Fundamental Concepts in the Embryogenesis of Dicotyledons ... · Embryogenesis represents the critical stage in the develop- ment of a plant at which its basic organization and body

The Plant Cell, Vol. 9, 1903-1919, November 1997 O 1997 American Society of Plant Physiologists

REVIEW ARTICLE

Fundamental Concepts in the Embryogenesis of Dicotyledons: A Morphological lnterpretation of Embryo Mutants

Donald R. Kaplanasl and Todd J. Cookeb aDepartment of Plant and Microbial Biology, University of California, Berkeley, California 94720-31 02 bDepartment of Plant Biology, University of Maryland, College Park, Maryland 20742-581 5

INTRODUCTION

Embryogenesis represents the critical stage in the development of a plant at which its basic organization and body plan have their inception. Therefore, in the effort to under- stand the regulation of plant development, molecular genet- icists have recently focused considerable attention on the search for genes expressed during the embryogenesis of plant model systems (Mayer et al., 1991 ; Meinke, 1991 b; Lindsey and Topping, 1993; West and Harada, 1993; Goldberg et al., 1994). This molecular approach toward plant embryogenesis has been given even greater impetus by the significant advances made in the isolation of embryonic genes in animal model systems, such as in Drosophila (Nüsslein-Volhard and Wieschaus, 1980; Nüsslein-Volhard, 1991) and zebrafish (Kane et al., 1996; Solnica-Krezel et al., 1996). As a consequence of these noteworthy successes in animal embryogenesis, there has been a natural tendency for plant biologists to follow suit and interpret molecular genetic data in terms of animal developmental models (Jürgens et al., 1991 ; Mayer et al., 1991). One problem arising from this parallel strategy toward the study of plant and animal embryogenesis is that plant embryos bear a significantly different relationship to their adult forms than do animal embryos. This situation is further complicated by the history of the field of plant embryology, in which certain perspectives have led to less fundamental, even ambiguous characterizations of embryo morphogenesis.

For example, traditional studies of plant embryogenesis have placed great emphasis on detailed characterizations of cell lineages in the earliest stages of plant embryogenesis (Johansen, 1950; Natesh and Rau, 1984). This kind of emphasis was usually at the expense of the more general features of embryo morphogenesis, such as the time at which the meristems of the shoot and root are first organized and begin to function. We demonstrate that it is the morphology of the embryo that determines the cell lineage patterns rather than vice versa.

'To whom correspondence should be addressed. E-mail kaplandrQ nature.berkeley.edu; fax 51 0-642-4995.

A second point of confusion is that plant embryologists traditionally described embryo morphogenesis by using a unique terminology, which makes this developmental process appear separate from the subsequent development of the plant. Consequently, students and newcomers to the field often obtain the impression that a plant's embryonic phase, like that of animals, represents a distinct and independent stage in the plant's life history.

This article reexamines the traditional concepts of plant embryology because we believe that as a whole, the studies of the past offer little framework for the evaluation of embryo mutants, particularly Arabidopsis embryo mutants. Model systems like Arabidopsis are selected for their ease of ex- perimental manipulation, with the implication that the data obtained can be generalized to all related organisms. How- ever, from the study of a single organism, it is impossible to determine whether the particular features of the structure or process being characterized are unique to that organism or generally applicable to other species.

By contrast, the comparative morphological approach (Hofmeister, 1867, 1868; Troll, 1937, 1939, 1943) establishes broad principles of plant organization through its study of a large number of taxa. Such an approach emphasizes the fundamental processes and structural features common to all plants. We believe that comparative morphology provides a broad interpretive framework for understanding the wild type of a model system and the mutants that deviate from it. Nevertheless, a comparative approach provides little if any insight into the underlying mechanisms operating to regulate a given process. Thus, these two investigative approaches are complementary and must be utilized together to expand our understanding of how genes act to control plant embryo development.

We begin by reexamining some basic concepts in plant embryogenesis from a broad comparative basis to construct a framework for making a more effective synthesis with plant molecular genetics in the interpretation of embryo mutants. In the process, we break away from the concepts provided by animal development, because in contrast to animal

1904 The Plant Cell

embryogenesis, which exhibits a defined beginning (fertilization) and end (pupation/birth), a definite ending cannot be designated for plant embryogenesis. Dormancy, which has been considered an endpoint, occurs arbitrarily during embryo development- its interpolation in the developmental process is correlated with the seed being the unit of dispersal. Thus, plant embryogenesis is not a separate stage in the life history of a plant; it heralds the initiation of an iterative process of meristematic activity that continues throughout the life of the plant.

After a discussion of these topics, we propose a new model for dicotyledon embryogenesis that emphasizes fundamental developmental processes rather than the traditionally employed shape designations. This new model allows the phenotypes of Arabidopsis embryo mutants to be interpreted in ways that are more consistent with the known mechanisms of plant development.

CONCEPTS OF EMBRYOGENESIS

Distinction between Plant and Animal Embryology

In contrast to animals, vascular plants exhibit a regular alter- nation between a haploid gametophytic phase and diploid sporophytic phase. Consequently, plant embryology has been classically defined as the study of the development of not only the new sporophyte generation, termed the embryo, but also of the gametophyte generation and the ante- cedent sporophytic structures of the flower (i.e., stamens and ovules; Maheshwari, 1950). Where the development of the embryo proper has been described, it tends to have been handled in a very descriptive way, usually divorced from the morphogenetic processes manifested in the sporophyte that germinates from the seed (Johansen, 1950; Natesh and Rau, 1984).

The best known description of plant embryogenesis is that for the shepherd’s purse (Capsella bursa-pastoris in the mustard family, Brassicaceae; Hanstein, 1870). Like most dicotyledonous angiosperms, the developing embryo of C. bursa-pastoris proceeds through a series of shape changes, going from filamentous to globular to heart to torpedo and then to later stages of intraseminal (within the seed) development (see below). Certain of these shape changes have been likened to stages in animal embryogenesis. For example, the hollow blastula of most animals and the solid globular embryo of C. bursa-pastoris acquire roughly spherical shapes, and polarized growth is said to generate both the animal gastrula and plant heart-shaped embryo. In addition, although it is not stated explicitly, the delimitation of the peripheral dermatogen at the globular stage of C. bursa-pastoris could be equated with the formation of the outer layer in the animal blastula. Similarly, the definition of the three primary meristematic tissues-protoderm, ground meristem, and procambium-which takes place between the globular

and heart phases of plant embryogenesis, has been likened to the origin of the primary germ layers during gastrulation in animals.

Thus, the use of shape designations not only establishes questionable parallels with animal development but also re- inforces the concept that plant embryogenesis, like animal embryogenesis, is a singular stage that is separate from subsequent development. Furthermore, considering embryo shapes only constructs a misleading distinction between the embryos of the two major groups of flowering plants. For example, the shape-based terms applied to dicotyledonous embryos cannot be used to characterize embryo development in monocotyledons because these embryos initiate only a single cotyledon and consequently do not proceed through a heart stage. As members of the division Magnolio- phyta, both groups might reasonably be expected to employ similar if not identical morphogenetic mechanisms. To make any model of embryogenesis applicable to the widest range of organisms, we must put emphasis on the developmental processes involved and not just arbitrary shape designations. Nevertheless, we continue to use shape terms in this article to avoid unnecessary confusion with descriptions in other papers.

Culmination of Embryogenesis

Dormancy has traditionally been used to define the endpoint of embryogenesis. However, dormancy does not always occur in the developing seed. Moreover, even when it does occur, dormancy can coincide with any phase of embryo development. For example, in the grasses, intraseminal development can be so extensive that the plant in the dormant seed will have severa1 sets of leaf primordia and shoot-borne roots. Conversely, in the orchids, embryo development is ar- rested so early that embryogenesis proceeds only as far as a rudimentary globular stage (Arditti, 1992). In addition, for most orchids, an endosperm is not established. All possible intermediate stages between these two extremes are manifested in angiosperm embryos.

Such variability is undoubtedly directly related to the critical role of the seed as the unit of dispersal. Orchids with their early dormancy produce abundant seeds; an obligatory mycorrhizal symbiont is required for subsequent growth of the seedlings. On the other hand, the grasses and plants that exhibit a delayed onset of dormancy produce seeds with sufficient storage reserves to help ensure a high proba- bility of seedling survival. For those plants in which the seed is not the unit of dispersal, embryo development is continu- o u ~ , and there is no intervening dormancy period that can be designated as the artificial boundary between embryogenesis and later plant development. An example is the viviparous mangroves of the genus Rhizophora (Rhizophoraceae), in which the embryo or seedling is the unit of dispersal; embryo development does not arrest (Sussex, 1975). The seedling germinates while still on the parent plant and falls

Concepts of Angiosperm Embryogenesis 1905

directly into the substrate, either establishing itself in the vi- cinity of the parent plant or floating for some time before taking root (Egler, 1948; Tomlinson, 1986). Similarly, because spores are the unit of dispersa1 in ferns and other pteridophytes, embryonic development in these plants also is continuous, without dormancy or arrest.

These examples demonstrate that seed dormancy is not an obligate stage of plant embryo development. Indeed, seed dormancy is best considered as an alternative physiological state that can be superimposed on the basic developmental program of seed 'plants at varied times. Somatic carrot embryos (Ammirato, 1983) and immature zygotic oat embryos (Triplett and Quatrano, 1982) cultured in vitro proceed to the seedling stage without an intervening dormant period. Adding abscisic acid (ABA) to the culture medium, however, initiates dormancy. Certain viviparous maize mutants, whose kernels germinate while attached to the ear, exhibit decreased sensitivity to ABA, thereby suggesting that the genetic lesions responsible for abnormal vivipary re- side in the signal transduction pathway for ABA (Robichaud et al., 1980). It is likely that in response to certain environ- mental selection pressures or nutritional habits, seed plants have evolved the capability to arrest different stages of embryo development by inserting an ABA-dependent dormancy period, which delays but does not disrupt the overall process of plant development.

Morphological Relationship between the Embryo and the Developing Plant

If dormancy is merely a temporary arrest in the continuous process of embryonic activity, then the architecture of the embryo should be continued in the body of the mature, elaborated plant. That is true. However, because the embryo has tended to be treated in isolation from the subsequent development of the plant, this fact has sometimes been ig- nored. This is not to imply that the plant embryo is a minia- ture of the adult. Rather, the embryo represents the first expression of the basic organography that is iterated throughout the life of the plant. Thus, plant embryo morphogenesis is best understood in reference to the organogenetic processes of postgermination plant development.

For example, the heart shape of a typical dicotyledonous embryo is not dueto an organogenetic process that is unique to embryo development. This shape arises because the first two appendages of the shoot are initiated in an opposite phyllotaxis from the dista1 pole of the embryo. Similarly, pro- phylls arise on a lateral bud in an opposite phyllotaxis (see later section).

Examples exist also for plants other than typically studied dicotyledons. The rooting relationship in ferns and other pteridophytic groups differs significantly from dicotyledons, which have bipolar embryos consisting of shoot and root systems that are directly opposite one another (Troll, 1943; Groff and Kaplan, 1988). Mature dicotyledonous plants also

have distinct root and shoot poles, with secondary roots developing from the primary root (allorhizy) and lateral branches from the shoot. In contrast, the ferns form roots exclusively from the shoot and do not develop a spatialiy separated root system (Figure 1A). This distinctive pattern of rooting is expressed in the unipolar embryos of ferns, with the first root

F

Figure 1. Morphological Relationship between the Embryo and the Subsequent Plant.

(A) Median longitudinal section of an idealized body of a fern. A fern undergoes unipolar development, with its roots being initiated from its shoot; hence, it lacks a spatially separate root system. Redrawn with permission from Troll (1959). (B) Median longitudinal section of the embryo of the fern Penta- gramma triangularis. The first root does not arise opposite the shoot pole but at an angle from the shoot, thus forecasting the initiation of subsequent roots from embryonic axis tissue beneath the shoot apex. Contributed by W. Hagemann. (C) Mature plant of Monotropa hypopitys. The root system produces shoot branches (inflorescences) that arise endogenously from those roots. Redrawn with permission from Troll (1943). (D) Median longitudinal section of a late gerrnination stage of the M. hypopitys embryo. This embryo produces a well-developed root cap on the apex of the primary root pole and an enlarged prirnary root surrounded by hyphae of a mycorrhizal fungus. The aborted shoot tip has barely initiated a pair of cotyledons. Redrawn with permission from Troll (1943). En, endosperm; F, fungus; LF, leaf primordium; Lf, adult frond; Lf l to Lf3, fern juvenile leaves; R, root; R I to R6, fern roots numbered in the sequence of decreasing age; RB, root bud; RC, root cap; SAM, shoot apical meristem; SP, shoot pole.

1906 The Plant Cell

arising at an oblique angle rather than diametrically opposite the shoot (Figure 1B; Groff and Kaplan, 1988). All subsequent roots in ferns arise from their shoot axis, below the shoot apical meristem (SAM; Figure 1A).

An even more dramatic expression of the correlation of embryo development with that of the fully developed plant can be seen in the dicotyledonous mycorrhizal heterotroph Monotropa hypopitys (Monotropaceae). The adult body of M. hypopitys consists exclusively of a root system bearing flowering shoots (Figure 1 C). Shoots are initiated endogenously from the roots (Francke, 1934; Troll, 1943). Studying the development of the embryo reveals that there is no shoot system in this plant. Before germination, the seed of M. hypopitys consists of an embryo with a small amount of cellular endosperm at the chalazal pole (Figure 1 D). Although one can recognize the root pole because of its rudimentary cap, the shoot pole has only the slightest suggestion of a pair of cotyledon primordia and consists of markedly vacu- olate, enlarged cells in contrast to the smaller, more densely cytoplasmic cells of the root apex (Figure 1 D). These developmental studies demonstrate that the SAM aborts during embryo development and that it is the root pole that devel- ops into the root system (Figure 1 D) and later initiates the individual shoots.

In conclusion, in contrast with the development of most animals in which morphogenesis and histogenesis are confined to embryogenic and specialized metamorphogenetic stages, comparable processes in plants are not confined to specific stages but extend for the entire life of the plant. There- fore, plant embryogenesis exhibits significant differences from animal embtyogenesis in that plant development is a continuum, albeit one that can be interrupted by dormancy. Moreover, the apical meristems of plants retain the organogenetic capability of the embryo throughout the plant’s life.

RELATIONSHIP BETWEEN CELL LINEAGES AND EMBRYO MORPHOGENESIS

In animal development, cell lineages play an important role in histogenesis and morphogenesis. How important are cell lineages for plant development? Most studies of early sporophyte development focused on the patterns of cell division in the embryo and the organographic and tissue fate of those lineages (reviewed in Johansen, 1950; Maheshwari, 1950; Davis, 1966). By contrast, only a handful of reports have been concerned with the more general features of embryo morphogenesis, such as the origin and activity of apical meristems (Nast, 1941 ; Miller and Wetmore, 1945; Reeve, 1948a, 1948b; Spurr, 1949; Hagemann, 1959; Mahlberg, 1960; Kaplan, 1969). Given that recent studies (Kaplan and Hagemann, 1991, 1992; Niklas and Kaplan, 1991; Cooke and Lu, 1992; Kaplan, 1992) have demonstrated that cell lineages are independent of morphogenesis, it is clear that this aspect of embryo development needs to be reevaluated.

Variability in the Orientation of Zygotic Cell Division

The orientation of the first cell division in the zygote has long been considered to be fundamental to the establishment of the basic polarity of the embryo; therefore, this division has been assigned an almost sacrosanct role in plant embryogenesis by most plant morphology texts (e.g., see Gifford and Foster, 1989). lnsofar as the first division of most land plant zygotes is transverse to the future embryonic axis, the basic polarity of the embryo has been interpreted as being fixed in either endoscopic (shoot pole oriented toward the center of the gametophytic tissue) or exoscopic (shoot pole oriented toward the periphery of the gametophytic tissue) direction, depending on whether the apical or basal cells arising from that first division become differentiated as the shoot pole. In particular, dicotyledonous embryos are said to have an endoscopic orientation, with the first division generating a suspensor initial adjacent to the micropyle and an embryo-proper initial toward the center of the embryo sac. Consequently, the first division has been proposed to establish the embryonic axis, with the result that the embryo proper and suspensor are destined to express different developmental fates (Goldberg et al., 1994).

However, a more extensive survey of the embryological literature turns up a minimum of 19 different dicotyledonous families in which the zygotes of at least one species undergo longitudinal, oblique, or even variable first divisions instead of the expected transverse division (T.J. Cooke and D.R. Kaplan, unpublished data). Measurements of such zygotes at the time of this first division lead to the general conclusion that the plane of the first division coincides with the minimum dimension of the cell. When the zygote is longer than wide, as in >90% of the dicotyledons, the minimum dimension will be the transverse dimension, and the first wall will also be transverse. For zygotes that are wider than they are long, the minimum dimension will be longitudinal, and consequently, the first wall will be longitudinal. In cases in which the zygote is isodiametric, the first wall tends to be oblique (T.J. Cooke and D.R. Kaplan, unpublished data). It is reasonable to conclude that the initial division, like most divisions in plant cells, is oriented according to the overall geometry of the zygote, thereby obeying the principles of minimum surface area: in the absence of other physical forces, an in- cipient wall occupies the plane within the dividing cell that represents the position of minimum surface area (Errera,

Two other significant conclusions can be drawn from these data. First, because growing zygotes acquire charac- teristic shapes before their first cell division, the division plane must be a consequence and not the cause of polarized zygotic growth. In fact, the zygotes of certain species, such as Downingia bacigalupii (Campanulaceae), exhibit marked elongation before they undergo their first cell division (Kaplan, 1969). Second, because almost all of the species described above establish a normal suspensor and embryo proper no matter along which plane the cell division

i 886).


occurs, it follows that the orientation of the initial division is not essential for establishing the embryonic axis. In other words, the underlying process of cytoplasmic polarization in the zygote and early embryo is independent of the initial plane of embryonic division. We thereby conclude that the initial zygotic division is purely coincidental and reflects but does not establish the underlying polarity of dicotyledon embryos. Interestingly, recent work involving the chemical manipulation of division planes in Fucus embryos has resulted in the same interpretation about how overall form is regulated in multicellular brown algae (Shaw and Quatrano, 1996).

Variability in Cell Lineages during Embryogenesis

The stages of cell lineage and division patterns traditionally have been described in terms of species such as C. bursa- pastoris (Hanstein, 1870), which exhibit regular patterns. There are, however, a whole host of species that undergo the same shape changes during embryo development as C. bursa-pastoris but which do not have the regularity of wall patterning that is found in the Brassicaceae. For example, early cell division patterns in the proembryos of cotton (Gos- sypium hirsutum, Malvaceae) are far more variable than are those in the Brassicaceae, and yet the same progression from spherical to heart shaped to torpedo morphology is exhibited as embryogenesis proceeds (Figure 2; Pollock and Jensen, 1964). If one compares the patterns of cell lineages in the proembryo of G. hirsutum (Figures 2A to 2L) and in the later stages of organogenesis after cotyledon emergence (Figures 2M and 2N), it is clear that cell alignments are irregular until the later stages of embryogenesis. One explanation for the change in cell alignment is that as the embryonic body expands in the later phases of morphogenesis, the lineages are pulled into alignment (Cooke and Lu, 1992). This irregularity of cell walls in early embryo development argues against the relevance of such lineages in establishing morphology. These deductions are compatible with contempo- rary biomechanical views of plant morphogenesis in which cell partitioning and alignment patterns are seen as the con- sequences of tensor forces along an organ (Hejnowicz and Hejnowicz, 1991 ; Hejnowicz and Karczewski, 1993).

The fass mutant of Arabidopsis is another example. Em- bryos of this mutant proceed through the same morphogenetic changes as those of the wild type, but the mutant exhibits far greater irregularity of wall patterning than is observed in the wild-type embryo (Torres-Ruiz and Jürgens, 1994). Its irregular pattern of cell lineages resembles that observed for the larger embryos of G. hirsutum (Pollock and Jensen, 1964) and Juglans regia (Nast, 1941). These observations argue that even the supposedly invariable pattern of cell division in the embryos of C. bursa-pastoris and Arabi- dopsis is related more to their small size and uniform growth than to any other fundamental process.

In conclusion, oriented cell divisions during dicotyledonous embryogenesis, as shown earlier for other aspects of

I P

Figure 2. Stages of Early Embryogenesis in G. hirsutum.

(A) Zygote. Redrawn with permission from Pollock and Jensen (1 964). (B) to (L) Filamentous to parenchymatous to early globular stages of embryo development. Note the irregularity of both the sequence and the orientation of early cell divisions in these embryos. Redrawn with permission from Pollock and Jensen (1964). (M) and (N) Heart stage of embryo development. The initiation of the cotyledons at the flanks of the shoot pole is shown in (M), and their subsequent growth is shown in (N). Note the seemingly greater regularity of cell lineages at these later stages. Redrawn with permission from drawings supplied by E.G. Pollock, which were based on photomicrographs originally published in Pollock and Jensen (1 964).

higher plant morphogenesis (Hofmeister, 1868; Kaplan and Hagemann, 1991,1992; Cooke and Lu, 1992; Kaplan, 1992), are not the basis of form generation. Although the characterization of cell division patterns may be a reasonable descrip- tor for embryos that exhibit very diagrammatic lineages (e.g., C. bursa-pastoris), it cannot be universally applied because embryos of many other dicotyledons (e.g., G. hirsutum) do not show such precise cell lineages. If we are to characterize developmental principles that are relevant to describing embryogenesis across the great range of angiosperm diversity, we cannot restrict our focus to cell lineage patterns and tissue geometry.

THE IMPORTANCE OF MERISTEMS: HOW AND WHEN DO MERISTEMS ARlSE DURING EMBRYOGENESIS?

The preoccupation with cell lineages in young embryos has resulted in far fewer studies of the inception of the apical

1908 The Plant Cell

meristems of the root and shoot (Nast, 1941; Miller and Wetmore, 1945; Reeve, 1948a, 1948b; Spurr, 1949; Hagemann, 1959; Mahlberg, 1960; Kaplan, 1969). This neglect and the aforementioned tendency to isolate processes occurring during embryogenesis from the subsequent development of the plant have led to some surprising conclusions, such as the idea that cotyledons are not products of the SAM (Hanstein, 1870; Soueges, 1919; von Goebel, 1933; Meinke, 1992; Barton and Poethig, 1993). That cotyledons indeed are products of the SAM is based on the following observations: (1) cotyledons produced on the embryo apex are homologous to leaves arising on the seedling apex; (2) the dista1 apex of the globular embryo represents the first manifestation of the SAM; and (3) apical zonation, which originates in the globular embryo and whose prominence is directly correlated to the size of the SAM, is a diagnostic marker for the activity of this meristem.

SAM Origin: Cotyledons and Leaves Are Homologous Organs

One interpretation of the leafy cotyledon mutation is that it triggers a homeotic switch in the fate of the first lateral structures emerging on the embryo apex from cotyledons to leaves (Meinke, 1992); this implies that cotyledons should not be considered as leaf homologs. An alternative interpretation is that cotyledons and foliage leaves are indeed homologous structures. What is the evidence that the latter interpretation is correct?

In his study of embryogenesis in finus strobus, Spurr (1949) reviewed the literature on cotyledon morphology to address the question of cotyledon homology as it relates to the origin of the SAM. The principal lines of evidence that Spurr (1949) cited in support of the homology between cotyledons and leaves are the following: (1) cotyledons occur in the same position on the shoot as do leaves; (2) cotyledons have buds in their axils as do leaves; (3) cotyledon development is very similar to leaf development; (4) cotyledons re- semble leaves in terms of mature morphology; and (5) intermediates occur between cotyledons and foliage leaves.

There can be little doubt of the positional equivalence of cotyledons and leaves. The existence of branch meristems in the axils of cotyledons is a widespread phenomenon, and in fact, the development of branches from cotyledonary buds is the basis for basitonic shoot development in many herbaceous annuals and perennials (Troll, 1937, 1964). Moreover, in those species such as J. regia, which have multiple or accessory buds in their foliage leaf axils &e., more than one bud per axil), the cotyledons also exhibit accessory buds (Nast, 1941).

The few studies that diverge from the usual characterization of embryonic cell lineages and provide some histogenetic detail (Nast, 1941; Miller and Wetmore, 1945; Reeve, 1948b; Spurr, 1949; Steffen, 1952; Mahlberg, 1960; Kaplan, 1969) are unanimous in their documentation of the equiva-

lente of the initial morphogenesis of cotyledons and leaves, as indicated by Spurr’s (1 949) third criterion. Although the later growth in breadth and thickness of cotyledons has received far less attention than these parameters have ac- quired in leaf development, those few investigations that have characterized these stages of cotyledon development have indicated that they are virtually identical with those of the subsequently formed leaves (Nast, 1941; Steffen, 1952).

Additional evidence of a leaf homology for cotyledons comes from species such as Cyclamen persicum (Primu- laceae), in which cotyledon morphology is virtually identical to that of the foliage leaves (Figures 3A to 3D; von Goebel, 1933; Hagemann, 1959). Von Goebel (1933) also cited examples of species of Ufricularia, Pinguicula, Viscum, and Spergula that exhibit the same morphological equivalence between the cotyledons and subsequent primary leaves of the shoot.

On comparative morphological evidence, we conclude that cotyledons and foliage leaves are homologous organs corresponding to the leaf or phyllome component of the shoot. This is not to say that there are not significant differences between cotyledons and foliage leaves. For example, in her interesting studies of the affects of phenylboric acid on embryogenesis in frânthis hiemalis (Ranunculaceae), Haccius (1 960) showed that, depending on the dosage (between 600 and 1200 ppm), she could obtain free cotyledons, un- equal cotyledons, monocotyledony, and finally, complete suppression of both cotyledons. Because phenylboric acid did not affect subsequent foliage leaves, there obviously are some developmental and/or physiological differences between cotyledons and foliage leaves that have yet to be defined.

There can also be very obvious differences between foliage leaf and cotyledon function. In addition to being photo- synthetic organs, cotyledons can be specialized as storage and/or haustorial organs for nutrient absorption. Neverthe- less, such functional specializations have nothing to do with the question of the homology of cotyledons with leaves (Kaplan, 1984).

SAM Origin: Cotyledons Arise from the Periphery of the Globular Embryo

If cotyledons are homologous with leaves and share the same morphogenetic pathways, then it follows that like leaves, cotyledons must also arise from a SAM. If this is so, then what is the explanation for the alternative viewpoint, which states that the cotyledons are not the products of a SAM (Hanstein, 1870; von Goebel, 1933; Wardlaw, 1955; Meinke, 1992; Barton and Poethig, 1993)? One reason for holding this conviction may be based on the artificial sepa- ration and isolation of the stages of embryo development from those of later organogenesis. An even more likely reason is that the boundaries of the SAM are usually defined relative to the positions of the adjacent leaf primordia


A B n

Figure 3. Cotyledon Morphology and Its Relationship to the Structure of Subsequent Leaves of the Shoot of C. persicum.

(A) Cotyledon. (B) Third leaf. (C) Fifth leaf. (D) Seventh leaf. Note the morphological similarity of the cotyledon (A) to the subsequent leaves [(B) to (D)]. Redrawn with permission from Hagemann (1959).

(Gifford, 1954). In particular, the SAM is described as a dome-shaped protuberance located between the cotyledons, which appears long after the latter are well developed rather than before their inception. However, if the SAM originates before the emergence of its first leaves &e., the cotyledons), then it becomes impossible to define SAM origin with reference to the positions of other leaves.

Alternatively, this protuberant meristem in fact could be described as a consequence of the SAM's broadening, which occurs before the initiation of the second leaf pair and at right angles to the cotyledons. We argue that the SAM is indeed present before the emergence of the cotyledons. This perspective is strengthened by the obvious parallels between axillary shoot development and embryo development (Troll, 1937). In the case of axillary meristems and the initiation of their first leaf products, the prophyll(s), there is general agreement that the axillary SAM is initiated before its leaf products (Troll, 1937; Kaplan, 1973). Then why should first leaf initiation from the embryonic SAM be treated so differently?

Given that cotyledons as leaf homologs must be the products of a preexisting SAM, what structural criteria can be used to define the origin of the embryonic SAM and to bol- ster our arguments that the SAM is already in place? Tradi- tionally, the criteria for defining the SAM at all stages of its existence have been shape, cell lineages, and cytohistological zonation. With reference to the first criterion, the SAM is usually considered as a convex dome that initiates leaf pri-

mordia from its periphery. However, depending on the geometry of the shoot and the degree of expression of axial shoot versus leaf components, shoot apical configuration ranges from convex to planar to concave (Wardlaw, 1965; Kaplan, 1967). Hence, the shape of the distal end of the shoot is not a hard and fast criterion for determining the timing of inception, much less the presence or absence of SAM activity. Nevertheless, reliance on the shape criterion is often largely responsible for defining the origin of the SAM as taking place after cotyledon initiation (Barton and Poethig, 1993; Nickle and Yeung, 1993).

Cell lineage patterns by themselves are also not decisive in defining the existence and activity of the SAM. For example, in a clonal analysis of embryonic cell lineages in G. hirsutum, Christianson (1 986) demonstrated that the cotyledons and the first two foliage leaves are derived from independent cell clones that are distinct from apical initial cells of the epicotylar SAM that produces all subsequent leaves. Because the boundary between the leaf homologs originating from independent embryonic lineages and those arising from the meristematic cells of the epicotylar apex does not fall as expected between the cotyledons and the first foliage leaves, the cell lineages cannot be considered relevant to defining the origin of the SAM. Moreover, if these very precise cell lineage analyses are put into the broader context of how SAMs generate their lateral structures, it becomes clear that the distal pole of the large cotton embryo is so volumi- nous that it is able to initiate the first four leaf homologs

'

191 O The Plant Cell

without requiring any divisions of summital initial cells to re- generate the meristem until the fourth plastochron.

We believe that the third criterion, cytohistological zonation, is the most indicative of the existence of the SAM because this is an expression of its organogenetic activity, namely, the initiation of leaves and buds. Zonation is manifested by the basic differences in cell and nuclear volume and chromatacity as well as by differences in vacuolation and wall thickness between cells in different positions in the shoot apex (Foster, 1938). Typically, cells at the apical summit, representing the promeristem or initial zone, exhibit the lowest mitotic index and consist of slightly larger, more vac- uolate cells with larger, less densely chromatic nuclei (Gifford and Corson, 1971). By contrast, cells at the meristem periphery, which are involved in organogenesis (leaf and branch initiation), are smaller with a higher mitotic index and with a denser cytoplasm and more chromatic nuclei (Gifford and Corson, 1971). From a functional perspective, Hagemann (1967) has called the former the initial zone (IZ) and the latter the morphogenetic zones (MZ) of the SAM (Figure 4).

This apical zonation is expressed in virtually all SAMs, regardless of their cell lineage patterns (Hagemann, 1967; Kaplan and Hagemann, 1991). Apical zonation is a fundamental property of vegetative shoot apices because the so-called MZ is actually the earliest expression of leaf inception and is apparent before the leaf primordium emerges from the meristem flank. Indeed, Hagemann (1960) has shown that the MZ takes the transectional outline of the leaf destined to emerge from that zone. Therefore, it is impossible to sepa-

Figure 4. Diagrammatic Representation of the Zonation of the Shoot Apex.

Note the summital initial zone (I) and the peripheral, eumeristematic morphogenetic zone (M) from which leaves with procambial strands arise. The dashed lines indicate the subapical cell files originating from the shoot apex. P, procambium. Redrawn with permission from Hagemann (1 959).

rate conceptually the expansion of the entire SAM at the maximal phase of the plastochron from the process of leaf formation.

Such perspectives reinforce the view that we are dealing with differentiation of the SAM, that is, the combination of leaf and stem, and not just the stem apical meristem. For this reason, when the SAM is first differentiated in developing embryos, as is described in the following section, the resulting zonation of the distal pole represents the simultaneous expression of the first leaf primordia as well as the apical initials per se.

SAM Origin: Variation in Expression of Apical Zonation

Because of the variable expression of apical zonation in different plants, we must first describe how the relative volumes of different regions within the SAM affect its cytohistological zonation. Then, we show how apical zonation is used to di- agnose the initial differentiation of the embryonic SAM by comparing its expression in developing embryos of three different species.

The relationship between meristem volume and degree of zonation expression is a reflection of the relationship between meristem volume and the size of the leaf primordium initiated during the course of a plastochron. In large vegetative shoot apices of cacti (Boke, 1941) and cycads (Foster, 1943), the volume of the leaf primordium is small relative to the volume of the entire meristem. As a result, the regeneration of meristem volume to the next maximal phase of the plastochron occurs largely at the peripheral MZ, with relatively little cell division activity occurring in the summital IZ. Consequently, the IZ is sharply delimited from the MZ by conspicuous differences in cytohistological features.

By contrast, the leaf primordium consumes a larger proportion of total meristem volume in the smaller vegetative shoot apices of such plants as Arabidopsis or C. bursa-pastoris. As a result, the IZ is more active in apical regeneration for the next plastochron, and greater fluctuations in the boundaries occur between the IZ and MZ. Consequently, apical zonation is much less prominent in smaller apices. Such differences in relative size of the distal (shoot) pole of embryos explain why apical zonation is more conspicuous in the embryonic shoot apices of some species than it is in others. Nevertheless, the following examples establish that apical zonation is an excellent diagnostic feature for demon- strating the origin of the SAM in all developing embryos. P. strobus exhibits a very conspicuous apical zonation at

the distal pole of its embryo before the cotyledons arise and at approximately the same time as its root apex is defined (Spurr, 1949). Hence, the SAM in P. strobus is clearly differentiated before its cotyledons, and like all of the subsequent leaves of the shoot (Sacher, 1954), the cotyledons are a product of the SAM. The IZ in this SAM is sharply set off from the peripheral MZ as a central conical protuberance. The maximal phase of the first plastochron is exhibited by

Concepts of Angiosperm Embryogenesis 191 1

the broad, shoulderlike protuberances of the MZ; these become elevated and form the cotyledon primordia. f . strobus in particular and gymnosperms in general exhibit very pro- nounced apical zonation in young embryos because their SAMs are significantly larger before cotyledon initiation than those of many angiosperms.

Nevertheless, the expression of shoot apical zonation also occurs in flowering plant embryos, even though its visibility depends on the relative volume of the distal shoot apical pole. The embryo of D. pulchella, for example, does not have a heart-shaped phase because the hypocotyl-root axis elongates significantly before the cotyledons are initiated (Figures 5A to 50). As a result, the shoot pole of this embryo is significantly wider when the cotyledons are initiated than in species such as C. bursa-pastoris (cf. Figures 5D and 5E with Figures 5H and 51). Apical zonation in the embryo of D. pulchella is expressed principally by the larger, less densely chromatic nuclei of the IZ in contrast to the eumeristematic nature of the MZ (Figures 5B to 5D). Tracing this cytodiffer- entiation back to earlier phases, it can first be detected at the late globular stage (Figure 5B). From that point on, the zonation becomes more manifest as the densely cytoplasmic shoulders of the MZ broaden and flatten, expressing the first plastochron maximum before cotyledon protuberance (Figures 5C and 5D). Using the criterion of the timing of zonation development, we conclude that the embryo of O. pulchella, like that of P. strobus, exhibits SAM differentiation and function in the spherical stage, at the same time the root apex is defined (Figure 5B).

Figures 5F to 5J show a closely graded series of median longitudinal sections of early embryogenesis in C. bursa- pastoris. This series shows the same plastochronic fluctuations noted for the other embryos and also for SAM activity in general. For example, in proceeding from the globular phase in Figure 5F, the distal pole of the embryo broadens and flattens, marking the meristem maximum of the first plastochron as noted above for D. pulchella (cf. Figures 5G and 5H with Figures 5C and 5D). After this apical broadening, the cotyledon primordia are evident as a pair of protuberances emerging simultaneously at the periphery of this SAM (Figure 51). This cotyledonary emergence is associated with periclinal wall insertions in the subjacent layers in a mode identical to leaf initiation in other angiosperm shoot apices (Figure 51). The only difference is that in the embryo of C. bursa-pastoris, the cotyledons consume so much of the SAM periphery that they leave only a small residue of the apical initials between the two vertically elongating cotyledons (Figures 51 and 5J). Otherwise, this stage has all the appearances of leaf initiation in a shoot with opposite and decussate phyllotaxis. The principal reason the IZ of the SAM appears only as a narrow residue between the two cotyledons is that expansion to the next plastochronic maximum occurs at right angles to the cotyledons (Figures 51, 6A, and 6B).

The later stages of C. bursa-pastoris embryogenesis are displayed in Figure 6. The emergent, apical dome between

the two cotyledons is not evident until the bending cotyledon stage, when the SAM becomes protuberant in associa- tion with the next plastochron maximum and the development of the epicotylar internode (Figure 6C). Despite this delay in meristem protuberance, the meristem widens between the cotyledons in the interval shown in Figures 6A and 6B, again indicative of meristem enlargement to the maximum of the second plastochron. The apical dome protruding at the last stage shown in Figure 6C represents the maximum of the second plastochron of meristem activity and not the first, as typically assumed. The first plastochron maximum is that evident before the cotyledons arise (Figure 5H).

In conclusion, the most significant difference between the embryos of C. bursa-pastoris and D. pulchella is that cotyledon inception in C. bursa-pastoris occurs so early in embryogenesis that the apex is not large enough to manifest a conspicuous cytological zonation. However, this difference in the relative expression of apical zonation merely reflects differences in embryo size and more importantly in the timing of the onset of cotyledon initiation. It does not result from different developmental processes in the two embryos.

A validation of our criterion of apical zonation as a marker of the SAM comes from a recent study on the expression of the SHOOTMfRlSTEMLESS (STM) gene in wild-type Arabi- dopsis embryos (Long et al., 1996). The earliest expression of this class 1 KNOTTED-like homolog is localized to the cells of the summit of the early to mid-globular stage embryo before cotyledon initiation, after which it is restricted to the summital initial cells in all subsequent stages of wild- type embryo and seedling development. The location and timing of STM expression also coincide exactly with the IZ that we have described for the embryo of D. pulchella. Be- cause class 1 KNOTTED-like genes are considered to play a central role in the perpetuation of meristematic activity (Hake et al., 1995), the localization of STM expression in this con- tinuing meristematic residue or promeristem not only verifies the function of that central zone of the SAM but also reveals the presence of apical zonation in the globular embryo of Ar- abidopsis. Thus, STM can be considered a molecular marker for the IZ. This interpretation is likely to be true for the closely related C. bursa-pastoris, even though the small volume of the SAM in this species makes its zonation less conspicuous by conventional staining procedures (Figures 5F to 51). Finally, it should be appreciated that apical zonation simultaneously defines the two regions of the SAM with separate functions: (1) leaf formation and (2) meristem perpetuation.

The stm mutant also provides a noteworthy example in which the timing of the SAM inception has molecular geneti- cal significance (Barton and Poethig, 1993; Long et al., 1996). This mutant proceeds through normal embryo development but exhibits a complete lack of epicotylar shoot development. Barton and Poethig (1993) originally named this mutant shoot meristemless because they believed that it is entirely devoid of a SAM. However, the ability of the stm mutant embryo to generate normal cotyledons provides diagnostic evidence that

Figure 5. Comparison of Median Longitudinal Sections of the Stages in Embryo Development of D. pulchella and C. bursa-pastoris.

(A) to (E) Stages of embryo development in D. pulchella.(A) The spherical embryo (58 (j.m long) exhibits no sign of apical zonation at the shoot pole, although the histology of the root apex is delineated.(B) The elongating globular embryo (73 p.m long) shows the earliest sign of shoot apex differentiation into the IZ and MZ.(C) A more elongated embryo (85 (xm long) displays more marked zonation and further definition of the root apex.(D) Just before cotyledon emergence, the distal pole of the embryo (140 p.m long) exhibits a broadened shoot apex in which the MZ appears asa pair of cotyledonary buttresses.(E) An older embryo (190 |j.m long) shows the emergence of the cotyledons from the more densely cytoplasmic MZ. The IZ cells with their lessdensely chromatic nuclei stand out clearly from the cells of the MZ with their more densely chromatic nuclei. Reproduced with permission fromKaplan(1969).(F) to (J) Stages of embryo development in C. bursa-pastoris.(F) The spherical embryo (40 urn long) does not exhibit any shoot apex differentiation.(G) The embryo (45 (xrn long) exhibits the early signs of apical pole enlargement.(H) The apical pole of the embryo (55 |j.m long) has broadened and flattened equivalent to the maximal phase of the first plastochron of the SAM.Note that root apex differentiation occurs at approximately the same time the SAM is defined.(I) The embryo (60 n.m long) shows cotyledonary buttresses emerging from the flanks of the SAM.(J) The embryo (~100 \Lm long) has undergone a small amount of axial (future hypocotyl) elongation and further lateral emergence of the cotyle-don primordia, which leave the remains of the shoot apex IZ between them.Cot, cotyledon; IZ, initial zone; MZ, morphogenetic zone; RA, root apex; Su, suspensor.


a functional SAM has arisen in the mutant embryo. Althoughthere can be no doubt that Barton and Poethig (1993) haveidentified a gene for meristem continuation, they have notidentified a gene for SAM origin. Given that meristem abor-tion is a well-documented phenomenon in a range of plants(e.g., Welwitschia, Monotropa, Arceuthobium), the investiga-tion of the sfm mutant is not without interest, especially be-cause the meristem aborts so early in development. However,to date, only the gurke (cucumber shaped) mutant would qual-ify as being truly meristemless because it does not initiate coty-ledons (Mayeret al., 1991).

In conclusion, we believe that a critical evaluation of thedevelopmental morphology of embryos indicates that thecotyledons are the first products of the SAM and that, in bi-polar embryos, the SAM originates in the globular stage be-fore the cotyledons arise, at virtually the same time that theroot apical meristem is defined.

Root Apical Meristem Origin

By contrast with the SAM, origin of the root apical meristem inembryogenesis is simpler and hence less controversial. Be-cause root apices tend to exhibit a regular alignment of celllineages that are often diagrammatic of the tissue arrange-ments, recent studies of root apical meristem origin have fo-cused on these lineage patterns (Dolan et al., 1993; Schereset al., 1994). There is some validity to this approach becausethe definition of the root apical meristem is based on thepericlinal cell lineages that delimit the initial cap from thebody of the root. Doubtless, the reason specific cell lineagesare evident early in the origin of the root is that the root'shistological definition occurs before its morphogenesis. His-togenesis is simply an endogenous partitioning of the proxi-mal pole of the embryo into separate regions (i.e., cap andbody) before that organ undergoes any significant growth.

Figure 6. Median Longitudinal Sections of Later Stages of Embryo Development in C. bursa-pastoris.

(A) Torpedo stage of embryo development. This embryo (350 urn long) exhibits elongated cotyledons and a SAM cut down to its minimal phaseconsisting of only the IZ.(B) Early bending cotyledon stage of embryo development. This embryo (~750 urn long) displays cotyledon curvature, which is conditioned bythe curved nature of the embryo sac. Here, the SAM derived from the IZ has broadened for the maximum of the second plastochron.(C) Mature embryo. This embryo (~1300 urn long) exhibits more advanced curvature and a protuberant SAM at the maximal phase of the sec-ond plastochron.Cot, cotyledon; IZ, initial zone; RA, root apex.

1914 The Plant Cell

Conversely, in the shoot, in which the morphogenesis of lateral organs is exogenous, histogenesis usually follows morphogenesis (Hagemann, 1967).

The most distinctive and earliest expression of root apex origin is the inception of the root cap (von Guttenberg, 1968). Characteristically, the cap originates acroscopically by periclinal subdivision of the preexisting protoderm or dermatogen lineages, regardless of whether or not it incorporates deriva- tives of the distal cell of the suspensor (hypophysis) as it does in Arabidopsis. Once the “cap” is differentiated, by definition, the “body” of the root is also delimited in accordance with the Korper-Kappe or body-cap concept of Schüepp (1 91 7). Because the initial differentiation of the body of the radi- cle must necessarily occur after the differentiation of the cap, there will be a delay in the inception and activity of the so- called initial cells of the body.

However, even though we recognize these lineage patterns in the basal part of the globular embryo, we view them simply as the markers of the root’s inception, not as its cause. This perspective is consistent with our deductions on the relationship of cell lineage patterns and embryo morphogenesis described earlier. Moreover, Van den Berg et al. (1995) have provided corroborating evidence from ablation experi- ments showing that cell lineages in Arabidopsis seedling roots are not correlated with cell fates. Their study supports not only our diminished emphasis on cell lineages for the characterization of plant meristems but also our lack of emphasis on the putative role of individual cells such as the hypophysis in the origin of the root apical meristem (Souèges, 1934). Indeed, if one examines the origin of the root apical meristem as it relates to the overall size of the developing embryo, it becomes clear that the specific site of the incipi- ent root apical meristem is critically dependent on the available space in the basal half of the globular embryo. Small embryos like those of C. bursa-pastoris and Arabidopsis uti- lize the hypophysis to construct a substantial portion of their root apical meristems, whereas intermediate-sized embryos position their root apical meristems somewhat distally from the hypophysis so that it serves, if at all, to generate only the columella (Souèges, 1920; Mahlberg, 1960). Large embryos organize their root apical meristems deeper within the basal region, away from the embryo-suspensor junction . (Nast, 1941).

Finally, with most of the recent attention on root apical meristem inception focusing on cell lineages, the occur- rence of cellular heterogeneity or zonation in the root apex comparable to that shown in the shoot apex has not been commented upon. For example, in the embryo of D. pulchella illustrated in Figure 5, cells at the summit of the body of the root are larger and have larger, less densely chromatic nuclei with smaller nucleoli. These root apical cells are virtually identical to those noted in the IZ of the shoot apex (com- pare root apex with I2 in Figure 5E). Because of their cell lineage alignment with the cortical tissues of the root, these four keystonelike apical cells were termed cortical initials (Kaplan, 1969) and can be traced to even earlier, precotyle-

donary stages of embryogenesis (Figure 58). Do these cells represent the earliest expression of a quiescent center? Fur- thermore, is this cell differentiation in the root apical meristem accompanied by localized patterns of gene expression comparable to the KNO77€D-type expression pattern observed in embryonic SAMs by Long et al. (1996)? Given that differing nuclear properties might be indicative of differential gene activity, this aspect of embryonic root apical meristem activity deserves more attention.

A NEW MODEL FOR DICOTYLEDOM EMBRYOGENESIS

The Model

In light of the deductions we have drawn in the preceding sections, we present a revised model for dicotyledon embryogenesis in Figure 7. This model focuses on the general morphogenesis of a dicotyledonous embryo from the zygote to the arrest of embryo development at dormancy. Because we have explained that the cell lineage patterns do not determine embryo form, we have omitted the lineages from these diagrams. Rather, we have outlined the major tissue tracts representing the primary meristematic tissues to show how embryo histogenesis correlates with its morphogenesis.

After fertilization, most dicotyledonous embryos undergo an initial phase of linear growth (linear proembryo; Figures 7A and 7B), which is followed by formation of a three-dimen- sional parenchyma at the chalazal end of the proembryo (Fig- ure 7C). Parenchyma formation differentiates the distal body of the embryo proper from the proximal, typically filamentous suspensor. During early embryogenesis, greater growth in length occurs in the proximal suspensor than in the distal embryo body because the suspensor plays an initial role in thrusting the embryo proper into a central position in the endosperm mass. Once it has reached its full length, elongation in the suspensor ceases, and morphogenesis is restricted to the distal parenchyma mass (Figures 7C and 7D). During this initial spherical or globular phase in bipolar dicotyledonous embryos, the shoot and root apical meristems become defined, and the first sign of tissue definition is expressed by the simultaneous differentiation of the procambial core and peripheral cortex as well as the protoderm (Figure 70). Defi- nition of the SAM is marked by cytohistological zonation, whereas root meristem delineation is by cap formation and zonation (Figures 7D and 7E). Once the SAM is differentiated, it initiates the first two appendages of the shoot, the cotyledons (Figure 7E), and may continue to initiate subsequent leaves before the onset of seed dormancy. The root pole, by contrast, shows relatively little growth until germination (Figures 7E and 7F).

The advantages of this revised model over the traditional characterization of embryogenesis are that (1) it takes the focus away from cell lineages and emphasizes the overall aspects of morphogenesis; (2) it eliminates the use of arbi-


A

O Zygote

B C

’r’ D E

Linear Embryo Proper lnitial Histogenesis lnitial Subsequent Growth, Proembryo Differentiation and Meristem Organogenesis Organogenesis

Organization and Histogenesls

Figure 7. New Model for the Description of Embryogenesis in Dicotyledonous Flowering Plants.

In this model, the processes that occur during embryo development rather than the overall shape of the embryo are emphasized. (A) to (C) lnitial morphogenesis of the embryo. (A) The zygote. (B) A linear proembryo. (C) The embryo proper. (D) lnitial histogenesis and meristem organization of the embryo. The embryo proper differentiates into the three primary meristematic tissues: (1) peripherai protoderm or epidermal precursor; (2) central procambium or vascular precursor; and (3) subepidermal ground tissue or cortical precursor. Simultaneously, the meristems of the shoot and root are first differentiated at the opposite poles of the embryo. (E) lnitial organogenesis of the embryo. The initial organogenetic event at the shoot pole is expressed by the inception of the first two leaves, the cotyledons. The root apex is defined by root cap formation via periclinai subdivision of proximal protoderm lineages. (F) Subsequent growth, organogenesis, and histogenesis. The major organ components, including cotyledons and hypocotyl, elongate at this stage.

trary shape designations in favor of the developmental processes; and (3) it emphasizes that embryogenesis is not an isolated event in the life history of the plant but the initiation of developmental processes that occur during the entire life of the plant. We now apply this model to the interpretation of the phenotypes of selected Arabidopsis embryo mutants.

Using the New Model for Evaluating Embryo Mutants

Recent interest in the process of plant embryogenesis has come from the application of molecular and genetic techniques and perspectives to the problems of embryo development and the interesting mutants these efforts have produced (Jürgens et al., 1991; Mayer et al., 1991; Meinke, 1991a). Although not the first to describe certain embryo

mutants, Jürgens and his colleagues were the first to present a conceptual model, derived largely from Drosophila developmental genetics, for interpreting the relationship between the mutants and the wild type.

Following a Drosophila-based terminology system and starting from a strict cell lineage concept of embryo development, Jürgens and his co-workers termed the laying down of the basic plant ground plan as “pattern formation” and referred to its genetically controlled, sequentially aligned subdivisions as “pattern formation elements” (Mayer et al., 1991). Furthermore, these authors distinguished two basic patterns: longitudinal (apical-basal) and transverse (radial). The apical-basal pattern is considered to reflect the partitioning of the embryonic axis into three major regions, apical, central, and basal (Mayer et al., 1991), which in turn have been traced back to the original octant tiers in the

191 6 The Plant Cell

proembryo. Thus, the earliest manifestation of the pattern formation process is considered to be expressed by the cell division pattern.

The phenotypes of the mutant seedlings were then interpreted as resulting. from hypothetical deletions from the wild-type plant body of certain longitudinal domains (Mayer et al., 1991). For example, a deletion encompassing the apical domain that forms the SAM and cotyledons in the wild-type seedling results in the gurke phenotype. The fackel or torch- shaped mutant is interpreted to be missing the central domain, that is, the hypocotyl region of the seedling. The monopteros mutant is said to be missing the basal region, that is, the hypocotyl and root pole, whereas the gnom (gn; dwarf or gnome) mutant is cone shaped or spherical because it exhibits terminal deletions of both the apical and basal poles.

Can these Arabidopsis mutants be described by using the model of embryogenesis we have just presented? Yes, they can, if we view these mutants not as being a result of the deletion of a particular embryo domain but rather as a consequence of a change in differential growth. From the vast literature of comparative morphology (Troll, 1937, 1939, 1943), virtually all variation in plant form and proportion can be described as a result of differential growth. Because the nature of multicellularity in plants represents the interna1 partitioning of a tubular cytoplasmic body (Kaplan and Hagemann, 1991), organ deletion is almost never observed as a mode of structural change in plants, except for rarely occurring abscission events (Kaplan et al., 1982).

For example, using our model for dicotyledon embryogenesis (Figure 7), the gurke mutant would be the result of the lack of differentiation of a SAM. If no SAM is differentiated, there will be no cotyledons. Similarly, we suggest that the fackel mutant results from the lack of hypocotyl extension and not from’a deletion of the central domain. Likewise, the root apical meristem has merely failed to differentiate in the monopteros embryo, whereas the gn embryo is not the result of deleting both terminal domains but rather is due to embryo development arresting at the spherical stage before the meristems are initiated.

These interpretations, based on the model in Figure 7, are further reinforced by recent efforts to clone the genes responsible for the mutant phenotypes. So far, none of the cloned genes for these “domain” mutants appears to be embryo specific. The deduced sequence of the GN protein is similar to yeast Sec7, a protein involved in vesicle trans- port between the endoplasmic reticulum and the Golgi (Shevell et al., 1994). The KNOLLE gene appears to encode a syntaxin-like protein (Lukowitz et al., 1996). Thus, there is no obvious strictly embryogenic function for GN and KNOLLE. Rather, they appear to act in general plant processes, such as secretion and cytokinesis, in which an impaired protein might reasonably be expected to cause developmental arrest. In other words, these mutants exhibit structural abnor- malities in embryo development not because they are missing a pattern formation element or domain but because the plant cannot proceed through the normal developmental

stages of embryogenesis because of a breakdown in some fundamental cellular process.

CONCLUDING REMARKS

In this article, we have used a broad comparative approach to assess critically the fundamental concepts of embryogenesis in plants. In particular, we have emphasized that the traditional focus on embryonic cell lineages must be redirected to the more central problem of meristem origin. We have shown that the concepts derived from animal developmental biology have artificially isolated the early stages of plant embryogenesis from the rest of the plant’s ontogeny and have also resulted in interpretations of embryo mutants that are divorced from our vast knowledge of plant morphology.

Although plant developmental biology has certainly been enhanced by the powerful tools that molecular biologists have brought from their studies on animals and microbes, we argue for the uniqueness of plant development and against the natural tendency to import animal-based concepts to plant systems. Plants represent a distinct eukary- otic lineage and long ago independently evolved their own morphogenetic mechanisms for constructing a multicellular body. Nevertheless, the similarity observed between GN or KNOLLE and various yeast proteins or between various sig- na1 transduction pathway components in plants and animals (Baker et al., 1997; Clark et al., 1997) no doubt reflects the conserved functions these proteins have at the cellular level.

The most fundamental event that occurs during embryogenesis in plants is the differentiation of meristems of the shoot and root and the onset of their activity in organogenesis. Thus, we see no difference between the origin of SAMs in typical embryos and their inception as lateral branches, in somatic embryos or in vegetative propagules, except that the latter structures often arise as part of the parent plant body instead of being isolated in a new, separate individual. In fact, the leaf-borne shoots in the dicotyledon Bryophy//um calycinum have been called “foliar embryos” (Yarbrough, 1932) because of their resemblance to plant embryos. Only in their origin as a product of fertilization are plant embryos comparable to animal embryos, and that is where the resemblance ends.

These perspectives become more significant in the efforts to identify unique genes directing embryogenesis. Given that the body plans of most animals are established by processes that occur only during early embryogenesis, it is reasonable to search for regulatory genes whose expression is restricted to the earliest periods of an animal’s development. However, embryogenesis in plants is not a special stage but the first expression of an iterative process that continues at the plant’s meristems during its entire life. It follows that the key regulatory genes, such as STM, and genes encoding signal pathway components, such as WUSCHEL (Laux et


al., 1996) or CLAVATA7 (Clark et al., 1997), that are expressed in embryos remain expressed in meristems to a greater or lesser extent postembryonically. The only genes isolated to date whose expression may be restricted to the earliest stages of plant development encode seed storage proteins that are critical for the nutrition of the embryo and dispersal of the plant but so far appear not to be essen- tia1 for seed development.

Nevertheless, progress in our understanding of the genetic basis of embryogenesis in plants depends as much on the proper conceptual framework as it does on technologi- cal advances. Without the former, there can be no advances in the latter, no matter how sophisticated the technology may be. In the case of plant embryogenesis and development in general, there will be no real progress until the fundamentals of the process are sorted out. These fundamentals cannot be discerned from single-species model systems, no matter how representative they may seem. Thus, there must be a dialog between the practitioners of comparative morphology and those of molecular genetics in order for progress to be made.

ACKNOWLEDGMENTS

An earlier version of this paper was presented at the 1993 Keystone Symposium on Evolution and Plant Development at Taos, New Mex- ico. We thank Linda Vorobik for the drawings in Figure I and Dr. Stephen Wolniak for his assistance with many of the other figures. We are indebted to Dr. Edward Pollock for the drawings of late stages of embryo development in cotton and the permission to re- produce them. We owe a special debt of thanks to Dr. Ann M. Hirsch for her invaluable editorial assistance in preparing this article for publication.

Received March 24,1997; accepted August 28,1997

REFERENCES

Ammirato, P.V. (1 983). The regulation of somatic embryo development in plant cultures: Suspension culture techniques and hor- mone requirements. Bio/Technology 1, 68-74.

Arditti, J. (1992). Fundamentals of Orchid Biology. (New York: John Wiley and Sons).

Baker, B., Zambryski, P., Staskawicr, B., and Dinesh-Kumar, S.P. (1997). Signaling in plant-microbe interactions. Science 276,

Barton, M.K., and Poethig, R.S. (1993). Formation of the shoot apical meristem in Arabidopsis thaliana: An analysis of development in the wild type and shoot meristemless mutant. Development

Boke, N.H. (1941). Zonation in the shoot apices of Trichocereus

726-733.

119,823-831.

spachianus and Opuntia cylindrica. Am. J. Bot. 28, 656-664.

Christianson, M.L. (1986). Fate map of the organizing shoot apex in Gossypium. Am. J. Bot. 73,947-958.

Clark, S.E., Williams, R.W., and Meyerowitz, E.M. (1997). The CLAVATA7 gene encodes a putative receptor-kinase that controls shoot and floral meristem size in Arabidopsis. Cell89, 575-585.

Cooke, T.J., and Lu, B. (1992). The independence of cell shape and overall form in multicellular algae and land plants: Cells do not act as building blocks for constructing plant organs. Int. J. Plant Sei.

Davis, G.L. (1 966). Systematic Embryology of the Angiosperms. (New York: John Wiley and Sons).

Dolan, L., Janmaat, K., Willemsen, V., Linstead, P., Poethig, R.S., Roberts, K., and Scheres, B. (1993). Cellular organization of the Arabidopsis thaliana root. Development 119, 71-84.

Egler, F.E. (1948). The dispersal and establishment of red mangrove, Rhizophora, in Florida. Caribb. For. 9, 299-319.

Errera, L. (1 886). Sur une condition fondamentale d'equilibre des cellules vivantes. C. R. Acad. Sci. Ser. 111 Sci. Vie 103, 822-824.

Foster, A.S. (1938). Structure and growth of the shoot apex in Ginkgo biloba. Bull. Torrey Bot. Club 65,531-556.

Foster, A.S. (1943). Zona1 structure and growth of the shoot apex in Microcycas calocoma (Miq.) A. Dc. Am. J. Bot. 30,56-73.

Francke, H.L. (1934). Beitrage zur Kenntnis der Mykorrhiza von Monotropa hypopitys L. Flora 129,142.

Gifford, E.M., Jr. (1954). The shoot apex in angiosperms. Bot. Rev.

Gifford, E.M., Jr., and Corson, G.E., Jr. (1971). The shoot apex in seed plants. Bot. Rev. 37, 143-229.

Gifford, E.M., Jr., and Foster, A S . (1989). Morphology and Evolu- tion of Vascular Plants, 3rd ed. (New York: W.H. Freeman).

Goldberg, R.B., de Paiva, G., and Yadegari, R. (1994). Plant embryogenesis: Zygote to seed. Science 266,605-614.

Groff, P.A., and Kaplan, D.R. (1988). The relation of root systems to shoot systems in vascular plants. Bot. Rev. 54,387-422.

Haccius, B. (1960). Experimente11 induzierte Einkeimblattrigkeit bei Eranthis hiemalis. 11. Monokotylie durch Phenylborsaure. Planta 54,482-497.

Hagemann, W. (1 959). Vergleichende morphologische, anatomische und entwicklingsgeschichtliche Studien an Cyclamen persicum Mill. sowie einigen weiteren Cyclamen-Arten. Bot. Stud. 9, 1-88.

Hagemann, W. (1960). Kritische Untersuchungen über die Organisation des Sprosscheitels dikotyler Pflanzen. Osterr. Bot. Z. 107,366-402.

Hagemann, W. (1 967). Die Gestaltung als determinierendes Prinzip bei der Entwicklung des Pflanzenkorpers. Math. Naturwiss. Unter- richt. 20, 289-297.

Hake, S., Char, B.R., Chuck, G., Foster, T., Long, J., and Jackson, D. (1 995). Homeobox genes in the functioning of plant meristems. Philos. Trans. R. SOC. Lond. Biol. Sci. 350, 45-51.

Hanstein, J. (1870). Die Entwickelung des Keimes der Monokotylen und Dikotylen. Bot. Abh. 1, 1-1 12.

Hejnowicz, Z., and Hejnowicz, K. (1991). Modeling the formation of root apices. Planta 184,l-7.

Hejnowicz, Z., and Karczewski, J. (1 993). Modeling of meristematic growth of root apices in a natural coordinate system. Am. J. Bot.

153, S7-S27.

20,477-529.

80,309-31 5.

191 8 The Plant Cell

Hofmeister, W. (1867). Die Lehre von der Pflanzenzelle. Handbuch der physiologischen Botanik Erster Band. (Leipzig, Germany: wilhelm Engelmann).

Hofmeister, W. (1 868). Allgemeine Morphologie der Gewachse. Hand- buch der physiologischen Botanik Erster Band. (Leipzig, Germany: Wilhelm Engelmann).

Johansen, D.A. (1950). Plant Embryology. (Waltham, MA: Chronica Botanica).

Jiirgens, G., Mayer, U., Torres-Ruiz, R.A., Berleth, T., and Míséra, S. (1991). Genetic analysis of pattern formation in the Arabidopsis embryo. Development 1 (suppl.), 27-38.

Kane, D.A., Maischein, H.-M., Brand, M., Van Eeden, F.J.M., Furutani-Seiki, M., Granato, M., Haffter, P., Hammerschmidt, M., Heisenberg, C.-P., Jiang, Y.-J., Kelsh, R.N., Mullins, M.C., Odenthal, J., Warga, R.M., and Nüsslein-Volhard, C. (1 996). The zebrafish early arrest mutants. Development 123, 57-66.

Kaplan, D.R. (1967). Floral morphology, organogenesis and interpretation of the inferior ovary in Downingia bacigalupii. Am. J. Bot.

Kaplan, D.R. (1969). Seed development in Downingia. Phytomor-

Kaplan, D.R. (1973). Comparative developmental analysis of the heteroblastic leaf series of axillary shoots of Acorus calamus L. (Araceae). Cellule 69, 251-290.

Kaplan, D.R. (1984). The concept of homology and its central role in the elucidation of plant systematic relationships. In Cladistics: Perspectives on the Reconstruction of Evolutionary History, T. Duncan and T.F. Stuessy, eds (New York: Columbia University Press), pp. 51-70.

Kaplan, D.R. (1992). The relationship of cells to organisms in plants: Problem and implications of an organismal perspective. Int. J. Plant Sci. 153, S28-S37.

Kaplan, D.R., and Hagemann, W. (1991). The relationship of cell and organism in vascular plants. BioScience 41, 693-703.

Kaplan, D.R., and Hagemann, W. (1992). The organism and plant cells in light of Goethe’s comparative morphological method. In der Mitte Zwischen Natur und Subjekt. Johann Wolfgang Goethes Versuch, die Metamorphosen der Pflanzen zu erklaren 1790-1 990. Sachverhalte, Gedanken, Wirkungen, G. Mann, D. Mollenhauer, and S. Peters, eds (Frankfurt, Germany: Waldemar Kramer), pp.

Kaplan, D.R., Dengler, N.G., and Dengler, R.E. (1 982). The mecha- nism of plication inception in palm leaves: Histogenetic observations on the palmate leaf of Rhapis excelsa. Can. J. Bot. 60,

Laux, T., Mayer, K.F.X., Berger, J., and Jürgens, G. (1996). The WUSCHEL gene is required for shoot and floral meristem integrity in Arabidopsis. Development 122, 87-96.

Lindsey, K., and Topping, J.F. (1993). Embryogenesis: A question of pattern. J. Exp. Bot. 44,359-374.

Long, J.A., Moan, E.I., Medford, J.I., and Barton, M.K. (1996). A member of the KNOTLED class of homeodomain proteins encoded by the STM gene of Arabidopsis. Nature 379, 66-69.

Lukowitz, W., Mayer, U., and Jürgens, G. (1996). Cytokinesis in the Arabidopsis embryo involves the syntaxin-related KNOLLE gene product. Cell84,61-71.

54,1274-1 290.

PhOlOgy 19,253-278.

93-1 17.

2999-301 6.

Maheshwari, P. (1950). An lntroduction to the Embryology of Angiosperms. (New York: McGraw-Hill).

Mahlberg, P.G. (1960). Embryogeny and histogenesis in Nerium oleander L. I. Organization of primary meristematic tissues. Phyto- morphology 10,118-1 31.

Mayer, U., Torres-Ruiz, R.A., Berleth, T., Míséra, S., and Jürgens, G. (1 991). Mutations affecting body organisation in the Arabidop- sis embryo. Nature 353, 402-407.

Meinke, D.W. (1 991 a). Embryonic mutants of Arabidopsis thaliana. Dev. Genet. 12,382-392.

Meinke, D.W. (1991 b). Perspectives on genetic analysis of plant embryogenesis. Plant Cell3, 857-866.

Meinke, D.W. (1992). A homeotic mutant of Arabidopsis thaliana with leafy cotyledons. Science 258, 1647-1650.

Miller, H.A., and Wetmore, R.H. (1945). Studies in the developmental anatomy of Phlox drummondii Hook. I . The embryo. Am. J.

Nast, C.G. (1941). The embryogeny and seedling morphology of Juglans regia L. Lilloa 6, 163-205.

Natesh, S., and Rau, M.A. (1984). The embryo. In Embryology of Angiosperms, B.M. Johri, ed (Berlin: Springer-Verlag), pp. 377-443.

Nickle, T.C., and Yeung, E.C. (1993). Failure to establish a functional shoot meristem may be a cause of conversion failure in somatic embryos of Daucus carofa (Apiaceae). Am. J. Bot. 80,1284-1291.

Niklas, K.J., and Kaplan, D.R. (1991). Biomechanics and the adap- tive significance of multicellularity in plants. In The Unity of Evolu- tionary Biology, Vol. 1, Proceedings of the Fourth lnternational Congress of Systematic and Evolutionary Biology, E.C. Dudley, ed (Portland, OR: Dioscorides Press), pp. 488-502.

Nüsslein-Volhard, C. (1 991). Determination of the embryonic axes of Drosophila. Development 1 (suppl.), 1-10,

Niisslein-Volhard, C., and Wieschaus, E. (1980). Mutations affecting segment number and polarity in Drosophila. Nature 287,795-801.

Pollock, E.G., and Jensen, W.A. (1 964). Cell development during early embryogenesis in Capsella and Gossypium. Am. J. Bot. 51,

Reeve, R.M. (1948a). The “tunica-corpus” concept and development of shoot apices in certain dicotyledons. Am. J. Bot. 35, 65-75.

Reeve, R.M. (1948b). Late embryogeny and histogenesis in Pisum. Am. J. Bot. 35, 591-602.

Robichaud, C.S., Wong, J., and Sussex, I.M. (1980). Control of in vitro growth of viviparous embryo mutants of maize by abscisic acid. Dev. Genet. 1,325-330.

Sacher, J.A. (1954). Structure and seasonal activity of the shoot apices of Pinus lambertiana and Pinus ponderosa. Am. J. Bot. 41, 749-759.

Scheres, B., Wolkenfelt, H., Willemsen, V., Terlouw, M., Lawson, E., Dean, C., and Weisbeek, P. (1994). Embryogenic origin of the Arabidopsis primary root and root meristem initials. Development

Schiiepp, O. (1917). Untersuchungen über Wachstum und Form- wechsel von Vegetationspunkten. Jahrb. Wiss. Bot. 57, 17-79.

Shaw, S.L., and Quatrano, R.S. (1996). The role of targeted secretion in the establishment of cell polarity and the orientation of the division plane in Fucus zygotes. Development 122,2623-2630.

Bot. 32, 588-599.

915-921.

120,2475-2487.


Shevell, D.E., Leu, W.-M., Gillmor, C.S., Xia, G., Feldrnan, K.A., and Chua, N.-H. (1994). EMB30 is essential for normal cell division, cell expansion, and cell adhesion in Arabidopsis and encodes a protein that has similarity to Sec7. Cell77, 1051-1062.

Solnica-Krezel, L., Stemple, D.L., Mountcastle-Shah, E., Rangini, Z., Neuhauss, S.C.F., Malicki, J., Schier, A.F., Stainier, D.Y.R., Zwartkruis, F., Abdelilah, S., and Driever, W. (1 996). Mutations affecting cell fates and cellular rearrangements during gastrulation in zebrafish. Development 123,67-80.

Soueges, R. (1 91 9). Les premieres divisions de I’oeuf et les différen- ciations du suspenseur chez le C. bursa-pastoris Moench. Ann. Sci. Nat. Sr. Bot. 10, 1-28.

SouBges, R. (1 920). Embryogenie des Solanacées. Développement de I’embryon chez les Nicotiana. C. R. Acad. Sci. Ser. 111 Sci. Vie

Souèges, R. (1 934). L‘hypophyse et I’épiphyse; les problèmes d’his- togenèse qui leur sont liés. I et II. Bull. SOC. Bot. Fr. 81, 737-748.

Spurr, A.R. (1949). Histogenesis and organization of the embryo in Pinus strobus L. Am. J. Bot. 36,629-641.

Steffen, K. (1 952). Die Embryoentwicklung von lmpatiens glandulig- era Lindl. Flora 139,394-461.

Sussex, 1. (1975). Growth and metabolism of the embryo and attached seedling of the viviparous mangrove, Rhizophora man- gle. Am. J. Bot. 62, 948-953.

Tomlinson, P.B. (1986). The Botany of Mangroves. (Cambridge, UK: Cambridge University Press).

Torres-Ruiz, R.A., and Jürgens, G. (1994). Mutations in the FASS gene uncouple pattern formation and morphogenesis in Arabi- dopsis development. Development 120, 2967-2978.

Triplett, B.A., and Quatrano, R.S. (1982). Timing, localization, and control of wheat germ agglutinin synthesis in developing wheat embryos. Dev. Biol. 91,491-496.

170,1125-1 127.

Troll, W. (1 937). Vergleichende Morphologie der hoheren Pflan- zen. Erster Band Vegetationsorgane. Erster Teil. (Berlin: Gebr. Borntraeger).

Troll, W. (1 939). Vergleichende Morphologie der hoheren Pflan- zen. Erster Band Vegetationsorgane. Zweiter Teil. (Berlin: Gebr. Borntraeger).

Troll, W. (1 943). Vergleichende Morphologie der hoheren Pflan- zen. Erster Band Vegetationsorgane. Dritter Teil. (Berlin: Gebr. Borntraeger).

Troll, W. (1 959). Allgemeine Botanik. (Stuttgart, Germany: Enke).

Troll, W. (1 964). Die Infloreszenzen. Erster Band. (Stuttgart, Ger- many: Gustav Fischer).

Van den Berg, C., Willemsen, V., Hage, W., Weisbeek, P., and Scheres, B. (1995). Cell fate in the Arabidopsis root meristem determined by directional signaling. Nature 378, 62-65.

von Goebel, K. (1 933). Organographie der Pflanzen. Dritte Auflage. Dritter Teil Samenpflanzen. (Jena, Germany: Gustav Fischer).

von Guttenberg, H. (1968). Der primare Bau der Angiospermenwur- zel. In Handbuch der Pflanzenanatomie. Spezieller Teil, Band VIII, Teil 5, W. Zimmermann, P. Ozenda, and H.D. Wulff, eds (Berlin: Gebr. Borntraeger).

Wardlaw, C.W. (1955). Embryogenesis in Plants. (New York: John Wiley and Sons).

Wardlaw, C.W. (1965). The organization of the shoot apex. Hand- buch der Pflanzenphysiologie, Band XV, Teil I, A. Lang, ed (Berlin: Springer-Verlag), pp. 966-1 076.

West, M.A.L., and Harada, J.J. (1 993). Embryogenesis in higher plants: An overview. Plant Cell5, 1361-1369.

Yarbrough, J.A. (1 932). Anatomical and developmental studies of the foliar embryos of Bryophyllum calycinum. Am. J. Bot. 19, 443-453.

DOI 10.1105/tpc.9.11.1903 1997;9;1903-1919Plant Cell

D. R. Kaplan and T. J. Cookeof Embryo Mutants.

Fundamental Concepts in the Embryogenesis of Dicotyledons: A Morphological Interpretation

This information is current as of January 24, 2021

Permissions 98X

https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw15322

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists

https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

http://www.plantcell.org/cgi/alerts/ctmain

http://www.plantcell.org/cgi/alerts/ctmain

http://www.aspb.org/publications/subscriptions.cfm

Documents

Fundamental Concepts in the Embryogenesis of Dicotyledons ... · Embryogenesis represents the critical stage in the develop- ment of a plant at which its basic organization and body