Simultaneous measurement of acetylene reduction and respiratory gas exchange of attached root nodules

A method was developed for the simultaneous measurement of acetylene reduction, carbon dioxide evolution and oxygen uptake by individual root nodules of intact nitrogen-fixing plants (Alnus rubra Bong.). The nodules were enclosed in a temperature-controlled leak-tight cuvette. Assay gas mixtures were passed through the cuvette at a constant, known flow rate and gas exchange was measured by the difference between inlet and outlet gas compositions. Gas concentrations were assayed by a combination of an automated gas chromatograph and a programmable electronic integrator. Carbon dioxide and ethylene evolution were determined with a coefficient of variation which was less than 2%, whereas the coefficient of variation for oxygen uptake measurements was less than 5%. Nodules subjected to repeated removal from and reinsertion into the cuvette and to long exposures of 10% v/v acetylene showed no irreversible decline in respiration or acetylene reduction. This system offers long-term stability and freedom from disturbance artifacts plus the ability to monitor continuously, rapidly and specifically the changes in root nodule activity caused by environmental perturbation.     

A Model of the Regulation of Nitrogenase Electron Allocation in Legume Nodules (II. Comparison of Empirical and Theoretical Studies in Soybean)

In N2-fixing legumes, the proportion of total electron flow through nitrogenase (total nitrogenase activity, TNA) that is used for N2 fixation is called the electron allocation coefficient (EAC). Previous studies have proposed that EAC is regulated by the competitive inhibition of H2 on N2 fixation and that the degree of H2 inhibition can be affected by a nodule's permeability to gas diffusion. To test this hypothesis, EAC was measured in soybean (Glycine max L. Merr.) nodules exposed to various partial pressures of H2 and N2, with or without changes in TNA or nodule permeability to gas diffusion, and the results were compared with the predictions of a mathematical model that combined equations for gas diffusion and competitive inhibition of N2 fixation (A. Moloney and D.B. Layzell [1993] Plant Physiol 103: 421-428). The empirical data clearly showed that decreases in EAC were associated with increases in external pH2, decreases in external pN2, and decreases in nodule permeability to O2 diffusion. The model predicted similar trends in EAC, and the small deviations that occurred between measured and predicted values could be readily accounted for by altering one or more of the following model assumptions: K1(H2) of nitrogenase (range from 2-4% H2), Km(N2) of nitrogenase (range from 4-5% N2), the allocation of less than 100% of whole-nodule respiration to tissues within the diffusion barrier, and the presence of a diffusion pathway that is open pore versus closed pore. The differences in the open-pore and closed-pore versions of the model suggest that it may be possible to use EAC measurements as a tool for the study of legume nodule diffusion barrier structure and function. The ability of the model to predict EAC provided strong support for the hypothesis that H2 inhibition of N2 fixation plays a major role in the in vivo control of EAC and that the presence of a variable barrier to gas diffusion affects the H2 and N2 concentration in the infected cell and, therefore, the degree of H2 inhibition.     

Respiration and nitrogenase activity in nodules ofCasuarina cunninghamiana and cultures ofFrankia sp. HFP020203: Effects of temperature and partial pressure of O2

The effects of time after exposure to acetylene and of nodule excision were examined using a flow-through system. After a transient depression in the rate of acetylene reduction that began about 1.5 min after exposure to acetylene, the rate recovered to 98% of the initial maximum value after 40 min. After nodule excision the rate stabilized to 90% of the initial maximum value observed in the intact plant.Excised nodules, measured at 6-min intervals in a closed system, with frequent changes of the gas mixture, were used for the remaining experiments. Acetylene reduction by the nodules increased rapidly as temperature was increased between 6 and 26°C. Between 26 and 36°C there was relatively little effect of temperature on acetylene reduction.Nodules and cultures ofFrankia were compared with respect to the effect of temperature and pO₂ (partial pressure of oxygen) on oxygen uptake. Cultures ofFrankia were grown on a nitrogen-free medium at either 0.3 kPa O₂ (vesicles absent) or 20 kPa O₂ (vesicles present). Oxygen uptake by nodules (vesicles absent) and by vesicle-containing cultures was strongly dependent on pO₂ at values below 20 kPa. This suggests the presence of a barrier to oxygen diffusion. Oxygen uptake was dependent on temperature as well as on pO₂, but the Q₁₀ was much larger for the cultures than for the nodules. This suggests that vesicles or related structures are not the source of the diffusion barrier in Casuarina nodules. Respiration by cultures ofFrankia lacking vesicles became O₂-saturated at low pO₂ values. Thus these cultures did not have a significant diffusion barrier. From these results it is concluded that nodules ofCasuarina cunninghamiana have a barrier to oxygen diffusion supplied by the host tissue and not byFrankia.     

Modeling and analysis of diffusion and reaction in legume nodules

Diffusion of gases through legume nodules is important for nitrogen fixation. A mathematical model is presented for diffusion and enzymatic reaction for legume nodules with a reactive core and an inert shell. The transient model is solved numerically for spherical geometry for acetylene reduction by nitrogenase enzyme. The results are used to estimate the diffusivities of acetylene and ethylene in the nodules by comparing predicted and experimental lag times. The experimental results are also analyzed using an effectiveness factor plot for spherical nodules with inert shells and reactive cores. The results show that the diffusivities are slightly higher than those for acetylene and ethylene in water because of some contribution of gas phase diffusion. Applications to oxygen diffusion through nodule tissue are suggested.     

Response to drought stress of nitrogen fixation (acetylene reduction) rates by field-grown soybeans

The effects of drought stress on soybean nodule conductance and the maximum rate of acetylene reduction were studied with in situ experiments performed during two seasons and under differing field conditions. In both years drought resulted in decreased nodule conductances which could be detected as early as three days after water was withheld. The maximum rate of acetylene reduction was also decreased by drought and was highly correlated with nodule conductance (r = 0.95). Since nodule conductance is equal to the nodule surface area times the permeability, the relationship of these variables to both whole-plant and unit-nodule nitrogenase activity was explored. Drought stress resulted in a decrease in nodule gas permeability followed by decreases in nodule surface area when drought was prolonged. Under all conditions studied acetylene reduction on a unit-nodule surface area basis was highly correlated with nodule gas permeability (r = 0.92). A short-term oxygen enrichment study demonstrated nodule gas permeability may limit oxygen flux into both drought-stressed and well-watered nodules of these field-grown soybeans.     

A rapid non-destructive assay to quantify soybean nodule gas permeability

The low gas permeability of a diffusion barrier in the cortex of soybean nodules plays a significant role in the protection of nitrogenase from oxygen inactivation. It may also set an upper limit on nodule respiration and nitrogen fixation rates. Two methods which have been used to quantify the gas permeability of leguminous nodules are reviewed and found to be unreliable. A new assay technique for determining both the nodule activity and gas permeability is developed and tested. This ‘lag-phase’ assay is based on the time nodules require to reach steady-state ethylene production after being exposed to acetylene. The technique is rapid, insensitive to errors in biochemical parameters associated with nitrogenase, and is non-destructive. The method was tested with intact aeroponically grown soybean plants for which the mean nodule gas permeability was found to be 13.3×10⁻³ mms⁻¹. This corresponds to a layer of cells approximately 35 um thick and is consistent with previously reported values.     

Effect of O2 on vesicle formation, acetylene reduction, and O2-uptake kinetics in Frankia sp. HFPCcI3 isolated from Casuarina cunninghamiana

The effect of the partial pressure of oxygen (PO2) on the formation of vesicles, which are thought to be the site of N2 fixation in Frankia, was studied in HFPCcI3, an effective isolate from Casuarina cunninghamiana. Unlike other actinorhizal root nodules, vesicles are not produced by the endophyte in Casuarina nodules. However, in culture under aerobic conditions, large, phase-bright vesicles are formed in HFPCcI3 within 20 h following removal of NH+4 from the culture medium and reach peak numbers within 72 to 96 h. In vivo acetylene reduction activity parallels vesicle formation. Optimum rates of acetylene reduction in short-term assays occurred at 20% O2 (0.2 atm (1 atm = 101.325 kPa] in the gas phase. O2 uptake (respiration) determined polarographically showed diffusion-limited kinetics and remained unsaturated by O2 until 300 microM O2. In contrast, respiration in NH+4-grown cells was saturated by O2 between 8 and 10 microM O2. These results indicate the presence of a diffusion barrier associated with the vesicles. Vesicle development was repressed in cells incubated in N-free media sparged with gas mixtures with PO2 between 0.001 and 0.003 atm. Nitrogenase was induced under these conditions, but acetylene reduction was extremely O2 sensitive. The kinetics of O2 uptake as a function of dissolved O2 concentration in avesicular cells were similar to those in NH+4-grown cells indicating the lack of a diffusion barrier. These results demonstrate that vesicle formation and the development of the O2 protection mechanisms of nitrogenase are regulated by ambient PO2 in HFPCcI3.     

A Model of the Regulation of Nitrogenase Electron Allocation in Legume Nodules (I. The Diffusion Barrier and H2 Inhibition of N2 Fixation)

A mathematical model is presented to explain the regulation of nitrogenase electron allocation to N2 fixation (EAC) in legume nodules. The model is based on two assumptions: (a) that H2 inhibits N2 fixation in a competitive manner; and (b) that O2, H2, and N2 move into and out of nodules by diffusion and their movement is impeded by a diffusion barrier, the permeability of which is controlled to maintain a very low infected cell O2 concentration. When the model was used to simulate nodules displaying a range of values for total nitrogenase activity (TNA), maximum EAC values were predicted to be between 0.69 and 0.71, and a negative correlation was predicted to exist between EAC and TNA. These predictions were in good agreement with empirically derived values reported in the literature and support the suggestion that H2 inhibition of N2 fixation is a major determinant in the regulation of nitrogenase EAC in legume nodules. Two versions of the model were constructed. A closed-pore model assumed that the diffusion barrier consisted of a solid shell of water of variable thickness in the nodule cortex. An open-pore model assumed that a small number of gas-filled intercellular spaces connected the nodule central zone with the root atmosphere and these pores were opened or closed by water to provide variations in the nodule's permeability to gas diffusion. Because of differences in the diffusivity of gases in the gaseous and aqueous phases, the model predicted that, at a given infected cell O2 concentration, an open-pore diffusion barrier would result in less H2 accumulation in the infected cells than a closed-pore diffusion barrier. Therefore, the model may be used to test specific hypotheses about the physical structure of the barrier to gas diffusion in legume nodules.     

豆科根瘤内气体交换和气体扩散模型研究

近年来的研究表明根瘤皮层内存在着可调节的气体扩散屏障,它是由根瘤皮层内的一层细胞及填充在胞间隙的水层构成的,而根瘤是通过改变填充该层胞间隙的水层厚度来调节对气体扩散的阻力。本文概述了关于模拟豆科根瘤内气体交换和气体扩散的数学模型研究,阐明调节根瘤内含类菌体细胞维持低氧分压的有关问题。模型研究使我们获得了对共生固氮根瘤内极为复杂的微生态环境的初步认识,有待于通过改进试验和借助其他理论进一步探索根瘤气体交换和气体扩散的本质。     

Oxygen regulation of a nodule-located carbonic anhydrase in alfalfa

Control of the permeability to oxygen is critical for the function of symbiotic nitrogen fixation in legume nodules. The inner cortex (IC) seems to be a primary site for this regulation. In alfalfa (Medicago sativa) nodules, expression of the Msca1 gene encoding a carbonic anhydrase (CA) was previously found to be restricted to the IC. We have now raised antibodies against recombinant Msca1 protein and used them, together with antibodies raised against potato leaf CA, to demonstrate the presence of two forms of CA in mature nodules. Each antibody recognizes a different CA isoform in nodule tissues. Immunolocalization revealed that leaf-related CAs were localized primarily in the nitrogen-fixing zone, whereas the Msca1 protein was restricted exclusively to the IC region, in indeterminate and determinate nodules. In alfalfa nodules grown at various O(2) concentrations, an inverse correlation was observed between the external oxygen pressure and Msca1 protein content in the IC, the site of the putative diffusion barrier. Thus Msca1 is a molecular target of physiological processes occurring in the IC cells involved in gas exchange in the nodule.     

Effect of nitrate on acetylene reduction activity and carbohydrate composition of Phaseolus vulgaris nodules

When Phaseolus vulgaris L. cv. Kentucky Wonder plants were supplied with various levels of nitrate for 34 days, nodule weight (plant)⁻¹, acetylene reduction activity (g nodule)⁻¹, and sugar concentration in nodules were depressed >60% (7.5 m M nitrate vs nil nitrate). Starch concentration in nodules was more than double the sugar concentration and declined only slightly in response to nitrate level. At the highest level of nitrate, sugar concentration in nodules was 50% greater than that in roots and nodule starch was about 6-fold greater than root starch on a fresh weight basis. When plants were grown with 1 m M nitrate and then supplied with 12 m M nitrate for 7 days, the rapid decline in acetylene reduction activity coincided with a decline in sucrose concentration. However, glucose and fructose concentrations declined only after the largest decrease in acetylene reduction had occurred, and the quantitative decrease in glucose and fructose in nodules was small relative to sucrose. Other results showed that the magnitude of the effect of nitrate on some nodule carbohydrate compounds depends on Rhizobium phaseoli strain and on whether plants were grown with or without nitrate prior to experimental treatments. Some of the results are consistent with the carbohydrate-deprivation hypothesis for inhibition of legume nodules by nitrate. However, there are several complications involved in the interpretation of results of this type, and other possible explanations for the results are suggested.     

Regulation of Soybean Nitrogen Fixation in Response to Rhizosphere Oxygen: II. Quantification of Nodule Gas Permeability

Nodule nitrogen fixation rates are regulated by a mechanism which is responsive to the rhizosphere oxygen concentration. In some legumes, this oxygen-sensitive mechanism appears to involve changes in the gas permeability of a diffusion barrier in the nodule cortex. In soybean evidence for such a mechanism has not been found. The purpose of this research was to make quantitative measurements of soybean nodule gas permeability to test the hypothesis that soybean nodule gas permeability is under physiological control and responsive to the rhizosphere oxygen concentration. Intact hydroponically grown soybean plants were exposed to altered rhizosphere oxygen concentrations, and the nodule gas permeability, acetylene reduction and nodule respiration rates were repeatedly assayed. After a change in the external oxygen concentration, nitrogenase activity and nodule respiration rates displayed a short-term transient response after which the values returned to rates similar to those observed under ambient oxygen conditions. In contrast to steady-state nitrogenase activity and nodule respiration, nodule gas permeability was dramatically affected by the change in oxygen concentration. Decreasing the external oxygen concentration to 0.1 cubic millimeter per cubic millimeter resulted in a mean increase in nodule gas permeability of 63%. Increasing the rhizosphere oxygen concentration resulted in decreased nodule gas permeability. These data are consistent with the hypothesis that soybean nodules are capable of regulating nitrogen fixation and nodule respiration rates in response to changes in the rhizosphere oxygen concentration and indicate that the regulatory mechanism involves physiological control of the nodule gas permeability.     

Tracing the Path of Oxygen into Birdsfoot Trefoil and Alfalfa Nodules using Iodine Vapor

Although physiological control of nodule 0₂ permeability is an active area of research, the gas diffusion pathway between the atmosphere and the infected zone has not been firmly established. Previous studies have used infiltration of ink or dyes to identify points of entry, but such water-soluble tracers could give a misleading picture of gas diffusion pathways. We therefore used iodine vapor (and its reaction with starch) to trace gas-phase pathways into the infected zone of determinate birdsfoot trefoil (Lotus corniculatus) and indeterminate alfalfa (Medicago sativa) nodules. We also used histochemical methods to identify suberized or lignified layers that could act as barriers to gas diffusion. Birdsfoot trefoil nodules were surrounded by a suberized periderm, but nonsuberized cells and intercellular spaces were observed in the periderm between lenticels and their associated vascular bundles. Iodine entered birdsfoot trefoil nodules only through lenticels. The periderm appears to provide a significant barrier to gas diffusion. Although airspaces were rare in the nodule parenchyma (also referred to as the “inner cortex”), we found some evidence that a few air-filled pathways cross this secondary barrier, also in the vicinity of vascular bundles. Alfalfa nodules were cylindrically surrounded by a suberized endodermis which ended near the meristematic tip; iodine entered principally at the end of the endodermis near the meristem. Future research on physiological control of nodule O₂ permeability should concentrate on strategic “choke points”, associated with lenticels in determinate nodules, or in the zone proximal to the meristem in indeterminate nodules.     

Physical and morphological constraints on transport in nodules

For active nodule nitrogen fixation, O₂, N₂, and carbohydrate must be transported throughout the nodule. No quantitative analysis of these transport processes in the nodules has been presented. By invoking several simplifying assumptions, a second-order differential equation for the various gradients and concentrations in the nodule was solved. Even though the nodule can only be approximated in this analysis, it indicates clearly that intercellular gas spaces must exist in nodules for adequate O₂ distribution. To preserve low O₂ concentrations and protect the nitrogenase, these gas spaces cannot be in direct contact with the ambient atmosphere. It is hypothesized that a gas barrier exists in the cortical region of the nodule to limit O₂ diffusion. This barrier would not substantially inhibit N₂ transport. Carbohydrate transport from the vascular tissue via diffusion in the liquid phase can adequately accommodate the requirements within the nodule.     

Correlation between ultrastructural differentiation of bacteroids and nitrogen fixation in alfalfa nodules.

Bacteroid differentiation was examined in developing and mature alfalfa nodules elicited by wild-type or Fix- mutant strains of Rhizobium meliloti. Ultrastructural studies of wild-type nodules distinguished five steps in bacteroid differentiation (types 1 to 5), each being restricted to a well-defined histological region of the nodule. Correlative studies between nodule development, bacteroid differentiation, and acetylene reduction showed that nitrogenase activity was always associated with the differentiation of the distal zone III of the nodule. In this region, the invaded cells were filled with heterogeneous type 4 bacteroids, the cytoplasm of which displayed an alternation of areas enriched with ribosomes or with DNA fibrils. Cytological studies of complementary halves of transversally sectioned mature nodules confirmed that type 4 bacteroids were always observed in the half of the nodule expressing nitrogenase activity, while the presence of type 5 bacteroids could never be correlated with acetylene reduction. Bacteria with a transposon Tn5 insertion in pSym fix genes elicited the development of Fix- nodules in which bacteroids could not develop into the last two ultrastructural types. The use of mutant strains deleted of DNA fragments bearing functional reiterated pSym fix genes and complemented with recombinant plasmids, each carrying one of these fragments, strengthened the correlation between the occurrence of type 4 bacteroids and acetylene reduction. A new nomenclature is proposed to distinguish the histological areas in alfalfa nodules which account for and are correlated with the multiple stages of bacteroid development.     

Mathematical modeling of oxygen diffusion and respiration in legume root nodules

The O₂ permeability of legume root nodules is under physiological control; decreases in permeability are triggered by various forms of stress. Two linked mathematical models were used to explore several hypotheses concerning the physical nature of the variable diffusion barrier in nodules. Respiration and diffusion of dissolved O₂ and oxygenated leghemoglobin were simulated for the nodule cortex and the nodule interior. Measured nodule permeabilities were shown to be inconsistent with the hypothesis that large numbers of air-filled pores penetrate the diffusion barrier. Changes in the affinity of leghemoglobin for O₂ or in the rate of cytoplasmic streaming in diffusion barrier cells did not result in the large changes in O₂ permeability reported for real nodules. The presence or absence, but not the thickness, of aqueous plugs in radial pores through the cortex was found to have a large effect on permeability. Flooding of intercellular spaces, either between layers of cells in the cortex or in the nodule interior, also caused large changes in simulated permeability. The unsteady-state O₂ method for determining nodule permeability was tested using data generated by the model. The accuracy of the method was confirmed, provided that certain assumptions (full oxygenation of leghemoglobin under pure O₂ and uniform conditions in the nodule interior) are met.     

Further Errors in the Acetylene Reduction Assay: Effects of Plant Disturbance

A flow-through gas system was used to study the effects of disturbanceon nitrogenase (acetylene reduction) activity of nodulated rootsystems of soyabean (Glycine max) and white clover (Trifoliumrepens). Detopping plus removal of the rooting medium (by shaking)produced a substantial decrease in maximum nitrogenase activity.This response is due to a reduction in oxygen flux to the bacteroidscaused by an increase in the oxygen diffusion resistance ofthe nodule. The decrease in maximum nitrogenase activity wasmuch smaller for roots subjected to detopping only. Thus, theeffect of root shaking is more important than that of shootremoval. The effect of detopping plus root shaking on nitrogenase activityoccurred whether the plants were equilibrated and assayed at25°C or 15°C. However, the effect of disturbance onthe oxygen diffusion resistance of the nodules, and thus onnitrogenase activity, was greater at the higher temperature.At the lower temperature the oxygen diffusion resistance ofthe nodules had already been increased in response to the reducedrequirement for oxygen. These nodules were less susceptibleto the effects of disturbance. Thus, comparisons of the effectsof equilibration temperature on nitrogenase activity produceddifferent results depending on whether intact or disturbed systemswere used. With intact systems activity was lower at the lowertemperature but with detopped/shaken roots the lowest activityoccurred at the higher temperature. It is concluded that the use of detopped/shaken roots can producesubstantial errors in the acetylene reduction assay, which makesthe assay invalid even when used for comparative purposes. However,comparisons with rates of ¹⁵N₂ fixation and H₂ production showthat accurate measurements of nitrogenase activity can be obtainedfrom maximum rates of acetylene reduction by intact plants ina flow-through gas system. The continued use of assay proceduresin which cumulated ethylene production from disturbed systemsis measured in closed vessels must be questioned. Key words: Nodules, acetylene, nitrogenase activity     

Direct Test for Altered Gas Exchange Rates in Water stressed Soybean Nodules

Water stress usually lowers the nitrogenase activity of soybeanroot nodules. This loss in activity might result from an increasedbarrier to nodular gas exchange, from a general reduction inbiochemical function, or both. To test for the possibility ofan increased barrier to gas diffusion, we measured the apparentlag time for initiation of acetylene reduction by intact soybeanplants, both before and after water stress. Mild nodular waterloss (i.e. 10% of fresh weight or less) seldom altered the apparentlag time, whereas severe water stress (20–40% f. wt loss)frequently produced a small increase in apparent lag time. Severewater stress also produced a large decrease (24%) in the externaldiameter of the nodules and a loss of the white lenticel traces.Water stress usually caused a decrease in the apparent K_m foracetylene. The data do not suggest a large change in the diffusiveresistance to acetylene of nodules subjected to water stress.Thus, the observed decrease in nitrogenase activity may resultprimarily from biochemical, rather than physical, changes. However,because of the relatively greater importance of gas-phase diffusionfor oxygen entry, we cannot exclude the possibility of a largechange in a small gas pathway that affects oxygen influx morethan acetylene influx. Diffusion, Glycine max, nitrogen fixation, water stress     

Dimensions and distribution of intercellular spaces in cryo-planed soybean nodules

The ability of legume nodules to regulate their permeability to gas diffusion has been attributed to physiological control over the size and distribution of gas-filed intercellular spaces within the nodule cortex. To examine the size and distribution of intercellular spaces and to determine whether they were filled with gas (high diffusion permeability) or liquid (low diffusion permeability), whole nodules were frozen in liquid nitrogen slush (-210°C), and then either cryo-fractured or cryo-planed before being examined by cold-stage scanning electron microscopy (SEM). The cryo-planed tissue was found to have many advantages over cryo-fractured nodules in providing images which were easier to interpret and quantify. Intercellular spaces throughout the nodule were examined in both tangential and medial planed faces. Since no differences were observed between views in either the size or shape of the open intercellular spaces, it was concluded that the intercellular spaces of nodules were not radially oriented as assumed in many mathematical models of gas diffusion. The inner cortex region in the nodules had the smallest intercellular spaces compared to other zones, and less than 10% of the intercellular spaces were occluded with any type of material in the central zone regions. Vacuum infiltration of nodules with salt solutions and subsequent cryo-planing for SEM examination showed that open and water-filled intercellular spaces could be differentiated. The potential is discussed for using this method to study the mechanism of diffusion barrier regulation in legume nodules.     

A new method to detect the presence of continuous gas-filled pathways for oxygen diffusion in legume nodules

A new method is presented which evaluates the contribution ofgas-filled and water-filled pathways across the barrier whichrestricts O₂ diffusion into infected cells of nodules. SinceO₂ will move through air about 104 times faster than throughwater any continuous gas-filled pathways which traverse thecortex would form a major route for O₂ transport. However, microscopicevidence for the existence of such direct connections is ambiguous. On theoretical grounds, O₂ should diffuse through a He atmosphereabout 3.7 times faster than through air or Ar, but O₂ flux acrossa liquid barrier should be unaffected by changes in the backgroundmixing gas. Thus, if O₂ influx to the infected cells is increasedwhen the ambient gas phase is changed from air to He/21% O₂this is evidence for a continuous pathway of gas-filled poresacross the whole width of the cortical barrier. This theoreticalapproach was validated by measuring the rates of diffusion ofO₂ through milli-pore filters in background atmospheres of eitherair, Ar or He. These membranes were used dry, to simulate the‘open pore’ model for nodule diffusion, or wettedwith water or gum Arabic (which is similar to the glycoproteinsassociated with the barrier) to simulate the ‘closed pore’situation. Knowing the diffusion constants for the gas- and water-filledpathways that may be involved in the nodule barrier, we thenevaluated the contribution of these components to the totaldiffusion resistance by examining the effect of Ar/O₂ and He/O₂gas mixtures on H₂ production and respiration of nodules. Theresults indicate that, in unstressed soyabean nodules, abouthalf of the O₂ flux to infected cells is via inter-connectedgas-filled pores, which ‘close’ to produce a liquid-filledbarrier as the diffusion resistance increases in response tostress. In unstressed lupin nodules all of the O₂ flux crossesa liquid-filled barrier. Key words: Oxygen diffusion, nodules, nitrogen fixation