Proce.dinRs'tl thr Rutal So.ie!l 4 Edmhurgh. l02B.5ll

521. la94

Polarographic oxygen electrodes and their use in plant aeration studies W. Armstrong Departmentof Applied Biology,Universityof Hull, Hull HU6 7RX, UK Synopsis 'unattackable' cathodic electrode of The electrolytic reduction of oxygen which occurs at a wetted platinum or gold when polarised in conjunction with a Ag/AgCl anode, forms the basis of most polarographic oxygen measurements in plant bioiogical work. Various types of polarographic electrode and their uses are reviewed. These include cylindrical sleeving 'bare', membrane-coated, and electrodes for quantifying localised oxygen fluxes from intact roots, Clark-type microelectrodes suitable for measuring concentrations and proiiles both inside roots and in the rhizosphere, and macro-Clark electrodes most frequently used in respiratory studies. Some details concerning equipment and electrode construction are given and some pitfalls in electrode application are discussed.

Background Polarography, first developed in the early 1920s by Jaroslev Heyrovsky, is an extremely versatile method of chemical analysis (Kolthoff & Lingane 1952). It is based upon the currents, and characteristicsof the current-voltage curves, generated by the electro-oxidation and electro-reduction of substancesin solution under an applied EMF in an electrolysiscell. The cathode in this cell may be one of a number of electrode types viz. the dropping mercury electrode (DME), hanging mercury 'unattackable' metal such as drop electrode (HDE) or one which employs an platinum or gold; a silver/silver chloride half-cell is most commonly used as the anode. Typically electrolysisbegins at a threshold EMF, the decomposition potential for the species concerned, and in operations involving the DME and Pt or Au electrodes,increaseswith increasingEMF and reactivity of the cathode until a point is reached, the limiting potential, at which current becomesindependent of voltage. The current value on this plateau is proportional to the concentration of the species undergoing electrolysis,and provided that the physical conditions around the cathode are reproducible from sample to sample, a plot of current verJrs speciesconcentration will generally follow a straight line. In terms of the numbers and types of substanceswhich may be analysed, the DME and HDE types are clearly the most versatile of theseelectrodes.For example, the DME may be used to identify and quantify most of the elements and many organic substances,sometimes several elements together in a mixture, at concentrations ranging from 10-6 to l0-2 molar, and in volumes of solution of micro-litre proportions. The HDE, by incorporating a concentrating step, is particularly suited for identifying and quantifying heavy metals in solution at concentrations as low as ppb. In contrast, the use of platinum or gold in place of mercury somewhat reduces the range of substanceswhich may be analysed, but simplifies some operations particularly as regards oxygen analysis. Oxygen is usually the most significant reacting species at applied potentials between zero and (-) 700 mV. This fact, together with stable electrode geometry, and becausegold and platinum lend them-

512 Table 1.

W. Armstrong Major types of oxygen electrode

Macro: Micro:

Naked types

Membrane types

Cylindrical Wire

Clark-type probe Clark-type Respirometer cell

Exposed Recessed

membrane coated: external anode Clark-type: internal anode

selvesfairly well to electrode fabrication, electrode variants to suit a multitude of purposes have been developed. Excellent and detailed reports on some of these can be found in Davies (1962), Kessler et al. (1973) and Gnaiger & Forster (1983). In biology and medicine the polarographic unattackable oxygen electrode has proved invaluable, not least by helping to remove the constraints so often imposed by the various manometric and volumetric methods of analysis. Electrodes are available which form the basis of respirometers,there are those which will provide for long-term monitoring of gas-phaseoxygen concentrations, for sampling in situ blood and plant animal tissues,for measuring oxygen fluxes and concentrations in soils and sediments. and from and into roots. Electrodes can now even be made small enough to probe individual cells. This paper is concerned with describing, and discussingthe use of, polarographic oxygen electrodesfor studying plant aeration, but particularly the aeration of roots under conditions of restricted radial oxygen supply from the rooting medium. A brief classification of polarographic electrode types used in plant aeration studies is set out in Table 1. Electrode reactions and the polarogram Oxygen is electrolytically reduced at the cathode in polarographic operations, and at pH>3.5 it is thought that the overall cathodic reaction is a two-stage process: O 2 + 2H2O I )s- --+H2O2 + 2OH-

(1)

ano HrO1t2e

-2OH

(2)

The complete reduction of one oxygen molecule thus normally requires four electrons, and produces alkalinity in the medium. For the DME the two reactions are clearly separated,the second requiring a higher applied EMF than the first. With gold or platinum no separation is obvious and the current voltage curve (the polarogram) exhibits a single plateau (Fig. l); it is thought that this occurs because the presenceof oxygen catalysesthe second reaction. Where a silver/silver chloride referenceelectrode is used as the anode the followins anode reactions occur:

Ag+Cl -AgCl+e-

(Eo---222 mY)

(3)

and 2Ag+2OH -AgrO+H2O+2e

(Eo: -35 mV)

(4)

It shouldbe noted that a thresholdlevelof indifferentsupportingelectrolytein the

Polarographic oxygen electrodes

513

(b)

Curve I II lll P

.j

o F

'

U

PhihlaleBulfer pH 3.0 AcelaleBuflc pl14.7 Btrate Bults pH9.0 00lNaOflO.lNKO pH t2.0 1

'a t

b H

2

2

8.

h X

1 .

-0r -o2 -0.3 -0.4 -05 -0.6 -0.7 -o8 -0.9 -r0 -r.l Applied Figure 1.

Negative

Potential

(Volts)

Oxygen polarograms at naked stationary platinum electrodes,and showing the drift of the plateau to more negative potentials with (a) increasing acidity (after Kolthoff & Lingane 1952), and (b) with decreasingsoil oxygen concentration: top curve aerated loam; bottom curve waterlogged loam. (After Armstrong 1967b.)

medium (5 25mu) is required to carry the current in all polarographic oxygen operations. The length of the current plateau for oxygen and its position relative to the applied EMF are important considerations, since it is only on the plateau that current is solely dependent on the rate of oxygen diffusion to the electrode;at lower potentials the rate of reaction is also a function of the applied EMF, and at higher potentials current increasesrapidly with increasing EMF as the hydrogen ion is reduced to hydrogen. Plateau length and position depend very much upon electrode type: gold electrodestend to give longer plateaux and with membrane electrodesplateaux tend to maintain a fixed position. Naked electrodes, where the gold or platinum are directly exposed to the sampling medium, are very prone to plateau shift towards less negative potentials with both increasing acidity (Fig. 1a) and decreasingoxygen levels in the medium (Fig. 1b). It is essentialthat the operator should be aware of such problems so that proper corrective measurescan be applied such as changing the applied analysis voltage to suit the circumstance. In some types of work it is imperative that polarograms are taken at frequent intervals to ensure that a correct analysis voltage is maintained. Polarograms may be obtained by increasing the applied EMF manually in a stepwise manner, with a short pause at each potential, but instruments can be obtained (the polarograph) which automate this procedure. At each voltage increment there will usually be a brief surge in the electrolysiscurrent followed by some decay which representsa stabilisation of the diffusion gradient between the electrode and the oxygen source. If an applied voltage in the plateau region is sustainedthe current equilibrates to a value which is related to the rate of diffusion to the electrode accordins to the Faraday equation:

4:nFAf,=o.,

(5)

514

W. Armstrong

where, /, is the electrolysis current in amperes at the equilibrium time l, r is the number of electrons required per molecule of oxygen reduced, l, is the area (cm2.; of exposedelectrodesurface,{ is the Faraday,96500 coulombs (amperess-1), and 2 1) f,=0., is the flux (mol cm s- at the electrodesurfaceat time l. The time necessary to establish an equilibrium varies according to the electrode type and properties in the sampling medium: with certain types of micro-electrode, response time may be measured in millseconds, with some large naked electrode types, however, equilibration might take up to l0 minutes or even considerably longer.

The polarograph As can be seenin Figure 2, the basic polarising circuit for polarographic operations is very simple, requiring only a source of potential in the form of a low voltage battery, an appropriate value of variable resistor, an appropriate ammeter and a voltmeter. Obtaining polarograms with such equipment is tedious work, however, and it is usually advisable to use an automated instrument - the polarograph. Specificationsvary according to the sensitivity required; for example microelectrode l' A) or even smaller, and work might require readings in the pico-amp range (10 the instrument must then incorporate some very sophisticatedcontrols to maintain stability and remove or mask out electrical interference; cylindrical- and Clark macro-electrodes usually require an instrument reading to 1.999 or 19.99 pA. Excellent polarographs built to a range of specifications may be obtained from Barman Electronics, Leys Lane, Skipsea,Driffield, E. Yorks., UK. The polarograph will usually ramp-up the applied potential from 0 to - 1.2 volts in a seriesof equal steps of for example 0.02 volts, pausing for a fixed period (e.g. 5 s at each step). The current may be recorded on a chart recorder or stored and displayed by computer. When the plateau becomesknown, the device can be re-set to a plateau potential, and this potential should be sustained until the equilibrium current is obtained, or indefinitely for continuous recording if desired.

Macro-electrodes The cylindrical electrode Not available commercially but fairly easily constructed, the sleeving cylindrical platinum electrode (Fig. 2) was developed specifically for the study of oxygen transport in and from root systemsgrowing in anaerobic media (Armstrong 1964, 1979). In these circumstancesthe only source of oxygen for the root is that transported from the shoot via the cortical gas-spacesystem. The reactive face of the electrode is the inner surface of the cylinder; the outer surface is insulated by an epon-epoxy resin coat and enclosedin acrylic tube. Celluloid guides affixed to each end of the electrode ensure that, for sampling purposes, roots are centralised along the longitudinal axis of the cylinder. This type of electrode is used in conjunction 'remote' anode, usually silver/silver chloride. Construction details of the with a platinum electrode are as follows; a reliable anode can be constructed as described by Armstrong & Wright (1975). The thermo-pure platinum cylinder can either be purchased as platinum tube or can be fabricated from platinum sheet using an appropriate former (e.g. the shaft

515

Polar ogr aphic oxy gen electro des l-5 Volts

NMEDTTIRI] TYPE

(+)

0, Co is the oxygen concentration in the medium at air saturation at the particular temperature (e.g. mol cm 3: obtained from tables e.g. Armstrong 1979), Z- is the volume of sampling medium in the electrode cell (c-t), and Z. is the volume of the tissue. Oxygen consumption rates estimated on a volume basis are particularly suitable for collecting data for the mathematical modelling of root aeration; for other purposes it may be more appropriate to substitute the fresh weight or dry weight for V, in Equation 8. The point at which the plot of oxygen concentration changes from linear to curvilinear, indicating a slowing down in the rate of oxygen consumption, represents a critical oxygen pressure for respiration (e.g. Armstrong & Gaynard 1976' Saglio et al. 1984). Critical oxygen pressuremeasured in this way can vary considerably, for a variety of reasons such as the thickness of the segments,porosity, the degree of secondary wall deposits, position along the root, temperature. Detailed reasons for variation in critical oxygen pressuresmeasured in this way have been given by Armstrong (1988). Since they are a function of radial diffusion into the roots they may greatly exceedeven the oxygen concentration of air-saturated water and must not be confused with what might be the respiratory critical oxygen pressure in the intact root. They probably represent the point at which the innermost cells of the root segments approach anoxia, and this is a function of oxygen demand and resistanceoperating from the edge of the boundary layer around the segmentsto their centres, i.e. incorporating the oxygen demands of extra-cortical tissue, the

522

W. Armstrong

cortex and the stele,and the resistancesof all three plus that of the boundary layers. In the intact root, particularly in wetland root systems.a major source of oxygen for root respiration comes from the shoot via the cortical gas space. In these circumstancesthe most relevant critical oxygen pressureis numerically that of the oxygen concentration drop across the stele (Armstrong et al. 1990), a much smaller value than if the whole root segmentwas supplied with oxygen only from the bathing medium. Thus critical oxygen pressuresmeasuredby means of the naked cylindrical electrode, in which only the oxygen concentration drop across the extra-cortical tissuesbecauseofoutward radial diffusion, generallyrevealscritical oxygen pressures much lower than those found by means of the Clark electrode respirometers (Armstrong & Gaynard 1976;Armstrong & Webb 1985). Clark electrodesof the probe-type illustrated in Figure 2 are sometimes used to measure critical oxygen pressuresfor whole attached root systemsor excisedroots. Again, careful interpretation of the results is necessary,but this type of study is one way of demonstratingthat in aeratedsolution culture oxygen transport to roots is bi-directional,i.e. from shoot and solution, and can highlight differencesbetween wetland and non-wetland plants (Laan et al. 1990). Caution is again required in interpreting data from a more recent application of Clark electrodes: the measurement of oxygen releasefrom whole-root systems of wetland plants. In this type of study a Clark electrode probe is used to sample the solution bathing whole root systemsof intact plants. Conclusions are based on the time course of oxygen uptake or release.Using such a system Gries el al. (1990), and Bedford et al. (1991), concludedthat only relativelysmall quantitiesof oxygen are released from wetland root systems. The Clark electrode, however, can only it does not measure oxygen release at the root sample from the bulk solution surface itself, and a feature of using whole-root systemsin solution cultures is that oxygen releasedfrom one part of the root systemcan be scavengedby another more demanding region without ever reaching the medium sampled by the Clark electrode. Subsequently( Kludze et al. 1993; Sorrell et al. 1993; Sorrell & Armstrong 1994), it has been shown that the use of an oxygen scavengingmedium in the form of titanium citrate buffer revealsoxygen releaserates of more than an order of magnitude greater than had been detected bv aforementioned Clark electrode studies. Microelectrodes A desirable goal of oxygen electrode development has been the construction of reliable electrodesfor in vivo measurementof oxygen concentrations within tissues, and microelectrodes with tip diameters of up to 50 pm have been in use now for more than 30 years. Much of the development work has been in the medical field, and a range of naked, recessedand membrane electrode types have been described (e.g. Fig.5). Considerabledetails concerning the design, and discussionof the difficulties of construction and application may be found in Kessler et al. (1973) and Gnaiger & Forstner (1983). The most commonly used electrode is based on Pt or Gold wire, taper-etchedin sodium nitrite or potassium cyanide to a long fine point, and subsequently glasscoated for insulation. Variants on this theme involve the etching back of the glasssleevedtip to give a recessedelectrode which in the case of Pt may be finally gold plated. Recessingthe electrodereducesits oxygen demand, and depending on recess

523

Polar ographic oxy gen electro des NAXED EI,ECISODES

CIARK_TYPf, T{ICROEI^ECTRODf, ExPosed tiP

-trire

to cathode

,-1r===-===:-l

I

-, ^ ,\

_ 1,,

,

tood's metal

Recessedtip f,ood's metal

YEXBRANE COATTD

Silicoue rubber piug

tt Tip diameters

Figure 5.

1 to 10 lm

Sectional plans of various types microelectrode types. N.B. Glass-insulatedetched platinum wire, with or without a gold tip, can substitute for the low melting point Wood's metal alloy. but this involves a different labrication orocedure.

length, can make it virtually indifferent to any stirring movements at the tip. Recessingdoes reduce the rate of response,however, although for many purposes this may acceptable,being only of the order of secondsor milliseconds. The spatial resolution (i.e. catchment range) of electrodesalso improves with decreasing size: catchment range, defined as that area which supplies 90% of the oxygen to the electrode,is given by Baumgiirtl & Liibbers (1983) as 6.3xradius of the exposed cathode surface. Hence a microelectrode with a tip diameter of 5 pm will influence the oxygen regime only to approx. 15 pm from its surface. Naked micro-electrodes, however, seem to be more susceptibleto poisoning than are macro-electrodes.They are particularly prone to protein poisoning and may be contaminated by heavy metal ions and other ions. Efforts to avoid such problems have centred on the production of various types of membrane-coveredelectrode (e.g. Fig. 5), and since in most casesthere is an external referenceanode, the cathode-coating membrane must be electrically conducting. The mounting fluid Depex may be used for this purpose. An alternative to the glass-sleevingof etched Pt or gold wire, which can present many construction difficulties, is to pre-pull glass microelectrode bodies using a commercial puller (e.g. Model p-87, Sutter Instrument Co., California), and to fiIl these (Whalen et al. 1967) with a low melting point alloy such as Wood's metal (Goodfellow Metals, Cambridge). Modern pullers may be programmed to produce many identical bodies in a short time; tip diameters of < I pm are possible. The Woods metal is first converted into wire by melting and drawing into a narrow bore

524

W. Armstrong

silicone rubber tube, where it cools and is removed by cutting away the silicone rubber. A small piece is then pushed as far as the shoulder of the electrode body, its apical portion melted by means of an external heating coil, and it is then 'piston'. Any mechanically forced into the tip of the electrode body by a metal wire extruded metal clinging to the electrode tip may be removed by gently stroking a cotton bud along the electrode tip. The Woods metal may then be etched back, giving a small recess(e.g. of 15 20 pm) which may subsequentlybe filled with gold by electroplating at about I Volt, using gold cyanide. The Woods metal is used as the cathode, a gold wire is provided as the anode. Electrodes of this types may be used naked, recessedor be membrane coated, but they can also form the core of a Clark-type micro-electrode. We have recently developed such an electrode using such a core, sleevedinto an outer body having a slightly larger tip diameter and filled with I M KCI as electrolyte (Armstrong & Darwent, unpublished, and Fig. 5). The tip of the outer body is plugged with approx. 10 l5 pm of silicone rubber, which is allowed to cure, before filling with electrolyte. A silver chloride coated wire sealed into the outer body forms the anode. An additional length of silver wire may also be included in the outer body of Clarktype microelectrodes(Revsbech 1989) to act as a guard cathode for removing oxygen from the electrolytein the outer body, so minimising residual current in the electrode. Although fewer in number, there have been developments in the design and application of microelectrodesin areas other than medicine, including the study of plant aeration: Bowling (1973) perhaps was first: using a commercial electrode (tip diameter I ,rm); he found only a shallow radial oxygen gradient acrossthe sunflower root cortex. Tjepkema & Yocum (1914) used glass-insulatedPt-microelectrodes, some with protruding naked tips l0pm long and2 pm diameter, to sample oxygen partial pressure gradients across soybean nodules. Their results indicated that the major resistanceto oxygen transport in root nodules lies in the cortex. More recently, Revsbech& Ward ( 1983) successfullyproduced a Clark-type microelectrodeto study oxygen microprofiles in natural sediments(Revsbech & Jorgensen 1986) and algal biofilms (Glud er al.1992), Hojberg & Sorensen(1993) studied oxygen in the barley rhizosphere, while Witty et al. (.1981)have used this electrode to study the oxygen distribution in the root nodules of pea and french bean. They also found that the major resistanceto oxygen transport in root nodules is in the cortex, and came to the conclusion that the resistancevaried rapidly in responseto ambient oxygen levels becauseof the disappearanceor re-appearanceof the intercellular spaces. A glass-sheathed,bare-tipped platinum microelectrode (tip diameter c. 5 trrm)has recently been used to study the radial oxygen distribution in maize roots. These roots were supplied with oxygen from the shoot, and to a lesserextent radially from an agar jacket surrounding the root (Armstrong et al. 1993). In keeping with modelling predictions (Armstrong et al. 1990), steep radial diffusion gradients were a characteristic of the non-porous epidermal/hypodermal shell and stele, while shallow profiles were generally found in the cortex (Fig. 6). Stelar anoxia, previously forecast by mathematical modelling, was inducible by manipulating the oxygen concentrations around the shoot, and examples of two-way and one-way diffusion in the epidermal/hypodermal cylinder, also predicted by mathematical modelling, were to be found. Profiles through the meristematic zone and the root cap indicated that, with roots in stagnant anaerobic media, the bulk of the root cap proper is likely to be anoxic, and that some anoxia may be the norm within the basipetal

525

Polarographic ox))gen elec tro des

10 6 i(

i

i

q

\P l io i i i :i i i

(d

i i

o

izi

() d O

X

6

io"ft

!

i r9"oa i u,", i fi

ii

i i

ii > ie

l

i

i t r 1i i o i i a ;

i :l l * ! i ,r !

!

i n i i

H

F b

li

i

i !'

i:

i2

J i

i=

ii

t l

l

r"i'aoo^g.p@o %

I i I

^o

t'-

I

srELE

i i

;

:

itr

:

i ^

i

i o i

i

i3

:

:

i

200

;

i xE ii +E

i

x :

iii\

i t

i > i

0

rr i\

Ii

1i5 lil M i I ii l l

i H '

i #

i iH 400

600

800

1000

1200

Dislance (g.m) Figure6.

embedded in agar (l% Radial oxygen distribution across a maize seedling root (l:85mm) w1v), at a position 35 mm from the base.Obtained by means of a bare-tipped Pt-microelectrode (d:5 pm) using a piezoelectric driver. The maize seedling was intact and the bulk of the oxygen supply came from the shoot by diffusion through the non-aerenchymatous cortical gas-space.M indicates position at which the eleclrode has moved the root for a few steps. (After Armstrong et al. 1993.)

extensionsof the cap which ensheaththe root apex. It was noted, however, that the damage caused by the electrode, even when the meristem and root cap were penetrated, was sufficiently localised that it did not obviously affect the oxygen regime within the root, nor interfere with growth. Up to now, microelectrodeshave been scarcelyused in plant aeration studies,not least becausethe construction and use of micro-electrodesare more difficult than for other electrode types, and the polarographic equipment is more expensivedue to the requirement to generatenoise-freecurrents in the nA, pA and even fA ranges. In addition, it is desirable,although not absolutely essential,that the micro-electrodes be servo-driven into the tissues.High impedancedevicessuitable for measuring such small currents have always been readily obtainable (Keithly Instruments, USA); now, however, complete polarographs with software and interfacing for 486 type PC-controlled microdriving (World Precision Instruments Micro-driver) and data collection may be obtained (Barman Electronics, Leys Lane, Skipsea, Driffield, E . Y o r k s . ,U K ) . Despite the constraints, there is clearly enormous opportunity for increasing the scope of plant aeration studies by the use of microelectrodes.

526

W. Armstrong Acknowledgements

I am indebted to the Royal Society, London, for financial support in connexton with our microelectrode work. I thank my wife, Dr J. Armstrong for her assistance during the preparation of this manuscript.

References Armstrong, J. & Armstrong, W. 1989. Light-enhanced convective throughflow increasesoxygenation in . ewPhytologistll4, l2l 8. r h i z o m e s a n d r h i z o s p h e r e o f P h r a g m i t e s a u s t r a l i s ( C a v . ) T r i n . e x S t e u dN Armstrong, W. 1964. Oxygen diffusion from the roots of some British bog plants. Nature 204, 801 2. Armstrong, W.1967a. The use of polarography in the assay of oxygen diffusing from roots in anaerobic media. Physiologia Plantarum 20, 540 53. Armstrong, W. 1967b. The relationship between oxidation-reduction potentials and oxygen-diffusion levels in some waterlogged organic solls. Journal oJ'Soil Science18,27 34. Armstrong, W. 1979. Aeration in Higher Plants. 1n Woolhouse, H. W. W. (Ed.) Advant'esin botanical re,seorchvol. 7, pp. 225 332. London: Academic Press. Armstrong, W. 1988. Aeration in roots. 1n Cherry, T. (Ed.) Environmental stressphl,siology, NATO ASI series,vol. Gl9, pp. 197 206, Berlin: Springer-Verlag. Armstrong, W. & Gaynard, T. J. 1976. The critical oxygen pressurefor root respiration in intact plants. Phy.siologia Plantarum 37, 200-6. Armstrong, W. & Webb, T. 1985. A critical oxygen pressure for root extension in rice. Journal oJ Experimental Botan"v 36, 1573 82. Armstrong, W. & Wright, E. J. 1915. The theoretical basis for the manipulation of flux data obtained by the cylindrical platinum electrode technique. Physiologitt Plantarum 35, 21 6Armstrong, W. & Wright, E. J. 1976. A polarographic assembly for the large scale sampling of soil oxygen in the field. Journal o/ Applied Ecobgy 13, 849 56. Armstrong. W., Booth, T. C., Priestley, P. & Read, D. J.19'76. The relationship between soil aeration, stability and growth of Sitka spruce, (Picea sitchensis Borg. Carr) on upland peaty-gleys. Journal oJ Applied Ecologl'13, 585 91. Armstrong, W., Healy, X., & Lythe, L. 1983. Oxygen diffusion in pea. II The oxygen status of the primary root as affectedby growth, the production oflaterals and radial oxygen loss. New Phytoktgist 94.549 59. Armstrong, W., Beckett, P. M., Justin, S. H. F. W. & Lythe, S. 1990. Modelling and other aspectsof root aeration. 1n Jackson. M. B., Davies, D. D. & Lambers, H. (Eds.) Plant li/e under oxygen stress, pp.267 82. The Hague: SPB Academic Publishing. Armstrong, W., Cringle, S., Brown, M. & Greenway, H. 1993. A microelectrode study of oxygen distribution in the roots of intact maize seedlings.In Jackson, M. B. & Black, C. R. (Eds) Interacting on plunts in a changing climate.NNlO ASI SeriesI; Global Change, Vol. 16, pp.287 304stres,\es Berlin: Springer-Verlag. Baumgiirtl, H. & Liibbers, D. W. 1983. Microcoaxial needle sensor for polarographic measurement of local oxygen pressure in the cellular range of living tissue. Its Construction and properties. 1ll Gnaiger, E. & Forster, H. (Eds) Polarographicox))gensensors,pp.35-65. Berlin: Springer-Verlag. Beckett, P. M., Armstrong, W., Justin, S. H. F. W. & Armstrong, J. 1988. On the relative importance of convective and diffusive gas-flows in plant aeration. Nev' Phytologisl 110, 463 68. Bedford, B. L., Bouldin, D. R. & Beliveau, B. D. 1991. Net oxygen and carbon dioxide balances in solutions bathing roots of wetland plants. Journal of Ecology 79,943 59. Blackwell, P. S. 1983. Measurements of aeration in waterlogged soils: some improvements of techniques and their application to experiments using lysimeters.Journal o.f Soil Stience 34,271 85. Bowling, D. J. F. 1973. Measurement of a gradient of oxygen partial pressure across the intact root. Planta lll. 323-28. Chance, B. & Williams, G. R. 1955. Respiratory enzymes in oxidative phosphorylation. I. Kinetics of oxygen utilisation. Journal of Biologital Chemistry 217,383-93. Davies, P. W. 1962. The oxygen cathode. In Physical Techniques in biological research, Vol IV: special methods. pp.137-179. New York: Academic Press. Davies, P. W. & Brink, Jr. F . 1942.Microelectrodes for measuring local oxygen tensions in animal tissue. Refiew oJ Scientific Instruments 13,524-33. Gaynard, T. J. & Armstrong, W. 1987. Some aspectsof internal plant aeration in amphibious habitats. 1n Crawford, R. M. M. (Ed) Amphibious and intertidal plants. British Ecological Society Special Symposium 5, pp.303-320. Oxford: Blackwell Scientific Publishers.

Polar ographic oxy gen electro des

527

Glud, R. N., Ramsing, N. B. & Revsbech, N. P. 1992. Photosynthesis and photosynthesis-coupled respiration in natural biofilms quantified with oxygen microsensors.Journal of Phycology 28,51 60. Gnaiger, E. & Forster, H. 1983. Polarographic oxygen sensors.Berlin: Springer-Verlag. Gries, C., Kappen, L. & Losch, R. 1990. Mechanism of flood tolerance in reed, Phragmites australis (Cav.) Trin. ex Steudel. New Phytologisr 114, 589-93. Hojberg, O. & Ssrensen. 1993. Microgradients of microbial oxygen consumption in a barley rhizosphere model system. Applied and Environmental Microbiology 59,431 7. Kolthoff, I. M. & Lingane, J. J.1952. Polarography.New York: IntersciencePublishers. Kessler, M. & Lubbers, D. W. et al (Eds). 1973. Oxygen supply: theoretital and practical aspe(:tsof oxygen supply and microcirculation oJ lrs.rre. Workshop Proceedings, Max-Planqk Institute, Dortmund. Baltimore, USA: University Park Press. Kludze, H. K., Delaune, R. D. & Patrick, W. H. Jr. 1993. Aerenchyma formation and methane and oxygen exchange in rice. Soil Science Society of America Journal 57, 386 91. Laan, P., Smolders,A., Blom, C. W. P. M. & Armstrong, W. 1989.The relative roles of internal aeration, radial oxygen losses, iron exclusion and nutrient balance in flood-tolerance in Rumex species.Acta Botanita Neerlandica 38. I31 45. Laan, P., Tosserams,M., Blom, C. W. P. M. & Armstrong, W. 1990.Internal oxygen transport in Rumer speciesand its significancefor respiration under hypoxic conditions. Plant & Soil 122,39-46. Lemon, E. R. & Erickson, A. E. 1952. The measurementof oxygen diffusion in the soil with a platinum electrode. Proteedings of the Soil Stience Society of America 16, t60 63. Lemon, E. R. & Erickson, A. E. 1955. Principle of the platinum microelectrode as a method of characterizing soil aeration. Soil Science79,383-92. Mclntyre, D. S. 1970. The platinum microelectrode method for soil aeration measurement.Advances in A g r o n L t m y2 2 . 1 3 5 2 8 3 . Revsbech, N. P. 1989. An oxygen sensor with a guard cathode. Limnology & Oceanography34,474 78. Revsbech, N. P. & Jorgensen, B. B. 1986. Microelectrodes: their use in microbial ecology. Adyances in Microbial Ecoktgy 9,293 352. Revsbech, N. P. & Ward, D. M. 1983. Oxygen microelectrode that is insensitive to medium chemical composition: use in an acid microbial mat dominated by Cyanidium caldarium. Applied Environmental Microbiology 45, 755 99. Saglio, P. H., Rancillac, M., Bruzan, F. & Pradet, A. 1984. Critical oxygen pressure for growth and respiration of excised and intact rooIs. Plant Physiobgy 76, 151 54. Sorrell, B. K. & Armstrong, W. 1994. On the difficulties of measuring oxygen releaseby root systemsof wetland plants. Journal o/ Ecology 82, 117 84. Sorrell, B. K., Brix, H. & Orr, P. T. 1993.Oxygen exchangeby entire root systemsof Cyperu,sinyctlucratus and Eleot:lruris sphacelata. Journal oJ Aquatic Plant Management 31,24 28. Thomson, C. J., Armstrong, W., Waters, I. & Greenway, H. 1990.Aerenchyma formation and associated oxygen movement in seminal and nodal roots of wheaL Plant. Cell und Environment 13, 395 403. Tjepkema, J. D. & Yocum, C. S. 1974. Measurement of oxygen partial pressurewithin soybean nodules by oxygen microelectrodes.Plunta l19, 351 60. Waters, I., Armstrong, W., Thomson, C. J., Setter, T. L., Adkins, S., Gibbs, J. & Greenway, H. 1989. Diurnal changes in radial oxygen loss and ethanol metabolism in roots of submerged and non-submerged rice seedlings.New Phytologisl 113,439 51. Whalen, W. J., Riley, J. & Nair, P. 1967. A microelectrode for measuring intracellular pOr. Journal oJ Applied Physiobgv-23, 798 801. Witty, J. F., Skot, L. & Revsbech, N. P. 1987. Direct evidence for changes in the resistanceof legume nodules to oxygen diffusion. Journal of Experimental Botany 38,1129-40.