. magrgf
phylla. Specimens have been placed in the herbaria of the W.K. Kellogg

Biological Station and of Michigan State university, East Lansing.

Plant Gas
Gases were normally withdrawn from the plants by means of a

remote-port system according to the protocol described in the caption

 

1Cores of plant tissue were taken from plants with a cork-
borer. The gas volume-percentages of the cores were determined by two
different methods. In one case, the cores were flushed with a gas mix-
ture of known composition (i.e., 10% CH ). The cores were then forced
into bottles of known volume, the bottles were stoppered, and the gases
were allowed to equilibrate inside the bottles. The equilibrium con-
centrations of CH in the bottles were used to calculate the volume of
CH in the tissue cores, and therefore, the total gas volume in the
cores.

The second method relied on the change in specific gravity
of fresh and water-flooded cores. The cores were weighed in air, then
placed under water and a vacuum.was applied (roughly 0.05 atm for
several hours). The pressure in the chamber was returned to atmos-
pheric, forcing water into the lacunae vacated by the expanding gas.
The weight increments were used to estimate the initial gas volume in
the cores.

23

to Figure 4, and the sample syringes were stuck into numbered butyl
rubber stoppers. Samples were normally held for no more than 4 hours
before analysis.

This system allowed repeated sampling over several days
from various points around the plant, without repeatedly puncturing
the plant whenever samples were to be taken. Since most sampling
sites were underwater, water leaks tended to be a problem, particularly
in the rhizome. The eventual flooding of tissue at the sampling point
was retarded by sealing the sampling needlewith silicone robber in

the manner illustrated in Figure 4.

Sediment Gas

 

Sampling of sediment gas was carried out in two distinct
fashions. The quicker and more direct method involved the trapping
of bubbles dislodged from the sediment. A large funnel was steppered
at its spout, filled with water, and inverted over the sediment. It
was then forced down into the sediment and vigorously agitated, there-
by dislodging bubbles from the sediment interstices. Samples were
withdrawn from the inverted funned by syringe and analyzed by gas
chromatography.

The second method involved the analysis of dissolved gases
collected from equilibrated dialysis samplers (Molongoski and Klug, in
prep.; Hesslein, 1976; Winfrey and Zeikus, 1977). As soon as the
dialysis sampler was withdrawn from the sediment and quickly rinsed,
the equilibration ports were covered with a molded silicone rubber
sheet (General Electric RTV-60) to stop gas exchange with the atmos-

phere. This operation was completed within 15 seconds of the sampler's

24

.ucmHa one cup“ nouns mo owoxmoa vumuou

mom ocean ozu cu owmamo ucm>oue ou A-I>Hm ..m.uv usuammm moons“ mooofiafim mam
m>ooam cow%H m up omuuoamom maamama mums osmmfiu ucmam oucw mouuomca meadows mnH
.owswuhm oaaamm msu omcfiu mam Eoumhm one :maam ou samumsuwz mp3 HE a ou a: oanfimm
humcfiawaoua m on .HB m.o zanwoou mos swumhm menu mo o8=Ho> Amoco may .momwuam
onu um upon unwwulwmw o co woaomwa moans» wuonouofia mom dammed owomw mm m

emoounu sown ow ocean one «0 mouse moowum> aoum cannonufis onus moHQEMm mow
.ECumhm woeamfiom upon ouoaomII.q muowfim

 

26

arrival at the lake surface. Since only those ports that contained
water used for atmosphere-sensitive analyses were covered by the rubber
sheet, water temperatures were quickly taken along the length of the
sampler. Water samples were then withdrawn through the rubber sheet

by means of "home-made vacutainers" (Figure 5), and the pH of the
water in the ports was measured by peeling back the rubber sheet from
each individual port at the time of analysis.

After several hours of equilibration at a known temperature,
the pressures of the flasks were measured by water manometry. The
composition of the gas in the headspace was determined by withdrawing
gas samples through the stopper, making sure to avoid atmospheric
contamination. The samples were analyzed by gas chromatography.

All headspace estimates were then used together with the
partitioning coefficients (Henry's law constants) of the various gases
at the equilibration temperature to calculate the total amount of each
gas in the original water sample.

The CO concentrations measured in the headspace cannot,

2

however, be used as direct estimates of the CO content of the pore

2

water. During the equilibration between the headspace and the water
sample, there is an equilibration of the bicarbonate buffer system in
the water which tends to force more CO2 into the gas phase.

The estimation of CO2 partial pressures in the sediment
water requires that the sample water be acidified in order to get an

estimate of the total dissolved inorganic carbon pool size, C The

T.

sample water was acidified by injecting 2 ml 3% H P04 into the serum

3
bottle. The bottle was allowed to equilibrate for several hours, the

pressure was again measured (this time by mercury manometer, since a

27

 

.mHuuon sea oucw csmpo hafimouom uwum3 mo meoHo> on» moaspmuom hamumuouom cu

coauomaaoo noumm omnwfimamu mam .cofiuooaaoo mHaEmm anewmn momma mums mumsfimusom> mnu .%oouw mesa
weapon .uoom3 mo m5=Ho> opossuouommua new as mamuo oauuon one ca asoom> mnu mom oHuuon Eamon
one so Homecom one monsoooon omam moamooo mmwnu mo moo .moaoom: osu up monouocoa mos uomnm
Hannah onu .uuoa mammamwm ecu Eouw coxmu we ou whoa mmaaamm cosz .uuoa ecu Eouw umHaEmm ago he
osmuo3uws on ou umuma mo mEoHo> onu wa«o«5umumm hnouosu .musmmoua moafisumuoooua m ou moumoom>w
coco mos oHuuon map as muosamoeuo Sofiamc och .mpmaaoum sweeps quan xuwnu one nwoounu moammo:
mo momma up sagas: sues momcfiu can mmuoaaouw mama moHuuon Epsom HE mm .cowumofiaamm mwnu aH
.moHeEmm umum3 m>wufimcomIouo£amoaum mo :oHuomHHoo osu mwfiaaafim mom woman cu mocwawmo mums wmonH

.muoaamuoom> mmmEIoEomII.m wuowfim

 

 

 

 

 

 

 

 

 

 

 

 

29

considerable amount of de-gassing takes place from the water). The
CO2 content of the headspace was then determined by gas chromatography

or IR analysis. The estimate of CT can then be used together with the
pH and temperature measured at the time of sampling from the dialysis
sampler to calculate an algebraic estimate of the partial pressure of
C02 in the pore water.

As suggested earlier, the two methods for determining gas
composition in the pore water have different applications. The

inverted funnel method yields good estimates of CH4, Ar, and N with-

2.
out any regard for the depth distribution of the gases. This method
consistently underestimates the pressures of CO2 because of its high
solubility in water, tending to dissolve inside the funnel.

The dialysis sampler gives good estimates of CH4 and C02.
Atmospheric gases, such as Ar and N2, can be estimated by this method,
but these values depend on complete rinsing of the bottle with helium
and an absence of leaks of atmospheric gas into the bottle after
evacuation (before sampling). The dialysis sampler has the general
advantage of giving precise depth profiles of the dissolved gases.
However, in this study these profiles are of limited utility, since the
distribution of the exchange surface presented by the plant in the
sediment is not known. The roots of the waterlily extend up to a
meter into the sediment, and it is reasonable to conclude that the

actual gas exchanges that occur are a weighted function of the depth

profile of the gases.

30

Chromatographic Analysis

Gas samples were withdrawn in k and 1 ml Glaspak tuberculin
syringes and were analyzed by gas chromatography. CH4 and CZH6
(Chapter 3) analyses were performed on a Varian 600 with a flame
ionization detector; gas separations were achieved in 1 meter of 1/8th
inch column packed with Porapak N.

Analyses of total gas composition (i.e., N2, 02, C02, CH4,
and Ar) were accomplished with a Carle BGC-8500 fitted with two
columns (each 1 meter long) in series, and four valves to facilitate
separation of component gases. C02 and H20 were separated from the
other constituent gases on the first (Porapak Q) column. The valving
arrangement allowed these gases to bypass the downstream column,
passing directly to the detector (thermistor TC). These gases could
also be backflushed to vent, as was usually done in the case of water
vapor. The downstream column (Mblecular Sieve 80/100) achieved the
separation of 02/Ar, N2, and CH4. This packing does not separate 02
and Ar, so an external column of hot copper filings was used in series
to remove 02 from samples, thereby allowing a determination of Ar
(and thus, by difference, 02). Since Ar was always present at levels
less than 1%, it was ignored in the determination of 02 when 02 con—
centrations exceeded 10%. When 02 concentrations were less than 10%,
the levels of Ar become a significant source of error, so the 02-
reduction technique was used.

Quantification of the partial pressure of water vapor was
achieved on the same chromatograph, using a 1 meter column of

Porapak Q3;

31

Room air was used as a standard for N2 (78%), O2 (21%), and
Ar (0.9%) when their composition was near atmospheric. 02 was the
only one of these gases that deviated significantly from atmospheric.
At concentrations less than 10% 02, a standard curve for 02 was con-
structed by the serial addition of room air (21% 02) to a helium?
rinsed bottle of known volume. CO2 standard curves were constructed
in like fashion using pure CO2 (Linde, 99.8% minimum). Methane
standards were supplied by Linde (10%, Custom Grade; 540 ppm, Custom
Grade) and standards outside these ranges were made by adding known
volumes of pure CH4 (Linde, 99% minimum) to a bottle of known volume.

Gas samples were injected into the chromatograph by two
different means: on-column injections of 0.4 ml and gas sample loop
injections of 0.2 m1 (via gas sampling valve). The latter method
was preferable as a result of its decreased variability — its co-
efficient of variation was found to be roughly half that for the
on-column injection. It has the disadvantage of requiring larger
volumes of gas (at least 1 ml) in order to rinse the sampling loop
prior to injection. The loop was used whenever possible in this

study.

Spring Emergence of the Waterlily

 

In Chapter 1, it was pointed out that gases tend to equili-
brate according to their partial pressures. As a result, a gas phase
in equilibrium with the gas-saturated sediment water will tend to
take on the composition and pressure of the bubbles in the sediment
water. This phenomenon, never before described in the ecology of

aquatic plants, was observed by measuring the pressure and composition

32

of the gas phase in the lacunae of rhizomes that had lain dormant in
the sediment throughout the winter. A water manometer was employed to
measure the pressure in plants situated in about 40 cm water (rhizome
depth). Gas samples were also taken from these rhizomes and bubbles
were taken from the sediment pore water; these data, all means of two

or more samples, are presented in Table 1.

 

Table 1 Late Winter Plant and Sediment Gases
Sample Source CO CH N 0 Pressure
2 4 2 2
(cm water over
barometric)
Petiole #1 6% 35% 56% 1% >10 cm
Petiole #2 3% 37% 53% 2.2% 23.5 cm
Petiole #3 4% 36% 53% 1.3% 27 cm
Petiole #3 5% 38% 53% 0.6% 27 cm
(after bubbling
overnight)
Sediment Bubbles 5% 40% 51% - 40 - 50 cm
Sediment Bubbles 4% 47% 51% - 4O - 50 cm

The most important point illustrated by the data in Table 1
is the fact that the pressurization of the gas phase in the rhizome
results from an equilibration of the lacunar gases with the interstitial
water gases. In all the samples shown, the pressure exerted by 02 in
the lacunar gas phase (1% is equivalent to 10 cm water) is less than
the pressure increment in that phase. This proves unequivocally that
photosynthesis does not play a significant role in this pressurization.
Except for carbon dioxide, the concentrations of the other
principal gases in the plant lie between the composition of the sedi-

ment and the atmosphere. This is exactly what would be expected in

33

such a case, where parts of the waterlily extend up into the water
column.

It is impossible to say from these data whether the apparent
concentration of carbon dioxide is due to plant respiration or simply
an artifact of the gas collection procedure. As described above, the
funnel collection method always underestimates CO2 concentrations in
comparison with the dialysis samples. It's likely, therefore, that
CO2 concentrations in the late winter sediment also exceed the con-
centrations found in the lacunae.

The highest pressures observed in plant tissues were seen

in the first emerging leaves of Nymphaea odorata. When such leaves

 

were torn, sustained streams of bubbles were forced from the torn
site, indicating that the pressure in the gas phase in the plant
exceeded the hydrostatic pressure at the depth of the leaf. Data on

such leaves are presented in Table 2.

 

Table 2 Gas Compositions in Late Winter Nymphaea
Sample Source CO CH N 0 Pressure
2 4 2 2

(cm water over
barometric)

Torn Leaf 5% 39% 53% 2% 70 cm

Torn Leaf 5% 35% 52% 2.5% 60 cm

Torn Leaf 4% 34% 57% 2.4% 40 cm

Sediment Bubbles 4% 40% 51% - 80 - 100 cm

Sediment Bubbles 3% 42% 60% 1% 70 - 100 cm

Tables 1 and 2 show that there is a strong tendency for the
gas in the lacunae to equilibrate with the gases dissolved in the

interstitial water while the plant is submerged in winter. As the

34

leaves grow towards the surface in spring, they encounter ever-
decreasing hydrostatic pressure, and the gases inside the leaf tend
to burst out through any tears in the leaf surface. (Figure 6)
Once such a leaf is torn, bubbles may stream from the opening for
several hours. Given the growth habit of Nuphar, it's likely that
the volume of bubbles that emerge from any such leaf reflect the
volume of the rhizome being vented. The submerged interconnections
of the various shoot apices also leads to the result that not every
emerging leaf exhibits this bubbling effect. Presumably, once one
leaf has vented to the atmosphere, bubbles will not stream in this
fashion from any other leaf connected by a common rhizome.

This pressurization phenomenon can be duplicated during the
summer by removing all the leaves from an isolated section of rhizome.
In Table 3, pressure and compositional data are shown for the gases
in one such rhizome growing in about 140 cm water. The dash marks

in the table mean that no data are available.

35

Figure 6.——Spring leaf emergence.
The gas composition of the sediment is expressed in
terms of the composition of equilibrated bubbles in
the sediment. During winter, the internal gas phase
of the rhizome has approached equilibrium with the
dissolved gases in the sediment. As the first leaves
rise towards the surface in spring, sustained streams
of bubbles escape from torn leaves.

 

 

 

 

VOL7¢
N2 48 - 60
CH4 4O - 41
C02 3 ' 4

 

 

 

36

 

 

 

 

CH4

VOI.%
55-75
10 - a7

 

 

 

 

 

37

 

Table 3 Pressurization of a Leafless Rhizome
Day # Hour 002 CH4 N2 02 Pressure Depth of
(cm water) shallowest
leaf
1 1200 9% 10% 74% 5% 0.0 cm -—
- (= not determined)
1 1800 9% 11% 72% 5% - -
2 1200 10% 14% 71% 2% 50 cm -
3 1800 11% 20% 63% 6% 118 cm —-
4 1800 11% 23% 61% 4% 111 cm -—
5 1100 12% 25% 60% 1% 64 cm 36 cm
6 1700 13% 27% 58% 1% 54 cm 30 cm
7 1500 —- 26% 54% 2% 38 cm 29 cm
9 1500 11% 32% 56% 2% 31 cm -—
10 1200 12% 31% 54% 1% 22 cm 6.5 cm
11 1100 12% 31% 52% 1% 0.0 cm at surface

It is important to realize that the rapid increase in gas
pressure is not simply due to the flooding of the plant tissue through
the cut surfaces of the petioles. The lacunae of the yellow waterlily
are constructed so as to restrict the flow of water into this tissue.
Furthermore, in a similar experiment on another rhizome, bubbling was
observed to flow from one of the cut-off petioles several days after
the leaves had been removed. This type of ebullition shows that the
pressurization mechanism lies in the buried parts of the plant, since
the lacunae of the petiole are occupied by gas phase.

Since the only way the pressure can be diminished in such a
rhizome is for it to lose gas to the water column, the pressure in the
rhizome correlates with the depth of the shallowest petiole or leaf.

In the case of Table 3, the petioles were all cut off very near the

38

rhizome. As the first leaf rose toward the lake surface, it allowed
the escape of gases, presumably by ebullition. Once the first leaf
reached the surface, the pressure in the rhizome dropped toward baro-

metric pressure.

Leaf Efflux of Methane

 

From the data in Table 3, it is clear that methane shows a
strong tendency to enter the lacunae of the roots and rhizomes of the
yellow waterlily. A cursory survey of the distribution of methane in
the lacunae during the summer showed it to be present throughout most
of the plant, with the highest concentrations in the roots and the
lowest in the leaves (Figure 7). This gradient from roots to leaves
suggests a diffusive flux of methane into the atmosphere.

The rate of flux was measured by tying gas-impermeable
Saran bags over individual leaves, tying them closed around the
petiole of the leaf (Figure 8). A study into the ecological signifi-
cance of this flux through the plant showed that it accounted for 75%
of the methane escaping from the littoral sediment of Duck Lake during
two weeks in August (Dacey and Klug, 1979). The remainder escaped by
ebullition directly from the sediment. Previously, investigators had
considered ebullition to be the major transport mechanism for the
escape of methane to the atmosphere, followed by a slower surfaCe
exchange (driven by a gradient of methane partial pressure between the
surface water and the atmosphere). The observed CH4 flux through the
plant introduces another pathway for methane transport. It's important
to understand all these pathways, since a loss of methane from an

ecosystem represents a loss of both carbon and energy from that system.

39

Figure 7.-Methane distribution throughout the plant during
summer growth.
Methane enters the roots and rhizome and escapes to the
atmosphere through emergent leaves. The concentration
of methane in the internal gas phase is highest in the
roots and older leafless parts of the rhizome, decreas-
ing towards the shoot apex.

40

  

-‘:-. “I".
_ "rec-s.

 

   

- in? f‘ ‘
W ‘a,§(o\.mt~
i‘ .. ' u ‘ ’ wt

\\

 

.2094

CH4 vol.% 3"“

.2676

 

 

41

Figure 8.--Bag on emergent leaf.
A gas-impermeable Saran bag was placed over a single
leaf and tied closed around the petiole. At the end
of a run, the composition of the gas in the bag was
determined by taking a sample by syringe through the
stoppered port at the top of the bag. The volume of
the bag was then determined by injecting an internal
standard into the bag, gently squeezing the bag to mix
the gas, and then re-sampling. In cases where the
changes in bag volume over a time interval were moni-
tored, the volume was also determined at the beginning
of the run by the same internal standard method.

 

 

43

These experiments with bagged leaves yielded some other
interesting results. Shading an individual leaf in a bag, or adding
CO2 (up to 3%) to the bag appeared to have no impact on the rate of
methane efflux from the leaf. This corroborates current ideas that
the function of these stomata is limited (Gessner, 1959).

It also became clear in the course of this study, that the
rate of methane efflux was directly related to the age of the leaf.
In Table 4, the rates of efflux from 6 leaves of a shoot apex are
presented along with data on the length of the petioles (from rhizome
to leaf) and the distance of insertion point of each leaf from the
shoot apex. Since the leaves arise at the shoot apex, the oldest

leaves are those farthest from the apex.

 

Table 4 Rate of CH4 Efflux as a Function of Leaf Age
Distance of petiole Length of Rate of CH efflux
insertion from apex petiole from the leaf
8.0 cm 61 cm Not detectable
9.0 cm 85 cm 1.0 ml/hr
9.5 cm 87 cm 1.4 m1/hr
10 cm 90 cm 1.6 ml/hr
13 cm 101 cm 2.6 ml/hr
15 cm 112 cm 3.0 ml/hr

The most immediate explanation for the differential rate of
methane efflux lies in an examination of the methane distribution data
shown in Figure 7. The oldest leaves insert into the more basal region
of the rhizome, and the concentrations of methane in the lacunae of

these regions are higher than in the more apical regions. As will be

44

shown in Chapter 3, there are other reasons for this pattern of efflux,
and the explanation lies in the pattern of fluid flow of gases within

the plant.

Fluid Flow

 

It was found that by tearing a leaf, there was an immediate
increase in the concentration of methane in the petiole of that leaf.

The data shown in Table 5 are representative of a number of such

 

experiments.
Table 5 Methane Increases in the Petiole of a Torn Leaf
Distance along petiole Time after leaf was torn
from leaf 0 min 4 min 14 min 22 min
3 cm 0.04% CH4 .08% .15% .15%
18 cm .29% .40% .52% .67%
33 cm .47% .61% .78% .84%

If the methane were moving from the rhizome up the petiole to
the atmosphere simply by means of a diffusive flux, damaging the leaf
should have exactly the opposite effect. It should decrease the methane
concentration at any point along the length of the petiole. The fact
that it increases means that the flux of methane is not simply diffu-
sive; it has a fluid flow component. The total pressure of the gases
in the petiole must have been slightly greater than barometric, and
when the leaf is torn, the pressure is vented to the atmosphere by an
accelerated fluid flow.

Other lines of evidence support this conception of fluid
flow. It had been casually noted throughout much of the study that the

volume of bags tended to increase during their period of attachment to

45

leaves losing methane. Data on the rates of change of bag volume will
be presented in Chapter 3.

The theoretical treatment of the lacunar "interphase" in
Chapter 1 demonstrated that a fluid flow of gas might be expected
simply on the basis of differential rates of exchange between the
plant and the sediment water. In order to understand these flows, it
was necessary to develop a method for studying them.

The method used most extensively involved the injection of
a tracer gas (usually ethane, C2H6) at a point along the length of a
petiole, followed by sampling of gas compositions at some distance from
the point of injection (Figure 9). By this means it was possible to
construct tracer concentration-time curves of the same form as the
elution curves for chromatography (Figure 10). As the tracer gas
moves along the petiole (carried by the fluid flow of the other gases
there), it diffuses into those gases and may also be absorbed by the
plant tissue. As a result, the initial spike of tracer gas at the
injection point broadens into a slightly tailing peak typical of gas
chromatograms as it is carried along in the petiole. The maximum in
the curve (Figure 10) can be taken to represent the point where the
tracer was injected into the flow.

The volume of gas in a given segment of petiole was estimated
by measuring its upper and lower circumferences and the distance between

the two.1 The rate of volume flow was then calculated on the basis of

 

1A regression was generated for the area predicted from the
circumference (measured by a string pulled tight around the petiole)
against the area measured of impressions of petiole sections pressed
onto graph paper. The volume of the petiole was calculated by assuming

46

Figure 9.—-Tracer movement through the petiole.
Tracer injected into a petiole during daylight moved
along the length of the petiole in a "plug". In the
case of efflux leaves, the rate of gas flow up the peti—
ole was determined by injecting tracer into the lower
end, and sampling sequentially at the upper end.
In the case of influx leaves (discussed in Chapter 3),
the direction of flow is reversed, so tracer was
injected into the upper petiole and sampled sequentially
at the lower end.

47

v
.5. - J
.Mc—ud"

_ ““I“t§

\-

I

I.
a.

 

 

 

 

48

Figure 10.—-Tracer elution curve.
This is the typical shape of the elution curve for the
tracer "plug" when the tracer injection was made
upstream of the sampling point. The time of the
maximum was taken to represent the time required for
the gas to be carried from the injection point to the
sampling point.

Tracer Concentration

 

49

TRACER ELUTION CURVE

 

Time

50

the volume of the segment under consideration and the time elapsed
between tracer injection and that time when the tracer maximum passed
the sampling site.

The curve in Figure 10 can be integrated, so that on the
basis of the rate of flow calculated for a particular experiment, the
total amount of tracer passing the sampling point can be estimated.
Typically 0.1 ml C2116 was injected into the petiole, and it appeared
that all (80% - 120%) of the tracer passes the sampling point whenever
the sampling point was downstream of the injection point.2 This proves
unequivocally that the gas in the petiole moved in a fluid flow.

Early attempts in this study to measure the pressures that
must be driving these flows were unsuccessful, so a simple lab experi-
ment was set up to further investigate the flow. The rate of flow up
a petiole was established for an intact plant growing in a tub in the
lab. The petiole was cut off at the base, and the length of the petiole
and its leaf were used to estimate the original pressure at the base of
the petiole.

The basal end of the petiole was inserted into a large bore
Tygon tube, and sealed with rubber (Rubbergel, L.D. Caulk Co.). The

other end of the Tygon tubing was attached to a four liter erlenmeyer

 

it takes the shape of a frustum of a right cone. Taking the gas spaces
as 50% of the petiole volume, the volume of gas in any segment can be
calculated without removing the segment from the plant.

2Rarely was tracer detected at a sampling point upstream from
the injection point. This only occurred when flow rates were extremely
slow, and the sampling point was near the injection point. Clearly,
integrations of the type described here are meaningless under those
circumstances.

51

bearing a manometer, a high pressure gas inlet, and a gas outlet. The
pressure in the flask was regulated by adjusting the flows of gas into
and out of the flask; the pressure was monitored on the manometer.

The tracer technique was applied in exactly the same way as it was for
the intact plant, with tracer injected at the base of the petiole,
followed by sampling at the upper end. The flow rates are shown as a
function of pressure in Figure 11. Extrapolating these data to lower
pressures, it appears that a pressure of 0.6 mm water forced the flow
in the intact plant.

These experiments also demonstrate the applicability of
Darcy's law for fluid flow through porous media (cf., Appendix C).

The flows that occur in the petioles of these plants behave in the
customary fashion — gas moves from a region of high pressure to a
region of low pressure at a rate proportional to the pressure gradient.

Direct pressure measurements were obtained later for gases
in the plants in sign, but only after certain improvements in the
manometric system. These observations will be discussed in the next
chapter.

It is impossible to conclude on the basis of the data pre—
sented thus far, that these flows are an example of net gas flow from
the sediment to the atmosphere, as postulated in Chapter 1. In fact,
there remains a major inconsistency in the data. The total volume
flow of gas up individual petioles (as measured by tracer experiments)
was found to be several orders greater than the volume of methane flow
up the same petiole. For example, a leaf that loses several hundred
milliliters of gas per hour may only lose several milliliters of methane

during the same time interval. This means that the influx of methane

52

Figure 11.--Flow as a function of applied pressure.
Since flow rate was determined by the tracer method,
and the same segment of petiole was studied, the flow
rates have been expressed as the inverse of the elution
time (of the peak maximum). In this way, the linear
rates of flow can be compared with that observed in the
petiole before it was removed from the plant.

53

I
1.0

I
_ ,5
Applied Pressure [cm H20]

 

 

_
3

Age: 26.va WES; 2.5 Less

_
2

—

1

P

54

from the interstitial water only contributes in a very small way to
the observed flows. The same is true for the other significant gas
in the interstitial water, C02. Clearly, there must be some other

mechanism driving these flows; that mechanism is the subject of the

next chapter.

CHAPTER 3

GAS CIRCULATION

An Apparent Contradicition

 

Methane was found in emerging (not yet at the lake surface)
leaves in concentrations approaching those in the rhizome (Figure 7)
and exceeding the methane concentration in all but the oldest emergent

leaves (Table 6).

 

Table 6 Methane in Submerged and Emergent Leaves1

Leaf type 0830 hrs 2000 hrs
Submerged Leaf 0.99% 0.87%
Emergent Leaf (mean) 0.38% 0.32%

Once such a leaf reached the surface, however, the methane
concentration was usually undetectable (by thermal conductivity detector)

during daylight, increasing dramatically at night (Table 7).

 

1The concentrations shown are the volume percentage of methane
in the petiole of an emerging (but still submersed) leaf and the mean
volume percentage of methane in the remaining seven emergent petioles
on the same rhizome. Some of the oldest emergent leaves had methane
concentrations greater than that in the submersed leaf, reflecting
their basal position on the rhizome (cf., Figure 7).

55

56

Table 7 Methane Concentrations in Newly:Emerged and Older Leavesl

Methane concentration

Leaf type 1530 — 1630 hrs 0630 hrs
Newly-emerged <:0.1% 1.7%
Older, efflux 0.3% 1.7%

The data in Table 7 strongly suggest that the drop in methane
concentration in the youngest emergent leaves is the result of a gaseous
exchange with the atmosphere. This idea is supported by other experi-
ments where methane concentrations were found to increase in the
petioles of young leaves after the upper surfaces of these leaves had
been either coated with silicone grease or resubmerged (under water).
During the investigation of methane efflux from leaves, it became evi-
dent that the young, newly-emerged leaves did not lose methane to the
atmosphere (Table 4). This raises an important question about the
distribution of methane throughout the plant. Hence the contradiction:
there appears to be a significant exchange between the atmosphere and
the lacunae of the young leaves, and yet there is no flux of methane
from these leaves.

An investigation of the distribution of the other major gases
in the lacunae corroborated this "anomaly" in methane distribution

and CO

(Table 8). In Table 9, the average daytime distribution of O2 2

 

1The volume percentages of methane are condensed from the
data in Table 8. Thus the estimates for the young leaves are the
means of twelve determinations; those for the older leaves are the
means of sixteen determinations. All samples were derived from a
single shoot apex.

57

.Boumam
nwsounuIaon one mo some one o« pea uo: pan ya on .mmoH umooao one ou Homes mos upon wcuaaamm mEONfinu ssh
.ooaa Houma ego 30am; uman mums muuoa Home: mnu “wommusm unoEHomw onu o>onm moamumwm 05mm ecu manwoou Ham

whoa meson mafiaaamm Hoaoa may .mo>mmH xoamwo mums snow umooHo mnu mmo>moa xoawcfi mums wo>mma umowsoom
mouse ecu newsm50xw mow ow oHou yucca Ou wawouooom momaoa who kmsu moanmu omega aH .uowawuomxm ago
«0 wcwocfiwmn ago on own ucoumaam ou wcfiouooom Auwooao cu umwwcsozv peaceabs mums mmaowuoa wSH

.ooCHEuoumo use was coauwwanoo one umnu momma woman hwaao om “mammawcm mawafim m mo adamou msu we
mumefiuwm comm .mowmn poem axe: onu so cm>aw mum Nz mam . on . o .qmu mo mowmucmuuoa oEoHo> any

 

mnmfi .HNImH umowo< .xom4 uOOSm m uaonwoounu macaufimomEou moo m manmfi

(vol %)

Methane

Table 8

Time

Petiole rank

0030 hrs 0615 hrs 1630 hrs 2115 hrs 0630 hrs 1530 hrs 2030 hrs

2030 hrs

by apparent age

0.1 0.3 < 0.1 < 0.1 0.4 < 0.1 < 0.1

< 0.1

upper

1

2.2 3.4 < 0.1 1.1 3.1 < 0.1 0.7

0.9

lower

0.1
0.6

1.0 < 0.1
0.6 2.6 < 0.1

< 0.1
< 0.1

0.5
2.8

0.2
1.6

< 0.1

upper
lower

xnIJuI

0.2 0.4 < 0.1 0.1 0.7 < 0.1 < 0.1

< 0.1

upper

0.9 2.5 < 0.1 0.3

< 0.1

1.8

1.1

lower

58

0.6 0.9 0.1 0.2 0.5 0.1 0.1

0.6

upper

4

0.5

0.1

1.5
2.7

1.8 0.4 0.5
0.1

2.8

1.0
1.9

0.9
1.2

middle
lower

1.0

0.1

0.9

0.6 0.9 0.2 0.5 0.6 0.2 0.3

0.6

upper

1.9 2.8 0.3 0.7 1.1 0.2 0.6

1.2

lower

0.5
1.1

0.3
0.3

2.6

1.1

0.3

0.4 0.5 0.3
2.0 3.0 0.4

0.5
1.4

upper
lower

6

XUIJJH

1.1 1.6 0.5 0.8 1.4 0.4 0.8
2.7 2.3

1.3
1.4

upper

7

1.0

0.7

1.1

0.5

1.8

lower

8.4 9.6 7.9 8.0 7.1 7.9 7.5

7.7

Rhizome

59

~.m m.w «.5 m.w w.w N.m «.0 m.w mEONH5m

m.ml ~.ma o.a m.s~ m.m_ m.o e.oa m.q~ umaoa
o.ma m.m_ ~.oa m.ma o.o~ m.oH a.m~ «.ma page: a
N.ma o.mz e.~ a.ma m.w~ “.0 m.o~ o.ma “mace
m.ma a.oN m.m~ o.o~ m.a~ ~.ma e.o~ page: a
a.ma ~.o~ w.ez m.~a m.ma a.e m.aa a.ca nmsoa
a.ma a.o~ 5.0a o.ma o.o~ m.ma m.aa o.o~ “was: m
m.ma a.oN m.o m.ma s.o~ N.m a.o o.ma pesos
m.ma m.o~ o.oa m.~H e.o~ o.m o.ma «.ma weaves
m.ma o._~ ¢.ea m.ma ~.a~ N.Ma “.ma o.ma Home: a
m.mz a.oN a.m m.sa w.o~ o.aa n.0a Hosea
m.o~ m.o~ o.ma o.o~ o.o~ a.ma e.o~ amass m
~.~a o.o~ m.e a.sa o.- w.e ~.aa m.ma pesos
~.o~ n.oN m.sz o.H~ q.ma page: N
e.ma m.oN w.m m.m~ a.o~ s.m o.a m.e~ 5530a
m.o~ m.o~ m.ea m.ai m.o~ m.a~ o.c~ owes: a

my: ooo~ was omma my; omoo was maaw mu: ones my: mace m5: once we: omou mwm acoumaam an
Moms ouoauom
mafia

AN Ho>v .mmwwmm. Am.u:oov w canny

xnIJJH

xnIJuI

(vol %)

Carbon Dioxide

Table 8 (cont'd)

 

Time

Petiole rank
by apparent age

0030 hrs 0615 hrs 1630 hrs 2115 hrs 0630 hrs 1530 hrs 2030 hrs

2030 hrs

0.6

2.5 0.1

1.2
3.7

2.4 0.1
14.6

7.7

0.4

upper

1

0.2 2.9

7.9

0.2

7.7

3.7

lower

0.5
1.6

2.4 0.0
0.1

5.4

1.3 0.1
11.1 0.1 1.4

0.9
4.6

0.6

upper
lower

2

xnt;u1

0.5

2.5 0.1

0.6

0.1

0.2 6.7 1.5

upper

3

1.2

6.7 0.2

3.3

0.1

9.5

3.2

lower

60

1.0 2.2 0.1 0.8

1.3
2.7

6.1 4.1 0.1

0.8

upper

4

0.8

0.2

4.0

6.0 5.7 0.3

1.0
3.5

middle

0.9 2.9

6.6

0.4

7.1

7.2

lower

In~o <3
CS—I F:
rsox c>
c>c> F4
oxc>
.303
varx In
.3—3 c5
uses h~
c>c> c>
€>O\ c:
o3—4 «a
0—0
qu- OI
oauw —~
Innn .q
0 O O
c>v1 C)
H H H
m m m
81% E
:3 PI 5
:n \o

xnIJJH

5.6 6.5 2.2 3.1 5.6 2.0 3.5

4.3

lower

2.9 4.4 1.4 3.2
2.5

1.3
1.9

4.1 10.4 5.1

upper

7

3.4

6.1

3.5

6.6

11.2

5.2

lower

9.8 14.7 7.7 7.4 6.4 8.2 7.4

9.3

Rhizome

61

«.05 «.mm ~.0n 0.05 ~.0m «.mn m.0n H.0m maowanm

~.om 5.05 o.ew a.55 5.55 5.sm m.aw 5.w5 55305
w.m5 5.55 w._m G.A5 n.55 m.~m m.om 5.05 5555: 5
m.m5 5.55 5.mw m.A5 w.55 ~.mw 5.5m o.m5 55305
5.55 m.55 o.w5 «.55 4.55 5.m5 m.w5 “was: a
e.m5 5.55 a.om «.m5 w.55 e.mw 5.5m w.m5 55305
5.55 m.55 5.55 m.w5 «.55 m.om m.m5 5.m5 55am: m
m.A5 5.55 a.mm 5.ow 5.55 «.5w c.5m m.m5 pesos
m.w5 3.55 w.aw m.m5 m.55 H.mm c.5m q.a5 unease
m.w5 «.55 «.55 5.m5 c.55 5.5m o.ow 5.m5 puma: a
m.55 5.55 e.mm a.w5 5.55 5.5m m.m5 gazed
a.m5 o.w5 o.om «.m5 o.m5 m.55 w.m5 G.m5 page: m
o.ow 5.55 «.mw «.55 o.m5 5.mm n.5m a.m5 Emacs
m.m5 m.55 c.5w c.m5 «.m5 a.w5 “can: 5
5.ow o.w5 o.qw 5.5m H.m5 5.sw m.~m o.ow 55305
a.m5 m.m5 m.m5 o.w5 5.55 5.55 a.m5 . o.m5 5555: a

we; omom my; omma mg: once my; maam my; omoa my; mace m5: omoo m5; anew mwm uaunmaam an
Mam..." UHOHumm
mass

55 Ho>v amwouuaz 5e.uaouV w «Hams

xnIJJH

xnIJUI

62

are shown for the young and old petioles. Oxygen concentrations were
essentially atmospheric (21%) along the entire length of the young
petioles, while the levels were significantly lower in the older
petioles. Carbon dioxide (like methane) was very low in the young

petioles, higher in the older petioles.

Table 9 .9 and CO

2 2 in Newly—Emerged and Older Leaves (Midafternoon)

(st. dev. in parentheses)

 

Leaf type 0 CO

2 2
Newly-emerged 20.7 (.3) 0.12 (.06)
Older, efflux 20.0 (.9) 1.05 (.75)

Clearly, the behavior of the gases in the lacunar "interphase"

is not as simple as the representation in Figure B2 (Appendix B). Con-

sidering the shading in Figure BZ to represent CH and CO the model

4 2’
must be adapted in the manner shown in Figure 12.

The representation in Figure 12 suggests that the lacunar gas
spaces may constitute a flOWbthrough system, with atmospheric gases
entering the young leaves (influx leaves), while gas rich in CO2 and
CH4 escapes to the atmosphere through the older, efflux leaves. This
model resolves the apparent contradiction, suggesting that the methane
escapes from the petioles of the young leaves by moving down to the
rhizome, rather than by escaping out to the atmosphere through the
young leaf. This model also solves the discrepancy noted at the end of
Chapter 2, that the volume of gas moving up the efflux petioles far

exceeds the volume of methane flux. It offers a mechanism for the

large volume flows observed to pass out through the efflux leaves.

63

Figure 12.-—Modified interphase.
The distribution of gases in the yellow waterlily
suggests that a distinction must be drawn between the
young emergent and the old emergent leaves. During
daylight there is very little CO or CH in the young
petioles, and a considerable amount of both in the
older petioles. This idealized drawing is a
modification of the hypothetical interphase in Figure
B2.

64

 

 

 

 

 

 

 

 

 

' o
a on... .‘n.
ADA-m'k'

 

 

65

A Flowahrough System

 

Tracer experiments were undertaken to study the movement of
gases throughout an isolated shoot apex. All the emergent leaves of a
shoot apex were enclosed in Saran bags (cf., Figure 8), and 5 ml
CZH6 was injected into the upper part of the petiole of the youngest
emergent leaf (with a 30 gauge needle). The composition of the gas
in the bags was sampled at intervals. Experiments with three plants
at various depths in the lake (21 leaves total) showed that an average
of 66% of the tracer had left the plant within several hours. As
shown in Table 10, virtually none of the tracer had entered the bags

on the youngest leaves.

 

Table 10 Tracer Efflux from a Shoot Apex

Petiole rank Amount of C H efflux

by apparent age from the leaf C2H6 in petiole
1 None detected 83 ppm

(0 H injected here)

2 6
2 0.03 ml 274 ppm
3 0.79 ml 422 ppm
4 1.01 ml 1154 ppm
5 1.26 ml 1238 ppm
2 = 3.09 ml

The distribution of C2H6 throughout the plant confirms the
pattern of leaf efflux. Even though the tracer (5 ml 02H6) was
injected into the top of petiole #1, the concentration of ethane in
that petiole is lower than in each of the others (Table 10). Further-

more, the concentration of ethane in each petiole correlates well with

66

the amounts of ethane lost from that leaf (Table 10).

These data show unequivocally that there is a fluid flow of
gas down the petiole of the youngest emergent leaves. The flow con-
tinues into and through the rhizome, up the petioles of the older
emergent leaves to the atmosphere.

Experiments with bags on these newly-emerged leaves showed
that the air moving down these petioles comes directly form the atmos-
phere. When a tracer was added to the bag on such a leaf, the gas
composition in the petiole of the leaf quickly approached that of the

gas in the bag (Table 11).

 

Table 11 Ethane Transport throughout a Shoot Apexl
Petiole rank Ethane Concentration as a Function of Time
by apparent age 5 min 15 min 30 min 75 min 140 min
1 —- 0.16% 0.18% 0.18% 0.18%
(< 0. 01%)
2 _ _ _ _ _
3 - —- 0.01% 0.04% 0.02%
4 ~ —- —- —- 0.03% 0.03%
5 - —- —- 0.01% 0.01%
Bag gas 0.19% 0.19%

The absence of tracer in the gas of petiole #2 shows that it
is also an influx leaf. Its presence on the shoot acts to dilute the
tracer being forced into the rhizome by leaf #1, thus the concentrations

of ethane remain low in the efflux petioles even after several hours.

 

1A bag was tied onto leaf #1, and ethane was injected into the
bag at time - 0. A dash means that the ethane concentration in the
sample was less than 0.01%.

67

The fact that the flowhthrough system draws its air from the
atmosphere is corroborated by another line of evidence. It was re-
peatedly observed that bags on the influx leaves tended to lose volume
over their period of attachment. Using volume estimates based on
internal standard injections, bags were monitored for several hours,
during which time their volume frequently decreased by several liters.
Better quantitative support for this influx will be presented below,
but these observations with bags vividly illustrated the capacity for
the leaves to withdraw air from the atmosphere. In some cases, bags
containing nearly 6 liters of air when attached to the leaf were
almost completely collapsed after several hours in sunlight (to
volumes of a few hundred milliliters).

Clearly, the flow-through system is not a mechanical process
of the sort exhibited by breathing animals. The waterlily is not
forcing air through its lacunae by collapsing and expanding some
section of its anatomy.

The most widely-known mechanism for such a flowbthrough
system has been described by Vogel (1972, 1978). He showed, for
example, that air is circulated from one end of a prairie dog tunnel
system to the other. His model depends on an external air current in
the atmosphere. According to the Bernoulli equation of fluid flow
(Appendix C), the pressure applied by a moving fluid on the walls of
its container decreases as the velocity of the flow increases. In
the case of the tunnel system, the fluid is the atmosphere, and the
ground is the surface along which it flows. By placing the openings
of their burrows at different heights above the ground surface, the

prairie dogs expose these openings to air moving at different velocities.

68

Since the pressure at an opening of the tunnel is a function of the
wind velocity there, a pressure gradient is set up along the length of
the tunnel system. Air is taken in through the lower holes and vented
to the atmosphere at the more elevated holes. (For further discussion
see Appendix C)

The similarities between the tunnel system and the waterlily
system are obvious, but the mechanisms underlying the two are very
different. For example, in the tracer experiment represented in Table
10, all the leaves of an isolated shoot apex were enclosed in bags.
Thus, there could have been no velocity differential to support the
kind of flowbthrough described by Vogel. Furthermore, the flows can
be stopped by damaging the influx leaf. In a representative experiment,
a leaf was torn inside its bag after it had been allowed to draw down
the volume of the bag from 3.0 liters to 0.4 l (in 80 minutes). Within
3 hours of tearing the leaf, the volume of the bag increased to more
than 4.5 l.

The important point demonstrated by that experiment is the
fact that the mechanism for the pump lies in the influx leaf. It has
already been pointed out that stopping the exchange across the leaf
surface (by silicone grease, for example) causes an increase in methane
and carbon dioxide concentrations in the petiole. Tearing the leaf has
exactly the same effect.

The capacity for these leaves to draw air into their lacunae
was vividly demonstrated by inverting a large beaker over an influx
leaf (Figure 13). The styrofoam collar around the beaker held it in
position, and the large tube allowed the escape of air from the beaker

as it settled into buoyant equilibrium over the leaf. Once the pressure

69

Figure 13.--Inverted beaker at t = 0.
A four-liter glass beaker was inverted over a young
emergent leaf. Air was allowed to escape from under the
beaker through a large rubber tube until the beaker
reached buoyant equilibrium maintained by the styrofoam
collar. The tube was then pinched shut, isolating the
atmosphere inside the beaker.

70

 

 

 

 

 

 

 

71

inside the beaker had equilibrated with the atmosphere, the tube was
pinched shut. With a healthy, undamaged leaf exposed to the sun's
illumination inside the beaker, the water level in the beaker could be
seen to rise up the inside wall of the beaker (Figure 14). This means
that the leaf drew a vacuum inside the beaker in the process of trans-
porting air into its lacunae. On a few occasions, the water level
rose nearly 2 cm above the lake level (meaning that the leaf had drawn

a vacuum 0.002 atm less than barometric).

Pressures and Flows

 

Just as the young leaf tended to draw a vacuum under the
beaker, it also tended to elevate the pressure in the lacunae of the
leaf. Using a micromanometer (Gilmont) connected to the petiole in
the manner shown in Figure 15, pressures were repeatedly measured in
the lacunar gases that were significantly higher than barometric (the
highest pressures being more than 2 cm water, or 0.002 atm greater
than barometric).

These elevated pressures in these leaves drove the fluid
flow of gas along a pressure gradient into the rhizome and subsequently
up the petioles of the older emergent leaves to the atmosphere. The
flow rates of gas down these petioles were measured by the same method
used to study the rates of volume efflux described in Chapter 2. In
this case, however, the tracer was injected into the upper end of the
petiole, and the sampling was performed at the lower end. As in the
case of efflux, integration of the tracer concentration-time curve
showed that all the tracer passed the sampling point at the base of

the petiole.

72

Figure 14.--Inverted beaker at t.
When a healthy young emergent leaf was exposed to the
sun's illumination under the inverted beaker, it
repeatedly drew a vacuum in the beaker (as manifested
by the water level rising up the inside wall of the
beaker). If the leaf were damaged, or old, it showed
no such capacity to evacuate the beaker.

There was no measurable change in the relative
proportions of the atmospheric gases in such an
evacuated beaker (except for water vapor).

73

 

 

 

 

74

.pHon ofiuuoEoems
osu on new: on? Aae\w 5.0 haemoopv manuUOIomH .Auoumsooma onu ou
aoaw mow oumuaafiomw ouv wcfinsu ouon owuma m can waves: mwsmw mm m we
momma he mHowumn no awulpaa mama onu Op peacoccoo mos Houosooms may
HoomEoomaouofiZII.mH muswam

76

In Figure 16, the volume rate of flow down the petiole of a
young leaf is shown to be a linear function of the pressure in the
upper petiole. This relationship proves that the gases inside the
lacunae behaved according to Darcy's law for fluid flow through porous
media. The rate of flow through the petioles of the waterlily depended
on the pressure gradient along the petioles. Furthermore, the dominant
process for pressurizing the internal gases during summer growth occurs

in the young, newly-emerged leaves.

A Pump

It certainly seems reasonable to call this uptake of atmos-
pheric gas a "pump". Since the gas entered the leaf against a pressure
gradient, the system has the capacity to do work: gas is forced out
through the older emergent leaves. The mechanism must depend on an
investment of energy.

The obvious source of energy for this pump is the sun. Shad-
ing an influx leaf had the immediate effect of decreasing the pressure
of its internal gases as monitored by manometry. In Figure 17, the
pattern of pressure exhibited by a single leaf during a day is drawn
as a function of light intensity. Contradicting the "first criterion
of a mediated-transport process" (Lehninger, 1970), the pump does not
show any tendency to become light-saturated in these data, or in any of
the other data collected during this study.

To further deny the possibility of the pump being a biological
process, it was found that the pump's dependence on the sun was for its
heat, not for its shorter wave radiation. Rapid changes in pressure in

the lacunar gas of a young leaf were initiated by holding a hot electric

77

.moofiufipooo unwfia wcawcmso ou mom pmwsmso musmmoua nfiulmfia may no hop ecu

ozonwsounu mavfionomu noumuu mnu me down :a mocfiauouop puma moons 30am may
.wmoH xoamaa om aw muommoua nfiulofia onu mo cowuoaow o no mums sloII.0~ ouswfim

 

78

2b

130
Pressure [cm H20]

 

 

60"

I l
O O

[ungw] 9123 Mo”

79

.:0Huoaomofi Esaaxma mo moowuoa weapon pounce yaamfiowmauum

mm3 wood can cons uneasy yHumHHsHm mos wmoa spams“ cm a“ whammoue may .oofium>

augmcmuow unwed ecu no how oawofim m usoswaounu oouooaaoo mama some can .ommo menu aH

.mmoH uoowuuao mosey m mo mommuam ego ou Hashes coauoouao one ca :oxmu mums

mmwuamoouow unwed .auwm mm.pou=mmme one: wofiufimomuoa unwwa 0am monommoua may
.yufimcmuofi uswfia mo cofiuocom o no ousmmmum naquHZII.5H onowfim

 

80

moomqu\m:_3m:_mqfi_ .3352. EMS

 

ooop own
LI 4
d
J
. “w.
o In s
n
I.
. a
4M...
. 13 w
o H
. .mu
. 5|.
0 Im.-.

 

81

frying pan at various distances from the leaf!

In a more detailed study, a young leaf was cut away from its
rhizome and held with its petiole's cut surface below water (to restrict
desiccation of the leaf tissues). A micromanometer was attached to the
upper part of the petiole, and a 23 gauge thermistor probe was inserted
into the mid-rib of the leaf. A heat lamp was directed towards the
upper surface of the leaf, and the consequent fluctuations of pressure
and temperature as irradiance was varied, are drawn in Figure 18.

A third line of evidence supporting the notion that the pump
is a purely physical process was supplied in a study of the gradients
in gas composition between the atmosphere and the lacunae in the mid-
rib of an influx leaf. In Table 12, these data are presented in terms
of chromatogram integration units, since compositional gradients (not
absolute quantities) are under investigation. Standard deviations for

the samples are in parentheses.

Table 12 Comparison of Ambient and Lacunar Gas CompOsitions

 

co 0 N o /N

2 2 2 2 2
Ambient Atmosphere
(Lake) n=7 272 (41) 92749 (629) 343765 (2629) .2698 (.0003)
Lacunar Atmosphere °
n=13 233(126) 91930 (612) 339592 (2124) .2700 (.0008)

The volume-percentages of O2 and N2 in the lacunar gas are
significantly (5% confidence) less than their respective concentrations
in the atmosphere. The concentration of CO2 is much more variable, and
does not appear to be different in the two phases. These data show

that there is no strong selection for one gas over another; 02/N2

82

Figure 18.-—Lacunar pressure as a function of leaf temperature.
The leaf was removed from the plant, and the cut surface
of the petiole was held underwater (to prevent gas loss
from the petiole and restrict leaf wilting). The leaf
temperature was monitored with a 23 gauge YSI thermistor
probe inserted into the mid-rib. The temperature of the
leaf was changed by holding a heat lamp at varying
distances from the leaf. The temperature of the room
air (about 17°C) was measured constantly; room air was
blown past the leaf surface by a fan in order to
maintain a steep temperature gradient between the leaf
and the air.

83

2'2 2'4 2'6

1'8 2'0
Leaf Temperature [°C]

16

 

2.4“

20-

.
8
1

q d u -

5 4. 2 O.
1

1 1L 1

hos: Eog 2:325

consume

14

 

O

84

ratios are not significantly different in the two phases. If the pump
were biologically mediated, it would most likely be highly specific
for the gas transported; such selectivity should lead to accumulation
of that gas in the lacunae.

The decrease in the volume-percentages of oxygen and nitrogen
can be explained by increased humidity (water vapor) in the samples
taken from the lacunae of the leaf. The depletion is about 1% (0.01
atm), which is well within the range of humidity gradients observed
between the leaf and the atmosphere outside the leaf. Estimates of the
water vapor pressure in lacunar gases were highly variable, but were
consistently higher than those for the atmosphere outside the leaf.

For example, in a representative case the total pressure in the lacunae
of the leaf was 0.002 atm greater than barometric and the water vapor
pressure in the leaf was 0.010 atm greater than barometric. This means,
therefore, that there was a 0.008 atm gradient in the partial pressures
of the constituents of dry air (N2, 02, etc.) from the atmosphere into
the leaf. This gradient supports the diffusion of these gases into the
lacunae, while water vapor tends to diffuse from the leaf to the atmos-
phere (along the 0.010 atm gradient). This process is described more
fully below.

Interestingly, some of the strongest evidence for the physical
basis of this pump was found in a body of literature more than a hun-
dred years old. This early work has been entirely ignored by modern
researchers; in fact, it has received very little attention since 1912.
Perhaps the lack of recognition afforded this work results from.the
timing of the original research. The literature is reviewed here in

detail, in an effort to put the work into the context of its time.

85

During the time span discussed below, two relevant conceptual frame-
works were evolving; presumably the ideas presented here "fell through
the cracks".

The kinetic theory of gases, which forms the basis of our
modern understanding of gases and their behavior,'was only developing
during the 19th century, and was only consolidated during the early
part of this century (cf., Loeb, 1934). Simultaneously, the fundamental
principles of plant physiology were being developed. Much of the
relevant discussion throughout this time period was directed at the
role of stomata in gas exchange (cf., Haberlandt, 1914). It is worth
noting that Haberlandt makes no mention of this body of literature in
his monumental review (1914) "Physiological Plant Anatomy", even though

he himself had investigated the stomata of waterlilies (1887).

IT
"Seerosenphanomen"

 

By the middle of the 19th century, it was widely recognized
that stomata are the principal opening to the atmosphere for the
internal gas system in plants, although it wasn't until the end of the
century that their precise role in gas exchange was established.
Dutrochet (1837) reported that gases streamed from the base of a cut
petiole of Nymphaea when the leaf was exposed to light. He observed
that this only occurred when the leaf was submerged, and concluded that
the stomata on the upper surface of the leaves allowed the escape of
oxygen produced during photosynthesis ("normal mode of respiration in
leaves"). When a leaf is submerged, this avenue for gas escape is cut
off, and the gas is forced down the petiole and out the cut end in a

stream of bubbles. He also reports in the same discussion that the

86

leaves could be asphyxiated when deprived of the oxygen produced in
photosynthesis, since 02 is needed for respiration ("the auxiliary
mode of respiration"). He does not comment on the possibility that
oxygen might also be available to the respiring plants from the atmos-
phere by way of the stomata.

Raffineau-Delile (1841) reported a similar bubbling process
from the leaves of Nelumbo (a related waterlily). Her claim was
fundamentally different than Dutrochet's, however. She found that the
bubbling stapped (rather than began) when the leaves were submerged.
This discrepancy precipitated a series of sharp exchanges between her
and Dutrochet; these remarks are published in the pages following
Raffineau-Delile's original report.

The discussion of plant gases continued, with Lechartier
(1867) and Barthelemy (1874) emphasizing the role of gases entering
the plant from the sediment water in determining the pressure rela-
tions of the plant gas. Barthelemy concluded that: "the stomata,
always in contact with (the air canals), have the purpose of allowing
the internal gases to pass to the outside; whereas they are generally
disposed so as to oppose their re-entry." Furthermore, "gas movements
are caused by evaporation, variable temperature distribution, atmos-
pheric variability, etc."

The first discussion of a mechanism for the pressurization
of gases in the leaf observed by Raffineau—Delile (and again in this
study) was put forth by Merget (1873). He explained the process in
terms of "thermodiffusion", a term borrowed from a physicist, Fedderson

(1873).

87

With the work of Maxwell outlining the velocity distribution
of gas molecules (ca. 1867), it became apparent that gases should
segregate between the two sides of a porous membrane that supports a
temperature gradient. Fedderson was apparently the first to investigate
these thermal effects, but according to Violle (1875) and Reynolds
(1879), his results were confounded by the presence of water vapor in
his experimental apparatus.

The phenomenon studied by Reynolds, and termed "thermal
transpiration" by him, is discussed in Appendix D. In short, the
thermal transpiration of gases through small pores leads to an equilib-
rium condition wherein the pressures on either side of the porous
partition are proportional to the square roots of their respective

temperatures. Accordingly,

PI '1'
_— . l?

where p1 and T1 are the pressures and absolute temperatures of the
respective gas mixtures at equilibrium.

The pressure differentials across such a porous partition
can be further accentuated when the gas on one Side of the partition
is in contact with a volatile liquid. Dufour (1872, 1874) studied
this effect of differential "humidity" on opposed sides of a porous
partition, and found that air tends to move from the drier side toward
the wetter side (cf., Appendix E). Such a flow of gas supported a
sustained stream of bubbles when a hose from the wet side was held
under water. This effect can be achieved without a temperature

gradient (as required by the related process of thermal transpiration).

88

Merget (1873) applied the principles of Dufour to explain the movement
of gas into the leaves of Nelumbo. He verified that the composition
of the external gas was not important in the process; it worked with
oxygen, nitrogen, hydrogen, carbon dioxide, carbon monoxide, and
nitrous oxide.

Merget (1874) further demonstrated the purely physical nature
of the gas transport system in experiments with dead leaves. Leaves of
Nelumbo were dried to brittleness, stored for five months, and remois—
tened by holding them between damp cloths.' Once remoistened, these
leaves again exhibited their capacity to support "thermo—diffusion".

Ohno (1910) repeated much of that work and also showed that
cooling the air outside the leaf had the effect of increasing the
pressure inside the leaf (cf., Appendix E).

The literature is well reviewed by Ursprung.(l912). The
only original English text that refers to the work is Arber's.(1922).
Her treatment is very superficial, and suggest that she failed to
understand the mechanisms proposed in the earlier literature.

This phenomenon is_again discussed by Gessner (1959), with
no new data added. Gessner accepted the purely physical nature of the
pump, and stated that its mechanism lies in some unspecified thermal
effect. He rejected the model proposed by Kundt (1877) which is
similar to the hygrometric diffusion component supported in this
thesis (Appendix E). Gessner went on to speculate briefly that the
effect must somehow be derived from the small size of the pores and
the changing character of flows through pores as their size decreases.

In his discussion of this process, Gessner made some brief comments

89

about the possible consequences of such pressures on plant metabolism,
and, in what appears to be the last published word on this topic to

date, he suggested that the question should be studied further.

Mechanism

It is evident from the work of Merget, Ohno, and this study
that the pump is the result of purely physical processes. As suggested
above and in Appendices D and E, there are two parallel mechanisms that
contribute to the generation of pressure in these leaves — thermal
transpiration and hygrometric diffusion. Thermal transpiration
(Appendix D) leads to pressure gradients simply of the basis of
temperature gradients across a nanoporous partition. "Hygometric
diffusion" leads to the same result wherever there are vapor gradients
across such a partition. Both these processes would lead to a pres-
surization of the lacunar gas when exposed to sunlight. Thermal
transpiration requires a continuous input of heat to maintain the
thermal gradient. Hygrometric diffusion requires a continuous input
of heat to vaporize water in the leaf. It also requires a steady
supply of water so the leaf won't dry out; this water is presumably
supplied by the plant's conductive tissue.

By the very nature of the system, it is impossible to
distinguish between these effects in practice. Both processes should
operate simultaneously and independently. Together, they should lead
to a steady state where the total pressure gradient is the sum of the
two independent gradients. Thus, under ideal conditions, the steady
state pressure is the sum of the thermal transpiration pressure

p2(Tl/T2)35 and the hydrometric gradient w-w2 (llwater vapor pressure).

1

90

For example, when the leaf in Figure 18 was 22°C, the air
being blown across its surface (by fan) was 17°C. Ideally, the
equilibrium thermal transpiration gradient would be 0.008 atm.
Assuming the room air was saturated with water vapor, the hygrometric
pressure would be 0.007 atm. Therefore, ideally, the total pressure
gradient would be 0.015 atm, or 15 cm.H20. The relative humidity in
the room was certainly considerably below 100% saturation, adding
still more to the expected pressure gradient. For example, at 50%
saturation for the same data, the ideal total pressure gradient would
be 24 cm H20. However, as shown in Figure 18, the actual pressure
measured in the petiole of the leaf was 1.4 cm. Clearly, the leaf is
far from being an "ideal" diffusive exchange partition.

In their normal state, these leaves are attached to a petiole
that allows the escape of gas down to the rhizome. Therefore, the
pressures measured in such leaves represent a steady state balance
between the net rate of diffusive entry into the leaf and the loss
from the leaf in fluid flow down the petiole. The leaf in Figure 18
had been cut from the rhizome and its cut petiole was sealed so gas
could not escape through the petiole.

There are, however, a number of other explanations for the
continued divergence of observed pressure from ideal behavior. For
example, temperature gradients across the surface of the leaf deny the
homogeneity of temperature that the ideal calculation assumes.

These calculations also require that the exchange between the
atmosphere and the lacunar gases be diffusive. These conditions are

only met when the average pore size through the exchange surface is

91

small relative to the free mean path of the molecules in the gas. For
a gas mixture at 1.0 atm, the free mean path is about 10_5cm, so the
pores in the leaf must be smaller than O.Olum.1

The fact that the leaf pressurizes at all means that the
influx system is predominantly a diffusive system. Over the surface
of the leaf there is presumably a range of pore sizes, and the pressure
in the petiole represents a net effect of the exchange processes over
the entire leaf.

As the leaves age, and become more porous (presumably by
increased stomatal apertures and increased porosity in the pallisade
parenchyma below the stomata), the leaves lose their ability to resist
fluid flow. The diffusive processes of gas uptake give way to the

fluid processes of gas efflux.2

 

1Direct measurement of pore sizes by surface impression

methods and scanning electron microscopy have been unsuccessful. It
is likely that the pores lie below the leaf surface in the unexpanded
parenchyma of the young leaves (cf., Esau (1953) for discussion of leaf
expansion). An estimate of pore dimensions can be made on the basis of
a few assumptions. Although these are likely several pores (in the
parenchyma) below each stoma, a calculation assuming one pore/stoma will
give a maximum pore size.

4 Since air molecules strike a surface at a rate of roughly
10 m1/cm sec (Danckwerts, 1971), they can be said to enter the pores
in the leaf surface at the same rate. Since there is roughly a .01 atm
gradient in the partial pressures of (dry) air molecules, the net flux
into the leaf (e.g., 30 ml/min) represents 1% of the total flux (or
100 ml/cmz sec). This means the total pore area per leaf is about
5 x 10-3th. With 1.5 x 107 stomata/leaf (on a leaf grown in the lab),
the average pore area (one pore/stoma) is therefore 3.3 x 10'10cm2.
Assuming these pores are circular in cross-section, the diameter of a
pore is 2 x 10' cm. This is certainly an overestimate of the average
pore size, but it serves to illustrate that the pores are very small.

2There is another relevant change that accompanies leaf aging.
The youngest leaves are commonly pigmented with anthocyanins, so they
appear red or reddish-green. This pigmentation presumably has the

92

Physiological Significance

 

It is actually this aging process of the leaves that makes
the pressurization phenomenon a physiologically significant process:
it gives the system its flowbthrough character. The newly-emerged
leaves are responsible for the increased pressure in the lacunar gas;
the older emergent leaves allow the escape of the pressurized gas to
the atmosphere (Figure 19). The flowbthrough system that results has
two physiologically significant effects: oxygen is transported to the
rhizome from the influx leaves, and carbondioxide is carried to the
efflux leaves.

The transport of oxygen into the plant is reflected in the
strong diurnal pattern of oxygen concentration throughout the plant
(Table 8). During daylight, 02 at atmospheric concentration was found
in the young leaves and petioles; its concentration is significantly
diminished as gas enters the base of the efflux petioles (Table 13).

Pressure data were collected throughout the day for the leaf
used in the flow rate/pressure regression (Figure 16). Integration of
the pressure throughout the day showed that more tha 16 liters of air
passed down that petiole during the day. That translates into several
liters of oxygen entering the rhizome from this one petiole. During the
summer growth period there are usually several of these influx leaves

on each shoot apex, so the rhizome may be receiving considerably more

oxygen.

 

effect of increasing leaf temperature during daylight, thereby facili-
tating the pressurization phenomena. The leaves gradually lose their
red color (turning green), reflecting the decreased importance of
elevated temperature in these older (efflux) leaves.

93

Figure 19.—- The pressurized gas flowbthrough system.
Air enters the youngest emergent leaves against a pres-
sure gradient. This elevated pressure forces a fluid
flow of air down their petioles to the rhizome, and
back out to the atmosphere through the older emergent
leaves.

95

Am.HvNy.~w An.vN<.N mm.~0N0.0 A0.~vnw.0 Ao.an.y5 Am.vum.o A0.VN¢.0~ “0.0N0.H Mason

Am.~vN~.ow Aq.vnm.o Aw.~vno.ma a0.~vum.m A~.vN¢.yy afi.va.o An.vum.o~ Am.vNy.o Home:
52 emu No N8 Nz emu N5. N8
oaauunwfiz «awoken oaoauon so coauunom

Anemonuoouma as .>oo .umv
oaHuummmz oco,oaauyon mousse moaoauom onmmu ow measuanomaoo mow ma canny

96

The pattern of CO flowing through the plant is also evident

2
in Table 8. A gas mixture with several percent CO2 is carried from the

rhizome toward the efflux leaves; the concentration depends on the time
of day, and presumably, the flow rate.

In Table 13, it is evident that the 00 level drops dramati-

2
cally as the gas moves up the petiole. All the evidence suggests that

the CO decline is due to photosynthetic fixation, in the leaf and

2

possibly in the upper petiole. Its unlikely that the CO is diffusing

2
from the petiole into the surrounding water, since its residence time
in the petiole is only a few minutes. At night, however, large con-
centrations of CO2 persist without diffusing away to the water.

The 002 gradient (0.7%) compares well with the opposite
oxygen gradient (1.1%) during daytime, suggesting photosynthetic
fixation as the gas mixture flows up the efflux petioles. Further-
more, there is no equivalent gradient in methane, which is a conserva-

tive gas within the plant, and like 00 it enters the petiole from the

2
rhizome.

These gradients in gas composition are further documented in
Table 14. using the ethane tracer method, the flow rate up the petiole
and O

was found to be 8.7 ml/min. Again, 00 exhibit the strongest

2 2
concentration gradients along the length of the petiole, suggesting
photosynthesis in the leaf and petiole. These data show that for a
plant in situ, three hundred microliters of CO2 (equivalent to all the

002 in 1 liter) were transported through the petiole each minute.

97

Table 14 Gas Composition Gradients along an Efflux Petiole (8.7 ml/min)

C02 02 CH4 N2
Leaf mid-rib 0.6% 19.6% 0.7% 77.7%
Upper petiole 1.6% 17.9% 1.4% 78.2%
Lower petiole 3.6% 16.3% 1.5% 78.7%

The relative contributions of the processes that produce the
CO2 concentrations in the rhizome were not established. The CO2 levels
found there must result form.both plant respiration and exchange with
the sediment water. Laing (1940b) has established that the rate of
respiration of the rhizome can be considerable, even under anaerobic
conditions.

As suggested in Chapter 1, there is also considerable
potential for the diffusion of CO2 into the roots and rhizome from
the interstitial water. A profile of the partial pressure of 002
(Figure 20) shows its range to be from 0.03 to 0.14 atm.in the upper
50 cm of the sediment. Since the depth profile of the roots and
rhizome surfaces for gas exchange are unknown, it's impossible to
estimate the steady state 002 concentration to be expected. The
average partial pressure of 002 over the upper 50 cm was 0.11 atm.
On this basis, the partial pressure in an equilibrated gas phase
would be 11% (at 1 atm). As shown in Table 8, the CO2 concentration
only exceeded this amount once in 8 samples. This suggests that there
is usually a diffusion gradient for 002 into the buried lacunar
spaces, but the kinetics remain uncertain.

It's important to realize that this flux of 002 up the

petioles from.the rhizome is not the only source of 002 available to

98

Figure 20.—-pH and P002 profile in Duck Lake sediments.
The values were determined by the methods described in
Chapter 2 for dialysis-equilibrated samples. The
shaded area in the P00 profile represents the area
integrated to give the 0.11 atm estimate used in the
text. These data were obtained from samples taken from
the lake during early October. Seasonal pH data for
the interstitial water of nearby Wintergreen Lake suggest
that the P002 profile does not change greatly through the
summer. These values may be slightly lower than the
summer average, since the pH tends to rise during winter
(lowering P002).

99

Partial Pressure

 

 

pH of CO, [atm]
6 7 8 .04 .08 .12 -16
1 l _ l l l I
0—
“ER
:2:
g 20-
u
D.
0
a
40"

 

 

 

DUCK lAKE SEDIMENTS

100

the plant. There is great potential for rapid exchange of 002 between

the atmosphere and the emergent leaves; its importance remains uncertain.

Nitrogen

In a sense, N2 acts as a carrier for the three gases of biolog-

ical interest (C02, 0 and CH4) —-it is a passive gas, and the dominant

2,
gas in the lacunae and the atmosphere. Even so, its dynamics cannot be
interpreted without understanding the dynamics of the other gases.
Contrary to a strictly diffusive model, there are considerable increases
in the concentration of N2 in the rhizome at night (Table 8). These
presumably result from the exchange (during respiration) of relatively
insoluble 02 for the more soluble (in plant tissue and sediment water)
002. This apparently leads to a slight decrease in the gas pressure in
the lacunae. It's unlikely this phenomenon has any significant
physiological effect, but it serves to make the point that the dynamics
of individual gases cannot be understood without knowing a good deal

about the other gases in the mixture.

Conclusions

 

By means of a purely diffusive exchange with the atmosphere,
the pressure of the gas phase in the lacunae of the youngest emergent
leaves increases over barometric. This elevated pressure drives a
fluid flow of gas down their petioles to the rhizome, and from the
rhizome up the petioles of the older emergent leaves to the atmosphere.
This pressurized flowbthrough system serves to deliver large volumes
of oxygen to the rhizome buried in anaerobic sediment. It also serves

to carry carbon dioxide from the rhizome to the photosynthesizing

101

tissue of the older emergent leaves}

In current theory, the transport of gases in the lacunae of
plants proceeds along diffusion gradients which are thought to be
primarily the product of gas metabolism in the plant--the varying
uptake and release of CO2 and 02, and the release of water vapor during
transpiration. These gases diffuse independently within the gas
mixture according to the gradients in their partial pressures. That is
the model of gas transport proposed by Laing (1940) for the yellow
waterlily. Except where external air currents may influence the
transport of gases internally (cf., ngel, 1978), this diffusion
of gases in a static gas phase is the model for gas flux in all plants.

Clearly, the generality of that model must be reexamined.

The gas flows in the waterlily do not depend on external air currents;
they are the product of diffusive exchange with the atmosphere. This
process is purely physical, but its significance lies in its biological

implications.

 

1Larvae of the moth Bellura sp. (cf., Levine and Chandler,
1974) were found frequently during this study. The larva burrows
through the surface of the emergent leaves of Nuphar and down into
their petioles. Such an attack on a young leaf effectively stops
its flow of gas down its petiole by allowing the pressurized gas to
escape to the atmosphere through the burrow opening. This efflux was
confirmed by covering the opening of a burrow'with a thin soap film-—
it quickly expanded into a convex bubble, and burst. Such termination
of the flowbthrough system must have short-term effects on plant
metabolism, but it's not clear what effect long term infestations might
have on the waterlily population. Clearly, the impact of the larva
may depend more on the location of its attack (i.e., young emergent
leaves) than on the actual amount of plant biomass consumed.

SUMMARY

The underlying principle in this study can be stated con-
sisely: although the gas mixture in the lacunae of plants manifests
biological processes, it obeys physical laws. Its behavior cannot be
adequately described in biological terms alone. The dynamics of the
gases can be completely understood only within a physical framework.

In the case of the waterlily, this approach has uncovered
a little-known physical process that has a profound impact on the
physiology of the plant. By means of what is nearly a "biological
steam engine", large volumes of oxygen are transported to the plant
parts that lie buried in the anaerobic sediment. Simultaneously, car-
bon dioxide and methane are carried from these sediments toward the
atmosphere. Much of the carbon dioxide is subsequently fixed by plant
photosynthesis, and the methane escapes to the atmosphere at rates that
are significant in terms of the carbon and energy budget of the lake
ecosystem. The fact that heat supplies the energy behind this gas

circulation system is a striking reminder of the economy of nature.

102

APPENDICES

APPENDIX.A

GAS SOLUBILITY

Ecologists find the concept of gas solubility easy to under-
stand intuitively, but rather harder to deal with in practice. Without
an understanding of the principles involved, a literature has arisen
which misapplies terms, and subsequently confounds the interpretation
of physiological and environmental conditions. Since the work in this
study is directed towards understanding the relations of gases (in gas
phase) in plants, this discussion will focus on the principles that
apply in such a study.

First consider the gas phase. Dalton, a pioneer of atomic
theory, observed that the total pressure exerted by a mixture of gases
is equal to the sum of the pressures which the several gases would
exert if each were to occupy the vessel alone (Dalton's law). Avagadro's
rule can be restated to say that: "under the same conditions of temper-
ature and pressure, equal volumes of different gases contain the same
number of molecules“ (Loeb, 1934). Combining these two statements, it
becomes clear that the volume-percentage of a gas in a.mdxture of known
total pressure can be directly related to the partial pressure of that
gas. This simple assertion is important because it allows a direct
relation between the volume estimate achieved by gas chromatography and

the partial pressure of the original sample.

103

104

Henry's law states that the solubility of a gas in a liquid
phase is proportional to the partial pressure of that gas in the equili-
brated gas phase. A gas dissolves in a liquid until the rate of escape
of gas molecules from.the surface of the liquid equals the rate at
which gas molecules enter the liquid. The converse also holds true,
as in the case of bubble generation. Gases leave the liquid until
equilibrium is achieved with the adjacent gas phase. At equilibrium,
the partial pressure of each gas in the bubble equals the partial
pressure of that gas in the aqueous phase. 3

Consider the case of a volume of water in equilibrium with
the atmosphere (i.e., 21% O2 and 78% N2 at 1.0 atm). A small bubble

at the surface of the water will contain 21% O and 78% N2 (or 0.21 atm

2
O2 and 0.78 atm N2, respectively). If this bubble were moved down into
the depths of the water, it will lose volume due to the hydrostatic
pressure. As the pressure in the bubble increases, so do the partial
pressures of the constituent gases. Since the partial pressures of
these gases exceed their partial pressures in the water column, the
bubble dissolves.

It is clear that in order for a bubble to remain at depth,
the partial pressures of the gases dissolved in the water must equal,
or exceed, the partial pressures of the gases in the bubble. This fact
has immediate application in the phenomenon of bubble formation in sub-

merged sediments. A.bubble will only be produced in this environment

when the sum of the partial pressures of the dissolved gases in the

105

interstitial water equals or exceeds1 the hydrostatic pressure at
bubble depth.2

The absolute solubility of gases in a function of temperature;
their behavior in solution is treated in thermodynamic theory as one of
liquid mixture (a gaseous liquid mixing with a liquid, cf., Glasstone,
1947). The solubility of the gas is then related to the vapor pressure
of the gas at the temperature of interest. This explains the changes
in gas solubility as a function of temperature. Since the heat of
vaporization is positive, the ideal solubility of gases decreases with
increasing temperature.

The use of partial pressures to express the solubility of
gases avoids the issue of changing solubility with temperature. Most
ecologists use terms of absolute solubility, such as wt.gas/vol.water,
vol.gas/vol.water, and moles gas/vol.water. These terms cannot be used
to discuss the condition of the dissolved gases relative to an adjacent
gas phase without a simultaneous estimate of temperature. Since this

investigation involves the interaction of a gas phase in the lacunae

(and in the sediment bubbles) with gases dissolved in the adjacent

 

1The pressure required for the initiation of a bubble under
hydrostatic pressure should exceed the latter pressure because of sur-
face tension effects. The actual pressure inside a small bubble at
depth is not only a function of its depth, but the surface tension of
the water and the diameter of the bubble. As the bubble grows in size,
the pressure increment decreases.

This effect can be very large when nucleation sites are
absent. However, the particles of sediment constitute such nucleation
sites, and the effect is presumably quite small.

2According to Trevelyan (1975), it was William Henry (Henry's
law) who first identified the combustible gas in lake sediments as
methane.

106

water, the concentrations of the latter are reported in terms of
equilibrium partial pressures (thus incorporating quantities and

temperature).

APPENDIX B

THE LACUNAR INTERPHASE

Consider the simple model presented in Figure Bl. As
described in Chapter 1, the water column and the atmosphere tend to
remain near equilibrium; the interstitial water of the sediment
diverges from atmospheric because the particulate matter of the sedi-
ment restricts turbulent and diffusive exchange with the overlying
water. Exchanges do, of course, exist; carbon dioxide and methane
diffuse from.the sediment water, and oxygen moves into the sediment
‘water. The flux of gas from the sediment is augmented by ebullition,
which occurs whenever the sum of the partial pressures of the gases
dissolved in the sediment water exceeds the hydrostatic pressure (+
surface tension; cf., footnote 1, Appendix A). The kinetics of the
various transports are a function of a great many variables (e.g.,
sediment porosity, concentration gradients, depth, turbulence, bio-
turbation, etc.). These are the transport processes that have
received the most attention in the literature of sediment-water
exchanges.

This research has shown that the gas in the lacunae of
emergent vascular plants can be important in the exchange processes
between the atmosphere and the interstitial water. The impact of

this interphase in the case of methane efflux has been demonstrated

107

108

Figure B1.-—Sedimentdwater exchanges.
This is a representation of the standard conception of
the gas exchange mechanisms between the sediment and the
overlying water. Gases diffuse, or are turbulently
mixed into and out of the sediment. Gases also escape
from the sediment by ebullition.

109

 

 

 

 

 

 

110

elsewhere (Dacey and Klug, 1979). The purpose of this appendix is
to develop a purely physical model of the consequences of this
extension of a gas phase into the sediment interstices.

During summer growth, the lacunae of the waterlilies are
essentially a continuous gas phase extending from the atmosphere
into the anaerobic sediment water. The lacunae are represented in
Figure B2 by a tube of variable porosity. Clearly, at the top of
the tube the exchange processes will bring about a gas composition
very hear atmospheric. The total pressure will be 1 atm.

The consequences of the exchange at the lower end of the
tube are not so obvious, and must depend somewhat on the porosity of
the tube leading to the surface. For example, as the porosity is
decreased, the rate of diffusion of atmospheric gases (most notably
02) to the bottom of the tube is limited. Similarly, the rate of
diffusion of the sediment gases away from the base of the tube is
also restricted. Under these conditions, the accumulation of sediment
gases would be expected in the buried portion of the lacunar gas phase.
If the porosity were so restricted so as to inhibit the fluid flow of
gases (Appendix C), the total pressure of the gases in these lacunae
might also increase above atmospheric. This happens in the case of
the waterlily when the rhizome is isolated from the atmosphere during
the winter months.

Such extreme conditions are absent during the summer growth
of the waterlilies. Their lacunae are definitely pervious, since 02
is present in significant quantities throughout the plant.

As the porosity increases, the influence of the internal

structure of the lacunae on diffusion decreases. The focus of attention

111

Figure B2.-The lacunar interphase.
The passive transport of gases through such an inter-
phase should reflect the gradients at its exchange
surfaces (the atmosphere and the interstitial water).
The stippling in the interphase is intended to represent

gases of sediment origin diffusing towards the atmos-
phere.

112

 

 

 

 

 

 

 

 

   

 

 

O
l
I
O I
o
0
I
I
'0
0
o
0.
'5
W
o
o
I

 

 

 

 

 

 

 

 

      
  
  
 
  

 

 

 

           
    

    

        

.0
o . . o
0': ‘ z’oi' 9. .:
0. 7.: 'f;'°l’:'.‘0,. 01.30%. .03
.. -. . .. . .: 0:30.. 9 .1. ....o .‘:.: .... .. ” I... ‘0 0 .‘.°:: I
"=="'"'-'-' - ,4 ...,.

 
 

113

moves from the lacunae to the interface of the lacunae with the
sediment water. This exchange might dominate the character of the
gas dynamics in the lacunae.

The simplest case to consider is one wherein the gases on
either side of the interface are identical with respect to solubility
in water and diffusivity in water and air. under these circumstances,
there would be a net pressure gradient into the plant, since the sum
of the partial pressures is 1.1 atm in sediment water (submerged l‘m)
and the pressure is 1 atm in the lacunae. ‘The resulting partial
pressure gradients would lead to a differential in the diffusive
exchanges resulting in a net flux of gas into the plant. This net
transport of gas could be observed as a fluid flow of gas up through
the plant to the atmosphere; the rate of flow would depend on the
differential in the exchange rates between the lacunae and the sediment
water.

In lake sediments, however, all the gases are not alike.

As a result, the character of the exchange between the two phases is
unpredictable. There are four main gases of interest: nitrogen,
oxygen, methane, and carbon dioxide. The former two are abundant in
the atmosphere; the latter two, in the sediment water. The resulting
fluxes of gases between the phases depend in part on the solubilities
and diffusivities of the gases in water. In actualility, there are
other processes that may have even greater significance —-the
biological production and consumption of gases near the interface.
These processes tend to steepen the diffusion gradients between the

lacunae and the water.

114

The oxygen gradient is certainly steep due to its consumption
by the plant surface tissue, and possibly by bacteria in the rhizo-
sphere. The methane gradient is also very likely steep due to
methanogenesis in the vicinity of the plant. The CO2 gradient is
also steep due to the strong buffering capacity of the sediment water.
Nitrogen, being a conservative gas (assuming denitrification has run
its course in the sediment water), is most likely to have a small
diffusion gradient into the sediment.

The net effect of all these simultaneous processes is clearly
difficult to predict. This treatment is intended to make the point that
this interface submerged in the sediment is bound to be a source and a
sink for different gases. The exchanges that occur here may lead to a
net flux of gas through the plant; such a fluid flow upwards from the

sediment would tend to inhibit the availability of O in the buried

2
plant parts. Whether or not such fluxes occur was not established in
this study. It was found that, in the case of the waterlily, the other

exchange processes between the plant and the atmosphere have influences

that dominate the dynamics of the lacunar gases.

APPENDIX C

FLUID FLOWS

The flows discussed in this appendix are described by the
body of physical theory known as fluid dynamics (hydrodynamics,
aerodynamics). The fluid flow of gas involves the mass movement of
gas from a region of high pressure to a region of lower pressure,
like a wind. The mass flux at any point along a tube carrying such

a flow is described by:

73% - pAv eq. C1

where p is the gas density, A is the cross-sectional area of the pipe,
and v is the speed of the flow through the cross-section. This
equation reveals the well-known relationship that the speed of flow
is inversely proportional to the cross-sectional area of the tube (or
inversely proportional to the square of the tube diameter).

The concepts of fluid flows are applied in this study to the
mass movement of gases along the length of the petioles. Since the
pressure gradients are very small, the density of the gas can be con-
sidered to remain constant. It is also apparent from this study that
there is very little exchange of gases with the water outside the
petiole, so at any given time, dm/dt can be said to remain constant
along the length of the petiole. Under these conditions, equation Cl
is called the equation of continuity.

115

116

The pressures that generate the flows in the waterlilies
result from diffusion processes through the upper surface of young,
newly-emerged leaves (cf., Appendices D and E). Once such a pressure
develops in these leaves, there results a flow of gas down their
petioles to the rhizome, and subsequently from the rhizome up the
petioles of the older emergent leaves. These flows operate in

accordance with Darcy's law for fluid flow through a porous medium:

u-'=B_LA‘2 eq. CZ

where Ap is the pressure drop along length L, and B is the permeability
coefficient of the petiole interior (cf., Carman, 1956).

Since the pressure gradients are extremely small, the flow
rates are also small in comparison with the rates of gas flow usually
measured by physicists. The standard means of measuring gas flow (e.g.,
Pitot tube, Venturi meter) are not applicable in this study because of
the slow flow rates and the inaccessibility of the flow channel. The
flow rate was studied by injecting a tracer gas (such as 02H6) into
the petiole. A fluid flow along the length of the petiole carries the
tracer gas much like the carrier gas in a chromatograph carries the
sample. As the tracer is carried along, it mixes somewhat with the
air as a result of diffusion and absorption processes. As a result,
when the tracer gas passes a point further along the petiole, the
original spike of tracer has broadened into a slightly tailing peak
(as is usually the case in chromatography). Such a peak is detected
by sampling the gas at intervals (of time) at a predetermined distance
from the injection point. The time when the maximum of the tracer

"elution" curve passes the sampling point is taken to be the time

117

taken for the whole gas mixture to move from the injection point to
the sampling point. Linear and volume flow rates are easily calculable
by methods in the text of Chapter 2. As described in that chapter, the
elution curve can be integrated, and knowing the volume rate of flow,
the total amount of tracer passing the sampling point can be estimated.
Comparison of this amount with the amount actually shows that all (80-
120%) of the tracer passed the sampling point. Since an insignificant
amount is lost to the surrounding water through diffusion, the equation
of continuity can be said to hold. Furthermore, this observation
supports the claim in Chapter 1 that the water column gases play no
significant role in the internal gas dynamics in the waterlilies.
Continuity is assumed in the treatment of fluid flow that
gives rise to Bernoulli's equation, a fundamental assertion of fluid

dynamics:

p +-kpv2 + pgh a constant eq. 03
It can be seen that the dimensions of the components are all energy/
volume. The Bernoulli equation is thus essentially a statement of the
energydwork theorem of fluid dynamics (Halliday and Resnick, 1960).
The last term, pgh, represents hydrostatic potential energy, and
therefore disappears from.the equation under the conditions of interest

in this study. Thus,

p +kpv2 - constant eq. C4
This equation describes the relationship between the pressure on the
inside walls of the tube and the velocity of the gas in the tube. This
relationship has no direct application in this study, but it is of

theoretical interest for two reasons.

118

This equation has been successfully applied in studies of
other fluid circulation systems of biological interest. ngel (1972,
1978) has applied these principles to show how a number of animals,
and possibly plants, use velocity gradients in their environment to
circulate the fluid environment to great benefit. A brief consider-
ation of the venturi meter in Figure C1 shows that there is a pressure
gradient within a tube connecting a region of low velocity to a region
of high velocity. Therefore, when the mercury column is removed from
the venturi meter, as in Figure CZ, there is a bulk flow of gas from
high to low pressure (or from low to high velocity). vogel has shown
that flows of this character circulate air through the burrows of
prairie dogs, and water through sponges. He even speculates on the
possibility of flows within leaves arising from.velocity (and therefore
pressure) gradients across the surface of the leaves.

There are many lines of evidence that prove the flows in
waterlilies are not derived in this way. The pressures in the water-
lilies exceed barometric. Vogel's model requires that the pressures
within the circulating system are less than barometric.

The second aspect of equation 4 that must be considered in-
volves the mode of pressure measurement. In this study, these measure-
ments were made within the gas spaces of the plant where gases typically
were moving, so that the manometric system represents half of a Venturi
meter (see Figure Cl), and as such measures an aggregate of the internal
static pressure and the flow rate. Measurements of both pressure and
flow rate establish that the velocity component in the Bernoulli
equation is small relative to the pressure component, and the pressures

measured by manometer are essentially static pressures (and therefore

119

.mofim “Ewan meu p.333 3on mauumaosms mo oomammmaamuo

men .3 omuummwsma ma Hmfiuomummmfio mun—momma many .Em umma on... mo 5an msu on menu

sum new: mfi no games man up mama m.“ Hams men so new meu .3 omuumxm muoummue on...

.umumaooma may we sum unwau may m>onm umummuw mu scam new no mums amused mew madam
.umuma «nousgIIAo muswam

 

 

 

 

 

 

 

 

 

 

121

.muommmua .333 no scammu map on Ayuauoamer umaod muommmue umnwfin mo 53.“me mnu
60.3 new 00 30.3 m gods 3 muomfim aoum mead owuumaosma msu mo HES—aux
.83 EBE>I.8 893a .

 

 

 

 

~(-—.

 

 

122

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

123

directly relevant to processes in kinetic theory (Appendix D and E).

APPENDIX D

THERMAL TRANSPIRATION

The character of the flow of gases through an opening
depends on the size of the opening. If the opening is large, the
flow accords with the conception of flows held by most readers:
that is, a wind from a region of high pressure towards a region of
lower pressure. Under these circumstances, the motion of the gas is
described by the theory of fluid dynamics (cf., Appendix C).

When the pore size is small relative to the free mean path
of the gas molecules, the flow of gas through the pore is called
molecular effusion, and is described by another body of physical
theory: the kinetic theory of gases.

The assumptions underlying the kinetic theory can be listed
as follows:

1) The volume of the gas under consideration is made up of
a large number of molecules, and in a given type of gas these molecules
are all alike.

2) The molecules are small relative to the distances between
them, and their motion as described by Newtonian mechanics.

3) The molecules are spherical, and undergo elastic
collisions.

4) There are no appreciable forces of attraction or repulsion

124

125

between the molecules themselves, or between the molecules and the
walls of their container.

On the basis of these assumptions, a system has been developed
that, in the words of Loeb (1934), "is as well-established a theory as
any theory accepted by science."

The phenomenon of thermal transpiration was first investigated
and named by Reynolds (1879):

"I have now ascertained that a difference of temperature may
be a very potent cause of transpiration through porous plates. So much
that with hydrogen on both sides of a porous plate, the pressure on one
sidg being that of the atmosphere, a difference of 1600 F (from52o to
212 ) in the temperature on the two sides of the plate secured a perma-
nent difference in the pressure on the two sides equal to an inch of
mercury; the higher pressure being on the hotter side. With different
gases and different plates, various results were obtained which are,
how2ver, connected by definite laws."

To understand the theory predicting thermal transpiration,
the reader should imagine a small plane of area A.drawn anywhere in a
gas space. It is possible to show (given the assumptions listed above)
that the rate at which molecules strike one side of A is equal to NEA/4,
where N is the number of molecules per unit volume and E is their
average velocity.

Now imagine that the surface A is a pore in a thin diaphragm.
When the linear dimensions of A are smaller than the free mean path of
the gas (L - 10-5cm at 1 atm), all molecules striking either side of A
will pass through it. In the case of such a plane within a gas at
equilibrium, the number of molecules moving through the plane (pore) is

equal in either direction. Therefore, at steady state:

Nlc1 - N2c2 eq. Dl

126

Taking NIEIINZEZ - 1, and multiplying both sides by El/E2 yields:

 

-2 -
N1°1 c1

_2 - :f- eq. D2
N2°2 c2

Pressure is an expression of the rate of change of molecular

momentum on a wall, and can be shown to be:

p = % 1.111.152 eq. D3

where m is the mass of a molecule. Since p is therefore proportional
to N32, pllp2 can be substituted into the left side of eq. D2.
For n moles of gas contained in volume V, the empirical

equation of state pV - nRT can be combined with eq. D3, yielding:

nRT --%-Nvm52 eq. D4
Eq. D4 shows that E varies with the square root of T, so T?/T§ can be

substituted into the right side of eq. D2. The substitution of eq. D3

and eq. D4 into eq. D2 yields:

P T8
1

7?; eq. D5
p2 2

This equation describes a steady state system, where the flows
of molecules through the pore in both directions are identical. It
shows that if there is a temperature gradient between the two sides,
there will arise a pressure difference between the two sides. When
there is no temperature gradient, pl/p2 - 1; then the pressures are
equal. As soon as a temperature gradient is set up, there is a net flow
of gas from.the cooler side to the warmer until the conditions in eq. D5

are met 0

127

The fact that the gas flows from cooler to warmer can be
demonstrated with the following mechanistic model (cf., Figures D1 and
D2). Recalling that the rate at which molecules enter the pore from
each side is equal to REA/4, it is instructive to study the differential
effects of temperature on the temperature-sensitive parameters of that
formula, N and E. Since N (the number of molecules per volume) varies
directly with density, it varies inversely with temperature in an open
system. On the other hand, E varies with the square root of temperature
(as shown in eq. D4). These opposing effeCts cause the rate of entry
of molecules into the pore to be proportional to the inverse square
root of the absolute temperature. Therefore, if T2 is greater than T1,
then N E A/4 (the rate at which molecules enter from side 1) is

1 1

(T2/T1)k times greater than the rate from side T This means that

2.
there is a net flow of gas from the cooler side to the warmer side.

This flow continues until the pressure differential in eq. D5 has been
achieved, or until the temperature gradient is removed.

Reynolds (1879) found that these laws apply, not only for
small pores, but also for membranes such as porous porcelain, stucco,
and meerschaum. The same principles apply; when a temperature gradient
is set up, a pressure gradient ensues.

Thermal transpiration is treated more thoroughly in texts on
kinetic theory, such as Loeb (1934), Kennard (1938), and Jeans (1940).
Interestingly, each author comments on the fact that if the chambers
were connected by a large tube, there would be a steady flow of gas
from the warmer to the cooler along this pipe as long as the temperature

gradient across the porous partition is maintained (see Figure D3).

Such a system can do work, and it superficially resembles a perpetual

128

Figure D1.—-Thermal transpiration (steady state).
Two gas phases are separated by a porous partition and
a temperature gradient is maintained across the parti-
tion. Since the stippled molecules are of higher temp-
erature, their density is less and their velocity is
greater than the cooler gas. The equilibrium pressure
of the hotter gas is also greater than the cooler.

129

 

 

 

 

 

 

130

Figure D2.-Porous partition.

This is an enlargement of the circled area in Figure D1.
When the total pressures on each side are the same, but
the temperatures are different, there results a net
movement of gas through the partition. Although the
cooler molecules have lower velocities, in terms of
diffusive exchange there are disproportionately more of
them per unit volume.l Therefore, they tend to enter the
hotter side faster than the hotter molecules leave.

 

132

Figure D3.-—Thermal transpiration (flow).
The thermal transpiration pressure derived from
diffusive exchange can be allowed to drive a fluid flow
of gas if a tube of moderate diameter (greater than
0.1um) is connected between the two phases. This flow
will continue as long as the temperature gradient is
maintained.

133

 

 

 

 

 

 

 

134

motion machine. However, Bridgman (1932) points out that the tendency
for the temperature gradient to break down fulfills the requirements
of the second law of thermodynamics.

Figure D3 is an accurate representation of the gas system in
Nuphar. The cooler side can be considered to represent the atmosphere.
The warmer side represents the gas spaces in the young, newly-emerged
leaves; the porous plug is the upper surface of these leaves. The
remainder of the plant —-petioles, rhizome, and older emergent leaves,
are represented by the external pipe connecting the two sides of the

l
porous plug.

It's interesting to note that Reynolds (1879, p. 842) anti-
cipated some significance for the dimensional properties of gas, of
which thermal transpiration is one:

"Although the results of the dimensional properties of gases
are so minute that it has required our utmost powers to detect them, it
does not follow that the actions which they reveal are of philosphical
importance only. The actions only become considerable within extremely
small spaces, but then the work of construction in the animal and
vegetable world, and the work of destruction in the mineral world, are
carried on within such spaces. The varying action of the sun must be

to cause alternate inspiration and expiration of air, promoting contin-
ual change of air within the interstices of the soil as well as within

 

1It might be argued that since the gases in the lacunae are

confined in volume by the size of the plant, the conditions for such a
system are not met. The warming of the gases in the leaves would tend
to increase the pressure there as a simple result of thermal expansion.

However, since the leaves are not closed systems, the gases
are free to escape down the petiole and out the other leaves. Certainly,
then, there would be an initial (and very small) flow caused by the
thermal expansion of the gases as one side is warmed. Such a flow con-
tinues until N has varied inversely with T . The flow continues down
the petiole for the rest of the day as a result of kinetic processes in
the upper surface of the leaf.

135

the tissue of plants. What may be the effects of such changes we do
not know, but the changes go on; and we may fairly assume that in the
processes of nature the dimensional properties of gas play no unimpor-

tant part."

APPENDIX E

HYGROMETRIC PRESSURE

Hygrometric pressure was first studied by Dufour (1873) who
found that: "when a porous wall separates two air masses of differing
humidity, there results two opposed and unequal diffusion flows across
the wall. The more abundant flow moves from the drier air to the
damper air. The difference in these currents depends principally on
the difference of vapor pressures from.one side to the other."

The explanation for this phenomenon was given by Kundt (1877).
In many ways the process is analogous to thermal transpiration, except
that the gradient across the porous partition is not thermal, but a
gradient in the vapor pressure of a volatile liquid. As in the case of
thermal transpiration (cf., Appendix D) the exchange of molecules
through the porous partition is a diffusive process, and the pressures
on either side of the wall can diverge in the process of seeking a
diffusive steady state.

As long as there is no temperature gradient and both sides of
the partition are dry (i.e., there is no volatile liquid on either side),
the gas pressures on both sides are equal. Calling the pressure of the
dry atmosphere on one side of the partition Pa’ these conditions will

lead to a pressure P of the "dry" constituents on the other side so

1
that Pl-Pa. As long as there is no temperature gradient between the

136

137

two sides, P1 will tend to equal Pa’ since the "dry" constituents of
the gas phase on both sides will tend towards diffusive equilibrium.

When a volatile liquid is added to one side of the partition
(Figure E1), the liquid will evaporate and tend to add its vapor pres-
sure Pv to the pressure of the gas on that side. If the gas on this
side occupies a closed volume, the total pressure of the gas mixture
will also tend to increase. According to Henry's law (cf., Appendix A),
the addition of the vapor does not affect the pressure of the "dry"

constituents, so P is not affected. The pressure on the wet side of

1
the partition therefore approaches the sum of the pressures of the two

components there, Pt - P + Pv.

1
As Pv increases, the vapor tends to diffuse through the

partition to the dry side. If this vapor is removed from the gas

phase on the dry side (for example, by absorbing it, or by diluting it

in a large volume of dry gas), it won't contribute significantly to the

gas pressure on the "dry" side. As it diffuses away from the "wet"

side, however, it tends to diminish Pv on that side. The liquid on

that side continues to evaporate in order to maintain saturation of the

vapor in the gas phase on that side. Once all the liquid has evaporated,

the vapor will eventually diffuse through the partition to the dry side,

Pv approaches zero, and the total pressure Pto on the previously "wet"

1:
side approaches P1 (=Pa).

Clearly, a liquid phase is required to support the increased
total pressure on the "wet" side. The evaporation of that liquid
replaces the vapor lost by diffusion through the porous partition. The

vaporization of liquid consumes energy and tends to cool the atmosphere

138

Figure E1.-Hygrometric pressure.
If no temperature gradient exists across the porous
partition, the dominant gas mixture (open circles)
equilibrates at equal density on either side cf the
partition. If a volatile liquid is present on'one side
of the partition, it tends to add its vapor pressure
to the total pressure on that side.

The vapor will also tend to diffuse through the parti-
tion, but as it does, more liquid evaporates in order
to maintain its vapor pressure on the "wet" side.

139

 

 

 

 

 

 

 

140

on that side of the partition, so if the system is to remain iso—
thermal, heat must be transported to that side. This process depends,
therefore, on the continual input of liquid and heat. This is
analogous to the continual input of heat required to maintain the
thermal gradient that drives thermal transpiration.

If a large tube is attached to this system so as to allow a
fluid flow of gas away from the region of high pressure (Figure E2),
the dynamics of the system are changed. The fluid flow of the gas
mixture (Vdry" components + vapor) away from the "wet" side decreases
the total pressure of the gas phase on that side. Since this flow
removes the various gas components in proportion to their abundance in
gas phase, both P1 and Pv decrease. Since P1 is now smaller than Pa’
there is a net influx of the "dry" constituents of the gas into the
"wet" side through the porous partition. This net influx of dry gas
is accompanied by evaporation of the liquid on the wet side, so the

pressure P I P

tot 1 +Pv on the wet side tends to return towards the

earlier steady state value. This tendency to increase the pressure
forces a continual flow along the large tube away from the region of
high pressure. As long as liquid remains on the wet side, such a steady
flow of gas (+ vapor) is generated.

In the foregoing discussion, it was assumed that the vapor
pressure of the liquid was negligible on the "dry" side. In fact, for
such a system to operate, the vapor pressure Of the liquid need only be
less than its vapor pressure on the other (Vwet") side. ‘In the Case of
the waterlily, the upper surface of the leaf is the porcus partition.
The volatile liquid is water. The vapor pressure gradient depends

very much on the ambient relative humidity, since the air inside the

141

Figure E2.-Hygrometric pressure (flow).
As in the case of thermal transpiration, a tube can allow
a continuous fluid flow of gas mixture (including vapor)
from high pressure to low pressure. As the pressure
drops in the "wet" chamber, there is a net rate of
diffusion into that side from the "dry" side, tending to

maintain pressure equilibrium. Simultaneously, liquid
evaporates.

Such a system can support the flow as long as liquid
remains on the wet side, and there is a humidity gradient
across the partition.

142

 

 

 

 

O ' O
O O O 0 ®
0 9
O O _______,""'_.__.____'-- ._.."'___O.'._._._I

 

 

 

143

leaf is presumably saturated at all times.

As is discussed in the text of this thesis, the system is
very rarely at thermal equilibrium. During daylight, the leaves are
usually warmer than the ambient air, so the vapor pressure of water
inside the leaf can be considerably higher than that of the environment.
(These temperature differentials also contribute to thermal

transpiration.)

REFERENCES CITED

REFERENCES CITED

Arber, A. 1920. Water plants - a study of aquatic angiosperms.
Cambridge University Press, Cambridge.

Armstrong, W. 1964. Oxygen diffusion from the roots of some British
bog plants. Nature. 204:801-802.

Armstrong, W. 1978. Root aeration in the wetland condition. Pages
269—297‘in_D.E. Hook and R.M;M. Crawford, eds. Plant life
in anaerobic environments. Ann Arbor Science Publishers,
Ann Arbor.

Barthelemy, A. 1874. De la respiration et de la circulation des gaz
dans les vegetaux. Annales des sciences naturelles, V Ser
19:131-175.

Beal, E.O. 1956. Taxonomic revision of the genus Nuphar Sm. of
North America and Europe. J. Elisha Mitchell Sci. Soc.
72:317-346.

Bridgman, P.W. 1932. Comments on the note by E.H. Kennard on
"Entropy, reversible processes and thermocouples". Proc.
Nat. Acad. Sci. 18:242-245.

Bristow, J.M., and M; Whitcombe. 1971. The role of roots in the
nutrition of aquatic vascular plants. Am. J. Bot. 58:8-13.

Carman, P.C. 1956. Flow of gases through porous media. Butterworths
Scientific Publications, London.

Crawford, R.M.M. 1971. Some metabolic aspects of ecology. Trans. Bot.
Soc. Edinb. 41:309-322.

Dacey, J.W.H. and M.J. Klug. 1979. Methane efflux from lake sediments
through waterlilies. Science 203:1253—1255.

Danckwerts, P.V. 1971. Diffusion and mass-transferlin.J.N. Sherwood,

A.V. Chadwick, W.M; Muir, F.L. Swinton, eds. Diffusion Pro-
cesses vol. 2. Gordon and Breach Science Publishers, London.

144

145

Denny, P. 1971. Sites of nutrient absorption in aquatic macrophytes.
J. Ecol. 60:819-829.

Dufour, L. 1872. Recherches sur la diffusion entre l'air sec et
l'air humide. Bull. Soc. Vaud. Sc. Nat. XIII 72:165-217.

Dufour, L. 1874. Sur la diffusion hygrometrique. Bull. Soc. Vaud.
Sc. Nat. XIII 74:608-641.

Dutrochet, H. 1837. Memoires pour servir a l'histoire anatomique
et physiologique des vegetaux et des animaux. Paris.

Esau, K. 1953. Plant anatomy. John Wiley and Sons. New York.

Feddersen, W. 1873. Uber Thermodiffusion von Gasen. Pogg 'Ann'
148-302.

Gessner, F. 1952. Der Druck in seiner Bedeutung fur das wachstums
submerser Wasserpflanzen. Planta 40:391-397.

Gessner, F. 1959. Hydrobotanik. Bd II Stoffhaushalt. VEB Deutscher
Verlag der Wissenschaften, Berlin.

Gessner, F. 1961. Hydrostatischer Druck and Pflanzenwachstum. Pages
668-690 in_W. Ruhland, ed. Encyclopedia of plant physiology
XVI. External factors affecting growth and development.
Springer-Verlag, Berlin.

Glasstone, S. 1947. Thermodynamics for chemists. VanNostrand Co.,
NeW’York.

Haberlandt, G. 1887. Zur Kenntris des Spaltoffnungsapparates. Flora
:97

Haberlandt, G. 1914. Physiological plant anatomy. Transln by
Drummond. Today and Tomorrow's Book Agency, New Delhi.

Halliday, D. and R. Resnick. 1960. Physics for students of science and
engineering. John Wiley and Sons, New York.

Hesslein, R.G. 1976. An in situ sampler for close internal pore water
studies. Limnol. Oceanogr. 21:912-914.

Hutchinson, G.E. 1975. A treatise on limnology. Vol. 3. Limnological
botany. Wiley Interscience, New York.

Jeans, J. 1940. An introduction to the kinetic theory of gases.
Cambridge university Press, Cambridge.

Kaplan, I.R. 1974. ed. Natural gases in marine sediments. Plenum
Press, New York.

146

Kennard, E.H. 1938. Kinetic theory of gases. McCraw-Hill Book Co.,
New'York.

Kundt, A. 1877. Zur Erklarung der Versuche Dufour's und Merget's
uber die Diffusion der Dampfe. Annalen der Physik, Leipzig
(Wiedmann Ann.) NF2:17-24.

Laing, H.E. 1940. The composition of the internal atmosphere of
Nuphar advenum.and other water plants. Am. J. Bot. 27:861-868.

Laing, H.E. 1940b. Respiration of the rhizomes of Nuphar advenum
and other water plants. Am. J. Bot. 27:574-581.

Lechartier, G. 1867. Sur 1e mouvement des gaz dans les plantes
aquatiques. Comptes rendus, Acad. sci. Paris 65:1087-1091.

Lehninger, A.L. 1970. Biochemistry. Worth Publishers Inc., New York.
Levine, E., and L. Chandler. 1974. Biology of Bellura ggrtynoides

(Lepidoptera: Nbctuidae), a yellow water lily borer, in
Indiana. Ann. Ent. Soc. Amer. 69:405-414.

 

Loeb, L.B. 1934. The kinetic theory of gases. Dover, New York.

MtRoy, C.P., and R.J. Barsdate. 1970. Phosphate absorption in eelgrass.
Limnol. Oceanogr. 15:6-13.

Merget, A. 1873. Sur des phenomenes de thermodiffusion gazeuse qui se
produisent dans les feuilles et sur les mouvements circula-
toires qui en resultant dans l'acte de la respiration chloro-
phyllienne. Comptes rendus, Acad. sci. Paris 77:1468-1472.

Merget, A. 1874. Sur la reproduction artificielle des phenomenes de
thermodiffusion gazuese des feuilles, pars les corps poreaux et
pulverulents humides. Comptes rendus, Acad. sci. Paris 78:
884-886.

Miller, M.C. 1972. The carbon cycle in the epilimnion of two Midhigan
lakes. Ph.D. Dissertation, Michigan State University,
East Lansing.

Mitsui, S. 1964. Dynamic aspects of nutrient uptake. pp53-62 $2.
The mineral nutrition of the rice plant. Johns Hopkins Press,
Baltimore.

Ohle, W. 1958. Die Stoffwechseldynamik der Seen in Abhangigkeit von
der Gasausscheidung ihres Schlammes. V m Wasser 25:127-140.

Ohno, N. 1910. Uber lebhafte Gasausscheidung aus den Blattern von
Nelumbo nucifera. Zeitschrift f. Pflanzenphys. 2:641-645.

Raffeneau-Delile, A. 1841. Evidence du mode respiratoire des feuilles
de Nelumbium. Annales sci. naturelles, II ser. 16:328-341.

147

Reeburgh, W.S. 1969. Observations of gases in Chesapeake Bay sedi-
ments. Limnol. Oceanogr. 14:368-375.

Reynolds, 0. 1879. On certain dimensional properties of matter in
the gaseous state. Phil. Trans. 170:727-845.

Scholander, P.F., E.D. Bradstreet, H.T. Hammel, E.A. Hammington. 1966.
Sap concentration in halophytes and some other plants. Plant
Physiol. 41:529-532.

Sculthorpe, C.D., 1967. The biology of aquatic vascular plants.
Edward Arnold, London.

Sifton, H.E. 1945. Air-space tissue in plants. Bot. Rev. 11:108.
Sifton, H.E. 1957. Air-space tissue in plants II. Bot. Rev. 23:303.

Sondergaard, M. and K. Sand-Jensen. 1979. Carbon uptake by leaves and
roots in Littorella'uniflora. Aq. Bot. 6:1-12.

Spence, D.H.N. and J. Chrystal. 1970a. Photosynthesis and zonation of
freshwater macrophytes. I Depth distribution and shade
tolerance. New Phytol. 69:205-215.

Spence, D.H.N. and J.Chrystal. 1970b. Photosynthesis and zonation of
freshwater macrophytes. II Adaptability of species of deep
and shallow waters. New Phytol. 69:217-227.

Strayer, R.F. and J.M. Tiedje. 1978. .22 situ methane production in a
small, hypereutrophic hardwater lake: loss of methane from
sediments by vertical diffusion and ebullition. Limnol.
Oceanogr. 23:1201-1206.

Trevelyan, W.E. 1975. The methane fermentation: a discussion paper.
Tropical Science 17:193-209.

Twilley, R.R., M.M..Brinson, and G.J. Davis. 1977. Phosphorus
absorption, translocation, and secretion in Nuphar luteum.
Limnol. Oceanogr. 22:1022-1032.‘

 

Tyler, P.D. and R.M;M. Crawford. 1970. The role of shikimic acid in
waterlogged roots and rhizomes of Iris pseudacorus L. J.
Exp. Bot. 21:677-682.

 

Ursprung, A. 1912. Zur Kenntnis der Gasdiffusion in Pflanzen.
Flora 4:129-156.

Violle, J. 1875. Expose des recherches entreprises sur la thermal
diffusion. J. Physiquez97-104.

ngel, S. 1972. Interfacial organisms: passive ventilation in the
velocity gradients near surfaces. Science 175:210—211.

148

ngel, S. 1978. Organisms that capture currents. Sci. Am. 239:
128-139.

Weidlich, W.E. 1976. The organization of the vascular system in the
stems of the Nymphaeaceae. I Nymphaea subgenera Castalia and

Hydrocallis. Am. J. Bot. 63:499-509.

Winfrey, M.R., and J.G. Zeikus 1977. Effect of sulfate on carbon and
electron flow during microbial methanogenesis in freshwater
sediments. Appl. Environ. Microbiol. 33:275-281.

Wium-Andersen, S. 1971. Photosynthesic uptake of free 00 by the roots
of Lobelia dortmanna. Physiol. Plant. 25:245-24 .