ABSTRACT
London Resin (LR) White is a commonly used resin for embedding specimens to be used for immuno- and/or cytochemical studies. In some instances, due to either the properties of the specimen or the availability of various reagents and equipment, it becomes necessary and/or more convenient to polymerize LR White using heat rather than chemical accelerators or UV light. It is known, however, that heat can reduce or even eliminate the anti genicity of the tissue being embedded. It is therefore desirable to polymerize specimens at the lowest temperature possible and to remove the specimens from the oven as soon as polymerization is complete. We have developed a technique that provides a visual marker that allows the exothermic polymerization of LR White to be monitored, thus minimizing the amount of time a specimen must stay in the oven while excluding oxygen from capsules of polymerizing LR White.
Subject(s)
Capsules/chemistry , Gelatin/chemistry , Hot Temperature , Immunohistochemistry/methods , Plastics/chemistry , Resins, Plant/chemistry , Tissue Embedding/methods , Arabidopsis/ultrastructure , Chloroplasts/ultrastructure , Gossypium/ultrastructure , Impatiens/ultrastructure , Plant Leaves/ultrastructure , Plant Roots/ultrastructure , Polyethylenes/chemistry , Polymers , Polypropylenes/chemistryABSTRACT
The tendrils of Virginia creeper (Parthenocissus quinquefolia) do not coil around their supports. Rather, they adhere to supporting objects by flattening against the support surface and secreting an adhesive compound which firmly glues the tendril to the support. In this study, microscopic and immunocytochemical techniques were utilized to determine the nature of this adhesive. Following touch stimulation, epidermal cells of the tendril elongate toward the support substrate, becoming papillate in morphology. Following contact with the support surface, an adhesive is produced at the base of the papillate cells. The adhesive appears as a highly heterogeneous, raftlike structure and consists of pectinaceous, rhamnogalacturonan (RG) I-reactive components surrounding a callosic core. In addition, more mobile components, composed of arabinogalactans and mucilaginous pectins, intercalate both the support and the tendril, penetrating the tendril to the proximal ends of the papillate cells. Following adherence to the support, the anticlinal walls of the papillate cells are devoid of RG I side-chain reactivity, indicating that extensive debranching of RG I molecules has taken place. Furthermore, a large amount of RG I backbone reactivity was observed in the contact area. These results may indicate that the debranched RG I molecules diffuse into and permeate the contact region, forming an integral part of the adhesive compound. These results indicate that Virginia creeper adheres to objects by a composite adhesive structure consisting of debranched RG I, callose, and other, less-well characterized mucilaginous pectins and that this structure subsequently becomes lignified and very weather-resistant upon the ultimate senescence of the tendril.
Subject(s)
Plant Structures/chemistry , Vitaceae/chemistry , Adhesiveness , Cell Wall/metabolism , Cell Wall/ultrastructure , Epitopes , Immunohistochemistry , Plant Structures/cytology , Plant Structures/ultrastructure , Vitaceae/cytology , Vitaceae/ultrastructureABSTRACT
Ginkgo biloba and the cycads are the only extant seed plants with motile sperm cells. However, there has been no immunocytochemical characterization of these gametes to determine if they share characteristics with the flagellated sperm found in bryophytes and pteridophytes or might give clues as to the relationships to nonflagellated sperm in all other seed plants. To determine characteristics of proteins associated with the motility apparatus in these motile sperm, we probed thin sections of developing spermatogenous cells of Ginkgo biloba with antibodies to acetylated and tyrosinated tubulin and monoclonal antibodies that recognize mammalian centrosomes and centrin. The blepharoplast that occurs as a precursor to the motility apparatus consists of an amorphous core, pitted with cavities containing microtubules and a surface studded with probasal bodies. The probasal bodies and microtubules within the blepharoplast cavities are labeled with antibodies specific to acetylated tubulin. Positive but weak reactions of the blepharoplast core occur with the centrosomereactive antibodies MPM-2 and C-9. Reactions to centrin antibodies are negative at this developmental stage. From this pre-motility apparatus structure, an assemblage of about 1,000 flagella and associated structures arises as the precursor to the motility apparatus for the sperm. The flagellar apparatus consists of a three-layered multilayered structure that subtends a layer of spline microtubules, a zone of amorphous material similar to that in the blepharoplast, and the flagellar band. Centrin antibodies react strongly with the multilayered structure, the transition zone of the flagella, and fibrillar material near the flagellar base at the surface of the amorphous material. Both the spline microtubules and all of the tubules in the flagella react strongly with the antibodies to acetylated tubulin. These localizations are consistent with the localizations of these components in pteridophyte and bryophyte spermatogenous cells, although the blepharoplast material surrounding and connecting flagellar bases does not occur in the seedless (nonseed) land plants. These data indicate that despite the large size of ginkgo gametes and the taxonomic separation between pteridophytes and Ginkgo biloba, similar proteins in gametes of both groups perform similar functions and are therefore homologous among these plants. Moreover, the presence of acetylated tubulin in bands of microtubules may be a characteristic shared with more derived non-flagellated sperm of other conifers and angiosperms.
Subject(s)
Cell Movement , Ginkgo biloba/cytology , Ginkgo biloba/ultrastructure , Flagella/ultrastructure , Ginkgo biloba/embryology , Ginkgo biloba/immunology , Immunohistochemistry , Microtubules/ultrastructure , Plant Proteins/metabolism , Tubulin/metabolismABSTRACT
Dodder (Cuscuta pentagona) hyphae are unique amongst the parasitic weeds for their ability to apparently grow through the walls of the host plant. Closer examination reveals, however, that the hyphae do not grow through the host but rather induce the host to form a new cell wall (or extend the existing wall) to coat the growing hypha. This chimeric wall composed of walls from two species is even traversed by plasmodesmata that connect the two cytoplasms. Compositionally, the chimeric wall is quite different from the walls of either the host or in other cells of the dodder plant, on the basis of immunocytochemical labeling. The most striking differences were in the pectins, with much stronger labeling present in the chimeric wall than in either the host or other dodder walls. Interestingly, labeling with monoclonal antibodies specific to arabinan side chains of rhamnogalacturonan I pectin fraction was highly enriched in the chimeric wall, but antibodies to galactan side chains revealed no labeling. Arabinogalactan protein antibodies labeled the plasma membrane and vesicles at the tips of the hyphae and the complementary host wall, although the JIM8-reactive epitope, associated with very lipophilic arabinogalactan proteins, was found only in dodder cells and not the host. Callose was found in the plasmodesmata and along the forming hyphal wall but was found at low levels in the host wall. The low level of host wall labeling with anticallose indicates that a typical woundlike response was not induced by the dodder. When dodder infects leaf lamina, which have more abundant intercellular spaces than petioles or shoots, the hyphae grew both intra- and extracellularly. In the latter condition, a host wall did not ensheath the parasite and there was clear degradation of the host middle lamellae by the growing hyphae, allowing the dodder to pass between cells. These data indicate that the chimeric walls formed from the growth of the host cell wall in concert with the developing hyphae are unique in composition and structure and represent an induction of a wall type in the host that is not noted in surrounding walls.
Subject(s)
Cuscuta/anatomy & histology , Cuscuta/chemistry , Cell Wall/ultrastructure , Chimera/anatomy & histology , Cuscuta/ultrastructure , Glucans/chemistry , Host-Parasite Interactions , Immunohistochemistry , Impatiens , Plasmodesmata/chemistryABSTRACT
The parasitic weed dodder (Cuscuta pentagona L.) invades a number of potential host species, but the mechanisms responsible for ensuring tight adhesion to the wide variety of host surfaces have yet to be identified. In this study, a battery of microscopy protocols is used to examine the host-parasite interface in an effort to deduce these mechanisms. As the dodder shoot approaches the host tissue, epidermal cells in the parasite shoot elongate and differentiate into secretory type trichomes. The trichome cell walls are malleable, allowing them to elongate towards the host and bend their walls to conform to the shape of the host cell surface. The presence of osmiophilic particles (probable cell-wall-loosening complexes) at far greater numbers than found in other species presages the expansion and malleable nature of the epidermal cells. In addition to the changes in cell shape, the dodder trichome cells secrete an electron-opaque cementing substance that covers the host-parasite interface. When probed with antibodies that recognize cell wall components, the cement reacted only with antibodies that recognize chiefly de-esterified pectins but not other common wall constituents. These data indicate that dodder utilizes both a cementing layer of pectin and a radically modified epidermal cell wall to secure the parasite to the perspective host.
Subject(s)
Cuscuta/ultrastructure , Cell Adhesion , Cuscuta/growth & development , Cuscuta/physiology , Host-Parasite Interactions , Immunohistochemistry , Microscopy, Confocal , Microscopy, Electron , Microscopy, Electron, Scanning , Plants/parasitology , Plastids/ultrastructureABSTRACT
Cotton fibers are often utilized as a model system to investigate cellulose biosynthesis and cell wall elongation. In this study, we grew cotton fibers in vitro, with ovules dissected at day zero post anthesis as the explant source, in the presence of three herbicides that inhibit cellulose biosynthesis. Cultures were sampled for electron microscopy and immunocytochemistry 1-2 days after beginning the treatments. After dichlobenil treatment, the fibers were much shorter than the controls and assumed a variety of abnormal shapes, from shortened versions of the control fiber to nearly spherical. The inner layers of the fiber wall often contained juxtaposed electron-translucent and -transparent areas; this layer reacted strongly with antibodies to callose. Cellulase-gold labeling in these newly developed fibers grown in the presence of dichlobenil was present at only about 3% of the control labeling. After treatment with either isoxaben or flupoxam, the fibers assumed spherical shapes and frequently (more than 60% of fibers) exhibited a new cell plate within the fiber, indicating that cell division had occurred, a process that rarely occurred in the controls. Unlike the dichlobenil-treated fibers, fibers grown in the presence of isoxaben or flupoxam contained an extensive accumulation of chiefly deesterified pectins, replacing the entire wall with an elaborated version of the pectin sheath found in control cotton fibers. These data indicate that all three herbicides are effective disrupters of cellulose biosynthesis and cause radical changes in cell wall structure and composition. Moreover, these data indicate that the composition of the walls may influence indirectly cell cycle kinetics, keeping these fiber cells in a more meristematic mode.
Subject(s)
Cellulose/biosynthesis , Gossypium/drug effects , Gossypium/growth & development , Herbicides/pharmacology , Benzamides/pharmacology , Cell Division/physiology , Cell Wall/chemistry , Cell Wall/drug effects , Cell Wall/ultrastructure , Cellulose/chemistry , Gossypium/metabolism , Gossypium/ultrastructure , Hexuronic Acids/metabolism , Immunohistochemistry , Nitriles/pharmacologyABSTRACT
Leaf and canopy photosynthesis of cotton (Gossypium hirsutum L.) declines as the crop approaches cutout, just as the assimilate needs for reproductive growth are peaking. Our objective with this study was to determine whether this decline is due to remobilization of leaf components to support the reproductive growth or due to some cue from the changing environmental conditions during the growing season. Field studies were conducted in 1995-1996 at Stoneville, Mississippi, using six cotton genotypes and two planting dates (early and late), which produced two distinctly different cotton populations reaching cutout at different times. Among the six genotypes were a photoperiod sensitive line (non-flowering) and its counter part which had photoperiod insensitive genes backcrossed four times to the photoperiod sensitive line (flowering). This pair was used to assess the degree that the photosynthetic decline could be attributed to reproductive sink development. Leaf CO(2)-exchange rate (CER) and chlorophyll (Chl) fluorescence measurements were taken in mid-August, a period corresponding to cutout for the early planted plots, and those leaves were collected. Leaf Chl level, soluble protein level, various soluble carbohydrate levels and Rubisco activities were assayed on those leaves. Averaged across years, leaf CER and soluble protein levels were reduced approximately 14% and 18%, respectively, for the early planted compared to the late planted cotton. Neither leaf Chl levels or Chl fluorescence Fv/Fm values for Photosystem II yield were altered by the planting date. In 1996, leaves from the non-flowering line had 12% greater Chl and 20% greater soluble protein levels than the flowering line. However, in 1996, the CER of the early planted non-flowering line was reduced 10% compared to the late planted. Although remobilization of leaf N to reproductive growth appears to be the principle component causing the cutout photosynthetic decline, the data also indicate that environmental factors can play a small role in causing the decline.
ABSTRACT
Microtubule-organizing centers (MTOCs) are morphologically diverse cellular sites involved in the nucleation and organization of microtubules (MTs). These structures are synonymous with the centrosome in mammalian cells. In most land plant cells, however, no such structures are observed and some have argued that plant cells may not have MTOCs. This review summarizes a number of experimental approaches toward the elucidation of those subcellular sites involved in microtubule nucleation and organization. In lower land plants, structurally well-defined MTOCs are present, such as the blepharoplast, multilayered structure, and polar organizer. In higher plants, much of the nucleation and organization of MTs occurs on the nuclear envelope or other endomembranes, such as the plasmalemma and smooth (tubular) endoplasmic reticulum. In some instances, one endomembrane may serve as a site of nucleation whereas others serve as the site of organization. Structural and motor microtubule-associated proteins also appear to be involved in MT nucleation and organization. Immunochemical evidence indicates that at least several of the proteins found in mammalian centrosomes, gamma-tubulin, centrin, pericentrin, and polypeptides recognized by the monoclonal antibodies MPM-2, 6C6, and C9 also recognize putative lower land plant MTOCs, indicating shared mechanisms of nucleation/organization in plants and animals. The most recent data from tubulin incorporation in vivo, mutants with altered MT organization, and molecular studies indicate the potential of these research tools in investigation of MTOCs in plants.
Subject(s)
Centromere/physiology , Centrosome/physiology , Cytoplasm/physiology , Plant Physiological Phenomena , Cytoskeleton/physiologyABSTRACT
Etched sections of methacrylate infiltrated plant tissue [Gubler (1989) Cell Biol. Int; Rep., 13:137-145; Baskin et al. (1992) Planta, 187:405-413] offer many advantages over the more traditional squash technique of Wick et al. [(1981) J. Cell Biol. 89:685-690] for immunofluorescence microscopic investigation of the plant cytoskeleton, especially during mitosis. These advantages include: (1) unimpeded access of antibody probes, (2) confocal-like imaging without the expense of confocal equipment, (3) maintenance of organ architecture as well as intracellular structure, (4) the ability to independently examine separate focal planes with the same or multiple antibody(s) or other labelling compounds, and (5) the ability to archive unetched sections, polymerized or non-polymerized infiltrated tissue. In this paper examples of staining of various microtubule cytoskeletal and mitotic proteins are shown in a variety of methacrylate embedded plant tissues.
Subject(s)
Methacrylates , Microscopy, Fluorescence/methods , Mitosis , Plant Roots/cytology , Antibodies, Monoclonal , Avena/cytology , Microscopy, Electron , Onions/cytology , Plant Roots/ultrastructure , Tissue EmbeddingABSTRACT
Polyphenol oxidase (PPO) appears to be ubiquitous in higher plants but, as yet, no function has been ascribed to it. Herein, we report on the localization of PPO based upon biochemical fractionation of chloroplast membranes in Vicia faba (broad bean) into various complexes and immunocytochemical electron microscopic investigations. Sucrose density gradient fractionations of thylakoid membranes after detergent solubilization reveals that PPO protein (by reactivity with anti-PPO antibody) and activity (based upon ability to oxidize di-dihydroxyphenylalanine) are found only in fractions enriched in photosystem II (PSII). Furthermore, of the PSII particles isolated using three different protocols utilizing several plant species, all had PPO. Immunogold localization of PPO on thin sections reveals exclusive thylakoid labeling with a distribution pattern consistent with other PSII proteins (80% grana, 20% stroma). These data strongly indicate that PPO is at least peripherally associated with the PSII complex.
ABSTRACT
The subperoxisomal distribution of glycolate oxidase (GO) in leaves and cotyledons of several plants was investigated using post-embedding immunogold labelling. In peroxisomes with amorphous nucleoids, all of the immunolabelling is associated with the matrix of the peroxisome, even in tissue embedded in Lowicryl, a resin that preserves antigenicity best. This same staining pattern was found after cytochemical staining for GO activity with cerium. In peroxisomes with crystalline inclusions, the inclusions are only lightly labelled, compared with the densely-labelled matrix. Cytochemical reactions are noted between the units of the crystal in these peroxisome types. Because cytochemical reactions for catalase are concentrated in the amorphous nucleoid and crystalline peroxisomal inclusions, the general lack of immunogold staining of GO and other peroxisomal proteins indicate that catalase may be the major (or in some cases the exclusive) constituent of these peroxisomal inclusions.
Subject(s)
Alcohol Oxidoreductases/metabolism , Plants/enzymology , Blotting, Western , Immunohistochemistry , Microbodies/enzymology , Glycine max/enzymologyABSTRACT
Mature maize leaf tissue (Zea mays L.) was immunolabeled using a pre-embedding protocol with specific antibodies for nitrate reductase and protein A-colloidal gold. Immunogold label was found exclusively in the cytoplasm of mesophyll cells; no reaction was detected in bundle sheath cells. Chloroplasts, which were sliced open during cryosectioning, had no labeling. Thus, it appears nitrate reductase is localized exclusively in the cytoplasm of maize leaf mesophyll cells. No gold labeling was found on tissue sections embedded in L. R. White's or Lowicryl resin, indicating that previous chloroplast localization utilizing these protocols may be artifactual, resulting from shared epitopes or nonspecific antibody binding.
ABSTRACT
Two immunological approaches were used to determine if ribulose bisphosphate carboxylase oxygenase (RuBisCo) is present in guard cell chloroplasts. Immunocytochemistry on thin plastic sections using tissue samples that were processed using traditional glutaraldehyde/osmium fixation and then restored to antigenicity with metaperiodate treatment, resulted in labeling over wild-type mesophyll and guard cell plastids of several green and white variegated Pelargonium chimeras. The density of immunogold labeling in guard cell chloroplasts was only about one-seventh of that noted in mesophyll chloroplasts on a square micron basis. Because guard cell chloroplasts are much smaller than mesophyll chloroplasts, and occur at lower quantities/cell, the relative differences in RuBisCo concentration between the cell types indicate that guard cells have only 0.48% of the RuBisCo of mesophyll cells. No reaction was noted over 70S ribosomeless plastids of these chimeras even though adjacent green chloroplasts were heavily stained, indicating the high specificity of the reaction for RuBisCo. Spurr's resin gave the most successful colloidal gold labeling in terms of low background staining and structural detail but L. R. White's resin appeared to be superior for antigen retention. In the white leaf edges of the white and green Pelargonium chimeras, the only green, functional chloroplasts are in the guard cells. When either whole tissue or plastid enriched extracts from this white tissue were electrophoresed, blotted, and probed with anti-RuBisCo a large subunit band was detected, identical to that in the green tissue. These data indicate that a low, but detectable, level of RuBisCo is present in guard cell chloroplasts.
ABSTRACT
A dinitroaniline-resistant (R) biotype of Eleusine indica (L.) Gaertner. (goosegrass) is demonstrated to be cross-resistant to a structurally non-related herbicide, amiprophosmethyl, and supersensitive to two other classes of compounds which disrupt mitosis. These characteristics of the R biotype were discovered in a comparative test of the effects of 24 different antimitotic compounds on the R biotype and susceptible (S) wild-type Eleusine. The compounds tested could be classified into three groups based upon their effects on mitosis in root tips of the susceptible (S) biotype. Class I compounds induced effects like the well known mitotic disrupter colchicine: absence of cortical and spindle microtubules, mitosis arrested at prometaphase, and the formation of polymorphic nuclei after arrested mitosis. The R biotype was resistant to treatment with some class I inhibitors (all dinitroaniline herbicides and amiprophosmethyl) but not all (e.g. colchicine, podophyllotoxin, vinblastine, and pronamide). Roots of the R biotype, when treated with either dinitroaniline herbicides or amiprophosmethyl, exhibited no or only small increases in the mitotic index nor were the spindle and cortical microtubules affected. Compounds of class II (carbamate herbicides and griseofulvin) cause misorientation of microtubules which results in multinucleated cells. Compounds of class III (caffeine and structually related alkaloids) cause imcomplete cell walls to form at telophase. Each of these last two classes of compounds affected the R biotype more than the S biotype (supersensitivity). The cross-resistance and high levels of resistance of the R biotype of Eleusine to the dinitroaniline herbicides and the structurally distinct herbicide, amiprophosmethyl, indicate that a mechanism of resistance based upon metabolic modification, translocation, or compartmentation of the herbicides is probably not operative.
ABSTRACT
Maize root tip cells were examined for the distribution of actin microfilaments in various cell types and to determine the effects of microfilament disrupters. Fluorescence microscopy on fixed, stabilized, squashed cells using the F-actin specific probe, rhodamine-labelled phalloidin, allowed for a three-dimensional visualization of actin microfilaments. Microfilaments were observed as long, meandering structures in root cap cells and meristematic cells, while those in immature vascular parenchyma were abundant in the thin band of cytoplasm and were long and less curved. By modifying standard electron microscopic fixation procedures, microfilaments in plant cells could be easily detected in all cell types. Treatment with cytochalasin B, cytochalasin D and lead acetate, compounds that interfere with microfilament related processes, re-organized the microfilaments into abnormal crossed and highly condensed masses. All the treatments affected not only the microfilaments but also the accumulation of secretory vesicles. The vivid demonstration of the effects of all of these microfilament disrupters on the number and size of Golgi vesicles indicates that these vesicles may depend on microfilaments for intracellular movement.
Subject(s)
Actin Cytoskeleton/drug effects , Actins/metabolism , Cytochalasin B/pharmacology , Cytochalasins/pharmacology , Cytoskeleton/drug effects , Organometallic Compounds/pharmacology , Zea mays/ultrastructure , Actin Cytoskeleton/ultrastructure , Cytochalasin D , Microscopy, Fluorescence , Rhodamines , Zea mays/metabolismABSTRACT
Morphometric, electrophoretic, and immunological procedures were used to probe the structural and physiological differences between triazine-resistant (R) and susceptible (S) isolines of canola (Brassica napus L.). The R biotype exhibited increased grana stacking and decreased amounts of starch compared to the S biotype. Likewise, characters associated with an increase in grana stacking (lower chlorophyll a/b ratio, increased chlorophyll a/b light-harvesting complex, and relatively lower amounts of the P700 chlorophyll a protein and chloroplast coupling factor) were all observed in the R isoline of canola. Proteins which occur with approximately equal frequency in grana and stroma lamellae (plastocyanin, cutochrome f) or present only in the stroma (ribulose 1,5-bisphosphate carboxylase/oxygenase) were not quantitatively different in the two biotypes. Gross anatomical parameters (volume of epidermis, palisade mesophyll, spongy mesophyll, and air space) were similar in the two isolines. Thus, the triazine-resistance mutation does not confer a shade-type anatomy despite the chloroplast changes that are characteristic of shade biotypes or shade adaptions. These data indicate that the differences in chloroplast structure noted previously in comparisons of nonisonuclear R and S weed biotypes reflect differences in the triazineresistance factor rather than characters unrelated to triazine resistance.
ABSTRACT
Previous studies in our laboratory have shown that tentoxin prevents the incorporation of polyphenol oxidase (PPO), a nuclearly-coded protein, into the chloroplasts of sensitive species. In this study, we show, by comparison of electrophoretically separated isozymes, that ferredoxin-NADP(+) reductase (FNR) is nuclearly coded in Nicotiana. Electrophoresis of FNR isozymes from tentoxin treated seedlings of a sensitive and a resistant species demonstrated that, unlike PPO, ferredoxin-NADP(+) reductase was unaffected by tentoxin treatment. These data indicate that tentoxin selectively inhibits transport of cytoplasmically synthesized proteins into the chloroplast, and does not produce a generalized disruption of cellular integration.
ABSTRACT
Cytochemical and immunocytochemical methods were used to localize photosystems I and II in barley (Hordeum vulgare L. cv Himalaya) chloroplasts. PSI activity, monitored by diaminobenzidine oxidation, was associated with the lumen side of the thylakoids of both grana and stroma lamellae. The P(700) chlorophyll a protein, the reaction center of PSI, was localized on thin sections of barley chloroplasts using monospecific antibodies to this protein and the peroxidase-antiperoxidase procedure. Results obtained by immunocytochemistry were similar to those of the diaminobenzidine oxidation: both grana and stroma lamellae contained immunocytochemically reactive material. Both the grana and stroma lamellae were also labeled when isolated thylakoids were reacted with the P(700) chlorophyll a protein antiserum and then processed by the peroxidase-antiperoxidase procedure. PSII activity was localized cytochemically by monitoring the photoreduction of thiocarbamyl nitroblue tetrazolium, a reaction sensitive to the PSII inhibitor, DCMU. PSII reactions occurred primarily on the grana lamellae, with weaker reactions on the stroma lamellae.
ABSTRACT
When African violet leaf explants are cultured in vitro, buds and shoots develop directly from the upper leaf surfaces. Three developmentally different African violet chimeras were cultured, and in each case adventitious shoots that developed into plants had the parent chimera pattern. A multicellular origin of the adventitious buds accounts for these results.