Printer Friendly

The question of cotyledon homology in angiosperms.

II. Introduction

The most fundamental difference between monocotyledons (monocots, Liliopsida) and dicotyledons (dicots, Magnoliopsida) expresses itself early in the ontogeny of the plant body, as the contrasting names of these great lineages imply. During early embryogenesis, from the quadrant and octant stages up to the globular proembryo [ILLUSTRATION FOR FIGURE 1A OMITTED], there is no significant difference between monocots and dicots (Natesh & Rau, 1984). Soon after a globular stage is formed, the developmental trajectory of the embryo takes one of two paths. In dicots a "heart-shaped" structure is precursor to two cotyledons flanking a central depression that becomes the site of the apical meristem [ILLUSTRATION FOR FIGURES 1B, C OMITTED]. In monocots the form following the globular stage is usually more elongate as a single cotyledon becomes the primary axis of the embryo [ILLUSTRATION FOR FIGURE 1E, F OMITTED]; a lateral invagination becomes the locus of the apical meristem and further upward growth. The single cotyledon of monocots is apical or lateral-apical, while the paired cotyledons of dicots are distally lateral and opposite. Monocot seedlings are bilaterally symmetrical from early in their development [ILLUSTRATION FOR FIGURE 1G OMITTED] - very different from dicot embryos [ILLUSTRATION FOR FIGURE 1D OMITTED], where two planes of symmetry arise early in development (Juguet, 1973; Titova & Batygina, 1996).

The distinction between monocot and dicot embryos represents a significant dichotomy that has served as an important taxonomic criterion for almost 200 years. The monocot-dicot distinction effectively divides the world's richest plant lineage into two usually easily distinguished groups of approximately 50,000 and 200,000 species, respectively. The general concordance of additional features such as root growth, vascular anatomy, sieve-tube plastids, prophyll number, leaf architecture, microsporogenesis, pollen types, and patterns of floral morphology supports the general view that both monocots and dicots are natural lineages. Recent phylogenetic studies using morphology and chloroplast DNA fragments also support monocot monophyly (Duval et al., 1993; Chase et al., 1995; Davis, 1995; Nickrent & Soltis, 1995; Stevenson & Loconte, 1995), though basal lineages can be difficult to align (Sytsma & Baum, 1996). Categorical morphological concepts, such as stems, leaves, and cotyledons are used here in the traditional sense, because they have proven useful in description, analysis, and communication. Metameric and continuum or process morphology approaches do not appear to be very helpful in these broadly comparative discussions (cf. Sattler & Rutihauser, 1997).

III. The Question of Cotyledon Homology

A major question, rarely discussed, is whether the structures so useful in separating monocots and dicots are in fact homologous. Nearly all studies of angiosperm embryos have assumed that the seed leaves (cotyledons) of monocots and dicots are indeed homologous, and then proceeded to develop evolutionary scenarios for loss or fusion to explain the difference (cf. Ryberg, 1959). I believe that such evolutionary hypotheses can be clarified if we first examine the fundamental question of homology. Homologous features share an underlying commonality of morphology because of a common evolutionary origin and a continuity of genetic information (Shubin, 1994; Raff, 1996). Before we compare structures to elucidate ancestry and relationship, we must first be sure that these structures are homologous. The use of synapomorphies in cladistic methodology has reaffirmed the significance of homology in phylogenetic hypotheses (Patterson, 1982; Stevens, 1984; Wagner, 1989a; Haszprunar, 1992; Nelson, 1994); and homology considerations have proven useful in a wide range of comparative botanical studies (Kaplan, 1984; Sattler, 1994; Burger, 1996).

A basic criterion for homology is that of equivalent position or topographic similarity (Kaplan, 1984; Riedl, 1978; Rieppel, 1993; Sattler, 1994). The two cotyledons of dicots arise from opposite sides of the hypocotyl and are similar in size and form. They are lateral, flanking a central apical growing point [ILLUSTRATION FOR FIGURE 1B, C OMITTED]. In contrast, the cotyledon of monocots is nearly always terminal [ILLUSTRATION FOR FIGURE 1E, F OMITTED], though it may arise as a crescent-shaped primordium on the side of the apex in larger embryos, as in the palms (Haccius & Philip, 1979). The generally consistent difference in position of the cotyledons in monocots and dicots contradicts this criterion (equivalent position) of homology.

A second criterion for homology is that of overall similarity, structure, or special quality (Kaplan, 1984; Riedl, 1978; Sattler, 1994). Unfortunately, the cotyledons are often very simple structures and their lack of complexity makes determination of homology more difficult (Donoghue & Sanderson, 1994). The cotyledons of monocots are usually narrow or linear with a clasping or sheathing base; a petiole is rarely developed. The cotyledons of dicots usually have a broad lamina with a narrow petiole-like base. In dicots the cotyledons are often very different in form from the next-formed leaves, and they are usually not in the same phyllotactic pattern. The first leaf or pair of leaves is usually oriented perpendicular to the two cotyledons in dicots (Ryberg, 1959). In contrast, the cotyledon of a monocot is usually similar in form to the next succeeding leaf and in the same phyllotactic series, suggesting that cotyledons in monocots are serially homologous (iteratively homologous) with the next-formed leaves (cf. Wagner, 1989b; Shubin, 1994). While variable over many taxa, these differences in fundamental form are sufficiently consistent among monocots and dicots to contradict the criterion of similarity. Cotyledons vary widely in how they function among both monocots and dicots, ranging from thin photosynthetic leaves to thick food-storage organs, and it does not appear that there are additional general differences between the two major groups as regards special physiological or structural qualities of the cotyledons. From these many comparisons it seems safe to conclude that the criterion of overall similarity fails to be supported.

A third criterion of homology is the occurrence of intermediate forms or transitions (Riedl, 1978; Stevens, 1984; Sattler, 1994). This criterion is not supported because the basic structure of monocot and dicot embryos is not bridged by families or genera having intermediate or transitional ontogenies. A few dicot embryos have one cotyledon modified as an absorbing organ or lost (Haccius, 1954; Ryberg, 1959; Eames, 1961: 343; Carlquist, 1981; Takhtajan, 1991), but these exceptions are found in only a few species that belong to genera and families in which typical dicotyledonous embryos are the norm. In addition, a few highly evolved groups form embryos without cotyledons, as in the Orobanchaceae, Santalales, Rafflesisles, and others among dicots, and Orchidaceae in monocots (Rangaswamy, 1967; Natesh & Rao, 1984). It should be noted that while presence of intermediate states supports a homology determination, absence of intermediates does not negate homology.

Rare embryos with two cotyledons in Agapanthus (Liliaceae) were the subject of an early study (Coulter & Land, 1914), but they arise from a tubular sheathing base, quite different from that of dicot cotyledons, and they may simply represent a split cotyledon; unfortunately, these observations have not been replicated. The interpretation that some species of Dioscorea (a genus of monocots in the Dioscoreaceae) has two cotyledons, one remaining in the seed as an absorptive organ and the other emerging later (Lawton & Lawton, 1967), does not seem comparable to the situation in dicots. It seems likely that the two structures referred to as cotyledons by Lawton and Lawton actually arise at different levels on the embryo, as is the case with the cotyledon and following leaf in other monocots. Likewise, the large terminal scutellum and the much smaller opposite first leaf of grasses have been interpreted as two cotyledons (Raju, 1990; Raju & Nambudiri, 1995), but such an interpretation is unconventional, and the two structures are not similar to the two cotyledons in dicots. Here again, it seems that the two so-called cotyledons do not arise opposite each other and simultaneously, as is the case with the two cotyledons of dicots. A study of the monocotyledonous Commelina (Commelinaceae) embryo concluded that there was no evidence for fusion or abortion of two cotyledons (Lakshmanan, 1978). Because many monocot cotyledons have a clasping base with two thick sides flanking the growing point, some thin longitudinal sections of a cotyledonary node can produce a two-lobed outline with a central growing point. Such a thin-section can be erroneously interpreted as representing an embryo with two cotyledons (cf. Solms-Laubach, 1878; Haccius & Philip, 1979). After carefully studying the seedlings of more than 650 species of monocots representing at least 60 families, Tillich (1992, 1995) concluded that the cotyledon in monocots corresponds to a single leaf and that there is no evidence for a second cotyledon or rudiment. Tillich's extensive study contradicts sporadic reports of dicotyledonous embryos in monocots over the last 100 years. Thus, a third criterion of homology (occurrence of intermediate forms) fails to support the general assumption that what we call cotyledons in dicot and monocot embryos are homologous structures.

IV. Co-occurrence in the Nymphaeales

A fourth criterion of homology is the logical assumption that the two contrasting homologous states can not co-occur in the same organism at the same developmental stage (Patterson, 1982). I believe that this criterion can be best analyzed by focusing on studies that have compared two not-too-distantly related lineages on either side of the monocot-dicot division. Because of their many similarities and the ease of accessibility to material, the Alismatales (monocots) and Nymphaeales (dicots) have been the subject of numerous comparative studies. Such studies avoid the great spectrum of variation found scattered through the angiosperms, and they deal with two lineages that have long been thought to be closely associated with the original divergence of monocots and dicots (Nitzschke, 1914; Takhtajan, 1969; Burger, 1977; Cronquist, 1981: 1034; Loconte & Stevenson, 1991; Bharathan & Zimmer, 1995). However, having been separated more than 100 million years ago, it is clear that they have diverged significantly over time (Les & Schneider, 1995). In both Alismatales and Nymphaeales there is a remarkably similar heteroblastic progression of linear to laminar to broadly ovate or peltate leaves. In fact, the embryos and early seedlings of Nymphaeales are so similar to those of the Alismatales, a few researchers have suggested that the Nymphaeales might be better placed in the monocots (Lyon, 1901; Schaffner, 1904; Cook, 1906; Guttenberg & Muller-Schroder, 1958; Haines & Lye, 1975; Titova & Batygina, 1996). These authors interpreted the early development of the Nymphaealean embryo as being more similar to that of the monocots than that of dicots (see further discussion below). Haines and Lye (1979) later changed their earlier interpretation. While the two thick cotyledons of Nymphaeales usually remain undeveloped as storage organs within the seed and never function as leaves, they do appear to be the homologs of the two cotyledons in other dicot lineages, based on their position, number, and general form.

After studying Alisma, Butomus, Sagittaria, and other Alismatales, Kudraishov (1964) concluded that their single cotyledon was initiated from the apical region of the embryo and that the cotyledon in these taxa was "obliquely terminal." He rejected previous hypotheses of fusion of cotyledons or loss of one cotyledon and supported the simpler hypothesis of loss of both cotyledons from a dicotyledonous ancestor. More important, Kudraishov agreed with an earlier interpretation by Meier (reference in Kudraishov, 1964; original not seen) that in Nymphaeaceae, beginning with the third leaf, development follows the pathway seen in the Alismatales. This interpretation claims that the cotyledon of monocots is identical to the third leaf in Nymphaeales. Simple visual support for such an interpretation can be seen in two illustrations of the early embryo of Nuphar luteum published by Guttenberg and Muller-Schroder (1958, and redrawn in Fig. 2A, B). Additional support for this view can be seen in a fine cross-section of the embryo of Barclaya longifolia published by Schneider (1978: fig. 13). When this figure is viewed with the seed operculum oriented downward (as redrawn in Fig. 2C, D), one can see a monocot-like embryo flanked by two large overarching cotyledons. All these studies lend credence to the interpretation that all three "cotyledons" co-occur within the embryo of Nymphaeales. Such a co-occurrence violates a fourth criterion of homology, and I believe we must conclude that the cotyledons of monocots and dieors are fundamentally different structures that do not share a common phylogenetic origin.

V. Other Interpretations of the Nymphaealean Embryo

In his early study, Lyon (1901) divided embryogenesis of Nelumbo (Nelumbonaceae) into four stages. Stage A was designated the spherical stage. He called stage B the "monocotyledonous stage," where the body of the embryo elongates slightly and the apex of the embryo develops a crescent-shaped mound on one side. Stage C he termed "the dicotyledonous stage," in which the lateral margin of the apex of the hypocotyl becomes two-lobed. Stage D he termed the "mature embryo" with two thick cotyledons. Though Nelumbo is now separated from the Nymphaeales and placed in its own order, Nelumbonales (Takhtajan, 1969; Moseley et al., 1993), the early stages of Castalia odorata and Nymphaea advena (Nymphaeaceae) were found to be very similar (Cook, 1906). Likewise, the early embryogeny of Nuphar luteum (Nymphaeaceae) was interpreted in a similar fashion by Guttenberg and Muller-Schroder (1958). In addition, these authors found that both root and shoot development in Nuphar luteum is virtually identical to that of monocots; and they concluded that these plants should be placed within the monocotyledons.

Recently, Titova and Batygina (1996) reviewed this earlier work and, with detailed studies of their own, came to the same conclusion. Following Lyon (1901), they argued that the Nymphaealean embryo is basically monocotyledonous because of its early bilateral symmetry; and they interpret the early lateral apex as representing two fused cotyledons which proceed to become two-lobed. Thus they view the early embryo of Nymphaeales as transitional to complete fusion of the cotyledons, and they interpret the origin of the single cotyledon in all other monocot lineages as the product of syncotyly. According to these authors, it is the change from an embryo with two planes of symmetry (dicots) to an embryo with only one plane of symmetry (monocots) that should be the defining difference between the two great assemblages (cf. Fig. 1D, G). However, a taxonomic realignment separating monocots from dicots on the basis of the early symmetry of their embryos does not coincide with traditional classification or with recent molecular studies that place Nymphaeales among the basal dicotyledons (Duval et al., 1993; Davis, 1995; Nickrent & Soltis, 1995). In addition, Nelumbo falls among the Hamamelids and outside the Nymphaeales in recent molecular phylogenies (Chase et al., 1993), which is consistent with an unusual floral morphology and a more derived pollen type in Nelumbo. These recent molecular studies confirm the traditional separation of monocots from Nymphaeales, and contradict placing Nymphaeales or Nelumbo within the monocots.

In contrast to Kudraishov, who assumed that the two cotyledons in monocots had been lost, Titova and Batygina (1996) claim that the single cotyledon is a fusion product of the two formerly distinct cotyledons. The evidence for this interpretation is not very strong, especially considering that the monocot-like form precedes the initiation of the two cotyledonary lobes in the early ontogeny of the embryo in Nymphaeales. But even if we were to accept syncotyly as the basis for monocotyledony, such a scenario, interposing a major transformation between monocot and dicot embryos, would be consistent with a major division between monocot and dicot embryos. Fusion, if it did result in a single cotyledon, is a significant change that, while supporting the argument for basic homology, does not conflict with the continued use of cotyledon-number to separate the two angiosperm classes. In contrast, the loss of both cotyledons in one lineage would clearly change homology relationships. As long ago as 1795, in his work on comparative anatomy, Goethe stated, "Anything that destroys the form of the part, dividing a muscle into fibres or turning bone into jelly, will not be applied here" - implying that such transformations resulted in a situation where homology inferences were no longer appropriate (Goethe quoted in Riedl, 1978: 246). Thus, while a fusion interpretation or the loss of one cotyledon does not conflict with the claim of cotyledon homology in angiosperms, the loss of both cotyledons in the ancestor of monocots would have resulted in a loss of homology between the structures we have been calling cotyledons in the two classes.

Interestingly, if Meier and Kudraishov are correct, there is a further problem regarding the embryos of monocots and Nymphaeales. If the single cotyledon of monocots is not homologous to the cotyledons of Nymphaeales but equivalent and homologous to their third leaf, what is the relationship of the two thick storage cotyledons of Nymphaeales to the succeeding leaves of the same plant? In short, are the two cotyledons of the Nymphaeales and other dicots really equivalent and homologous to other leaves of the dicot stem?

VI. The Cotyledons of Dicots as Sui Generis Organs

It is well known that the cotyledons of dicots are quite variable. For example, there are three or four whorled leafy cotyledons in Degeneria (Swamy, 1949), and three or four massive storage cotyledons in Idiospermum (Wilson, 1979), primitive members of the Magnoliales and Laurales, respectively. Occasional, tricotyledonous embryos also occur in Acer, Brassica, Juglans, Pittosporum, Raphanus, and Sesamum (Palser, 1975; Gupta & Jain, 1980; Dube et al., 1981; Pillai & Goyal, 1983); while monocotyledonous seedlings have been found in some Apiaceae (Palser, 1975), Stylidiaceae (Carlquist, 1981), and other dicot families. In addition, the cotyledons of dicots are usually not in the same phyllotactic spiral as the succeeding leaves. Such variation casts serious doubt on the assumption that the two cotyledons in dicots are simply the first leaves of an as yet undifferentiated stem apex. More significantly, the disparity of phyllotactic position and usual difference in form contradict the likelihood of homology between cotyledons and succeeding leaves within dicots.

Intensive recent studies in the developmental genetics of Arabidopsis thaliana (a dicot in the Brassicaceae) has led to the discovery of the laterne mutant. This unusual mutation produces seedlings lacking cotyledons but with continued growth that produces the succeeding leaves in a normal fashion (Goldberg et al., 1994). Such an unusual transformation in the early embryogeny of Arabidopsis is unexpected. It had been earlier suggested that mutations expressed early in an ontogenetic progression are much less likely to be successfully incorporated into the genome than mutations occurring at the terminal points of ontogeny (Stebbins, 1974: 128; Carlquist, 1981). Wimsatt and Schank (1988) express this same idea as generative entrenchment, where earlier gene cascades have more wide-ranging effects than genes expressed late in development. Nevertheless, in animal development both the.earliest and later stages of development appear to have been quite labile over evolutionary time and exhibit considerable diversity even in closely related lineages, in contrast to the more critical intermediate stages where the stereotypical body plan of each phylum takes shape (Raff, 1996). These generalities of animal development may not bear directly on early plant growth, but they are suggestive in helping us assess the significance of the laterne gene. This newly discovered mutation, if it can be shown to be widespread among dicots, implies that the two cotyledons of dicots are part of a developmental program separate from the further growth of the stem axis.

Concordant with the laterne gene, a number of other mutations discovered in Arabidopsis result in embryos that fail to produce a shoot-meristem but"have no other obvious phenotypic defects; in particular, the cotyledons are completely normal, suggesting that cotyledons are not formed by an embryonic shoot meristem" (Laux & Jurgens, 1994: 255). Such mutations halt the further growth of the embryo. These mutations are consistent with the observations that the shoot meristem is undeveloped until after the bent-cotyledon stage in Arabidopsis, when it becomes visible as a mound of cells between the bases of the cotyledons (Barton & Poethig, 1993). Together, these various mutants and developmental observations indicate that the two cotyledons of Arabidopsis have a degree of genetic and developmental independence that supports the hypothesis that they are different from the next succeeding leaves. The fact that the two cotyledons of dicots often appear to arise as outgrowths of the hypocotyl and that their vascular traces are derived from the hypocotyl (Eames, 1961: 364) also lends credence to the notion that they are independent of the shoot meristem. This is consistent with Christianson's (1986) finding in Gossypium that "the developmental history of the first two leaves differs in a fundamental way from all the other leaves." In a parallel finding, the cotyledons of Aceraceae were found to have a flavonoid profile very different from that of the true leaves (Delendick, 1990). These various observations strongly suggest that the cotyledons of dicots are organs sui generis and fundamentally different from the normal foliage leaves.

The genetic data from Arabidopsis and variability among dicots make a strong case for the argument that the cotyledons of dicots are not homologous to the succeeding leaves of the same stem. Also, these genetic and developmental differences between the cotyledons of dicots and the succeeding leaves of the same plant violate Wagner's (1989b) definition of biological homology, wherein structures are "homologous if they share a set of developmental constraints caused by locally acting self-regulating mechanisms of organ differentiation." Such a view is consistent with the interpretation that each embryo of Nelumbo and the Nymphaeales possesses homologs for both the single cotyledon of monocots as well as the two cotyledons of other dicots. If the cotyledon of monocots is homologous with the third leaf in Nelumbo and Nymphaeales, then it is logical to conclude that this set of structures is homologous with the early stem leaves of Arabidopsis and all other dicots.

VII. Possible Origins of the Monocot-Dicot Difference

The strong inference that the two cotyledons of dicots are not homologous with the single cotyledon of monocots brings us back to the question of how this fundamental difference between monocots and dicots arose. Since dicotyledonous embryos are common and widely distributed throughout seed plants (Singh, 1978; Loconte & Stevenson, 1990), it has been generally agreed that dicots retain a primitive and widely shared form of embryo. The fact that many gymnosperms have embryos with two cotyledons and that these cotyledons may differ greatly from succeeding leaves of the same plant lends credence to the view that the situation we observe in dicots is the plesiomorphic state for all seed plants (Loconte & Stevenson, 1990). This view is further supported by the fact that the living sister-group of angiosperms, the Gnetopsids, also have dicotyledonous embryos.

Concluding that the dicotyledonous embryo is a plesiomorphy among seed plants implies that the monocotyledonous embryo of monocots is a distinctive autapomorphy for this class of angiosperms. Following this line of reasoning, explanations for the origin of the monocot-dicot distinction have included the loss of both cotyledons (Kudraishov, 1964), the loss of one cotyledon (Coulter & Land, 1914), and a fusion of the two original cotyledons in the earliest monocots (Sargent, 1903; Ryberg, 1959; Titova & Batygina, 1996). Our determination, that the cotyledons of dicots are not homologous with the cotyledons of monocots, forces us to reject both the hypothesis of cotyledon fusion and the hypothesis of loss of one cotyledon, and we can conclude that the dicotyledonous character state was simply lost or became suppressed in the early origin of monocots. This is a simple and straightforward conclusion. But is there a way to test this conclusion, and are other scenarios possible?

The wide range of variability in morphology and number found among the cotyledons of dicots, and their apparently separate and distinct genetic control, leave open the possibility that the dicotyledonous condition in angiosperms has had an independent and convergent origin when compared to other seed plants. Since the two cotyledons of dicots do not appear to be homologous to the next-developed leaves of the same stem and are very likely the product of an independent genetic program, it is possible that they are also not homologous with the two cotyledons found in most other seed plants, and that dicotyledonous seeds in more-distant gymnosperm lineages may prove to be convergent character states, rather than a single shared symplesiomorphic trait among all seed plants. Such a possibility allows for two alternative scenarios. Again, the most parsimonious scenario, consistent with prevailing opinion, would be that dicotyledony arose in the ancestor of Gnetopsids and angiosperms (or earlier), with monocots exhibiting a derived state by loss of the two cotyledons. However, while this view is consistent with sister-group morphology and all recent studies of angiosperm phylogeny, there is an alternative possibility that deserves consideration.

A seemingly unlikely hypothesis is that the monocot embryo represents the primitive condition for angiosperms and retains the plesiomorphic state. There are several reasons for putting forth a suggestion that is so contrary to prevailing opinion. First, the general form of the monocot embryo and its early development are very similar to the form and growth patterns found in a number of pteridophytic embryos (Campbell, 1930; Schaffner, 1934; Burger, 1981; Jacques-Felix, 1982a, 1982b, 1988; Mestre & Guedes, 1983; Vallade et al., 1993; Raju & Nambudiri, 1995). The embryos of pteridophytes have a single terminal first-leaf, a usually lateral growing point, and a root that fails to develop, followed by the formation of adventitious roots. The embryo of many monocots is essentially identical. In addition, the cotyledon of monocots is similar in form to the succeeding leaves and usually in the same phyllotactic pattern, suggesting a uniform developmental continuity which argues for a primitively simple pattern.

The second significant point is that studies of the early developmental stages in Nymphaeales and Nelumbo have indicated that a monocot-like configuration precedes the dicotyledonous state in the ontogeny of these taxa (Lyon, 1901; Guttenberg & Muller-Schroder, 1958; Titova & Batygina, 1996). This ontogenetic sequence implies that the monocot-like form is the antecedent state and the more primitive condition. In such ontogenetic or heterobiastic sequences, the primitive condition almost always precedes the derived condition. Ontogenetic progressions have played an especially important role in determining character-state polarity in zoological studies (Kraus, 1988; Williams et el., 1990; Rieppel, 1990, 1993). The ontogenetic directionality seen in the early Nymphaealean embryo is consistent with a later and independent origin for the two cotyledons, as implied by the laterne mutant and development of the Arabidopsis embryo. The significance of the laterne mutation is that it can eliminate the development of cotyledons without seriously altering the further growth of the plant, suggesting that the two cotyledons may have been a later addition to the developmental programs of the young embryo.

There are additional reasons for suggesting that the monocot embryo might be a plesiomorphic condition. An important point is that there is no well-documented evidence of "dicotyledony" in living monocots (Dahlgren & Rasmussen, 1983; Gifford & Foster, 1988:611), whereas in the dicots monocotyledony and pleiocotyledony are well known (Haccius, 1954; Ryberg, 1959; Haccius & Lakshmanan, 1967; Palser, 1975; Gifford & Foster, 1988: 608). Haccius and Lakshmanan (1967) were able to make the normally monocotyledonous embryos of Anemone apennina and Ranuneulus ficaria (dicots in the family Ranunculaceae) acotyledonous using a phenylboric acid treatment; such a treatment had no special effect on the embryos of monocots. This suggests that the monocot embryo is less responsive to alteration than is the dicot embryo. Also, if the monocotyledony of monocots was due simply to the loss of two ancestral cotyledons, one might expect occasional reversions, but the invariance of monocot embryos contradicts this expectation. In addition, since loss of structures has occurred frequently in the history of life, one might expect considerable polyphyly among monocots (Kudraishov, 1964), but that is not apparent. These varied aspects of monocot embryos and their development should give pause to those who would dismiss the possibility that monocotyledonous embryos might be plesiomorphic for angiosperms. This interpretation focuses renewed attention on the strong similarity between the form and growth of monocot embryos with that of the embryos of many pteridophytes. Morphological studies have not compared monocot and pteridophyte embryos because pteridophytes are seen to be completely unrelated to angiosperms and an inappropriate group for comparative analysis. In addition, such comparisons might imply deep polyphyly among living seed plants, a possibility that recent land-plant phylogenies have rejected (but see Smith, 1964). In contrast, those holding the generally agreed view that seed plants are monophyletic (with dicotyledony as a symplesiomorphic state) could propose an evolutionary hypothesis for the origin of the monocot embryo based on the loss of the two cotyledons, coupled with a reversion to an earlier and simpler morphology. Such a scenario would also be consistent with the pteridophyte-like form and structural invariance seen in monocot embryos. How can these alternative hypotheses be further analyzed?

The laterne mutant and the later development of the two cotyledons in early Nymphaealean embryos support the contention that the two cotyledons may have been a later addition to the embryo, distinct from the formation of the stem and stem-borne leaves. The question then becomes: Did this evolutionary step arise in the ancestors of angiosperms and Gnetopsids (or even earlier among cycads) or was it a later innovation in the origin of dicots? If dicotyledony was present in the common ancestor of Gnetopsids and angiosperms, then all Gnetopsids and angiosperms should have the laterne gene or, in the case of monocots, its nonfunctional homolog. Alternatively, if the two cotyledons were an independent innovation in the earliest dicots, then the laterne gene is likely to be restricted to dicots. This second hypothesis predicts that monocots do not have homologs for those genes that are associated with cotyledon production in the early developmental stages of Arabidopsis and other dicots.

Almost all angiosperms begin with an early embryo that is close to being spherical. In dicots the embryo soon becomes distally two-lobed, apparently initiated by at least one gene represented by the laterne mutant. The next-formed leaves, if Arabidopsis is a typical example, are then initiated by a group of genes that include stem formation and the production of further leaves. A critical genetic question is whether the intensively studied Zea mays (a monocot in Poaceae) or other species of monocots contain homologs for the laterne mutant. If the single cotyledon of monocots is either the fusion product of what were once two cotyledons, the result of having lost one of the two original cotyledons, or is the first stem leaf with cotyledon development having been suppressed, then one should expect a laterne homolog to be present in Zea and other monocots. If a laterne homolog is absent in monocots (and assuming it is present in all dicots), it would suggest that the two cotyledons in dicots are an innovation, unlikely to be homologous with the two cotyledons in all other seed plants, and that those lineages probably lack the laterne gene.

VIII. Conclusion

Clearly, the cotyledons borne by the embryos of monocots and dicots do not appear to be homologous structures. Likewise, there is strong evidence that the two cotyledons of dicots are not homologous with the succeeding leaves of the same plant. These conclusions add significance to the distinction that so clearly separates the embryos of the only two generally recognized classes of angiosperms (Cronquist, 1981; Takhtaj an, 1969). Despite these differences in the embryo, the two classes share many other distinctive characteristics, such as the carpel, stamen, pollen structure, embryo sac, double fertilization, triploid endosperm, and various anatomical traits. Because of these many distinctive similarities, there should be no doubt that monocots and dicots originated from a common ancestor and are members of a monophyletic Magnoliophyta.

Our conclusions suggest a limited number of possibilities with regard to the origin of the difference between monocot and dicot embryos. The most parsimonious scenario is that two cotyledons were present in the ancestors of angiosperms; that they are retained in dicots; and that they are homologous with the two cotyledons seen in Gnetopsids, and perhaps in many other seed plants as well. In this case, monocots could have simply lost cotyledon expression or had it suppressed. Our conclusion that dicot and monocot cotyledons are not homologous render unlikely the hypotheses of loss of one cotyledon or the fusion of two cotyledons in monocot ancestors.

A less parsimonious possibility is that the dicotyledonous embryo was a later evolutionary development in early angiosperm evolution and is convergent, rather than symplesiomorphic, with dicotyledonous embryos in Gnetopsids and other gymnosperms. Such a possibility is consistent with the invariance of monocot embryos and their strong similarity to the form and growth pattern of many pteridophytic embryos. This interpretation implies that monocots are not likely to have homologs for the laterne gene associated with cotyledon production in the early developmental stages of Arabidopsis. With many mutants known to affect early embryogenesis in Zea and in Arabidopsis, perhaps developmental genetics can clarify these differences in early embryogenesis (cf. Meinke, 1995). In the future, comparisons with the genetic controls of embryogenesis in the Gnetopsids, other gymnosperms and pteridophytes may place the striking differences between monocot and dicot embryos in a broadly comparative context. Such investigations, we hope, will be able to answer the question, How did the distinctive difference between monocot and dicot embryos arise?

IX. Acknowledgments

Fred Barrie, Andrew Douglas, Thomas Lammers, Edward Schneider, and Dennis Stevenson provided many helpful comments and useful criticisms on earlier drafts of this paper.

X. Literature Cited

Barton, M. K. & R. S. Poethig. 1993. Formation of the shoot apical meristem in Arabidopsis thaliana: an analysis of development in the wild type and in the shoot-meristemless mutant. Development 119: 823-831.

Bharathan, G. & E. Zimmer. 1995. Early branching events in monocotyledons - partial 18s ribosomal DNA sequence analysis. Pp. 81-107 in P. J. Rudall et al. (eds.), Monocotyledons: systematics and evolution. Vol. 1. Royal Botanic Gardens, Kew.

Burger, W. 1977 [1978]. The Piperales and the Monocots: alternate hypotheses for the origin of monocotyledonous flowers. Bot. Rev. (Lancaster) 43: 345-393.

-----. 1981. Heresy revived: the Monocot theory of angiosperm origin. Evol. Theory 5: 189-225.

-----. 1996. Are stamens and carpels homologous? Pp. 111-117 in W. D'Arcy & R. Keating (eds.), The anther: form, function, and phylogeny. Cambridge University Press, New York.

Campbell, D. H. 1930. The phylogeny of monocotyledons. Ann. Bot. 44: 311-331.

Carlquist, S. 1981. Studies in Stylidiaceae: monocotyly in the family; nomenclatural change. Aliso 10: 35-38.

Chase, M. W., D. E. Soltis, R. G. Olinstead, D. Morgan, D. H. Les, B. D. Mishler, M. R. Duvall, R. A. Price, H. G. Hills, Y.-L. Qui, K. A. Kron, J. H. Rettig, E. Conti, J. D. Palmer, J. R. Manhart, K. J. Sytsma, H. J. Michaels, W. J. Kress, K. G. Karol, W. D. Clark, M. Hedren, B. S. Gaut, R. K. Jansen, H.-J. Kim, C. F. Wimpee, J. F. Smith, G. R. Furnier, S. H. Strauss, Q.-Y. Xiang, G. M. Plunkett, P. S. Soltis, S. Swensen, S. E. Williams, P. A. Gadek, C. J. Quinn, L. E. Eguiarte, E. Golenberg, G. H. Learn, Jr., S. W. Graham, S. C. H. Barrett, S. Dayanandan & V. A. Albert. 1993. Phylogenies of seed plants: an analysis of nucleotide sequences from the plastid gene rbcL. Ann. Missouri Bot. Gard. 89: 528-580.

Chase, M. W., D. W. Stevenson, P. Wilkins & P. J. Rudall. 1995. Monocot systematics: a combined analysis. Pp. 685-730 in P. J. Rudall et al. (eds.), Monocotyledons: systematics and evolution, Volume 2. Royal Botanic Gardens, Kew.

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

Cook, M. T. 1906. The embryology of some Cuban Nymphaeaceae. Bot. Gaz. 42: 376-392.

Coulter, J. M. & W. J. G. Land. 1914. The origin of monocotyledony. Bot. Gaz. 57: 509-519.

Cronquist, A. 1981. An integrated system of classification of flowering plants. Columbia University Press, New York.

Dahlgren, R. & F. N. Rasmussen. 1983. Monocotyledon evolution: characters and phylogenetic estimation. Evol. Biol. 16: 255-395.

Davis, J. I. 1995. A phylogenetic structure for the Monocotyledons, as inferred from chloroplast DNA restriction site variation, and a comparison of measures of clade support. Syst. Bot. 10: 503-527.

Delendick, T. K. 1990. A survey of foliar flavonoids in the Accrateac. Mem. New York Bot. Gard. 54: 1-29.

Donoghue, M. J. & M. J. Sanderson. 1994. Complexity and homology in plants. Pp. 394-421 in B. K. Hall (ed.), Homology: the hierarchical basis of comparative biology. Academic Press, New York.

Dube, V. P., D. K. Awasthi & V. P. Singhal. 1981. Comparative anatomical observations on the dicotylous and tricotylous seedlings of Raphanus sativus L. (Brassicaceae). Acta Bot. Indica 9: 134-137.

Duvall, M. R., M. T. Clegg, M. W. Chase, W. D. Clark, W. J. Kress, H. G. Hills. L. E. Eguiarte, J. F. Smith, B. S. Gaut, E. A. Zimmer & G. H. Learn Jr. 1993. Phylogenetic hypotheses for the Monocotyledons constructed from rbcL data. Ann. Missouri Bot. Gard. 80: 607-619.

Eames, A. J. 1961. Morphology of Angiosperms. McGraw-Hill, New York.

Gifford, E. M. & A. S. Foster. 1988. Morphology and evolution of vascular plants. Ed. 3. W. H. Freeman, New York.

Goldberg, R. B., G. de Paiva & R. Yadegari. 1994. Plant embryogenesis: zygote to seed. Science 266: 605-614.

Gupta, M. L. & V. K. Jain. 1980. Comparative anatomy of the normal and tricotylous seedlings in Brassica oleracea var. capitata. Curr. Sci. 49: 277-278.

Guttenberg, H. & R. Muller-Schroder. 1958. Untersuchnngen uber die Entwicklung des Embryos und der Keimpflanze von Nuphar luteum Smith. Planta 51: 481-510.

Haccius, B. 1954. Embryologische und histogenetische Studien an "monokotylen Dikotylen." I. Claytonia virginica. Oesterr. Bot. Z. 106: 373-389.

----- & K. K. Lakshamanan. 1967. Experimental studies on monocotyledonous dicotyledons: phenylboric acid - induced "dicotyledonous" embryos in Cyclamen persicum. Phytomorphology 17: 488-494.

----- & V. J. Philip. 1979. Embryo development in Cocos nucifera L.: a critical contribution to a general understanding of palm embryogenesis. Pl. Syst. Evol. 132: 91-106.

Haines, R. W. & K. A. Lye. 1975. Seedlings of Nymphaeaceae. J. Linn. Soc., Bot. 70: 255-265.

----- & -----. 1979. Monocotylar seedlings: a review of evidence supporting an origin by fusion. J. Linn. Sot., Bot. 78: 123-140.

Haszpruner, G. 1992. The types of homology and their significance for evolutionary biology and phylogenetics. J. Evol. Biol. 5: 13-24.

Jacques-Felix, H. 1982a. Les Monocotyledones n'ont pas de cotyledon. Bull. Mus. Hist. Nat. (Paris), 4e ser., 4, sect. B, Adansonia 1-2: 3-40.

-----. 1982b. Embryologic vegetal - les idees meconnues de Henri Baillon sur las structure de l'embryon des Monocotyledones. Compt. Rend. Hebd. Seances Acad. Sci. ser. III: 255-257.

-----. 1988. Les Liliopsida (ex Monocotyledones) n'ont pas de cotyledon. II. La prefeuille de la plantule: ses rapports avec celles des axes feuilles. Bull. Mus. Hist. Nat. (Paris), 4e ser. 10, sect. B, Ariansonia 3: 275-333.

Juguet, M. 1973. Expression precoce de la monocotyledonie et raise en place du point vegetatif de la tige chez quelques Monocotyledones, avec quelques remarques sur les types de symetrie des Monocotyledones et de Dicotyledones. Bull. Soc. Bot. France, Mem. 23: 337-354.

Kaplan, D. R. 1984. The concept of homology and its central role in the elucidation of plant systematic relationships. Pp. 51-70 in T. Duncan & T. F. Stuessy (eds.), Cladistics: perspectives on the reconstruction of evolutionary history. Columbia University Press, New York.

Kraus, F. 1988. An empirical evaluation of the use of the ontogeny polarization criterion in phylogenetic inference. Syst. Zool. 37: 106-141.

Kudraishov, L. V. 1964. The origin of monocotyledony (as illustrated by the example of Helobiae). Bot. Zhurn. (Moscow & Leningrad) 49: 473-486.

Lakshmanan, K. K. 1978. Studies on the development of Commelina benghalensis III. Cotyledon. Phytomorphology 28: 253-261.

Laux, T. & G. Jurgens. 1994. Establishing the body plan of the Arabidopsis embryo. Acta Bot. Neerl. 43: 247-260.

Lawton, J. R. S. & J. T. Lawton. 1967. The morphology of the dormant embryo and young seedling of five species of Dioscorea from Nigeria. Proc. Linn. Soc. London 178: 153-159.

Les, D. H. & E. L. Schneider. 1995. The Nymphaeales, Alismatidae, and the theory of an aquatic monocotyledon origin. Pp. 23-42 in P. J. Rudall et al. (eds.), Monocotyledons: systematics and evolution. Vol. 1. Royal Botanic Gardens, Kew.

Loconte, H. & D. W. Stevenson. 1990. Cladistics of the Spermatophyta. Brittonia 42: 197-211.

----- & -----. 1991. Cladistics of the Magnoliidae. Cladistics 7: 267-296.

Lyon, H. L. 1901. Observations on the embryogeny of Nelumbo. Minnesota Bot. Stud. 2: 643-656.

Meinke, D. W. 1995. Molecular genetics of plant embryogenesis. Annual Rev. Pl. Physiol. Molec. Biol. 46: 369-394.

Mestre, J.-Ch. & M. Guides. 1983. Nature et signification phylogenetique du cotyledon. Bull. Soc. Bot. France, Actual. Bot. 130, 3/4: 7-22.

Moseley, M. F., E. L. Schneider & P. S. Williamson. 1993. Phylogenetic interpretations from selected floral vascular characters in Nymphaeaceae sensu lato. Aquat. Bot. 44: 325-342.

Natesh, S. & M. A. Rau. 1984. The embryo. Pp. 377-443 in B. M. Johri (ed.), Embryology of Angiosperms. Springer Verlag, Berlin.

Nelson, G. 1994. Homology and systematics. Pp. 101-149 in B. K. Hall (ed.), Homology: the hierarchical basis of comparative biology. Academic Press, New York.

Nickrent, D. L. & D. E. Soltis. 1995. A comparison of Angiosperm phylogenies from nuclear 18S rDNA and rbcL sequences. Ann. Missouri Bot. Gard. 82: 208-234.

Nitzschke, J. 1914. Beitrag zur Phylogenie der Monokotyln, gegrundet auf der Embryosackentwicklung apokarper Nymphaeaceen und Helobien. Beitr. Biol. Pflanzen 12: 223-268.

Palser, B. 1975. The bases of Angiosperm phylogeny: embryology. Ann. Missouri Bot. Gard. 62: 621-646.

Patterson, C. 1982. Morphological characters and homology. Pp. 21-74 in K. A. Joysey & A. E. Friday (eds.), Problems of phylogenetic reconstruction. Systematics Association Special Vol. 21. Academic Press, London.

Pillai, A. & S.C. Goyal. 1983. Developmental anatomy of some oil-yielding plants. IV. Normal and tricotylous seedlings of Sesamum indicum L. Feddes Repert. 94: 87-90.

Raff, R. 1996. The shape of life: genes, development, and the evolution of animal form. University of Chicago Press, Chicago.

Raju, M. V. S. 1990. The wild oat inflorescence and seed - anatomy, development and morphology. Canadian Plains Center, Regina.

----- & E. M. V. Nambudiri. 1995. The embryo of the eusporangiate vascular plants and the probable antiquity of the monocotyledonous embryo. Phytomorphology 45: 191-206.

Rangaswamy, N. S. 1967. Morphogenesis of seed germination in angiosperms. Phytomorphology 17: 477-487.

Riedl, R. 1978. Order in living organisms. R. P. S. Jefferies, trans. John Wiley, New York. [Originally published as Die Ordnung des Lebcndigen. 1975. Paul Parey, Hamburg.]

Rieppel, O. 1990. Ontogeny - a way forward for systematics, a way backward for phylogeny. Biol. J. Linn. Soc. 39: 177-191.

-----. 1993. The conceptual relationship of ontogeny, phylogeny and classification: the taxic approach. Evol. Biol. 27: 1-32.

Ryberg, M. 1959. A morphological study of Corydalis nobilis, C. cava, C. solida and some allied species, with special reference to their underground organs. Acta Horti Berg. 19:15-119.

Sargent, E. 1903. A theory of the origin of monocotyledons, founded on the structure of their seedlings. Ann. Bot. 17: 1-92.

Sattler, R. 1994. Homology, homeosis and process morphology in plants. Pp. 423-475 in B. K. Hall (ed.), Homology. The hierarchical basis of comparative biology. Academic Press, New York.

----- & R. Rutishauser. 1997. The fundamental relevance of morphology and morphogenesis to plant research. Ann. Bot. 80: 571-582.

Schaffner, J. H. 1904. Some morphological peculiarities of the Nymphacaceae and Helobiae. Ohio Naturalist 4: 83-92.

-----. 1934. Phylogenetic taxonomy of plants. Quart. Rev. Biol. 9: 129-160.

Schneider, E. L. 1978. Morphological studies of the Nymphaeaceae. IX. The seed of Barclaya longifolia Wall. Bot. Gaz. 139: 223-230.

Shubin, N. H. 1994. The phylogeny of development and the origin of homology. Pp. 201-225 in L. Grande & O. Rieppel (eds.), Interpreting the hierarchy of nature. From systematic patterns to evolutionary process theories. Academic Press, New York.

Singh, H. 1978. Embryology of Gymnosperms. Handbuch der Pflanzenanatomie. Vol. 10, pt. 2: 1-302. Gebruder Borntraeger, Berlin.

Smith, D. L. 1964. The evolution of the ovule. Biol. Rev. Cambridge Philos. Soc. 39: 137-159.

Solms-Laubach, H. 1878. Uber monocotyle Embryonen mit Scheitelburtigen Vegetationspunkt. Bot. Zeitung (Berlin) 36: 65-74, 81-93.

Stebbins, G. L. 1974. Flowering plants: evolution above the species level. Harvard University Press, Cambridge.

Stevens, P. F. 1984. Homology and phylogeny; morphology and systematics. Syst. Bot. 9: 395-409.

Stevenson, D. W. & H. Loconte. 1995. Cladistic analysis of monocot families. Pp. 543-576 in P. J. Rudall et al. (eds.), Monocotyledons: systematics and evolution. Vol. 2. Royal Botanic Gardens, Kew.

Swamy, B. G. L. 1949. Further contributions to the morphology of the Degeneriaceae. J. Arnold Arbor. 30: 10-38.

Sytsma, K. J. & D. A. Baum. 1996. Molecular phylogenies and the diversification of Angiosperms. Pp. 314-340 in D. W. Taylor & L. J. Hickey (eds.), Flowering plant origin, evolution and phylogeny. Chapman & Hall, New York.

Takhtajan, A. 1969. Flowering plants, origin and dispersal. Oliver & Boyd, Edinburgh.

-----. 1991. Evolutionary trends in flowering plants. Columbia University Press, New York.

Tillich, H.-J. 1992. Bauprinzipin und Evolutionslinien bei monocotylen Keimpflanzen. Bot. Jahrb. Syst. 114: 91-132.

-----. 1995. Seedlings and systematics in monocotyledons. Pp. 303-352 in P. J. Rudall et al. (eds.), Monocotyledons: systematics and evolution. Vol. 1. Royal Botanic Gardens, Kew.

Titova, G. E. & T. B. Batygina. 1996. Is the embryo of Nymphaealean plants (Nymphaeales s.l.) a dicotyledonous? Phytomorphology 46: 171-190.

Vallade, J., F. Bugnon & Z. Ibannain. 1993. Interpretation morphologique de l'embryon ches les Embryophytes, avec application au cas des Graminees (Poaceae). Canad. J. Bot. 7 l: 256-272.

Wagner, G. P. 1989a. The origin of morphological characters and the biological basis of homology. Evolution 43:1157-1171.

-----. 1989b. The biological homology concept. Ann. Rev. Ecol. Syst. 20: 51-69.

Williams, D. M., R. W. Scotland & S. Blackmore. 1990. Is there a direct ontogenetic criterion in systematics? Biol. J. Linn. Soc. 39: 99-108.

Wilson, C. L. 1979. Idiospermum australiense (Idiospermaceae) - aspects of vegetative anatomy. Amer. J. Bot. 66: 280-289.

Wimsatt, W. C. & S.C. Schank. 1988. Two constraints on the evolution of complex adaptations and the means for their avoidance. Pp. 231-273 in M. H. Nitecki (ed.), Evolutionary progress. University of Chicago Press, Chicago.

Yamashita, T. 1976. Uber die Embryo- und Wurtzelentwicklung bei Aponogeton madagascariensis (Mirbel) van Bruggen. J. Fac. Sci. Univ. Tokyo, Sect. 3, Bot. 12: 37-63.
COPYRIGHT 1998 New York Botanical Garden
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Burger, William C.
Publication:The Botanical Review
Date:Oct 1, 1998
Words:7541
Previous Article:Numerical and comparative analyses of the modern systems of classification of the flowering plants.
Next Article:A critical review of concepts and methods used in classical genome analysis.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |