Printer Friendly

Root hairs: specialized tubular cells extending root surfaces.

II. Introduction

Root hairs are present on roots of representatives of all the major groups of vascular plants, indicating their long evolutionary history and presumably their significance to the success of various plant groups in adapting to changing environmental conditions (Peterson, 1992). Considerable literature has accrued on root hair structure, development, and physiology and, more recently, on the interaction between root hairs and rhizosphere microorganisms. Although several reviews have dealt with aspects of root hair structure and function (Cormack, 1949, 1962; Robards, 1983; Hofer, 1991), there has not been a comprehensive synthesis of information pertaining to all aspects of the biology of root hairs.

This review integrates new information on root hairs and indicates areas of current interest in the study of these important root structures.

III. Sites of Root Hair Initiation

The most striking example of cellular differentiation in the root epidermis is the initiation and development of root hairs. In the majority of plant species, any protoderm cell has the potential to form a root hair. In a few other species, however, root hair formation occurs only in certain cells that are specialized early in epidermis differentiation (Cormack, 1949, 1962). These cells, termed "trichoblasts" (Leavitt, 1904), have been the focus of considerable research, because they have many distinctive physiological and structural characteristics and they represent potential model systems to study the bases of cellular differentiation.

It has been known since Leavitt's observations in 1904 that trichoblasts are the result of an asymmetric cytokinesis that occurs during early epidermis differentiation. This has been confirmed by more recent light microscopical observations (Cutter & Feldman, 1970a) and ultrastructural studies (Avers, 1963; Cutter & Hung, 1972; Gunning et al., 1978). In festucoid grasses such as Phleum, the smaller cell of the asymmetric cytokinesis is located at the apical end of the initial cell (i.e., toward the root apex) (Avers, 1963), whereas in Hydrocharis morsus-ranae, a member of the Hydrocharitaceae, it is at the proximal end (Cutter & Hung, 1972). Prior to cell plate formation, the nucleus migrates either to the apical end of the cell, as in Phleum (Avers, 1963), or to the proximal end of the cell, as in Hydrocharis morsus-ranae (Cutter & Hung, 1972), and vacuoles may appear at the opposite end of the initial (Cutter & Hung, 1972). The protoderm cell also develops a further polarity, in that a pre-prophase band of microtubules determines the site of cell plate formation (Gunning et al., 1978). The subsequent asymmetric cytokinesis is critical to the divergence in development of the sibling cells, since the longer cell can undergo one or more mitoses, with the progeny all developing as hairless epidermal cells (Avers, 1963). Subsequent to cell plate formation, numerous physiological and structural differences distinguish the trichoblasts from adjacent epidermal cells.

Avers and her co-workers [see Cormack (1962) for a summary of early work] have shown that in the festucoid grass species Phleum pratense, there is a differential activity of many enzymes (e.g., succinic dehydrogenase, peroxidase, acid phosphatase) between trichoblasts and hairless cells within a defined zone of the root, with the trichoblasts usually showing enhanced activity over hairless initials. This was confirmed for phoshatase activity using a variety of enzyme substrates (Czernik & Avers, 1964). Of interest in this study is that enzyme activity was present in only one of the two sibling cells, with distribution dependent on the region of the root in which they were located. In the 100-200 [[micro]meter] region of the meristem proximal to the apex, hairless cell initials showed enzyme activity and trichoblasts did not; in the 200-300/[[micro]meter] region, the reverse was true.

Using Elodea canadensis roots, Dosier and Riopel (1977) studied 19 enzymes for differential activity between trichoblasts, and hairless initials. All except phosphorylase showed the same pattern, with a higher activity developing in trichoblasts immediately subsequent to the asymmetric cytokinesis. With phosphorylase, activity was higher in trichoblasts, but only immediately prior to hair formation. These authors argue that this latter protein is the best candidate of those studied to qualify as a "root-hair"-specific protein. This system could be explored further with current molecular biology techniques.

Differences in other macromolecules have also been observed between trichoblasts and hairless initials. Trichoblasts in H. morsus-ranae show higher nuclear and cytoplasmic RNA, more nucleohistone, and more total protein (Cutter & Feldman, 1970b); similar results were obtained with Elodea canadensis (Dosier & Riopel, 1978). Trichoblast nuclei of H. morsus-ranae and E. canadensis continue to synthesize DNA after the asymmetric cytokinesis and become endopolyploid (Cutter & Feldman, 1970b; Dosier & Riopel, 1978), whereas this does not occur in Phleum pratense. For this reason, it is difficult to assign a functional significance to endopolyploidy in terms of cell differentiation; endopolyploidy may be a result of trichoblast differentiation rather than a controlling factor (Cutter & Feldman, 1970b). In a comparison of DNA levels in various root cell types of Equisetum hyemale, Tradescantia Clone 02, and Hordeum vulgare, Murry and Christianson (1987) found that nuclei of all epidermal cell types, including trichoblasts, showed an increase in DNA levels during differentiation. In Equisetum hyemale the increases were at levels that were powers of 2, indicating endoreduplication (endopolyploidization), whereas in the other two genera the increased levels of 2.5C and 3C indicated some amplification rather than endoreduplication. In a subsequent study (Murry et al., 1987) it was shown experimentally that the nucleus of differentiating Hordeum vulgare root hairs does contain amplified extra chromosomal DNA sequences, thus supporting their earlier conclusions. A salt-stress prevented the amplification of DNA but did not inhibit root hair development, indicating that enhanced DNA levels are not a prerequisite for the differentiation of root hairs.

Trichoblasts may be quite different structurally from adjacent epidermal cells. Obvious differences occur in nuclear size in those species in which endopolyploidy occurs (Cutter & Feldman 1970b; Dosier & Riopel, 1978), and significant increases in nucleolar size in trichoblasts have also been noted (Lowary & Avers, 1965; Rothwell, 1964; Lewis & Rothwell, 1964). In H. morsus-ranae, a floating aquatic plant, chloroplasts differentiate in root cells; those in hairless epidermal cells are large and have well-developed thylakoids, whereas those in trichoblasts are small and poorly developed (Cutter & Hung, 1972).

Equisetum hyemale roots have trichoblasts characterized by numerous cell organelles and a thick two-layered outer tangential cell wall (Harris, 1979). Trichoblasts in Sinapis alba (Peterson, 1967) and in Lepidium sativum (Volkmann & Peters, 1995) can be recognized very early in protoderm differentiation by their position overlying intercellular spaces in the cortex and by their abundance of cellular organelles; this latter feature was also noted in rice by Kawata and Chung (1979).

Although panicoid grasses have been distinguished from festucoid grasses in part by the lack of trichoblasts in the root epidermis (Row & Reeder, 1957), Rothwell (1966) found evidence that there was some specialization of root epidermal cells in Panicum virgatum. Certain cells accumulated large numbers of protein bodies and had nuclei with enlarged nucleoli; these were usually the cells that formed hairs. Further work is needed, however, to determine if species supposedly lacking trichoblasts do show some specialization in protoderm cells.

IV. Structure and Growth of Root Hairs

A. INTRODUCTION

Root hairs, like pollen tubes and fungal hyphae, show a phenomenon called "tip growth." This specialized type of growth, in which wall synthesis and extension occur at the tip of the cell, has been studied and reviewed extensively (Sievers & Schnepf, 1981; Heath, 1990). The purpose here is not to review the mass of information that has accrued on tip growth in general but to summarize the main findings that pertain to root hairs.

B. CYTOLOGY

The cytological features of developing root hairs have been of considerable interest as they relate to the growth and functioning of these cells. Early ultrastructural studies (Sievers, 1963a, 1963b; Newcomb & Bonnett, 1965; Bonnett & Newcomb, 1966) described the general organization of organelles and subcellular structures and features of the cell wall using chemically fixed samples. Numerous more recent studies have focused either on general cytology (Meekes, 1985; Prin & Rougier, 1986; Ridge, 1988) or on the organization of the cytoskeleton and the relationship of this to patterns of cell wall synthesis, organelle positioning, or cytoplasmic streaming (see Heath, 1990; Lloyd, 1991). In this section the overall cytological features of root hairs will be summarized, followed by a discussion of the cytoskeleton in these cells.

A diagrammatic representation of root hair structure based on information obtained from living, conventionally fixed, and freeze-substituted hairs is presented in Figure 1. Much of the information, however, is derived from freeze-substituted root hairs of Vicia hirsuta (Ridge, 1988) and Equisetum hyemale (Emons & Derksen, 1986; Emons, 1987), since this method of preservation is superior to other methods for ultrastructural studies.

Various microscopical studies have shown the polarized organization of these tubular cells with a differential concentration of cytoplasmic structures at the tip (Sievers, 1963a, 1963b; Bonnett & Newcomb, 1966; Meekes, 1985; Ridge, 1988). Figure 1 is drawn to reflect this. The cytology of the extreme tip of root hairs has been of particular interest, since it is here that matrix substances for the forming cell wall originate and are carried to the plasma membrane. Although it has been known for years that a large population of vesicles is located in this region (Sievers, 1963a, 1963b; Bonnett & Newcomb, 1966; Volkmann, 1984) it is with freeze-substituted samples that the complexity of this population has been realized (Emons, 1987; Ridge, 1988). In V. hirsuta root hairs, Ridge (1988) identified three vesicle types in the sub-apical region: pyriformis vesicles found primarily in the apical dome; coated secretory vesicles located anywhere between the nucleus and the tip; coated vesicles [these were also described by Bonnett and Newcomb (1966) and Newcomb and Bonnett (1965)] associated with dictyosomes and the plasma membrane, in regions of cytoplasm away from the apical dome. Emons (1987), on the other hand, described only electron-dense spherical secretory vesicles in freeze-substituted root hairs of Equisetum hyemale, suggesting that species differences in this characteristic may occur.

The role of some of the vesicles at the extreme tip is presumably to carry substances to the plasma membrane, where, through a process of exocytosis, these substances are released into the matrix of the developing cell wall (Volkmann, 1984; Ridge, 1988). The origin of this population of vesicles is the endomembrane system, specifically the dictyosomes (Derksen & Emons, 1990). Peroxidase activity has been localized in dictyosomes, associated secretory vesicles, and ribosomes associated with endoplasmic reticulum and in the cell wall of Lepidium sativum root hairs (Zaar, 1979). Zaar (1979) suggests that peroxidase may protect the root hairs against soil pathogens. It has been proposed that the coated vesicle/coated pit system, positioned adjacent to the plasma membrane along the length of V. hirsuta and E. hyemale root hairs, may function in retrieving excess membrane as a result of fusion of various vesicles with the plasma membrane at the tip (Emons, 1987; Ridge, 1988).

The plasma membrane at the tip and along the length of root hairs appears smooth and adherent to the cell wall in freeze-substituted samples (Emons, 1987; Lloyd et al., 1987; Ridge, 1988), as opposed to convoluted in conventionally fixed hairs (Newcomb & Bonnett, 1965; Volkmann, 1984). The plasma membrane at the tip of Lepidium sativum root hairs consists of numerous "blisters," sites of vesicle fusion, which are largely free of intramembranous particles in freeze-fractured samples; randomly arranged intramembranous particles are present outside these regions (Volkmann, 1984). Rothberg and Cunningham (1978) reported similar results with freeze-fractured root hairs of Raphanus sativus. Particle rosettes are evident on the protoplasmic fracture face of the plasma membrane at the tip of E. hyemale root hairs (Emons, 1985); these are thought to represent sites of cellulose synthase for cellulose assembly.

The nucleus, usually seen as a spherical or spindle-shaped body in living and conventionally fixed root hairs (Sato et al., 1995), appears to have deep furrows and grooves in freeze-substituted samples (Ridge, 1988). The nucleus migrates into the developing hair and is positioned at some distance from the tip as long as the hair is growing (Meekes, 1985; Sato et al., 1995). In hairs that have completed elongation, the nucleus either assumes a more or less random position (Meekes, 1985) or, frequently, migrates to the base (Sato et al., 1995). In Vicia the position of the nucleus in relation to the tip is controlled by centrally located microtubules; the basipetal migration of the nucleus, however, appears to be controlled by actin microfilaments (Lloyd et al., 1987). In Raphanus sativus (radish) root hairs, microtubules are anchored to the nucleus and appear to be unaffected by either colchicine treatment or cold temperatures which trigger movement of the nucleus to the base (Sato et al., 1995). Although these authors report the presence of actin microfilaments in these hairs, their relationship to nuclear migration was not determined.

Other organelles that are present in the sub-apical region of the root hair - i.e., between the position of the nucleus and the tip (but not in the extreme tip) - in larger numbers than in more basipetal regions are dictyosomes, mitochondria, and cisternae of endoplasmic reticulum (Sievers 1963a; Meekes, 1985; Ridge, 1988). Small vacuoles may also be present in this region (Meekes, 1985), and ribosomes are prevalent in the dome region of root hairs (Ridge, 1988).

Basipetal to the nucleus in developing root hairs, the cytoplasm becomes confined to the periphery of the cell, the central zone being occupied by a large vacuole (Sievers, 1963a, 1963b). The peripheral cytoplasm contains ribosomes, mitochondria, dictyosomes, endoplasmic reticulum, and plastids (Meekes, 1985; Ridge, 1988).

Plasmodesmata are present at the base of root hairs, i.e., between the epidermal cell from which the hair has developed and contiguous epidermal and cortical cells (Harris, 1979). If root hairs take up ions into the symplast, then these plasmodesmatal connections would play a pivotal role in the transport of ions into the cortex (Clarkson, 1985). Some difference has been reported in numbers of plasmodesmata in the base of root hairs compared to the base of adjacent hairless cells, in that the former have more plasmodesmata per surface area (Vakhmistrov & Kurkova, 1979; Vakhmistrov et al., 1981). Recently, Lew (1994), using a voltage clamp technique, has determined that plasmodesmatal connections between the bases of adjacent Arabidopsis thaliana root hairs show similar cell-to-cell electrical coupling characteristics to gap junctions in animal cells.

C. CYTOSKELETON: CELL WALL FORMATION

There is a large volume of literature on the relationship between the cytoskeleton and the orientation of cellulose microfibrils in tip-growing cells, including root hairs, and much of this has been critically reviewed recently (Heath & Seagull, 1982; Lloyd, 1984; Seagull, 1989; Derksen & Emons, 1990; Giddings & Staehelin, 1991). The following, therefore, is a synopsis of the main features of cell wall synthesis in root hairs and the relationship of the cytoskeleton to this process.

Considerable progress has been made in studying the cytoskeleton in tip-growing cells with the development of various fluorescence techniques (Lloyd, 1987) that enable whole mounts to be viewed by conventional epifluorescence microscopy, or more recently by confocal laser scanning microscopy. The cell wall at the tip of growing root hairs consists of randomly arranged cellulose microfibrils, and since elongation occurs only by tip growth, so does the entire primary cell wall (the outer wall layer, or [Alpha] layer) of the tubular portion of root hairs (Belford & Preston, 1961; Newcomb & Bonnett, 1965; Seagull & Heath, 1980a).

Root hairs synthesize an extra wall layer (a secondary cell wall, or [Beta] layer) internal to the primary cell wall beginning basipetal to the immediate tip (Belford & Preston, 1961; Newcomb & Bonnett, 1965). The orientation of the cellulose microfibrils in this layer is species dependent - i.e., species such as radish (Newcomb & Bonnett, 1965; Seagull & Heath, 1980a), Lepidium sativum, and several others (Emons, 1987) have a net axial orientation, whereas species such as onion, Urtica dioica, and Equisetum have a helicoidal pattern (Lloyd, 1983; Emons, 1982, 1987; Emons & Wolters-Arts, 1983; Emons & van Maaren, 1987; van Amstel & Derksen, 1993). There seems to be a correlation between the presence of trichoblasts and the subsequent development of a helicoidal cellulose microfibril pattern in hairs derived from these specialized cells (Emons, 1987). Many aquatic plant species also have root hairs with helicoidal cellulose microfibril orientation (Emons, 1987), but the significance of these two observations is not known.

If there is a congruence between the pattern of microtubules (MTs) and the orientation of cellulose microfibrils as shown for many plant cell types (see Lloyd, 1984), then there should be at least three different MT patterns in root hairs. As early as 1965, Newcomb and Bonnett reported that in radish root hairs MTs had an axial orientation from the base to within 2-3 [[micro]meter] from the tip. This was confirmed by Seagull and Heath (1980a) using serial reconstructions of chemically fixed material. The total number of MTs was shown to increase linearly with distance from the tip up to 25 [[micro]meter] (Seagull & Heath, 1980a). The orientation of MTs remained constant along the length of the hair and was consistent only with cellulose microfibril orientation of the secondary cell wall, i.e., axial orientation (Seagull & Heath, 1980a). Since then, various patterns and positioning of MTs in reference to the extreme tip and the length of root hairs have been reported (see Derksen & Emons, 1990; van Amstel & Derksen, 1993). While a random pattern of organization of MTs occurs in the extreme tip of Equisetum and Limnobium root hairs (Traas et al., 1985; Emons, 1989), helically organized MTs in the tip of Allium and Nigella root hairs have been reported (Lloyd, 1983; Lloyd & Wells, 1985; Traas et al., 1985), and in other species axial orientation seems to be the case. The predominant pattern of MTs along the length of root hairs is axial (Seagull, 1989), even though in many species cellulose microfibrils are helically arranged (Emons, 1987; van Amstel & Derksen, 1993). Allium root hairs are exceptional in that the MTs are arranged in a helical pattern matching the helical arrangement of cellulose microfibrils (Lloyd, 1983; Lloyd & Wells, 1985). The variation in apparent MT orientation pattern can be accommodated if the helical model of MT organization (Lloyd, 1984) is invoked and it is assumed that the helices are able to shift their pitch (Lloyd & Wells, 1985). The lack of congruence between the pattern of MT and cellulose microfibrils in root hairs of many species examined is used as evidence to suggest that MT organization is not involved in cellulose microfibril patterns (Seagull, 1989; van Amstel & Derksen, 1993); but until more is known about the dynamic aspects of MTs during initiation and growth of these cells, this conclusion may be premature. It is evident that there is still much to be learned about microtubule-microfibril relationships in root hairs.

Experiments to release protoplasts by enzyme digestion have yielded indirect evidence that the cell wall at the tip of growing root hairs has different properties than that of the rest of the hair. Treatment of root hairs of Lotus corniculatus with a cocktail of enzymes to digest cell wall components resulted in a release of protoplasts from the tips of root hairs within 30-40 seconds (Rasheed et al., 1990). These protoplasts were totipotent and were able to divide, form callus, and, under the appropriate cultural conditions, initiate plants. Recently, protoplasts have been released from tips of growing Medicago sativa root hairs following laser microsurgery undertaken to perforate the cell wall at the extreme tip (Kurkdjian et al., 1993). Cooper and Brown (1981) have shown that treatment of radish root hairs with cellulose biosynthesis inhibitors caused a decrease in growth rate followed by bursting of the tips. They also noted that, at certain concentrations, Calcofluor White (a fluorochrome specific for cellulose) induced the production of large deposits of what might be callose at the tips of root hairs.

Most of the research emphasis on root hair cell wall composition has been either on the deposition of cellulose microfibrils in relation to the cytoskeleton or simply the organization of the microfibrillar component of root hair walls. In contrast, few studies have dealt with the other components of root hair walls. Most recently, Webster and Stone (1994) have analyzed the chemical constituents of root hair walls of the marine monocotyledonous species Heterozostera tasmanica, and Mort and Grover (1988) analyzed hairs from two monocot and six dicot species. Webster and Stone (1994) localized polyanionic polymers (pectins), proteins, callose, lipids, and [Beta]-glucans in root hair walls by histochemical techniques. They also reported high amounts of apiose and uronic acid as well as lower amounts of glucose and other monosaccharides by chemical analysis. Details of the findings of Mort and Grover (1988) are discussed in the section on Rhizobium interaction with root hairs.

D. CYTOSKELETON: CYTOPLASMIC STREAMING

Root hairs are favourite objects for studying cytoplasmic streaming, for they are easily observed single-cell systems. Also, a series of cells of varying length and age can be assessed simultaneously along the length of a growing root. The velocity of streaming is dependent on the length of the root hair up to a particular stage of development (Sattelmacher et al., 1993) and on the growth rate (Soran and Lazar-Keul, 1978), so that the effect of various treatments on the rate of streaming must be compared using root hairs that are comparable in length or growth rate. One of the difficulties in assessing changes in velocity of streaming stems from the lack of any reliable method to carry this out. Particles of various size move along obvious cytoplasmic tracks that parallel the long axis of the hair; this is less evident at the extreme tip (Seagull & Heath, 1980b; Emons, 1987). Some method of measuring the velocity of these particles is usually used as a representation of the rate of cytoplasmic streaming (Seagull & Heath, 1980b; Emons, 1987; Clarkson et al., 1988; Sattelmacher et al., 1993; Ayling & Butler, 1993; Ayling et al., 1994).

The role of actin-microfilaments in cytoplasmic streaming has been well documented in a number of plant systems (Seagull, 1989), including root hairs (Seagull & Heath 1980b; Emons, 1987). Two populations of microfilaments have been identified in radish root hairs, microfilament bundles located throughout the cytoplasm except at the extreme tip, and single cortical microfilaments associated with microtubules (Seagull & Heath, 1980b). Equisetum root hairs also have two populations of microfilaments (Emons, 1987). Cytochalasin B, a microfilament-disrupting chemical, affects cytoplasmic streaming throughout the length of radish root hairs, but the differential effects on the movement of small and large particles is interpreted to mean that both populations of microfilaments are affected (Seagull & Heath, 1980b).

The regulatory role of [Ca.sup.2+] in cytoplasmic streaming in a variety of plant tissues has been actively researched (Seagull, 1989), including some information available on root hairs. It is evident that a certain minimum concentration of cytosolic [Ca.sup.++] is required for cytoplasmic streaming (Clarkson et al., 1988) and that various exogenously applied substances can affect this level and influence streaming. Calcium channel blockers such as verapamil (Clarkson et al., 1988), and calcium ionophores such as A23187 (Felle et al., 1992) and m-N-ethylmaleimide (Clarkson et al., 1988), have been used in attempts to alter cytosolic [Ca.sup.2+] levels and to relate these changes to cytoplasmic streaming.

Verapamil applied to root hairs of tomato and oilseed rape caused a cessation in cytoplasmic streaming within approximately 4 minutes but a recovery if no additional verapamil was added; addition of m-N-ethylmaleimide caused membrane depolarization and irreversible cessation of cytoplasmic streaming in these same species (Clarkson et al., 1988). Verapamil caused a significant decrease in cytosolic [Ca.sup.2+] levels within 2 minutes of application, although levels recovered to the initial readings; m-N-ethylmaleimide, however, induced a rapid rise in cytosolic [Ca.sup.2+]. Results with verapamil are difficult to explain (Clarkson et al., 1988) if the dogma is correct concerning the effect of elevated levels of [Ca.sup.2+] on decreasing cytoplasmic streaming (Seagull, 1989).

Application of the ionophone A23187 leads to higher levels of cytosolic [Ca.sup.2+] in root hairs and a decrease in cytoplasmic streaming (Felle et al., 1992; Sattelmacher et al., 1993), supporting the concept that higher free cytosolic levels of [Ca.sup.2+] are inhibiting to cytoplasmic streaming.

If [Ca.sup.2+] is involved in cytoplasmic streaming, then other compounds known to influence this process may act through controlling [Ca.sup.++] levels. For example, it has been known for years that auxin affects cytoplasmic streaming in root hairs (Sweeney, 1944); if auxin action is mediated by [Ca.sup.2+], then applications of this growth regulator may alter cytosolic [Ca.sup.2+] levels. In a recent study (Ayling et al., 1994), indole-acetic acid (IAA) at concentrations above [10.sup.-6] M inhibited cytoplasmic streaming in tomato root hairs; concentrations between [10.sup.-7] and [10.sup.-8] M stimulated streaming. These authors found, however, that changes in [Ca.sup.2+] levels were not rapid enough to correlate with the effects of IAA on streaming.

Sattelmacher et al. (1993) studied the effects of ammonium, nitrate, and aluminum on cytoplasmic streaming in wheat (Triticum aestivum) root hairs. Only ammonium had any effect, stopping cytoplasmic streaming if the pH was high.

E. OUTGROWTH OF ROOT HAIRS

Cormack (1962) discussed in detail the possible mechanisms involved in the outgrowth of a root hair papilla and hypothesized that it is a differential hardening of portions of the wall of the initiating cell by [Ca.sup.2+] that accounts for this; he also discussed the considerable opposition to this hypothesis. Curiously, there have been few studies since his review that directly address this phenomenon. As with all plant cells, the driving force for extension growth of root hairs is turgor pressure. Schroter and Sievers (1971) treated Tradescantia flumiensis roots with various concentrations of glucose and monitored changes in root hair development under different turgor pressures. With reduction of turgor pressure in developing root hairs, extension growth was reduced but cell wall synthesis continued, resulting in a thickened cell wall at the apex. As will be discussed in more detail later, Lew (1991) has shown that the net influx of [K.sup.+] ions measured in Arabidopsis thaliana root hairs is sufficient to drive cell expansion by turgor. Weisenseel et al. (1979) have demonstrated the occurrence of electrical fields in Hordeum vulgare root hairs largely due to [H.sup.+]; the current appears to enter the growing root hair tip and, according to the authors, may play a role in maintaining tip growth in these specialized cells.

V. Genetic Control of Root Hair Initiation and Development

The initiation and development of root hairs, like all developmental processes in plants, are ultimately controlled by the plant genome. Until recently, however, there have been few documented cases of gene control of root hair development. Caradus (1979) screened 300 genotypes of Trifolium repens cv. Tamar for root hair length variability and, through selection, was able to obtain lines with significantly longer hairs. He calculated that roots with the longest root hairs were able to explore an increase of 11% of soil volume, suggesting that this may lead to a more efficient root system for nutrient acquisition. Hochmuth et al. (1985) determined by crosses that the "cottony root" (crt) phenotype in tomato, characterized by having a large number of root hairs, had a recessive mode of inheritance.

One of the most important plant species in terms of genetic analysis of various aspects of development is maize (Zea mays) because of the number of well-characterized mutants. Recently, three mutants displaying altered root hair development have been described (Wen & Schnable, 1994). The root hair-defective mutants were isolated from transposon-induced mutations, i.e., selfed mutator transposon stocks. The mutants affect root hair development in various ways: rht 1 results in normal initiation but a lack of elongation; rht 2 shows normal initiation but reduced elongation; rht 3 shows abnormal initiation and elongation. Interestingly, the only mutant that showed a marked decrease in growth in soil was rht 1 (Wen & Schnable, 1994); the authors conclude that under some nutrient conditions, root hairs are less important to plant growth than previously suggested.

Most of the advances in genetic control of root hair development have come, however, from chemically induced mutants of Arabidopsis thaliana. Auxin-resistant mutants (dwf, axr 2) almost completely lack root hairs (Mirza et al., 1984; Wilson et al., 1990; Okada & Shimura, 1992); the axr 2 mutant gene is dominant and located on chromosome 3 (Wilson et al., 1990). When grown on high concentrations of auxin, plants that are homozygous or heterozygous for this gene develop root hairs, indicating that absence of root hairs in the mutant is due to a reduction in auxin sensitivity (Wilson et al., 1990).

The most extensive screening of mutagenized A. thaliana seedlings for root hair mutants has been done by Schiefelbein and Somerville (1990). They identified more than 40 mutants with defects in root hair initiation/development. Four of these were characterized in detail; all were nuclear recessive mutations. The gene product of one of these, RHD 1, is involved in the early stage of root hair initiation, the phenotype of which is a hair with a bulbous base and limited elongation. Gene products of the other three (RHD 2, DHD 3, and RHD 4) are involved in hair elongation. A further discussion of the use of Arabidopsis mutants in studies of root development can be found in Schiefelbein and Benfey (1991).

One of the unique characteristics of root hair development is tip growth, a phenomenon shared with pollen tubes and fungal hyphae. An Arabidopsis mutant, tip 1, has been isolated that shows defects in elongation of both root hairs and pollen tubes (Schiefelbein et al., 1993). Root hairs of the mutant are initiated from the apical end of an epidermal cell as in the wild type, but then the hair remains short and branches, indicating some defect in normal tip growth.

Mutants such as those described will be invaluable in unravelling the complexities involved in the control of root hair development. For example, the role of the cytoskeleton in hair initiation, hair elongation, and wall formation can be explored further with some of these mutants.

VI. Nutrient Acquisition by Root Hairs and Effects of Minerals on Root Hairs

As discussed by Clarkson (1985) the development of cylindrical root hairs theoretically should increase the absorbing surface of the root. In actuality, for ions with low mobility in the soil, such as phosphorus, root hairs may develop into preexisting depletion zones caused by the absorption of these ions by epidermal cells prior to root hair initiation. Thus, unless root hairs are of sufficient length to reach beyond the depletion zone, they may be largely ineffective in absorbing these ions. Root hairs may, however, play an important role in the uptake of more mobile ions, and therefore they have been studied quite extensively in this regard. There has also been some interest in ascertaining the effects that various minerals have on the development and function of root hairs.

A. STUDIES ON ROOT HAIR PLASMA MEMBRANES

Since most ions appear to enter the symplast of root hairs, characteristics of the plasma membrane and the cytoplasm related to uptake have been explored experimentally (e.g., Grabov & Bottger, 1994). Root hairs of some species are large enough to allow microelectrodes (Bertl & Felle, 1985; Ullrich & Novacky, 1990) and micropipettes to be inserted without disturbing either cytoplasmic streaming or cell elongation (Lew, 1991). The latter author demonstrated two ion transport processes in Arabidopsis root hairs using a [K.sup.+] channel blocker and two inhibitors (one direct and one indirect) that interfere with plasma membrane [H.sup.+]-ATPase in combination with double-barreled micropipettes (Lew, 1991). He concluded from his measurements that the net [K.sup.+] influx, controlled by the plasma membrane [H.sup.+]-ATPase pump, was sufficient to drive cellular expansion due to turgor. Samuels et al. (1992) have confirmed the existence of high levels of plasma membrane [H.sup.+]-ATPase in protoderm cells and root hairs of Hordeum vulgare roots by labelling sections with antibodies to [H.sup.+]-ATPase conjugated to FITC. Grabov and Bottger (1994) have shown that redox reactions are not involved directly with [K.sup.+] channels in root hairs.

Microelectrodes have been used effectively to study intracellular pH of root hairs (Bertl & Felle, 1985; Felle, 1987; Ullrich & Novacky, 1990), the effects of fusicoccin on membrane polarization and intracellular pH (Felle, 1982; Bertl & Felle, 1985; Ullrich & Novacky, 1990), and the effect of uptake of various ions on membrane potential and pH (Felle, 1987; Ullrich & Novacky, 1990). The pH of root hair cytoplasm measured in several species is slightly alkaline (7.33-7.5), and this is increased or decreased depending on the ion being absorbed.

B. WATER

It is a natural supposition that root hairs function in absorption of water. Controversies exist, however, concerning this idea (see Cailloux, 1972, for a discussion of early work), and very little experimental data is available in its support, primarily because of the difficulty in working with these small cells (Jones et al., 1983). Cailloux (1972) designed a microchamber including various types of microporometers to measure water uptake and excretion by individual root hairs. He concluded from various experiments that root hairs are capable of either absorbing or excreting water depending on external factors such as [CO.sub.2] concentration and degree of anaerobiosis. He also concluded that only the tip portion of root hairs is normally involved in absorption or excretion of water. Ahmad and Cailloux (1971, 1972), using molinate or other respiratory inhibitors, concluded that water uptake into root hairs is not simply through osmosis; rather, it involves cellular respiration. These experiments are difficult to assess, since condensation onto root hairs may have occurred during the experiments.

Jones et al. (1983) used a miniaturized pressure-probe to compare various parameters of root hairs, hairless epidermal cells, and cortical cells of wheat roots. Root hairs and other epidermal cells had a significantly lower turgor pressure than did cortical cells. The hydraulic conductivity values for the three cell types were not significantly different, however, indicating that the membranes of root hairs are not particularly specialized for water movement.

C. CALCIUM

Considerable early work on the effects of calcium on root hair outgrowth was reviewed by Cormack (1962); the emphasis of this work, particularly that of Cormack, was on the role that calcium may play in combining with cell wall matrix material (pectins) in differentially firming up portions of the epidermal cell wall so that root hair protuberances form.

Several studies since Cormack's review (1962) have determined the effects of calcium on root hair length without considering the mechanism of control. Tanaka and Woods (1972) found a significant reduction in root hair length in Avena sativa (oats) when seedlings were grown in calcium-deficient solutions. Interestingly, atomic absorption determinations demonstrated that calcium was not limiting within the root system itself; the authors attributed the deformation and lack of elongation of root hairs to the toxicity of other ions rather than to a limitation of calcium. Subsequent experiments (Tanaka & Woods, 1973) showed that strontium could not replace calcium in the control of root hair growth and that therefore [Ca.sup.2+] ions are essential. In an experiment to determine the effects of various components of the nutrient solution on root hair growth, Ewens and Leigh (1985) confirmed the importance of calcium in root hair elongation in wheat (Triticum aestivum).

The role that [Ca.sup.2+] plays in root hair growth and functioning has been approached using a variety of techniques. Treating root hairs with the antibiotic chlorotetracycline (CTC) was used to show CTC-[Ca.sup.2+] patterns of fluorescence (Reiss & Hearth, 1979); an intense fluorescence occurred at the root hair tip with fluorescence decreasing toward the base. The authors concluded that this gradient in [Ca.sup.2+] (CTC is known to bind with [Ca.sup.2+] and then show fluorescence) indicates its importance in the phenomenon of tip growth, but the precise mechanism was not discussed. This same antibiotic was used by Sethi and Reporter (1981) to show changes in [Ca.sup.2+] distribution in clover root hairs subsequent to infection by Rhizobium. A proportion of uninfected root hairs had fluorescence confined to the tip, i.e., [Ca.sup.2+] concentrated there; this fluorescence was eliminated by prior treatment with the [Ca.sup.2+] ionophone A23187. Fluorescence was localized near Rhizobium infection threads in some infected cells, suggesting that [Ca.sup.2+] may be essential for the formation and growth of these threads.

Clarkson et al. (1988) used microinjected fura-2 combined with image analysis to determine [Ca.sup.2+] concentrations in root hairs of tomato (Lycopersicon esculentum) and oilseed rape (Brassica napus). Fura-2, when bound to [Ca.sup.2+], shows a fluorescence spectrum shift that allows the ratio of emission fluorescence at the two wavelengths to be used as a measure of [Ca.sup.2+] concentration. Levels at the tip were on the order of 30-90 nM, whereas in the vacuole they were greater than 5/[[micro]molar]. These authors did not consistently see a gradient in [Ca.sup.2+] along the length of root hairs and got variable results when correlating [Ca.sup.2+] concentrations and cytoplasmic streaming.

A calcium-specific vibrating microelectrode was used to determine extracellular [Ca.sup.2+] gradients around growing root hairs of Arabidopsis thaliana (Schiefelbein et al., 1992). A distinct gradient was present at the tip of root hairs but not along the sides. The direction of the gradient at the tip indicated a net influx of [Ca.sup.2+]. This gradient was disrupted when root hairs were treated with the [Ca.sup.2+] channel blocker nifedipine.

D. PHOSPHORUS

Phosphorus (P) is one of the most immobile essential elements for plant growth; therefore, modifications to root systems to increase access to the pool of P in the soil have certainly increased the success of vascular plants (Peterson, 1992). One modification to enhance nutrient uptake made by many vascular plants is the formation of root hairs (but see Clarkson, 1985); a second significant modification is the association with mycorrhizal fungi (Peterson, 1992). It is important to realize that, when divorced from a consideration of mycorrhizal associations, discussions of P uptake by roots are quite artefactual, but a discussion of this is beyond the scope of this review.

Clarkson (1991) claims that root systems with many root hairs would effectively eliminate immobile ions such as P from the epidermal surface; i.e., each root hair would have a depletion zone, and the presence of many root hairs would result in depletion zones overlapping. He also points out that the perimeter of the root interfacing with the soil in terms of ion uptake would be just beyond the tips of the collection of root hairs (Nye, 1966). Several studies have shown the importance of root hairs in P uptake (Lewis & Quirk, 1967; Barley & Rovira, 1970; Bhat & Nye, 1973; Ithoh & Barber, 1983; Misra et al., 1988). In a comparison of P uptake between roots of rape (Brassica napus cv. Wesvoona), a species with numerous, long root hairs, and cotton (Gossypium hirsutum cv. Sicot 3), a species with short root hairs, Misra et al. (1988) showed that P uptake per unit length of root was greatest in rape. Also, most of the 33p taken up was from this zone. Curiously, in an earlier study (Bole, 1973), rape (Brassica napus) was reported to have negligible root hair development but yet a higher P uptake than wheat roots with abundant root hairs.

Various parameters of root hairs, including length, radius, and density, have been shown to be important in affecting P ptake, with length being particularly important (Ithoh & Barber, 1983).

Phosphate concentration can have an effect on root hair development in some species (rape, spinach, tomato) in that low concentrations can lead to more root hairs of greater length as compared to higher P concentrations (Foehse & Jungk, 1983). In wheat, P concentration did not affect root hair length (Ewens & Leigh, 1985).

E. IRON

Plant roots can respond to iron deficiency by changes in several morphological and physiological features, some of these involving root hairs. Iron stress can lead to an increase in the number of root hairs (Kramer et al., 1980; Romheld & Marschner, 1981; Landsberg, 1982, 1986; Romheld, 1987) and to the triggering of wall ingrowth formation within root hairs (Landsberg, 1982, 1986, 1989). The developmental features of wall ingrowths, and the accompanying increase in the number of mitochondria and other organelles observed in root hairs as a response to iron stress, are typical of transfer cells that have been implicated in intensive short-distance transport of solutes (Landsberg, 1986). It is primarily dicotyledonous angiosperms that seem to have the ability to form these transfer cells as a response to iron deficiency (Romheld & Kramer, 1983).

Physiological changes induced in roots by iron deficiency include an increase in proton excretion (Romheld & Kramer, 1983) and therefore acidification of the rhizosphere (De Vos et al., 1986), elevated levels of ferric chelate reductase (Buckhout et al., 1989; Moog et al., 1995), and an increase in the reduction of chelated iron (Bienfait et al., 1982; Chaney & Bell, 1987), much of which occurs in the region of root hairs (Bell et al., 1988). Recently, ferric chelate reductase has been localized on isolated plasma membranes from root cells, and the level of this enzyme was higher under conditions of iron stress (Buckhout et al., 1989).

Since there is a close temporal correlation between an increase in root hair number and the induction of wall ingrowths with one or more of the physiological changes noted above, it has been concluded that they are functionally related (e.g., Landsberg, 1982, 1989). Chaney et al. (1992) concluded, however, that modifications to root structure in tomato are the result of the chlorosis caused by iron stress rather than a result of the stress itself. Recently, an Arabidopsis thaliana root mutant that lacks root hairs and fails to form transfer cells under iron stress nevertheless showed elevated levels of ferric chelate reductase, as reported by Moog et al. (1995). These authors concluded that the appearance of the typical morphological features of roots during iron stress are coincidental to the physiological changes but are not necessary.

F. POTASSIUM

Considerable research has been published on potassium ([K.sup.+]) uptake into roots, but much less on uptake into root hairs specifically (see section VI.A). Gassmann and Schroeder (1994) have used protoplasts from root hairs of wheat combined with the patch-clamp technique to study [K.sup.+] uptake. These authors showed that inward-rectifying [K.sup.+] channels provide a mechanism by which [K.sup.+] ions can be taken up by root hairs at physiological concentrations of [K.sup.+]. These channels might act to regulate membrane potential.

G. SODIUM CHLORIDE

The effects of NaCl on root hairs have been studied in terms of salt stress, i.e., elevated levels of NaCl provided to root systems. In Atriplex hastata, a halophyte (Kramer et al., 1978), and Nicotiana tabacum, a non-halophyte (Tyerman et al., 1989), the response to high-saline conditions was the same in that "bladder-like" root hairs developed. In Medicago sativa, levels of 70-100 mol [m.sup.-3] NaCl reduced the number of root hairs initiated and the number of root hairs infected by Rhizobium (Lakshmi-Kumari et al., 1974). In Vicia faba, NaCl at 50 and 100 mol [m.sup.-3] again reduced the number of root hairs infected by Rhizobium (Zahran & Sprent, 1986).

VII. Secretion and Soil Sheath Formation

Roots exude or secrete a number of soluble and insoluble compounds that interact with microorganisms and soil particles to help establish the rhizosphere. The literature dealing with the rhizosphere is voluminous and will not be reviewed; only that work considering the roles of root hairs in exudation and secretion and as an integral part of the soil sheath (rhizosheath) will be discussed.

Mucilage layers in general on root surfaces have been studied extensively (e.g., Greaves & Darbyshire, 1972; Oades, 1978; Foster, 1982) as has the secretion from root cap cells (see Rougier, 1981; Rougier & Chaboud, 1985). Less is known about secretions from epidermal cells (see Abeysekera & McCully, 1993) and even less about secretions from root hairs. Although mucilage or thin films of unidentified material have been reported on root hairs (Dart, 1971; Greaves & Darbyshire, 1972; Sprent, 1975; Prin & Rougier, 1986), these papers do not show that the materials originated from the root hairs. In a detailed study of the epidermis of maize root tips, Abeysekera and McCully (1993) described a complex three-layered boundary at the surface of meristematic epidermal cells: the helicoidal primary cell wall (L1), an adjacent inner fibrillar layer (L2), and an outer fibrillar layer (L3). These authors suggest that the term "pellicle" be used to designate the two layers (L2, L3) that are structurally and chemically distinct from each other and from the cell wall. Interestingly, neither layer of the pellicle extended over the root hair surface; in fact, the authors report that the root hair papillae breached the pellicle and the original primary cell wall. Root hairs were nor followed during later stages of development to determine if they subsequently secrete a surface covering. In a cryo-SEM study of Lepidium sativum roots, root hair papillae were also shown to break through epidermal mucilage (Sargent, 1986). Root hairs of various Sorghum species (Werker & Kislev, 1978) and apple (Head, 1964) produce viscous droplets. In Sorghum, the droplets stained for pectin but not for proteins or lipids (Werker & Kislev, 1978). Although these authors attempted to determine the origin of this exudate ultrastructurally, the fixation of the material was not of a quality to make decisive conclusions; freeze-substituted tissue would be best for this material.

Many plant species form soil sheaths (rhizosheaths) over portions of their root system (Wullstein & Pratt, 1981; Vermeer & McCully, 1982; McCully & Canny, 1985, 1988; Duell & Peacock, 1985; Huang et al., 1993). These sheaths consist of aggregations of soil particles, mucilage, isolated cells (most likely sloughed root cap cells), root hairs (Vermeer & McCully, 1982), and microorganisms. Root hairs may occur in large numbers (Wullstein & Pratt, 1981) and remain alive within the sheath (Vermeer & McCully, 1982). Although soil particles adhere to the root hairs (Wullstein & Pratt, 1981; Buckley, 1982; Vermeer & McCully, 1982; Fyson et al., 1988; Huang et al., 1993), the large quantity of mucilage present within the sheath, at least in Zea mays, has histochemical features of root cap mucilage and not epidermal cell mucilage (Vermeer & McCully, 1982). Mucilage could be identified on the surface of epidermal cells within the soil sheath, but it did not extend beyond the base of root hairs. McCully and Canny (1985) suggest that the mucilage of maize soil sheaths could have its origin from both root cap cells and associated bacteria. Various microorganisms, including nitrogen-fixing bacteria, have been reported within and on soil sheaths (Wullstein et al., 1979; Wullstein, 1980; Wullstein & Pratt, 1981; Gochnauer et al., 1989) so that secretions from these may certainly contribute to the mucilage within the sheath. Indeed, other authors (Wullstein & Pratt, 1981) suggest various sources of soil sheath mucilage, including root cap cells, epidermal cells, root hairs, and microorganisms. Whatever the source of the mucilage, the root hairs appear to be instrumental in anchoring the soil sheath to the root and in this regard the hairs are often gnarled and sometimes branched (McCully & Canny, 1985). There is no direct evidence yet that the root hairs within the sheath are involved in the uptake of water and ions; this needs to be determined. Soil sheaths have been shown to reduce water loss from roots as soil dries (Huang et al., 1993) and to provide a unique syndrome of root characteristics compared to adjacent bare regions (McCully & Canny, 1985). Root sheaths are thicker and more compact when they form in dry soils than when they form in wet soils (Watt et al., 1994). The methods used in this latter paper lend themselves to studies designed to determine the role of root hairs in sheath formation.

VIII. Interactions with Growth Regulators

As might be expected, various growth regulators have been studied in terms of their effects on the development and functioning of root hairs, often in the broader context of their influence on root development in general. A few studies have dealt with the effects of growth regulators on root hairs specifically.

Ethylene, known to accumulate in anaerobic or partially anaerobic soils, has various effects on root development, including a stimulation of root hair production (Smith & Robertson, 1971; Crossett & Campbell, 1975). Roots can respond to water stress in a number of ways, including the development of short, bulbous root hairs (Schnall & Quatrano, 1992). In Arabidopsis thaliana this response can be triggered by abscisic acid, a growth regulator known to be involved in a number of stress responses in plants; mutants of this species that are insensitive to abscisic acid fail to develop modified root hairs when treated with this growth regulator (Schnall & Quatrano, 1992).

Kinetin, a growth regulator that is usually inhibitory to root extension, causes an increase in both root hair length and potassium accumulation in root hairs of the aquatic species Trianea bogotensis (Abutalybov & Akhundova, 1982). In radish, on the other hand, kinetin was reported to increase the number of root hairs (Bittner & Buschmann, 1983). The latter study suffers, however, in failing to give details of how root hair number was determined.

As mentioned previously (see section IV.D), auxin has been shown to influence cytoplasmic streaming in root hairs, but this does not seem to be mediated via changes in [Ca.sup.2+] levels. Recently, Tretyn et al. (1991) used microelectrodes to determine the effect of a number of auxins on membrane potential and other physiological parameters of Sinapis alba root hairs. Indole-acetic acid was more effective than a number of synthetic auxins in altering membrane potential. Cytoplasmic [Ca.sup.2+] levels and pH were also altered by auxin, but these stabilized at control levels after auxin removal. A presumptive auxin-binding protein of molecular mass 18.6 Kd has been isolated from the cytosolic fraction of root hairs of Zea mays (Radermacher & Klambt, 1993), but auxin binding to this protein was not shown.

IX. Interactions with Microorganisms

A. RHIZOBIUM

Bacteria of the genus "Rhizobium" (also including Bradyrhizobium and Azorhizobium) establish symbioses exclusively with roots of leguminous plant species, with the exception of one tropical non-leguminous plant genus. The symbiosis that is formed in all cases is characterized by the formation of a root nodule in which fixation of atmospheric nitrogen occurs. With legumes, development of this nodule is regulated initially by the interaction of root hairs with the symbiotic bacteria. The wealth of studies on the Rhizobium-legume symbiosis (see reviews by Brewin, 1991; Hirsch, 1992; Kijne, 1992) has revealed, at least for certain legume species, the important roles the root hairs play in host-specific recognition and compatibility (Diaz et al., 1989) and formation of infection threads (Puhler et al., 1991).

Although generally confined to members of the Leguminoseae, Rhizobium bacteria have also been shown to naturally form symbiotic nitrogen-fixing nodules on non-legumes in the tropical genus Parasponia (Ulmaceae), as reported first by Trinick (1973). Unlike the development of a typical legume root nodule, initiation of this symbiosis does not appear to be dependent on root hair involvement, although root hair responses to inoculation have been observed (Trinick, 1981; Lancelle & Torrey, 1984). The ensuing nodule development in Parasponia also appears to differ from the Rhizobium-legume association (see Ridge et al., 1992, for review). Unfortunately, a thorough understanding of this symbiosis and, specifically, root hair involvement during its establishment is lacking due to the limited distribution and economic importance of the phytobiont.

In view of the differing responses of root hairs of leguminous versus non-leguminous plants to Rhizobium inoculation, the involvement of root hairs in the development of each symbiosis is discussed separately.

B. LEGUMINOUS SPECIES

Prior to physical contact with root hairs, Rhizobium bacteria in the rhizosphere are chemotactically directed by plant-produced flavonoids toward the root, where subsequent interactions may lead to bacterial infection and nodule formation (Caetano-Anoles et al., 1988). This initial chemotactic response may well represent the first of a series of events that determine host specificity of Rhizobium bacteria to legume root hairs rather to those of other plant species (Nap & Bisseling, 1990).

Among legumes there is great variation in both the site and manner of root hair involvement in the infection process. Environmental factors, such as soil moisture, and physical and chemical properties can affect the production and size of the root hairs (Sprent & McInroy, 1984), but plant species is probably the most important determinant. In soybean (Bhuvaneswari et al., 1980) and the tropical legume Macroptilium (Ridge & Rolfe, 1985), it has been determined that the infective zone is an acropetally moving region located between the root tip and the smallest emerging root hairs. In these species, it is the short emergent root hairs that are involved in recognition and infection by the bacteria. In fact, Gloudemans et al. (1989) found the yield of root hair RNA from inoculated pea plants to be consistently 10-40% higher than uninoculated plants, leading them to suggest that the bacteria might stimulate new root hair growth in some manner. In soybean, it is speculated that the developing trichoblast, when less than 5 hours old, is activated by the bacteria in some way to become a subsequent site of infection and nodulation (Bhuvaneswari et al., 1980; Turgeon & Bauer, 1982). Indeed, in both soybean and Macroptilium, root hairs older than 5 hours after initiation from the root meristem are no longer infectible. This is in contrast with root hairs of clover, which not only developed tip infections in the emergent root hair zone but also formed lateral or basal infections (initiated at a branch near the base of a hair) in the mature, fully elongated root hair zone (Callaham & Torrey, 1981). In the legume species Arachis (Chandler, 1978) and Stylosanthes (Chandler et al., 1982), root hairs developed at the junction of lateral roots, and colonization by Rhizobia occurred only in these regions. With these particular legume species, Rhizobia-induced root hair curling, although sometimes present, did not lead to bacterial penetration and infection-thread formation. Rather, bacteria penetrated intercellularly between the base of root hair cells and adjacent epidermal and cortical cells to trigger nodule formation. It is of interest that the cytoplasm of these root hairs was more cytoplasmically active than adjacent uncolonized root hair cells, and the walls occasionally had internal projections of wall thickening (Chandler, 1978; Chandler et al., 1982). Since Rhizobia seem capable of invading primary wall material only (Sprent & de Faria, 1988), this variability in the site and manner of the infection process among legume species may be due, in part, to differences in secondary cell wall deposition and the location of differentiated plasma membranes (Volkmann, 1984).

Following attraction to a susceptible root hair, adhesion of the bacteria may occur. This attachment is hypothesized by Smit et al. (1989) as following a two-step model. In the first step, a protein termed "rhicadhesin" (a [Ca.sup.2+]-binding adhesin) on the surface of the bacteria mediates the attachment of single rhizobial cells to the surface of the root hair tip. This initial binding appears to be non-specific, since the bacteria are able to attach to root hair tips of dicotyledonous and monocotyledonous non-leguminous plants (Smit et al., 1989). Following a single adhesion, bacteria accumulate in a cluster on the root hair tip by adhering to the solitary root hair-bound Rhizobium cell by means of bacterial cellulose fibrils with or without the aid of host plant lectins (Smit et al., 1989). There is evidence that specific recognition by the plant is determined, at least in part, by legume lectins which are localized at the surface of the cytoplasmic membrane on susceptible root hair tips (Lerouge et al., 1990; Franssen et al., 1992), the sites of which are correlated with areas of infection (Dazzo et al., 1978; Law & Strijdom, 1984; Diaz et al., 1986). This role of lectins is supported by the ability of a pea-specific Rhizobium strain to effectively nodulate white clover, following introduction of the pea lectin gene into clover roots (Diaz et al., 1989).

During this attachment stage, root hairs must provide carbon (C) and nitrogen (N) for the endophyte by means of exudates. Given the obligate nature of these bacteria in order to fix nitrogen within the plant, the root may have to provide large amounts of C and N to support growth prior to and during infection (Sprent & Raven, 1985). Included in these exudates are flavonoids that are released by the root. These chemicals act not only as chemo-attractants but also induce Rhizobium nod genes which activate the nod gene products or "Nod factors" (Zaat et al., 1987; Maxwell et al., 1989; Hartwig et al., 1990). These Nod factors, extracellular polysaccharides produced by the bacteria, serve a variety of functions. Of these Nod factors, a family of sulphated lipopolysaccharides (Schultze et al., 1992; Truchet et al., 1991) appears to be involved in Rhizobium-legume specificity as shown by their affinity to bind with root hair lectins (Truchet et al., 1986). Dazzo et al. (1991) treated root hairs with bacterial lipopolysaccharide (LPS) and subsequently observed, using immunolocalization, LPS that was bound not only to the external surface of the root hair but also within and on its inner surface. The mechanism by which these LPSs are transferred across the root hair wall and plasma membrane has yet to be elucidated. Not surprisingly, cultural conditions have been found to markedly influence root hair attachment and infection by Rhizobia (Callaham & Torrey, 1981). In specific, limitation of C caused the bacteria to lose their infective capabilities (Smit et al., 1987) whereas manganese-limited bacteria were still infective (Kijne et al., 1988).

Following inoculation and bacterial adhesion, changes in root hair morphology, including curling, branching, swelling, twining, and all manners of deformation have been observed (Higashi & Abe, 1980; Callaham & Torrey, 1981). Computer simulation of root hair curling suggests that the attached bacterial cell induces localized growth in the root hair, causing a curling or branching at the region of attachment (van Batenburg et al., 1986). Corroborating this, Wood and Newcomb (1989) observed that root hair deformation, curling, and branching occurred opposite the position of the root hair nucleus that corresponded directly to bacterial flocs on the root hair surface. It was suggested by Kijne (1992) that only those root hairs that deform in some fashion lead to a successful interaction with a bacterial cell - this deformation response being indicative of successful chemical signalling between root and bacterium. It has been proposed that root hair deformation marks the first visible step in the Rhizobium-legume interaction (Krause et al., 1994).

Root hair deformation (Had) and curling (Hac) are accomplished through the activity of "hadulins," which are root hair-specific proteins induced by the same group of Nod factors implicated in binding and host specificity (Krause & Broughton, 1992). Strong evidence supporting the production of such root hair proteins has been provided by Krause et al. (1994) through their construction of a cDNA library for Vicia unguiculata root hairs inoculated with Rhizobium. From this library they have isolated a cDNA clone that corresponded directly to a hadulin, the protein sequence of which had strong homology to a non-specific lipid transfer protein (LTP) of Hordeum vulgare (Jakobsen et al., 1989). The homology to an LTP is of interest because of the functional implications it suggests for this induced hadulin. LTPs, in general, are capable of transferring different kinds of lipids between lysosomes and mitochondria (Arondel & Kader, 1990), leading Krause et al. (1994) to suggest a role for these proteins in membrane biogenesis. These researchers hypothesized that this lipid transfer-like protein (homologous to a hadulin) that is induced in the inoculated root hairs of V. unguiculata is responsible for the synthesis of the new membrane required during root hair curling and in advance of the growing infection thread (Turgeon & Bauer, 1985; Ridge & Rolfe, 1986). Further evidence to support the homology between the lipid transfer-like protein (hadulin) and an LTP is provided by their similar distribution and accumulation; each protein is present or induced in cell walls of epidermal cells (Molina & Garcia-Olmedo, 1993; Krause et al., 1994). Furthermore, the dramatic increase in LTP-like transcripts following root hair inoculation with Rhizobium (Krause et al., 1994) is analogous to the induction of LTP-encoding genes by pathogens (Molina et al., 1993). Not surprisingly, this amplification in LTP-like transcripts corresponded directly with the expression of hadulins (Krause & Broughton, 1992). It is interesting to note that hadulin expression which resulted in root hair deformation and curling could be induced by flavonoid-pretreated rhizobia regardless of the host symbiont relationship; root hair changes, therefore, are dependent on initial Nod factor induction of hadulins rather than later chemical signalling from the root (Scheres et al., 1990; Terouchi & Syono, 1990).

Following deformation of the root hair, bacterial infection is initiated by a localized hydrolysis of the plant cell wall in the region of the curl (Ridge & Rolfe, 1985). The specificity of Rhizobium bacteria for legume root hairs rather than those of other plant species may be due, in part, to the chemical makeup of the legume root hair walls. These walls have been determined to have a low percentage of sugar by weight and a high pectin content. They are specifically high in arabinose, galacturonic acid, galactose, and rhamnose; features that may facilitate bacterial penetration and that have been shown to be strikingly different from other dicotyledonous and monocotyledonous species that do not harbor Rhizobia (Mort & Grover, 1988). It is not known whether the enzymes involved in the localized cell wall degradation are of plant or bacterial origin (see review by Sprent, 1989), although it has been hypothesized by Ridge and Rolfe (1985) that host root polygalacturonase may be responsible for the localized depolymerization of hair cell wall microfibrils that occurs adjacent to the attached bacteria (Bauer, 1981; Callaham & Torrey, 1981).

Subsequent to wall dissolution, the plasma membrane of the root hair invaginates, around which a tube-like infection thread is formed (Callaham & Torrey, 1981). Its formation is mediated by a complex series of signals initiating with the plant flavonoids (Nap & Bisseling, 1989; Fisher & Long, 1992). As discussed previously, these flavonoids induce bacterial nod genes, causing a secretion of signal molecules called Nod factors which, in turn, induce a variety of plant molecules including the expression of plant "early" nodulin genes called ENOD genes (Nap & Bisseling, 1989; Fisher & Long, 1992). In infected pea roots, two specific ENOD genes were expressed in root hairs and cortical cells containing a growing infection thread, as shown by in situ hybridization (Scheres et al., 1990). Of these ENOD proteins, one was suggested as being part of the plasma membrane of the infection thread and the other was conjectured to be a component of the additional cell wall formed during infection-thread development (Horvath et al., 1993). Although the mechanism of infection-thread formation still requires a great deal of investigation, it is evident that both ENOD proteins and hadulins are expressed in infected root hairs and are involved in the processes that occur therein.

Studies of the cytoskeleton involvement in infection-thread development (Lloyd et al., 1987; Ridge, 1992) suggest that rhizobia are able to repolarize the microtubules and vesicle delivery system toward the invasion site. Ridge (1992) proposes that tip growth is then carried on "inside-out" from normal hair growth, with maintenance of growth possibly being provided by helically arranged cortical microtubules. This "inverted" tip growth of an infection thread is similar to normal tip growth of a root hair (Wood & Newcomb, 1989) in that the nucleus remains a constant distance from the tip (in this case, of the infection thread) (Ridge, 1992). Lloyd et al. (1987) provided evidence that microtubules may be responsible for "anchoring" the nucleus to the root hair tip, and micro filaments may mediate the basipetal movement of the nucleus during infection-thread growth.

Subsequent to the basipetal growth of an infection thread through a root hair, contact is made with the root cortical cells where distinct morphological changes and cell divisions have been initiated by nodulin genes as a prelude to nodule formation (Kijne, 1992). The adjacent cells are thought to form cytoplasmic bridges, termed "pre-infection thread structures" (Pit) (van Brussel et al., 1992), which allow the transfer of bacteria from cell to cell. Nodule formation then inhibits further infection-thread formation in root hairs (Bauer, 1981).

Researchers are continually exploring the mechanisms, genes, and gene products involved in this very sophisticated symbiotic interaction. Through the use of many Rhizobium mutant strains with altered symbiotic properties, and transgenic legume plants which can express reporter genes coupled to the gene of interest (Pichon et al., 1992), the identification of plant genes involved in these intricate molecular- and cellular-communication mechanisms involving legume root hairs and Rhizobium can continue to be elucidated.

C. NON-LEGUMINOUS SPECIES (ULMACEAE)

The only known naturally occurring non-legume/Rhizobium association is that between Parasponia spp. and Rhizobium or Bradyrhizobium (Trinick & Galbraith, 1980). Despite involving the same bacterial symbiont as that with legumes, the symbiosis is developmentally very different. Root hairs of this plant genus do not appear to be directly involved with the infection process (Lancelle & Torrey, 1984; Bender et al., 1987a, 1987b). Bacterial penetration of the cortex is facilitated by cell divisions in subjacent cortical cells which rupture the overlying epidermal cells, thus providing sites for entry for the ensuing nodulation. The susceptible region of the growing root is that area immediately basipetal of the tip, where bacteria have been observed to attach both directly to the root and to root hairs (Bender et al., 1987a). Unlike the Rhizobium-legume symbiosis, however, there is no subsequent root hair curling or deformation following attachment, suggesting either that Nod factors produced by the bacteria are not capable of expressing root hair proteins involved in curling or that Nod factors are not initially induced by plant flavonoids. Indeed, genetic analysis of this association has revealed that infection, unlike with legumes, does not require the specific genes involved in root hair curling, although there is the necessity that Rhizobium carry a host-specific nodulation region for Parasponia which is not required in legume infection (Bender et al., 1987b; McIver et al., 1989).

Root hair response to inoculation is unique to this association and involves the development of multicellular root hairs (MCRH) that form in clumps surrounded by unicellular hairs. The structure of these MCRHs varies from being swollen and distorted with cell divisions in many planes to elongate and septate with mainly anticlinal divisions (Lancelle & Torrey, 1984). Although these hairs are always associated with subjacent swellings of the outer cortical cells, their presence does not appear to be a prerequisite for root cell proliferation (Bender et al., 1987a).

Intercellular penetration of the swollen root region by Rhizobium appears to be facilitated by a bacterial erosion of the mucilage of the epidermal primary cell walls at the point of attachment (Bender et al., 1987a). Lancelle and Torrey (1984) observed an accumulation of loose fibrillar material in the subdivided cortical cells in areas adjacent to Rhizobia and speculated that an enzymatic loosening of the wall fibrils had occurred, similar to bacteria-mediated wall softening in legume infection of root hairs (Callaham & Torrey, 1981; Ridge & Rolfe, 1985). The lack of any root hair erosion at the location of adhering bacteria on Parasponia (Bender et al., 1987a), suggests that there are structural or chemical barriers that are present or elicited in the root hair wall of this particular genus which prevent such activity.

In view of the features that make this association different from the legume-Rhizobium symbiosis, it suggests a potentially useful system for comparison in the study of such mechanisms as root hair development, deformation, and infection-thread formation during Rhizobium infection.

D. ACTINORHIZA

A number of species of mostly woody dicotyledonous plants form root nodules as a result of a symbiotic association with an actinomycete, Frankia, and are then able to fix nitrogen (Torrey, 1978). Entrance of Frankia into the roots of these species is primarily through modified root hairs but in some species hyphae may penetrate directly into epidermal cells (Miller & Baker, 1986; Liu & Berry, 1991).

Typical changes in infected root hairs induced by Frankia, perhaps in association with various soil bacteria (Knowlton et al., 1980; Berry & Torrey, 1983), include curling and various types of branching (Berry & Torrey, 1983; Burggraaf et al., 1983; Shayra et al., 1987; Mansour & Torrey, 1991). Berry and Torrey (1983) have shown in Alnus rubra that a root hair has to be at a specific: stage of development in order for deformation to occur.

The form that Frankia assumes during root hair entrance has been debatable. Earlier descriptions by Lalonde (1977, 1980) of free-living bacteria being encapsulated to form an "exo-encapsulation thread" have not been confirmed in subsequent work. Instead, Frankia, present in a septate hyphal form in the substrate, makes direct contact with an infectible, deformed root hair (Callaham et al., 1979; Berry & Torrey, 1983; Berry et al., 1986; Mansour & Torrey, 1991) to initiate the infection process. The site of entrance is at a folded region of the root hair (Callaham & Torrey, 1977; Berry & Torrey, 1983; Berry et al., 1986); various changes in the root hair cell wall occur at this site. For example, the primary cell wall shows some disorganization of cellulose microfibrils (Berry et al., 1986), and wall ingrowths reminiscent of those in transfer cells have been described in Myrica gale, Comptonia peregrina (Callaham et al., 1979), and Alnus rubra (Berry et al., 1986; Berry & McCully, 1990); these may be a consistent feature of this association. The role that these ingrowths may play in the infection process has been discussed by Berry et al. (1986); these authors suggest that the loosely organized cell wall of these ingrowths may be more accessible to enzymes produced by Frankia hyphae, thereby allowing easier access to the interior of the hair.

Infected root hairs of Alnus rubra elaborate a multilamellar secondary cell wall, a feature not seen in adjacent uninfected root hairs (Berry et al., 1986). In this same species, callose-containing papillae are deposited in root hairs showing arrested infection by Frankia (Berry & McCully, 1990); callose was not present in the transfer cell-like wall ingrowths.

Subsequent to entry of a deformed root hair by Frankia, the hypha is encapsulated by fibrillar wall material synthesized by the host cell (Lalonde, 1977; Callaham et al., 1979; Berry et al., 1986). Encapsulated hyphae penetrate into a cortical cell through the base of the root hair. There has been less emphasis on the cytological changes in root hairs subsequent to infection. The plasma membrane does elaborate around encapsulated hyphae (Lalonde, 1980); and the cytoplasm of Alnus rubra root hairs has been described as containing numerous ribosomes, mitochondria, Golgi bodies, plastids, and endoplasmic reticulum (Berry et al., 1986). Microtubules were also numerous throughout the cytoplasm of infected root hairs. Tannin deposits are present in infected root hairs of Alnus glutinosa (Lalonde, 1980), and moderately electron-dense deposits occur in infected root hairs of A. rubra (Berry et al., 1986). The nucleus of infected root hairs has been reported to either migrate with the growing hypha or to be positioned at the base of the hair and not associated with the penetrating hypha (Callaham et al., 1979; Berry et al., 1986).

Sequerra et al. (1994) has shown that the fungus Penicillium nodositatum, which induces myconodules on various Alnus species (Valla et al., 1989), enters the root of Alnus incana via root hairs. Interestingly, this fungus triggers changes in the root hairs that are essentially identical to those induced by Frankia.

E. ASSOCIATIVE NITROGEN FIXERS

Inoculation of cereals, grasses (Patriquin et al., 1983), and a few other plant families (Crossman & Hill, 1987; Bashan et al., 1989; Puente & Bashin, 1993) with associative nitrogen-fixing bacteria of the genera Azospirillum (Tarrand et al., 1978; Krieg & Dobereiner, 1986), Enterobacter, and Klebsiella (Haahtela et al., 1990) produces significant changes in plant growth parameters, including root hairs. Unlike Rhizobium, these rhizosphere bacteria do not elicit infection thread or nodule formation but, rather, adhere to root hair and epidermal cell wall surfaces and then simply penetrate intercellularly between cortical cells (see review by Bashan & Levanony, 1990). Root hair involvement is thus less complex than that in a Rhizobium symbiosis although there are significant changes in root hair development and morphology resulting from inoculation. These modifications include an enhancement of root hair density, length, and development (Tien et al., 1979; Kapulnik et al., 1985; Hadas & Okon, 1987; Morgenstern & Okon, 1987), and these are often accompanied by changes in other root parameters such as length of the elongation zone (Kapulnik et al., 1985; Levanony & Bashan, 1989) and number and length of lateral roots (Morgenstern & Okon, 1987; Barbieri et al., 1988); these may, in turn, increase root hair density and numbers.

Root hair growth promotion is correlated with a number of factors including the bacterial strain (Patriquin et al., 1983), inoculum density and age (Hadas & Okon, 1987; Morgenstern & Okon, 1987; Fallik et al., 1988; Bashan et al., 1989), and plant species (Boddey & Dobereiner, 1982; Reynders & Vlassak, 1982; Kapulnik et al., 1983; Sarig et al., 1984). Increased development of root hairs, and branching patterns unique to inoculation, have been obtained using purified hormones isolated from bacterial culture (Jain & Patriquin, 1984; Kapulnik et al., 1985; Harari et al., 1988). Indole-acetic acid is frequently an important component (Van de Geijn & Van Maaren, 1986; Harari et al., 1988; Haahtela et al., 1990).

Inoculation of field-grown wheat and maize seedlings with Azospirillum brasiliense resulted in the development of unique root hair branching patterns termed "tuning-fork deformations," i.e., branches of equal length. These root hair-branching patterns differed from those in uninoculated plants or plants inoculated with non-homologous strains of bacteria, which developed branches of unequal length only at a low frequency (Patriquin et al., 1983). The significant increase in the number and branching of root hairs, and thus the increase in root hair surface area in the inoculated plants, may be related to observations that root hairs seem to be the preferential site for bacterial adhesion (Haahtela et al., 1986). Interestingly, addition of nitrogen to these inoculated plants resulted in higher numbers of tuning-fork deformations (Patriquin et al., 1983); in contrast, root hair development and bacterial adhesion of inoculated pearl millet was suppressed in the presence of nitrogen (Umali-Garcia et al., 1980). Adhesion of enterobacteria to grass roots is proposed by Haahtela et al. (1986) to be facilitated by the presence of fimbriae on the bacterium surface. These bacterium-binding proteins recognize specific carbohydrate receptors which are suggested to be localized in the mucilages exuded from the root hairs and root cap (Haahtela et al., 1986). In grasses, the contrasting appearances of the fibrillar epidermal mucilage versus the granular root hair cell mucilages (Umali-Garcia et al., 1980) may indicate distinct chemical compositions that favor bacterial adhesion to root hairs. The exact mechanism of attachment, however, and its universality among associative [N.sub.2]-fixing bacteria-plant interactions is still unclear.

Azospirillum bacteria have been shown to alter the membrane potential of cells in the elongation zone of soybean and cowpea roots (Bashan et al., 1992; Bashan, 1991), which is the region where root hairs initiate and preferential colonization of bacteria occurs. Concomitantly, there is a proton efflux in this zone that is suggested to be responsible for an increase in the uptake of nitrogenous compounds (Bashan, 1991), a well-documented phenomenon in these bacteria-plant associations (Bashan & Levanony, 1990). Further studies are needed to determine the detailed physiological interactions involving root hairs in this and other bacteria-plant associations.

F. MYCORRHIZAL FUNGI

1. Introduction

The involvement of root hairs in the establishment of various mycorrhizal symbioses has not been studied extensively. From the limited information available, root hairs appear to have a less integral role during a mycorrhizal association than that displayed in the Rhizobium-legume and the Actinorhizal symbioses. At present, only two of the seven mycorrhizal types (see Peterson & Farquhar, 1994, for review) have been identified as having significant root hair involvement; what follows is a review of this information.

2. Ectomycorrhizas

These symbiotic associations that form mainly between woody angiosperms or gymnosperms and basidiomycetous or ascomycetous fungi (Malloch et al., 1980) can have root hair involvement. Although ectomycorrhizal roots characteristically lack root hairs, unlike non-mycorrhizal roots which often have many (Scales & Peterson, 1991), their absence is usually due not to an inhibition of development but rather to their incorporation into the mantle. During mantle development in a variety of gymnosperm and angiosperm ectomycorrhizas, elongating hairs at the base of lateral roots or along the primary root can become partially enveloped by hyphae, which proliferate, change their orientation of growth, and wrap around the base of the hairs (Massicotte et al., 1987, 1989b). In some associations, numerous root hairs initiated just behind the lateral root apex become colonized quickly by hyphae that subsequently branch extensively and cover the root hair entirely (Massicotte et al., 1987). As the mantle develops, root hairs frequently collapse and degrade as they become covered with hyphae (Brown & Sinclair, 1981).

Colonization of Picea mariana roots by the ectomycorrhizal fungus Pisolithus tinctorius includes a mantle-like proliferation of fungal hyphae on developing root hairs (Thomson et al., 1989). The root hair walls tend to invaginate in contact with these hyphae, which display altered morphology typical of Hartig net hyphae (Kottke & Oberwinkler, 1986a, 1986b). This altered fungal growth can occur very early in development of the root hair papillae near the root apex, and subsequent root hair elongation may be associated with fungal hyphal growth in an acropetal direction so that the hair becomes completely surrounded. At this stage, root hairs become vacuolated, growth is often suppressed, and the hairs become incorporated into the mantle (Thomson et al., 1989). In a manner similar to fungal colonization of root epidermal cells (Piche et al., 1983), extracellular fibrils are present between the root hair surface and surrounding hyphae during mantle formation on root hairs (Thomson et al., 1989). Whether these fibrils are of fungal origin and have a role in fungal-root binding (Bonfante-Fasolo et al., 1987; Lei et al., 1991) or are simply root hair exudates to be used as a carbon source by the ectomycorrhizal fungus (Thomson et al., 1989) is not known. To date, this is the only account in the literature of such a unique root hair-fungal interaction.

In some species, there are regions of the root where root hair differentiation has progressed to such an extent that colonization seems to be prevented (Massicotte et al., 1989a; Thomson et al., 1989). This finding has led to the suggestion that there may be a limit to the differentiation of epidermal and root hair cells beyond which hyphal interaction is inhibited. Altered structural and physiological characteristics of the root hairs may be responsible for this phenomenon, but this has yet to be determined.

3. Vesicular-Arbuscular Mycorrhizas (VAM)

Most herbaceous terrestrial plants are in mutualistic association with fungi of the family Endogonaceae (Trappe, 1982). Typically termed "vesicular-arbuscular" mycorrhizas (although some authors now refer to them simply as "arbuscular mycorrhizas"), these symbioses are characterized by inter- and intracellular fungal penetration of plant cortical cells and subsequent formation of fungal vesicles (these are absent with some fungal species) and arbuscules in the root tissues (see review by Peterson & Farquhar, 1994). Unlike symbioses involving ectomycorrhizal fungi in which a network of hyphae elaborates on the root's exterior to form a mantle, VAM fungi penetrate inward immediately following root contact and the formation of appressoria usually on the surface of epidermal cells. Since VAM hyphae only occasionally enter via root hairs no attention has been given to the chemical and physiological interactions between root hairs and VA mycorrhizal fungi.

It was suggested by Hoveler as early as 1892 (see Baylis, 1975) that plant species without root hairs were the most consistently VA-mycorrhizal, a hypothesis that has since been corroborated by a few workers (Metsavainio, 1931; Baylis, 1975). It was suggested that these VA-mycorrhizal plants with few or no root hairs actually obtain greater benefits from the extraradical hyphae than could be provided by the root hairs alone. This is because of the decreased hyphal diameter, which allows entry into smaller soil pores than root hairs allow, thereby utilizing more soil nutrients. In addition, the mutualistic fungus favorably changes the microbial population in the rhizosphere, to increase nutrient availability (Linderman, 1992). Although this correlation between root hair absence and colonization is not understood, it is possible that there is a dependence of the plant upon the fungus for its function as root hairs.

There are many VA-mycorrhizal plants, however, that do have abundant root hairs. In such situations, the extent of root hair and epidermal cell exudation of polysaccharides may play a role in the establishment of the VAM fungus. Although there is great variation among plant species, root hairs have been shown to exude a sizeable quantity of material (Head, 1964) that could provide abundant metabolites for fungi. Indeed, there is a direct correlation between the phosphorus nutrition of the plant, which greatly affects exudation patterns, and VAM fungal colonization. Specifically, an increased availability of phosphorus to the plant decreases root cell-membrane permeability and root exudation of carbohydrates and amino acids (Graham et al., 1981), limiting the resources available to the fungus and thereby inhibiting the association.

Recently, in a study of half-sib families of Medicago sativa inoculated with Glomus versiforme, Lackie et al. (1988) suggested that subtle differences in root hair length and number might be partially responsible for observed variations in percent colonization among the families; those with shorter root hairs had less colonization than did families with longer root hairs. The reason for this was hypothesized to be because plant species that inherently develop fewer or shorter hairs would be at a disadvantage in providing adequate metabolites to harbor the fungus. It was suggested that these particular plants would need to employ other mechanisms for fungal attraction and sustenance in order to attain high levels of VAM colonization similar to that of plants with longer root hairs.

In view of a recent study by Baon et al. (1994), however, this suggested direct correlation between root hair length and number, and mycorrhizal colonization may only be a plant species-specific phenomenon. Baon et al. (1994) compared phosphorus (P) uptake of short versus long root hair-selected (SRH and LRH, respectively) rye plants with VAM colonization. The LRH plants were less responsive to mycorrhizal colonization than were the SRH plants, yet were more efficient in P uptake than VAM-colonized SRH plants. These results do not directly support the hypothesis by Lackie et al. (1988) but, rather, suggest that certain plant species may be capable of establishing an optimum level of VAM colonization correlating with root hair characteristics, in order to facilitate maximum nutrient uptake by the plant.

The differing results from these two studies reflect the vast variation present in VAM associations, and the requirement for further studies employing a range of plant and fungal species and root hair-free plant mutants.

X. General Summary

Root hairs continue to be studied in terms of their structure and function because they show a specialized mode of extension growth (tip growth), are accessible for experimental studies of membrane potential and ion uptake, and are central to the interaction between many soil microorganisms and plants. With the identification of genes involved in the control of some aspects of root hair initiation and outgrowth, it is expected that rapid progress will be made in understanding factors controlling these processes.

XI. Acknowledgments

We thank Lewis Melville for drawing Figure 1 and for reviewing the manuscript and Laurie Winn and Kathi Kennedy Baxter for typing the manuscript. We acknowledge ongoing support from the Natural Sciences and Engineering Research Council of Canada.

XII. Literature Cited

Abeysekera, R. M. & M. E. McCully. 1993. The epidermal surface of the maize root tip. I. Development in normal roots. New Phytol. 125: 413-429.

Abutalybov, M. G. & T. S. Akhundova. 1982. Cytokinin participation in regulation of potassium ion activity in root epidermal cells. Soviet Pl. Physiol. 29: 395-402.

Ahmad, M. & M. Cailloux. 1971. The effects of malonate on absorption of water by root hairs of Avena sativa. Canad. J. Bot. 49: 521-528.

----- & M. Cailloux. 1972. Effects of some respiratory inhibitors on water flux in root hairs of Avena sativa. Canad. J. Bot. 50: 575-579.

Arondel, V. & J. C. Kader. 1990. Lipid transfer in plants. Experientia 46: 579-585.

Avers, C. J. 1963. Fine structure of Phleum root meristem cells. II. Mitotic asymmetry and cellular differentiation. Amer. J. Bot. 50: 140-148.

Ayling, S. M. & R. C. Butler. 1993. Time-series analysis of measurements on living cells illustrated by analysis of particle movement in the cytoplasm of tomato root hairs. Protoplasma 172: 124-131.

-----, C. Brownlee & D. T. Clarkson. 1994. The cytoplasmic streaming response of tomato root hairs to auxin; observations of cytosolic calcium levels. J. Pl. Physiol. 143: 184-188.

Baon, J. B., S. E. Smith & A. M. Alston. 1994. Growth response and phosphorus uptake of rye with long and short root hairs: Interactions with mycorrhizal infection. Pl. & Soil 167: 247-254.

Barbieri, P., A. Bernardi, E. Galli & G. Zanetti. 1988. Effects of inoculation with different strains of Azospirillum brasiliense on wheat roots development. Pages 181-188 in W. Klingmuller (ed.), Azospirillum. IV. Genetics, physiology, ecology. Springer-Verlag, Berlin.

Barley, K. P. & A. D. Rovira. 1970. The influence of root hairs on the uptake of phosphate. Commun. Soil Sci. Pl. An. 1: 287-292.

Bashan, Y. 1991. Changes in membrane potential of intact soybean root elongation zone cells induced by Azospirillum brasiliense. Canad. J. Microbiol. 37: 958-963.

----- & H. Levanony. 1990. Current status of Azospirillum inoculation technology: Azospirillum as a challenge for agriculture. Canad. J. Microbiol. 36: 591-608.

-----, L. Alcarez-Melendez & G. Toledo. 1992. Responses of soybean and cowpea root membranes to inoculation with Azospirillum brasiliense. Symbiosis 13: 217-228.

-----, Y. Ream, H. Levanony & A. Sade. 1989. Nonspecific responses in plant growth, yield, and root colonization of noncereal crop plants to inoculation with Azospirillum brasiliense Cd. Canad. J. Bot. 67: 1317-1324.

Bauer, W. D. 1981. Infection of legumes by rhizobia. Annual Rev. Pl. Physiol. 32: 407-449.

Baylis, G. T. 1975. The magnolioid mycorrhiza and mycotrophy in root systems derived from it. Pages 373-389 in F. E. Sanders, B. Mosse, & P. B. Tinker (eds.), Endomycorrhizas. Academic Press, London.

Belford, D. S. & R. D. Preston. 1961. The structure and growth of root hairs. J. Exp. Bot. 12: 157-168.

Bell, P. F., R. L. Chaney & J. S. Angle. 1988. Staining localization of ferric reduction on roots. J. Pl. Nutr. 11: 1237-1252.

Bender, G. L., M. Nayuda, W. Goydych & B. Rolfe. 1987a. Early infection events in the nodulation of the non-legume Parasponia andersonii by Bradyrhizobium. Pl. Sci. 51: 285-293.

-----, W. Goydych, B. Rolfe & M. Nayuda. 1987b. The role of Rhizobium conserved and host specific nodulation genes in the infection of the non-legume Parasponia andersonii. Molec. Gen. Genet. 210: 299-306.

Berry, A. M. & M. E. McCully, 1990. Callose-containing deposits in relation to root-hair infections of Alnus rubra by Frankia. Canad. J. Bot. 68: 798-802.

----- & J. G. Torrey. 1983. Root hair deformation in the infection process of Alnus rubra. Canad. J. Bot. 61: 2863-2876.

-----, L. McIntyre & M. E. McCully. 1986. Fine structure of root hair infection leading to nodulation in the Frankia-Alnus symbiosis. Canad. J. Bot. 64: 292-305.

Bertl, A. & H. Felle. 1985. Cytoplasmic pH of root hair cells of Sinapis alba recorded by a pH-sensitive microelectrode. Does fusicoccin stimulate the proton pump by cytoplasmic acidification? J. Exp. Bot. 36: 1142-1149.

Bhat, K. K. S. & P. H. Nye. 1973. Diffusion of phosphate to plant roots in soil. I. Quantitative autoradiography of the depletion zone. Pl. & Soil 38: 161-175.

Bhuvaneswari, T. V., B. G. Turgeon & W. D. Bauer. 1980. Early events in the infection of soybean (Glycine max L. Merr) by Rhizobium japonicum. I. Localization of infectible root cells. Pl. Physiol. 66: 1027-1031.

Bienfait, H. F., J. Duivenvoorden & W. Verkerke. 1982. Ferric reduction by roots of chlorotic bean plants: Indication for an enzymatic process. J. Pl. Nutr. 5: 451-456.

Bittner, A. & C. Buschmann. 1983. Uptake and translocation of [K.sup.+], [Ca.sup.2+] and [Mg.sup.2+] by seedlings of Raphanus sativus L. treated with kinetin. Z. Pflanzenphysiol. 109: 181-189.

Boddey, R. M. & J. Dobereiner. 1982. Association of Azospirillum and other diazotrophs with tropical graminae. Pages 28-47 in 12th International Congress on Soil Science symposia papers. Vol. 1, 8-16 Feb. 1982. New Delhi. International Society of Soil Science, Food and Agriculture Organization, Rome.

Bole, J. B. 1973. Influence of root hairs in supplying soil phosphorus to wheat. Canad. J. Soil Sci. 53: 169-175.

Bonfante-Fasolo, P., S. Perroto, B. Testa & A. Faccio. 1987. Ultrastructural localization of cell surface sugar residues in ericoid mycorrhizal fungi by gold-labelled lectins. Protoplasma 139: 25-35.

Bonnett, H. T. Jr. & E. H. Newcomb. 1966. Coated vesicles and other cytoplasmic components of growing root hairs of radish. Protoplasma 62: 59-75.

Brewin, N. 1991. Development of the legume root nodule. Annual Rev. Cell Biol. 7: 191-226.

Brown, A. & W. A. Sinclair. 1981. Colonization and infection of primary roots of Douglas-fir seedlings by the ectomycorrhizal fungus Laccaria laccata. Forest Sci. 27(1): 111-124.

Buckhout, T. J., P. F. Bell, D. G. Luster & R. L. Chaney. 1989. Iron-stress induced redox activity in tomato (Lycopersicum esculentum Mill.) is localized on the plasma membrane. Pl. Physiol. 90: 151-156

Buckley, R. 1982. Sand rhizosheath of an arid zone grass. Pl. & Soil 66: 417-421.

Burggraaf, A. J. P., J. Van der Linden & T. Tak. 1983. Studies on the localization of infectible cells on Alnus glutinosa roots. Pl. & Soil 74: 175-183.

Caetano-Anoles, G., D. K. Crist-Estes & W. D. Bauer. 1988. Chemotaxis of Rhizobium meliloti to the plant flavone luteolin requires functional nodulation genes. J. Bacteriol. 170: 3164.

Cailloux, M. 1972. Metabolism and the absorption of water by root hairs. Canad. J. Bot. 50: 557-573.

Callaham, D. A. & J. G. Torrey. 1977. Prenodule formation and primary nodule development in roots of Comptonia (Myricaceae). Canad. J. Bot. 55: 2306-2318.

----- & -----. 1981. The structural basis for infection of root hairs of Trifolium repens by Rhizobium. Canad. J. Bot. 59: 1647-1664.

-----, W. Newcomb, J. G. Torrey & R. L. Peterson. 1979. Root hair infection in actinomy-cete-induced root nodule initiation in Casuarina, Myrica and Comptonia. Bot. Gaz. 140: 51-59.

Caradus, J. R. 1979. Selection for root hair length in white clover (Trifolium repens L.). Euphytica 28: 489-494

Chandler, M. R. 1978. Some observations on infection of Arachis hypogaea L. by Rhizobium. J. Exp. Bot. 29: 749-755.

-----, A. Date & R. J. Roughly. 1982. Infection and root-nodule development in Stylosanthes species by Rhizobium. J. Exp. Bot. 33: 47-57.

Chaney, R. L. & P. F. Bell. 1987. Complexity of iron nutrition: Lessons for plant-soil interaction research. J. Pl. Nutr. 10: 963-994.

-----, Y. Chen, C. E. Green, M. J. Holden, P. F. Bell, D. G. Luster & J. S. Angle. 1992. Root hairs on chlorotic tomatoes are an effect of chlorosis rather than part of the adaptive Fe-stress response. J. Pl. Nutr. 15: 1857-1875.

Clarkson, D. T. 1985. Factors affecting mineral nutrient acquisition by plants. Annual Rev. Pl. Physiol. 36: 77-115.

-----. 1991. Root structure and sites of ion uptake. Pages 417-453 in Y. Waisel, A. Eshel, & U. Kafkai (eds.), Plant roots, the hidden half. Marcel Dekker, New York.

-----, C. Brownlee & S. M. Ayling. 1988. Cytoplasmic calcium measurements in intact higher plants cells: Results from fluorescence ratio imaging of fura-2. J. Cell Sci. 91:71-80.

Cooper, K. M. & R. M. Brown Jr. 1981. Growth of radish root hairs in substances which alter cellulose synthesis. J. Cell Biol. 91: 146a.

Cormack, R. G. H. 1949. The development of root hairs in angiosperms. Bot. Rev. (Lancaster) 15: 583-612.

-----. 1962. The development of root hairs in angiosperms II. Bot. Rev. (Lancaster) 28: 446-464.

Crossett, A. N. & D. J. Campbell. 1975. The effects of ethylene in the root environment upon the development of barley. Pl. & Soil 42: 453-464.

Crossman, S. M. & W. A. Hill. 1987. Inoculation of sweet potato with Azospirillum. Hort. Sci. 22: 420-422.

Cutter, E.G. & L. J. Feidman. 1970a. Trichoblasts in Hydrocharis I. Origin, differentiation, dimensions and growth. Amer. J. Bot. 57: 190-201.

----- & -----.1970b. Trichoblasts in Hydrocharis II. Nucleic acids, proteins and a consideration of cell growth in relation to endopolyploidy. Amer. J. Bot 57: 202-211.

----- & C.-Y. Hung. 1972. Symmetric and asymmetric mitosis and cytokinesis in the root tip of Hydrocharis rnorsus-ranae L. J. Cell Sci. 11: 723-737.

Czernik, C. A. & C. J. Avers. 1964. Phosphatase activity and cellular differentiation in Phleum root meristem. Amer. J. Bot. 51: 424-431.

Dart, P. J. 1971. Scanning electron microscopy of plant roots. J. Exp. Bot. 22: 163-168.

Dazzo, F., G. Truchet, R. Hollingsworth, E. Hrabak, H. S. Pankratz, S. Philip-Hollingsworth, J. Salzwedel, K. Chapman, L. Appenzeiler, A. Squartini, D. Gerhoid & G. Orgambide. 1991. Rhizobium lipopolysaccharide modulates infection thread development in white clover root hairs. J. Bacteriol. 17: 5371-5384.

-----, W. E. Yanke & W. J. Brill, 1978. Trifoliin: A Rhizobium recognition protein from white clover. Biochim. Biophys. Acta 539: 276-286.

Derksen, J. & A. M. Emons. 1990. Microtubules in tip growth systems. Pages 147-181 in I. B. Heath (ed.), Tip growth in plant and fungal cells. Academic Press, San Diego.

De Vos, C. R., H. J. Lubberding & H. F. Bienfait. 1986. Rhizosphere acidification as a response to iron deficiency in bean plants. Pl. Physiol. 81: 842-846.

Diaz, C., P. van, R. Bakhuizen, G. J. Logman, E. J. Lugtenberg & J. Kijne. 1986. Correlation between infection by Rhizobium leguminosarum and lectin on the surface of Pisum sativum L. roots. Planta 168: 350-359.

-----, L. Melchers, P. Hooykaas, B. Lugtenberg & J. Kijne. 1989. Root lectin as a determinant of host-plant specificity in the Rhizobium-legume symbiosis. Nature 338: 579-581.

Dosier, L. W. & J. L. Riopel. 1977. Differential enzyme activity during trichoblast differentiation in Elodea canadensis. Amer. J. Bot. 64: 1049-1056.

----- & -----. 1978. Origin, development, and growth of differentiating trichoblasts in Elodea canadensis. Amer. J. Bot. 65:813-822.

Duell, R. W. & G. R. Peacock. 1985. Rhizosheaths on mesophytic grasses. Crop Sci. 25: 880-883.

Emons, A.M. C. 1982. Microtubules do not control microfibril orientation in a helicoidal wall. Protoplasma 113: 85-87.

-----. 1985. Plasma-membrane rosettes in root hairs of Equisetum hyemale. Planta 163: 350-359.

-----. 1987. The cytoskeleton and secretory vesicles in root hairs of Equisetum and Limnobium and cytoplasmic streaming in root hairs of Equisetum. Ann. Bot. 60: 625-632.

-----. 1989. Helicoidal microfibril deposition in a tip-growing cell and microtubule alignment during tip morphogenesis: A dry cleaving and freeze substitution study. Canad. J. Bot. 67: 2401-2408.

----- & J. Derksen. 1986. Microfibrils, microtubules and microfilaments of the trichoblast of Equisetum hyemale. Acta Bot, Neerl. 35: 311-320.

----- & N. van Maaren. 1987. Helicoidal cell wall texture in root hairs. Planta 170: 145-151.

----- & A.M. C. Wolters-Arts. 1983. Cortical microtubules and microfibril deposition in the cell wall of root hairs of Equisetum hyernale. Protoplasma 117: 68-81.

Ewens, M. & R. A. Leigh. 1985. The effect of nutrient solution composition on the length of root hairs of wheat (Triticum aestivum L.). J. Exp. Bot. 36: 713-724.

Fallik, E., Y. Okon & M. Fisher. 1988. Growth response of maize roots to Azospirillum inoculation: Effect of soil organic matter content, number of rhizosphere bacteria and timing of inoculation, Soil. Biol. Biochem. 20: 45-49.

Felle, H. 1982. Effects of fusicoccin upon membrane potential, resistance and current-voltage characteristics in root hairs of Sinapis alba. Pl. Sci. Lett. 25: 219-225.

-----. 1987. Proton transport and pH control in Sinapis alba root hairs: A study carried out with double-barrelled pH microelectrodes. J. Exp. Bot. 38: 340-354.

-----, A. Tretyn & G. Wagner. 1992. The role of the plasma-membrane [Ca.sup.2+]-ATPase in [Ca.sup.2+] homeostasis in Sinapis alba root hairs. Planta 188:306-313.

Fisher, R. F. & S. R. Long. 1992. Rhizobium-plant signal exchange. Nature 357: 655-660.

Foehse, D. & A. Jungk. 1983. Influence of phosphate and nitrate supply on root hair formation of rape, spinach and tomato plants. Pl. & Soil 74: 359-368.

Foster, R. C. 1982. The fine structure of epidermal cell mucilages of roots. New Phytol. 91: 727-740.

Franssen, H., I. Vijn, W. Yang & T. Bisseling. 1992. Developmental aspects of the Rhizobium-legume symbiosis. Pl. Molec. Biol. 19: 89-107.

Fyson, A., P. Kerr, J. N. A. Lott & A. Oaks. 1988. The structure of the rhizosphere of maize seedling roots, a cryogenic scanning electron microscopy study. Canad. J. Bot. 66: 2431-2435.

Gassmann, W. & J. I. Schroeder. 1994. Inward-rectifying [K.sup.+] channels in root hairs of wheat. A mechanism for aluminum-sensitive low affinity [K.sup.+] uptake. Pl. Physiol. 105: 1399-1408.

Giddings, T. H. Jr. & L. A. Staehelin. 1991. Microtubule-mediated control of microfibril deposition: A reexamination of the hypothesis. Pages 85-99 in C. W. Lloyd (ed.), The cytoskeletal basis of plant growth and form. Academic Press, London.

Gloudemans, T., T. V. Bhuvaneswari, M. Moerman, T. van Brussel, A. van Kammen & T. Bisseling. 1989. Involvement of Rhizobium leguminosarum nodulation genes in gene expression in pea root hairs. Pl. Molec. Biol. 12: 157-167.

Gochnauer, M. B., M. E. McCully & H. Labbe. 1989. Different populations of bacteria associated with sheathed and bare regions of roots of field-grown maize. Pl. & Soil 144: 107-120.

Grabov, A. & M. Bottger. 1994. Are redox reactions involved in regulation of [K.sup.+] channels in the plasma membrane of Limnobium stoloniferum root hairs? Pl. Physiol. 105: 927-935.

Graham, J. H., R. T. Leonard & J. A. Menge. 1981. Membrane-mediated decrease in root exudation responsible for phosphorus inhibition of vesicular-arbuscular mycorrhiza formation. Pl. Physiol. 68: 548-552.

Greaves, M.P. & J. F. Darbyshire. 1972. The ultrastructure of the mucilaginous layer on plant roots. Soil Biol. Biochem. 4: 443-449.

Gunning, B. E. S., A. R. Hardham & J. E. Hughes. 1978. Pre-prophase bands of microtubules in all categories of formative and proliferative cell division in Azolla roots. Planta 143: 145-160.

Haahtela, K., T. Laakso & T. Korhonen. 1986. Associative nitrogen fixation by Klebsiella spp.: Adhesion sites and inoculation effects on grass roots. Appl. Environm. Microbiol. 52: 1074-1079.

-----, R. Ronkko, T. Laakso & P. Williams. 1990. Root-associated Enterobacter and Klebsiella in Poa pratensis: Characterization of an iron-scavenging system and a substance stimulating root hair production. Molec. Pl.-Microbe Interact. 3(6): 358-365.

Hadas, R. & Y. Okon. 1987. Effect of Azospirillum brasiliense inoculation on root morphology and respiration in tomato seedlings. Biol. Fertil. Soils 5: 241-247.

Harari, A., J. Kigel & Y. Okon. 1988. Involvement of IAA in the interactions between Azospirillum brasiliense and Panicum miliaceum roots. Pl. & Soil 110: 275-282.

Harris, W. M. 1979. Ultrastructural observations on the trichoblasts of Equisetum. Amer. J. Bot. 66: 673-684.

Hartwig, U. A., C. A. Maxwell. C. M. Joseph & D. A. Phillips. 1990. Chrysoeriol and luteolin released from alfalfa seeds induce nod genes in Rhizobium meliloti. Pl. Physiol. 92: 116-122.

Head, G. C. 1964. A study of "exudation" from the root hairs of apple by time-lapse cine-photomicrography. Ann. Bot. 28: 495-498.

Heath, I. B. 1990. Tip growth in plant and fungal cells. Academic Press, San Diego.

----- & R. W. Seagull. 1982. Oriented cellulose fibrils and the cytoskeleton: A critical comparison of models. Pages 163-182 in C. W. Lloyd (ed.), The cytoskeleton in plant growth and development. Academic Press, London.

Higashi, S. & M. Abe. 1980. Scanning electron microscopy of Rhizobium trifolii infection sites on root hairs of white clover. Appl. Environm. Microbiol. 40(6): 1094-1099.

Hirsch, A. 1992. Developmental biology of legume nodulation. New Phytol. 122: 211-237.

Hochmuth, G. J., W. H. Gabelman & G. C. Gerloff. 1985. A gene affecting tomato root morphology. Hort. Science 20: 1099-1101.

Hofer, R.-M. 1991. Root hairs. Pages 129-148 in Y. Waisel, A. Eshel, U. Kafkai (eds.), Plant roots, the hidden half. Marcel Dekker, New York.

Horvath, B., R. Heidstra, M. Lados, M. Moerman, H. Spaink, J-C. Prome, A. van Kammen & T. Bisseling. 1993. Lipo-oligosaccharides of Rhizobium induce infection-related early nodulin gene expression in pea root hairs. Pl. J. 4: 727-733.

Huang, B., G. B. North & P.S. Nobel. 1993. Soil sheaths, photosynthate distribution to roots, and rhizosphere water relations for Opuntia ficus-indica. Intl. J. Pl. Sci. 154: 425-431.

Itoh, S. & S. A. Barber. 1983. A numerical solution of whole plant nutrient uptake for soil-root systems with root hairs. Pl. & Soil 70: 403-413.

Jain, D. K., & D. G. Patriquin. 1984. Root hair deformation, bacterial attachment, and plant growth in wheat-Azospirillum associations. Canad. J. Microbiol. 48: 1208-1213.

Jakobsen, K., S. Klemsdal, R. B. Aalen, M. Bosnes, D. Alexander & O. Olsen. 1989. Barley aleurone cell development: Molecular cloning of aleurone-specific cDNAs from immature grains. Pl. Molec. Biol. 12: 285-293.

Jones, H., A.D. Tomos, R. A. Leigh & R. G. Wyn Jones. 1983. Water relation parameters of epidermal and cortical cells in the primary root of Triticum aestivum L. Planta 158: 230-236.

Kapulnik, Y., Y. Okon & Y. Henis. 1985. Changes in root morphology of wheat caused by Azospirillum inoculation. Canad. J. Microbiol. 31: 881-887.

-----, S. Sarig, I, Nur & Y. Okon. 1983. Effect of Azospirillum inoculation on yield on field-grown wheat. Canad. J. Microbiol. 29: 895-899.

Kawata, S. & W. Chung. 1979. On the formation of root hairs in crown roots of rice plants especially on the fine structure of epidermal cells. Jap. J. Crop Sci. 48: 115-122.

Kijne, J. W. 1992 The Rhizobium infection process. Pages 349-398 in G. Stacey, R. H. Burris & H. J. Evans (eds.), Biological nitrogen fixation. Chapman and Hall, New York.

-----, G. Smit, C. L. Diaz & B. J. Lugtenberg. 1988. Lectin-enhanced accumulation of manganese-limited Rhizobium leguminosarum cells on pea root hair tips. J. Bacteriol. 170: 2994-3000.

Knowlton, S., A. Berry & J. G. Torrey. 1980. Evidence that associated soil bacteria may influence root hair infection of actinorhizal plants by Frankia. Canad. J. Microbiol. 26: 971-977.

Kottke, I. & F. Oberwinkler. 1986a. Mycorrhiza of forest trees - structure and function. Trees 1986: 1-24.

----- & -----. 1986b. Root-fungus interactions observed on the initial stages of mantle formation and Hartig net establishment in mycorrhizas of Amanita muscaria on Picea abies in pure culture. Canad. J. Bot. 64: 2348-2354.

Kramer, D., W. P. Anderson & J. Preston. 1978. Transfer cells in the root epidermis of Atriplex hastata L. as a response to salinity: A comparative cytological and x-ray microprobe investigation. Austral. J. Pl. Physiol. 5: 739-747.

-----, V. Romheld, E. Landsberg & H. Marschner. 1980. Induction of transfer-cell formation by iron deficiency in the root epidermis of Helianthus annuus L. Planta 147: 335-339.

Krause, A. & W. J. Broughton. 1992. Proteins associated with root-hair deformation and nodule initiation in Vigna unguiculata. Molec. Pl.-Microbe Interact. 5(1): 96-103.

-----, C. Sigrist, I. Dehning, H. Sommer & W. J. Broughton. 1994. Accumulation of transcripts encoding a lipid transfer-like protein during deformation of nodulation-competent Vigna unguiculata root hairs. Molec. Pl.-Microbe Interact. 7(3): 411-418.

Krieg, N. R. & J. Dobereiner. 1986. The genus Azospirilium. Pages 96-104 in N. R. Krieg & J. G. Holt (eds.), Bergey's manual of systematic bacteriology. Vol. 1. Williams and Wilkins, Baltimore.

Kurkdjian, A., G. Leitz, P. Manigautt, A. Harim & K. O. Greulich. 1993. Non-enzymatic access to the plasma membrane of Medicago root hairs by laser microsurgery. J. Cell Sci. 105: 263-268.

Lackie, S. M., S. R. Bowley & R. L. Peterson. 1988. Comparison of colonization among half-sib families of Medicago sativa L. by Glomus versiforme (Daniels & Trappe) Berch. New Phytol. 108: 477-482.

Lakshmi-Kumari, M., C. S. Singh & N. S. Subba-Rao. 1974. Root hair infection and nodulation in lucerne (Medicago sativa) as influenced by salinity and alkalinity. Pl. & Soil 40: 261-268.

Lalonde, M. 1977. Infection process of the Alnus root nodule symbiosis. Pages 569-589 in W. Newton, J. R. Postgate & C. Rodriquez-Barrueco (eds.), Recent developments of nitrogen fixation. Academic Press, London.

-----. 1980. Techniques and observations of the nitrogen-fixing Alnus root nodule symbiosis. Pages 421-434 in N. S. Subba Rao (ed.), Recent advances in biological nitrogen fixation. Edward Arnold, London.

Lancelle, S. & J. G. Torrey. 1984. Early development of Rhizobium-induced root nodules of Parasponia rigida. I. Infection and early nodule initiation. Protoplasma 123: 26-37.

Landsberg, E.-C. 1982. Transfer cell formation in root epidermis: A prerequisite for Fe-efficiency? J. Pl. Nutr. 5: 415-432.

-----. 1986. Function of rhizodermal transfer cells in the Fe stress response mechanisms of Capsicum annuum L. Pl. Physiol. 82: 511-517.

-----. 1989. Proton efflux and transfer cell formation as a response to Fe deficiency of soybean in nutrient solution culture. Pl. & Soil 114: 53-61.

Law, I. & B. Strijdom. 1984. Role of lectins in the specific recognition of Rhizobium by Lotononis bainesii. Pl. Physiol. 74: 779-785.

Leavitt, R. G. 1904. Trichomes of the root in vascular cryptogams and angiosperms. Proc. Boston Soc. Nat. Hist. 31: 273-313.

Lei, J., K. Wong, & Y. Piche. 1991. Extracellular concanavalin A-binding sites during early interactions between Pinus banksiana and two closely related genotypes of the ectomycorrhizal basidiomycete Laccaria bicolor. Mycol. Res. 95: 357-363.

Lerouge, P., P. Roche, C. Faucher, F. Maillet, G. Truchet, J.-C. Prome & J. Denarie. 1990. Symbiotic host-specificity of Rhizobium meliloti is determined by a sulphated and acylated glucosamine oligosaccharide signal. Nature 344: 781-784.

Levanony, H. & Y. Bashan. 1989. Enhancement of cell division in wheat root tips and growth of root elongation zone induced by Azospirillum brasiliense Cd. Canad. J. Bot. 67: 2213-2216.

Lew, R. R. 1991. Electrogenic transport properties of growing Arabidopsis root hairs. The plasma membrane proton pump and potassium channels. Pl. Physiol. 97: 1527-1534.

-----. 1994. Regulation of electrical coupling between Arabidopsis root hairs. Planta 193: 67-73.

Lewis, D. G. & J.P. Quirk. 1967. Phosphate diffusion in soil and uptake by plants. III. [P.sup.31]-movement and uptake by plants as indicated by [P.sup.32]-autoradiography. Pl. & Soil 26: 445-453.

Lewis, R. F. & N. V. Rothwell. 1964. Implications of nucleolar differences in the root epidermis among several grass species. Amer. J. Bot. 51: 1107-1113.

Linderman, R. G. 1992. Vesicular-arbuscular mycorrhizae and soil microbial interactions. Pages 45-70 in G. J. Bethlenfalvay & R. G. Linderman (eds.), Mycorrhizae in sustainable agriculture. Amererican Society of Agronomy.

Liu, Q. & A.M. Berry. 1991. The infection process and nodule initiation in the Frankia-Ceanothus root nodule symbiosis: A structural and histochemical study. Protoplasma 163: 82-92.

Lloyd, C. W. 1983. Helical microtubular arrays in onion root hairs. Nature 305: 311-313.

-----. 1984. Toward a dynamic helical model for the influence of microtubules on wall patterns in plants. Intl. Rev. Cytol. 86: 1-51.

-----. 1987. The plant cytoskeleton: The impact of fluorescence microscopy. Annual Rev. Pl. Physiol. 38: 119-139.

-----. 1991. The cytoskeletal basis of plant growth and form. Academic Press, London.

----- & B. Wells. 1985. Microtubules are at the tips of root hairs and form helical patterns corresponding to inner wall fibrils. J. Cell Sci. 75: 225-238.

-----, K. J. Pearce, D. J. Rawlins, R. W. Ridge & P. J. Shaw. 1987. Endoplasmic microtubules connect the advancing nucleus to the tip of legume root hairs, but F-actin is involved in basipetal migration. Cell Motility & Cytoskel. 8: 27-36.

Lowary, P. A. & C. J. Avers. 1965. Nucleolar variation during differentiation of Phleum root epidermis. Amer. J. Bot. 52: 199-203.

Malloch, D. W., K. A. Pirozynski & P. H. Raven. 1980. Ecological and evolutionary significance of mycorrhizal symbioses in vascular plants (a review). Proc. Natl. Acad. Sci. U.S.A. 77: 2113-2118.

Mansour, S. R. & J. G. Torrey. 1991. Frankia-spores of strain HFPCgI4 as inoculum for seedlings of Casuarina glauca. Canad. J. Bot. 69: 1251-1256.

Massicotte, H. B., R. L. Peterson & L. H. Melville. 1989a. Hartig net structure of ectomycorrhizae synthesized between Laccaria bicolor (Tricholomataceae) and two hosts: Betula alleghaniensis (Betulaceae) and Pinus resinosa (Pinaceae). Amer. J. Bot. 76: 1654-1667.

-----, ----- & -----. 1989b. Ontogeny of Alnus rubra-Alpova diplophloeus ectomycorrhiza. I. Light microscopy and scanning electron microscopy. Canad. J. Bot. 67: 191-200.

-----, L. H. Melville & R. L. Peterson. 1987. Scanning electron microscopy of ectomycorrhizae: Potential and limitations. Scan. Microscop. 1: 1439-1454

Maxwell, C. A., U. A. Hartwig, C. M. Joseph & D. A. Phillips. 1989. A chalcone and two related flavonoids released from alfalfa roots induce nod genes of Rhizobium meliloti. Pl. Physiol. 91: 842-847.

McCully, M. E. & M. J. Canny. 1985. Localization of translocated 14C in roots and root exudates of field-grown maize. Physiol. Pl. 65: 380-392.

----- & -----. 1988. Pathways and processes of water and nutrient movement in roots. Pl. & Soil 111: 159-170.

McIver, J., M. Djordjevic, J. Weinman, G. Bender & B. Rolfe. 1989. Extensions of host range of Rhizobium leguminosarum bv. trifolii caused by point mutations in nodD that result in alterations in regulatory function and recognition of inducer molecules. Molec. Pl.-Microbe Interact. 2: 97-106.

Meekes, H. T. H. M. 1985. Ultrastructure, differentiation and cell wall texture of trichoblasts and root hairs of Ceratopteris thalictroides (L.) Brongn. (Parkeriaceae). Aquatic Bot. 21: 347-362.

Metsavainio, K. 1931. Untersuchungen uber das Wurzelsystem der Moorpflanzen. Ann. Soc. Zool.-Bot. Fenn. Vanamo 1: 1-417.

Miller, I. M. & D. D. Baker. 1986. Nodulation of actinorrhizal plants by Frankia strains capable of both root hair infection and intercellular penetration. Protoplasma 131: 82-91.

Mirza, J. I., G. M. Olsen, T.-H. Iversen & E. P. Maher. 1984. The growth and gravitropic responses of wild-type and auxin-resistant mutants of Arabidopsis thaliana. Physiol. Pl. 60: 516-522.

Misra, R. K., A.M. Alston & A. R. Dexter. 1988. Role of root hairs in phosphorus depletion from a macrostructured soil. Pl. & Soil 107: 11-18.

Molina, A. & F. Garcia-Olmedo. 1993. Developmental and pathogen-induced expression of three barley genes encoding lipid transfer proteins. Pl. J. 4: 983-991.

-----, A. Segura & F. Garcia-Olmedo. 1993. Lipid transfer proteins (nsLTPs) from barley and maize leaves are potent inhibitors of bacterial and fungal plant pathogens. FEBS Lett. 316: 119-122.

Moog, P. R., T. A. W. van der Kooij, W. Bruggemann, J. W. Schiefelbein & P. J. C. Kuiper. 1995. Responses to iron deficiency in Arabidopsis thaliana: The turbo iron reductase does not depend on the formation of root hairs and transfer cells. Planta 195: 505-513.

Morgenstern, E. & Y. Okon. 1987. The effect of Azospirillum brasiliense and auxin on root morphology in seedlings of Sorghum bicolor x Sorghum sudanense. Arid. Soil Res. Rehabil. 1: 115-127.

Mort, A. J. & P. B. Grover. 1988. Characterization of root hair cell walls as potential barriers to the infection of plants by rhizobia. The carbohydrate component. Pl. Physiol. 86: 638-641.

Murry, L. E. & M. L. Christianson. 1987. Phylogenetic comparison of large nuclear DNA contents of differentiated cells in the roots of Equisetum, Tradescantia and Hordeum. Amer. J. Bot. 74: 1772-1778.

-----, -----, S. H. Alfinito & S. J. Garger. 1987. Characterization of the nuclear DNA of Hordeum vulgareroot hairs: Amplification disappears under salt stress. Amer. J. Bot. 74: 1779-1786.

Nap, J.-P. & T. Bisseling. 1989. Nodulin function and nodulin gene regulation in root nodule development. Pages 181-229 in P. M. Gresshoff (ed.), The molecular biology of symbiotic nitrogen fixation. CRC Press, Florida.

----- & -----. 1990. Developmental biology of a plant-prokaryotic symbiosis: The legume root nodule. Science 250: 948-954.

Newcomb, E. H. & H. T. Bonnett Jr. 1965. Cytoplasmic microtubule and wall microfibril orientation in root hairs of radish. J. Cell Biol. 27: 575-589.

Nye, P. H. 1966. The effect of nutrient intensity and buffering power of a soil, and the absorbing power, size and root hairs of a root, on nutrient absorption by diffusion. Pl. & Soil 25: 81-105.

Oades, J. M. 1978. Mucilages at the root surface. J. Soil Sci. 29: 1-16.

Okada, K. & Y. Shimura. 1992. Aspects of recent developments in mutational studies of plant signalling pathways. Cell 70: 369-372.

Patriquin, D. G., J. Dobereiner & D. K. Jain. 1983. Sites and processes of association between diazotrophs and grasses. Canad. J. Microbiol. 29: 900-914.

Peterson, R. L. 1967. Differentiation and maturation of primary tissues in white mustard root tips. Canad. J. Bot. 45: 319-331.

-----. 1992. Adaptations of root structure in relation to biotic and abiotic factors. Canad. J. Bot. 70: 661-675.

----- & M. L. Farquhar. 1994. Mycorrhizas - Integrated development between roots and fungi. Mycologia 86: 311-326.

Piche, Y., R. L Peterson & C. A. Ackerley. 1983. Early development of ectomycorrhizal short roots of pine. Scan. Elec. Microscop. 3: 1467-1474.

Pichon, M., E.-P. Journet, A. Dedieu, F. de Billy, G. Truchet & D. Barker. 1992. Rhizobium meliloti elicits transient expression of the early nodulin gene ENOD12 in the differentiating root epidermis of transgenic alfalfa. Pl. Cell 4: 1199-1211.

Prin, Y. & M. Rougier. 1986. Cytological and histochemical characteristics of the axenic root surface of Alnus glutinosa. Canad. J. Bot. 64: 2216-2226.

Puente, M. & Y. Bashan. 1993. Effect of inoculation with Azospirillum brasiliense strains on the germination and seedling growth of the giant columnar cardon cactus (Pachycereus pringle). Symbiosis 15: 49-60.

Puhler, A., W. Arnold, A. Buendia. Claveria, D. Kapp, M. Keller, K. Nehaus, J. Quant, A. Roxlau & W. Weng. 1991. The role of Rhizobium meliloti exopolysaccharides EPSI and EPSII in the infection process of alfalfa nodules. Pages 189-194 in H. Hennecke & D. P.S. Verma (eds.), Advances in molecular genetics of plant-microbe interactions, vol. 1. Kluwer Academic, Dordrecht.

Radermacher, E. & D. Klambt. 1993. Auxin-dependent growth and auxin-binding proteins in primary roots and root hairs of corn (Zea mays L.). J. Pl. Physiol. 141: 698-703.

Rasheed, J. H., M. K. Al-Mallah, E. C. Cocking & M. R. Davey. 1990. Root hair protoplasts of Lotus corniculatus L. (birdsfoot trefoil) express their totipotency. Pl. Cell Rep. 8: 565-569.

Reiss, H.-D. & W. Hearth. 1979. Calcium gradients in tip growing plant cells visualized by chlorotetracycline fluorescence. Planta 146: 615-621.

Reynders, I. & K. Vlassak. 1982. Use of Azospirillum brasiliense as biofertilizer in intensive wheat cropping. Pl. & Soil 66: 217-223.

Ridge, R. W. 1988. Freeze substitution improves the ultrastructural preservation of legume root hairs. Bot. Mag. Tokyo 101: 427-441.

-----. 1992. A model of legume root hair growth and Rhizobium infection. Symbiosis 14: 359-373.

----- & B. Rolfe. 1985. Rhizobium sp. degradation of legume root hair cell wall at the site of infection thread origin. Appl. Environm. Microbiol. 50: 717-720.

----- & -----. 1986. Sequence of events during the infection of the tropical legume Macroptilium atropurpureum Urb. by the broad-host-range, fast growing Rhizobium ANU240. J. Pl. Physiol. 122: 121-137.

-----, -----, Y. Jing & E. Cocking. 1992. Rhizobium nodulation of non-legumes. Symbiosis 14: 345-357.

Robards, A. W. 1983. General and molecular cytology. Pages 1-17 in K. Esser, K. Kubitzki, M. Runge, E. Schnepf & H. Zeigler (eds.), Progress in botany Vol. 45. Springer-Verlag, Berlin.

Romheld, V. 1987. Existence of two different strategies for the acquisition of iron in higher plants. Pages 353-374 in D. van der Helm, J. B. Neilands & E. Winkelmann (eds.), Iron transport in microbes, plants and animals. Verlag Chemie, Weinhem.

----- & D. Kramer. 1983. Relationship between proton efflux and rhizodermal transfer cells induced by iron deficiency. Z. Pflanzenphysiol. 113: 73-83.

----- & H. Marschner. 1981. Iron deficiency stress induced morphological and physiological changes in root tips of sunflowers. Physiol. Pl. 53: 354-360.

Rothberg, K. G. & W. P. Cunningham. 1978. Membrane transformations in the root hairs of Raphanus sativus. J. Cell Biol. 79: 373a.

Rothwell, N. V. 1964. Nucleolar size differences in the grass root epidermis. Amer. J. Bot. 51: 172-179.

-----. 1966. Evidence for diverse cell types in the apical region of the root epidermis of Panicum virgatum. Amer. J. Bot. 53: 7-11.

Rougier, M. 1981. Secretory activity of the root cap. Pages 542-574 in W. Tanner & F. A. Loewus (eds.), Encyclopedia of plant physiology, new series. Vol. 13B, Plant carbohydrates II. Springer Verlag, Berlin.

----- & A. Chaboud. 1985. Mucilages secreted by roots and their biological function. Israel J. Bot. 34: 129-146.

Row, H. C. & J. R. Reeder. 1957. Root-hair development as evidence of relationships among genera of the Gramineae. Amer. J. Bot. 44: 596-601.

Samuels, A. L., M. Fernando & A.D. M. Glass. 1992. Immunofluorescent localization of plasma membrane [H.sup.+]-ATPase in barley roots and effects of K nutrition. Pl. Physiol. 99: 1509-1514.

Sargent, J. A. 1986. Cryo-preservation of roots for scanning electron microscopy. Amer. J. Bot. 58: 183-185.

Sarig, S., Y. Kapulnik, I. Nur & Y. Okon. 1984. Response of non-irrigated Sorghum bicolor to Azospirillum inoculation. Exp. Agric. 20: 59-66.

Sato, S., Y. Ogasawara & S. Sakuragi. 1995. The relationship between growth, nucleus migration and cytoskeleton in root hairs of radish. Pages 69-74 in F. Baluska, M. Ciamporova, O. Galparikova & P. W. Barlow (eds.), Structure and function of roots. Kluwer Academic, Dordrecht.

Sattelmacher, B., I. Heinecke & K. H. Muling. 1993. Influence of minerals on cytoplasmic streaming in root hairs of intact wheat seedlings. Pl. & Soil 155/156: 107-110.

Scales, P. & R. L. Peterson. 1991. Structure of ectomycorrhizae formed by Wilcoxina mikolae var. mikolae with Picea mariana and Betula alleghaniensis. Canad. J. Bot. 69: 2149-2157.

Scheres, B., C. van de Wiel, A. Zalensky, B. Horvath, H. Spaink, H. van Eck, F. Zwartkruis, A.-M. Wolters, T. Gloudemans, A. van Kammen & T. Bisseling. 1990. The ENDO12 gene product is involved in the preinfection process during the pea-Rhizobium interaction. Cell 60: 281-294.

Schiefelbein, J. W. & P. N. Benfey. 1991. The development of plant roots: New approaches to underground problems. Pl. Cell 3: 1147-1154.

----- & C. Somerville. 1990. Genetic control of root hair development in Arabidopsis thaliana. Pl. Cell 2: 235-242.

-----, M. Galway, J. Masucci & S. Ford. 1993. Pollen tube and root-hair tip growth is disrupted in a mutant of Arabidopsis thaliana. Pl. Physiol. 103: 979-985.

-----, A. Shipley & P. Rowse. 1992. Calcium influx at the tip of growing root-hair cells of Arabidopsis thaliana. Planta 187: 455-459.

Schnall, J. A. & R. S. Quatrano. 1992. Abscisic acid elicits the water-stress response in root hairs of Arabidopsis thaliana. Pl. Physiol. 100: 216-218.

Schroter, K. & A. Sievers. 1971. Wirkung der Turgorreduktion auf den Golgi-Apparat und die Bildung der Zellwand bei Wurzelhaaren. Protoplasma 72: 203-211.

Schultze, M., B. Quiclet-Sire, E. Kondorosi, H. Virelizier, J. Glushka, G. Endre, S. Gero & A. Kondorosi. 1992. Rhizobium meliloti produces a family of sulfated lipo-oligosaccharides exhibiting different degrees of plant host specificity. Proc. Natl. Acad. Sci. 89: 192-196.

Seagull, R. W. 1989. The plant cytoskeleton. C. R. C. Crit. Rev. Pl. Sci. 8: 131-167.

----- & I. B. Heath. 1980a. The organization of conical microtubule arrays in the radish root hair. Protoplasma 103: 205-229.

----- & -----. 1980b. The differential effects of cytochalasin B on microfilament populations and cytoplasmic streaming. Protoplasma 103: 231-240.

Sequerra, J., A. Capellano, M. Faure-Raynard & A. Moiroud. 1994. Root hair infection process and myconodule formation on Alnus incana by Penicillium nodositatum. Canad. J. Bot. 72: 955-962.

Sethi, R. S. & M. Reporter. 1981. Calcium localization pattern in clover root hair cells associated with infection process: Studies with aureomycin. Protoplasma 105: 321-325.

Sharaya, L. S., S. N. Novik, A. N. Pariiskaya & L. V. Kalakutskii. 1987. Specific deformation of alder root hairs upon interaction with infectious Frankia cultures. Microbiology 56: 110-114.

Sievers, A. 1963a. Beteiligung des Golgi-Apparates bei der Bildung der Zellwand von Wurzelhaaren. Protoplasma 56: 188-192.

-----. 1963b. Uber die Feinstruktur des Plasmas wachender Wurzelhaare. Z. Naturforsch. 18: 830-836.

----- & E. Schnepf. 1981. Morphogenesis and polarity of tubular cells with tip growth. Pages 265-299 in O. Kiermayer (ed.), Cytomorphogenesis in plants. Springer-Verlag, Vienna.

Smit, G., J. Kijne & B. Lugtenberg. 1987. Involvement of both cellulose fibrils and a [Ca.sup.2t]-dependent adhesin in the attachment of Rhizobium leguminosarum to pea root hairs tips. J. Bacteriol. 169: 4294-4301.

-----, T. Logman, M. Boerrigter, J. Kijne & B. Lugtenberg. 1989. Purification and partial characterization of the Rhizobium leguminosarum biovar viciae Ca2+-dependent adhesin, which mediates the first step in attachment of cells of the family Rhizobiaceae to plant root hair tips. J. Bacteriol. 171(7): 4054-4062.

Smith, K. A. & P. D. Robertson. 1971. Effect of ethylene on root extension of cereals. Nature 234: 148-149.

Soran, V. & G. Lazar-Keul. 1978. Relationship between cell growth and rate of protoplasmic streaming. Cytologia 43: 265-271.

Sprent, J. I. 1975. Adherence of sand particles to soybean roots under water stress. New Phytol. 74: 461-463.

-----. 1989. Which steps are essential for the formation of functional legume nodules? New Phytol. 111: 129-153.

----- & S. M. de Faria. 1988. Mechanisms of infection of plants by nitrogen fixing organisms. Pl. & Soil 110: 157-165.

----- & S. G. McInroy. 1984. Effects of salinity on growth and nodulation of Arachis hypogaea. Page 546 in C. Veeger & W. E. Newton (eds.), Advances in nitrogen fixation research. Martinus Nijhoff/Junk, Pudoc.

----- & J. A. Raven. 1985. Evolution of nitrogen-fixing symbioses. Proc. Roy. Soc. Edinburgh 85B: 215-237.

Sweeney, B. M. 1944. The effect of auxin on protoplasmic streaming in root hairs of Avena. Amer. J. Bot. 31: 78-80.

Tanaka, Y. & F. W. Woods. 1972. Root and root hair growth in relation to supply and internal mobility of calicum Bot. Gaz. 133: 29-34.

----- & ----- 1973. Root and root hair growth of oats: Replaceability of calcium. Canad. J. Bot. 51: 1655-1659.

Tarrand, J. J., N. R. Krieg & J. Dobereiner. 1978. A taxonomic study of the Spirillium lipoferum group, with descriptions of a new genus, Azospirillum gen. nov. and two species, Azospirillum lipoferum (Beijerink) comb. nov. and Azospirillum brasiliense sp. nov. Canad. J. Microbiol. 24: 967-980.

Terouchi, N. & K. Syono. 1990. Hair curling induced in heterologous legumes and monocots by flavonoid-treated rhizobia. Pl. Cell Physiol. 31: 113-118.

Thomson, J., L. H. Melville & R. L. Peterson. 1989. Interaction between the ectomycorrhizal fungus Pisolithus tinctorius and root hairs of Picea mariana (Pinaceae). Amer. J. Bot. 76: 632-636.

Tien, T., M. H. Gaskins & D. H. Hubbell. 1979. Plant growth substances produced by Azospirillum brasiliense and their effect on the growth of pearl millet (Pennisetum americanum L.). Appd. Environm. Microbiol. 37: 1016-1024.

Torrey, J. G. 1978. Nitrogen fixation by actinomycete-nodulated angiosperms. BioScience 28: 586-592.

Traas, J. A., P. Braat, A. M. C. Emons, H. Meekes & J. Derksen. 1985. Microtubules in root hairs. J. Cell Sci. 76: 303-320.

Trappe, J. M. 1982. Synoptic keys to the genera and species of zygomycetous mycorrhizal fungi. Phytopathology 72: 1102-1108.

Tretyn, A., G. Wagner & H. H. Felle. 1991. Signal transduction in Sinapis alba root hairs: Auxins as external messengers. J. Pl. Physiol. 139: 187-193.

Trinick, M. J. 1973. Symbiosis between Rhizobium and the nonlegume, Trema aspera. Nature 244: 459-460.

-----. 1981. The non-legume-Rhizobium association. Page 255 in A. Gibson & W. Newton (eds.), Current perspectives in nitrogen fixation. Australian Academy of Science, Canberra.

----- & J. Galbraith. 1980. The Rhizobium requirements of the non-legume Parasponia in relation to the cross-inoculation group concept of legumes. New Phytol. 86: 17-26.

Truchet, G. L., J. E. Sherwood, H. S. Pankrantz & F. B. Dazzo. 1986. Clover root exudate contains a particulate form of the lectin, trifolin A, which binds Rhizobium trifolii. Physiol. Pl. 66: 575-582.

-----, P. Roche, P. Lerouge, J. Vasse, S. Camut, F. de Billy, J-C. Prome & J. Dinarie. 1991. Sulphated lipo-polysaccharide signals of Rhizobium meliloti elicit root nodule organogenesis in alfalfa. Nature 351: 670-675.

Turgeon, B. G. & W. D. Bauer. 1982. Early events in the infection of the soybean by Rhizobium japonicum. Time course cytology of the initial infection process. Canad. J. Bot. 60: 152-161.

----- & -----. 1985. Ultrastructure of infection-thread development during the infection of soybean by Rhizobium japonicum. Planta 163: 328-349.

Tyerman, S. D., P. Oats, J. Gibbs, M. Dracup & H. Greenway. 1989. Turgor volume regulation and cellular water relations of Nicotiana tabacum roots grown in high salinities. Austral. J. Pl. Physiol. 16: 517-531.

Ullrich, C. I. & A. J. Novacky. 1990. Extra- and intracellular pH and membrane potential changes induced by [K.sup.+], Cl, [H.sub.2]P[O.sub.4-], and N[O.sub.3-] uptake and fusicoccin in root hairs of Limnobium stoloniferum. Pl. Physiol. 94: 1561-1567.

Umali-Garcia, M., D. Hubbell, M. Gaskins & F. Dazzo. 1980. Association of Azospirillum with grass roots. Appl. Environm. Microbiol. 39: 219-226.

Vakhmistrov, D. B. & E. B. Kurkova. 1979. Symplastic connections in the rhizodermis of Trianea bogotensis. Soviet Pl. Physiol. 26: 763-771.

-----, ----- & I. F. Zlotnikova. 1981. Symplastic connections and intracellular activity of potassium in the rhizodermis of Raphanus sativus. Soviet Pl. Physiol. 28: 826-833.

Valla, G., A. Capellano, R. Hugueney & A. Moiroud. 1989. A new species inducing myconodules on Alnus roots. Pl. & Soil 114: 142-146.

van Amstel, A. N. M. & J. Derksen. 1993. The complex helical texture of the secondary cell wall of Urtica dioica root hairs is not controlled by microtubules: A quantitative analysis. Acta Bot. Neerl. 42: 141-151.

van Batenburg, F. H., R. Jonker & J. W. Kijne. 1986. Rhizobium induces marked root hair curling by redirection of tip growth: A computer simulation. Physiol. Pl. 66: 476-480.

van Brussel, A., R. Bakhuizen, P. C. van, H. P. Spaink, T. Tak, B. J. Lugtenburg & J. Kijne. 1992. Induction of pre-infection thread structures in the leguminous host plant by mitogenic lipo-oligosaccharides of Rhizobium. Science 257: 70-72.

Van de Geijn, S. & J. M. Van Maaren. 1986. A fast screening method for bacterial isolates producing substances affecting root-growth. Symbiosis 2: 67-75.

Vermeer, J. & M. E. McCully. 1982. The rhizosphere in Zea: New insight into its structure and development. Planta 156: 45-61.

Volkmann, D. 1984. The plasma membrane of growing root hairs is composed of zones of local differentiation. Planta 163: 392-403.

----- & P. Peters. 1995. Structural basis of root hair formation: Early development of trichoblasts and atrichoblasts. Pages 61-67 in F. Baluska, M. Ziamporova, O. Gasparikova & P. W. Barlow (eds.), Structure and function of roots. Kluwer Academic, Dordrecht.

Watt, M., M. E. McCully & M. J. Canny. 1994. Formation and stabilization of rhizosheaths of Zea mays L. Effect of soil water contact. Pl. Physiol. 106: 179-186.

Webster, J. & B. A. Stone. 1994. Isolation, histochemistry and monosaccharide composition of the walls of root hairs from Heterozostera tasmanica (Martens ex Aschers.) den Hartog. Aquatic Bot. 47: 29-37.

Weisenseel, M. H., A. Dorn & L. F. Jaffe. 1979. Natural [H.sup.+] currents traverse growing roots and root hairs of barley (Hordeum vulgare L.). Pl. Physiol. 64: 512-518.

Wen, T.-J. & P.S. Schnable. 1994. Analyses of mutants of three genes that influence root hair development in Zea mays (Gramineae) suggest that root hairs are dispensable. Amer. J. Bot. 81: 833-842.

Werker, E. & M. Kislev. 1978. Mucilage on the root surface and root hairs of Sorghum: Heterogeneity in structure, manner of production and site of accumulation. Ann. Bot. 42: 809-816.

Wilson, A. K., F. B. Pickett, J. C. Turner & M. Estelle. 1990. A dominant mutation in Arabidopsis confers resistance to auxin, ethylene and abscisic acid. Molec. Gen. Genet. 222: 377-383.

Wood, S. & W. Newcomb. 1989. Nodule morphogenesis: The early infection of alfalfa (Medicago sativa) root hairs by Rhizobium meliloti. Canad. J. Bot. 67: 3108-3122.

Wullstein, L. H. 1980. Nitrogen fixation associated with rhizosheaths of Indian ricegrass used in the stabilization of Slick Rock, Colorado, tailings pile. J. Range Managem. 33: 204-206.

----- & S. A. Pratt. 1981. Scanning electron microscopy of rhizosheaths of Oryzopsis hymenoides. Amer. J. Bot. 68: 408-419.

-----, M. L. Bruening & W. B. Bollen. 1979. Nitrogen fixation associated with sand grain root sheaths (rhizosheaths) of certain xeric grasses. Physiol. Pl. 46: 1-4.

Zaar, K. 1979. Peroxidase activity in root hairs of cress (Lepidium sativum L.). Cytochemical localization and radioactive labelling of wall bound peroxidase. Protoplasma 99: 263-274.

Zaat, S., A. van Brussel, T. Tak, E. Pees & B. Lugtenberg. 1987. Flavonoids induce Rhizobium leguminosarum to produce nodDABC gene-related factors that cause thick, short roots and root hair responses on common vetch. J. Bacteriol. 169(7): 3383-3391.

Zahran, H. H. & J. I. Sprent. 1986. Effects of sodium chloride and polyethylene glycol on root-hair infection and nodulation of Vicia faba L. plants by Rhizobium leguminosarum. Planta 167: 303-309.
COPYRIGHT 1996 New York Botanical Garden
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1996 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Peterson, R. Larry; Farquhar, Melissa L.
Publication:The Botanical Review
Date:Jan 1, 1996
Words:20958
Previous Article:The cytological and genetical mechanisms of plant domestication exemplified by four crop models.
Next Article:Strategies for 'minimal growth maintenance' of cell cultures: a perspective on management for extended duration experimentation in the microgravity...
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters