Two theories of origin of the land-plant sporophyte: which is left standing?
Two somewhat conflicting theories (the homologous theory and antithetic theory) of the origin of alternating generations, specifically the origin of the sporophyte, in embryophytes (land plants) have had respective supporters for approximately a century. The question of initial sporophyte development in the land-plant life cycle resides at a much more fundamental level than the question of whether to interpret the sporophytes we observe today as fundamentally "axial" or "phytonic" structures (Wardlaw, 1968). Although one theory or the other of sporophyte origin (homologous vs. antithetic) has at times been favored by various authors (some authors remaining "neutral"), any real resolution of this question (i.e., which theory should actually be considered correct) has not been overwhelmingly apparent. This seeming indecision, or lack of clarity, has continued in spite of the accumulation of much pertinent knowledge (cytological, ultrastructural, biochemical, molecular-genetic) during the last four decades on the particular group of Chlorophyta, namely the Charophyceae, thought to be most representative of immediate land-plant ancestors. Green algae (chlorophytes) s.l., including charophytes, and green plants make up the large but generally related lineage "Viridiplantae" (Cavalier-Smith, 1981; Blackwell & Powell, 1995; Nakayama et al., 1998).
There is convincing evidence that, among Chlorophyta (s.l.), members of the Charophyceae (e.g., Coleochaete, Nitella, Chara) possibly offer critically important clues to landplant origin (Prescott, 1968; Graham, 1984, 1993); some authors (e.g., Bold et al., 1987) recognize the Charophyceae as "Charophyta," a division distinct from Chlorophyta, perhaps more closely related to embryophyte plants (than other green algae). In any event, in the context of an improved knowledge of putative land-plant ancestors, we may now appropriately ask again the question, which theory of alternation of generations (and sporophyte origin) in land plants is more plausible, the homologous theory or the antithetic theory? Based on an increased knowledge and/or further scrutiny of the morphology, cytology, biochemistry, and life cycle of charophycean algae (in particular), only the antithetic theory may be considered presently tenable--that is, still both logically and evidentially supported (as discussed herein).
III. The Alternating Generations of a Land Plant
Alternating generations (the sexual plant or gametophyte, and the spore-producing plant or sporophyte) in the life cycle of land plants has been appreciated since the work of Hofmeister (1851); see the discussions in Wardlaw (1952), Cronquist (1961), and Kaplan (2001). This alternation occurs in all major groups of land plants (i.e., in groups of nonvascular plants and in groups of vascular plants), although the "balance" (comparative dominance in the life cycle) of the two generations may vary greatly among groups (cf. Bold et al., 1987; Niklas, 1997). A land-plant life cycle may be viewed as encompassing what actually amounts to "two types of organisms" (Niklas, 1994): one promoting sexual reproduction and genetic diversity; the other, organismal proliferation and dissemination via asexual, olden one-celled, propagules (spores). Since, in evolution, the development of sex surely preceded alternation of generations, the gametophyte generation is considered, necessarily, to be older than the sporophyte generation (cf. Scagel et al., 1984; South & Whittick, 1987). The gametophyte generation, through most of plant (including land-plant) evolution, was tied into an aquatic environment, or at least the presence of water for a motile sperm (angiosperms and some gymnosperm groups being exceptions, cf. Bold et al., 1987). The sporophyte in most land plants, by contrast, effects aerial dispersal, and the development of this generation is thus the "key" to terrestrial plant development (Niklas, 1997). These statements concerning alternating generations are relatively noncontroversial. The bone of contention has been, just how, precisely, might the sporophyte generation have originated in the land-plant life cycle?
IV. Putative Origins of the Sporophyte
A. HOMOLOGOUS THEORY
Pringsheim (1878) proposed the "homologous" theory (later called the "transformation" or "modification" theory) for the origin of alternating generations, that is, fur the occurrence of the sporophyte, in the land-plant life cycle (Fig. 1). In this interpretation, the sporophyte is considered to be a direct modification of the gametophyte, in effect a "transformed" gametophyte with the specific function of spore production. The gametophytes and sporophytes of certain algae (e.g., Ulva) are similar (isomorphic), obviously related, structures. Such gametophytes and sporophytes are thus considered "homologous," the sporophyte being a kind of "diploid" version of the "haploid" gametophyte. (Since various levels of ploidy may be encountered, it is perhaps more accurate to say, simply, that the sporophyte usually has twice the chromosome number of the gametophyte.) Algal ancestors are considered to have given rise to land plants, with both a gametophyte and a sporophyte already present in the ancestral algal life cycle (meiosis already being sporic) prior to land invasion. Actually, algae with either isomorphic or heteromorphic generations have been given attention in theorizing land-plant origins, but the greater consideration has gone to algae with isomorphic alternation--in alleged corroboration of the homologous theory. The homologous theory of sporophyte origin in land plants found support in the writings of Goebel (1930), Eames (1936), Bold (1948), and Ditmer (1964).
[FIGURE 1 OMITTED]
B. ANTITHETIC THEORY
In "The Origin of a Land Flora," Bower (1908) suggested an alternative hypothesis for the origin of the sporophyte of land plants (Fig. 2.). In his "antithetic" theory, as in the homologous theory of other authors, algal ancestors of land plants were also envisioned (but not the same kinds of algae in the antithetic theory as in the homologous theory and not the same kind of algal life cycle). In algal ancestors, suggested in the antithetic theory by Bower, only the gametophyte was thought to have been present initially in the algal life cycle, the sporophyte arising subsequently and in correspondence with (or soon after) land occupancy by the gametophyte. Bower suggested that the sporophyte appeared in the life cycle by "amplification of the zygote," i.e., by delay of meiosis, during which a series of mitotic divisions produced a mass of cells (the new sporophyte). Meiosis would thus shift naturally with this new development, from a zygotic "position" to a sporic "position" in the life cycle. The main point is that, in this theory, the sporophyte is viewed as something "new," added or intercalated into the life cycle. Bower later (1935), wishing to downplay some of the connotation of "antithetic" (i.e., as something truly "different" or "foreign" in the life cycle), preferred to refer to the "antithetic theory" as the "interpolation theory," indicating that the sporophyte was indeed "added" to the life cycle, but not that it need be envisioned as something profoundly different from the gametophyte. Smith (1938), Campbell (1940), and Haupt (1953) were among those who supported Bower's antithetic theory of sporophyte origin in land plants.
[FIGURE 2 OMITTED]
Some have looked at the question of sporophyte origin with different conclusions, or else without deciding in favor of one theory or the other. In a somewhat novel idea (though not accepted as a whole at present), Frisch (1916) suggested that the bryophyte sporophyte originated antithetically but that the sporophyte of pteridophytes ("higher" plants) was of homologous origin. This is confusingly referred to as Frisch's "pseudo-homologous" alternation theory (cf. Graham, 1993). Brown (1935) favored independent origins of alternating generations "in various lines," his meaning being somewhat elusive. Brown (1935: 657-658) clearly believed, however, that "it is doubtful if any alternation of generations ... in any ... algae has any relationship to the Bryophyta and other land plants," seeming in this statement to favor an antithetic origin of the land-plant sporophyte but (perhaps overzealously) ruling out the role any alga might have played in this. Land-plant origins have in fact been viewed as monophyletic (e.g., Zimmerman, 1930) or decidedly polyphyletic (e.g., Church, 1919; see also the discussion in Wardlaw, 1952).
At least as numerous as those taking sides, a number of authors have been "on the fence" of the issue; such authors have usually just presented the ideas of both the homologous theory and antithetic theory, without apparent preference (e.g., Smith, 1938; Wardlaw, 1955; Foster & Gifford, 1959; Burns, 1974; Delevoryas, 1977; Gifford & Foster, 1989). Bold, though earlier (e.g., 1948, 1957) favoring the homologous theory, in later writings (e.g., Bold et al., 1987), gave equal credence to the homologous theory and to the antithetic theory. Graham first (1984, 1985) supported the antithetic origin of the land-plant sporophyte but later wavered to an extent by stating, "it is evident that consensus has not yet been reached among plant scientists regarding the origin of plants and their life cycle" (1993: 38).
V. A Priori Arguments
Questions of logical content and consistency of theories of land-plant origin should be addressed before a posteriori (evidential) arguments. A priori arguments are important because of the matter of scientific plausibility and coherence of theories, and are advisedly taken into account prior to the launching of such into the literature of science. The theories of sporophyte origin should (until now, perhaps) have been regarded as "hypotheses," rather than "theories," because of their relatively limited scope (compared with evolution as a whole) and lack of confirmation (cf. Sattler, 1986).
Regardless, in the homologous theory the assumption is made that both alternating generations of an algal ancestor would be carried over, essentially intact, to land, that both would adapt (the sporophyte having presumably the greater adaptative burden), and that the sporophyte would become (at least to an extent) dependent (structurally and physiologically) on the gametopbyte (this, given our knowledge of the dependent or semidependent nature of extant sporophytes of bryophytes and, in initial stages, vascular cryptogams). No real mechanism has been suggested in the homologous theory per se for exactly how all of this would occur or for how motile spores of the algal sporophyte would become nonmotile and adapted to aerial (often wind) dispersal. It is instructive here to think of ontogenetic possibilities (Arber, 1950). Whether the origin and certain changes in the sporophyte occurred before, during, or after invasion of land, it would seem reasonably patent that such changes would necessarily involve altered development of the zygote; if so, then one is actually suggesting an antithetic mechanism, that is, theory reduction (cf. Ruse, 1988) to the antithetic theory, for incipient sporophytes.
A further complication is that two lines of subsequent land-plant evolution (cf. Cronquist, 1961) would be required in the homologous theory. Postulated are both a "downgrade" and a limited "upgrade" development sequence leading to bryophyte (s.l.) sporophytes (some bryophyte sporophytes are inexplicably much more complex, or "less reduced," than others); by contrast, a significantly "upgrade" sequence is envisioned as leading to vascular plant sporophytes. Thus, both reduction and elaboration sequences of the sporophyte are enigmatically proposed in the homologous theory (even within one group, e.g., the Hepatophyta), occurring in what was probably the same, or a very similar, environment (again, without a defining mechanism to account for the two different paths). To this scenario, Bold et al. (1980) added the questionable speculation that homworts (e.g., Anthoceros) evolved (devolved?) by reduction from Psilophyte sporophytes and (without explanation of "how"), in the process, lost useful land-plant adaptations (including vascular tissue)--hence, a putative "downgrade," after an "upgrade," development. A speculation such as that by Bold et al. plays into the larger question, long debated (cf. Minkoff, 1983), as to whether the tracheophyte and bryophyte sporophytes actually had any direct connection to one another.
In contrast to the homologous theory, the antithetic theory presents none of the theoretical difficulties discussed so far in this section, because all bryophyte and vascular plant sporophyte patterns (and assumptions about these) are based on a progressive, "vegetative" development (i.e., by mitosis) of the zygote, retained on the gametophyte, with meiosis obviously delayed. Over time (many generations), a subsequent, generally "upgrade," evolutionary development would be envisioned to take place, by which one could (eventually) account for virtually any land-plant sporophyte type (see Zimmerman, 1952; Niklas, 1992). Bower's (1908, 1935) antithetic (interpolation) theory thus projects both a definite ontogenetic mechanism and a plausible phylogenetic sequence. In not necessarily requiring major morphological reductions (reversals), the antithetic theory is a much more parsimonious interpretation of land-plant evolution (i.e., the evolutionary development of the sporophyte) than is the homologous theory.
VI. A Posteriori Assessments (the "Evidence")
Various arguments, based on evidence (real or supposed), have been laid down in favor of either the homologous theory or the antithetic theory of alternation of generations; that is, of land-plant sporophyte origins. Following are an enumeration and brief discussion of some of the main propositions set forth so far.
A. ALGAE WITH ISOMORPHIC GENERATIONS
An argument often put forward in favor of the homologous theory of alternation of generations (and sporophyte origin) is the alleged "evidence" of algae with isomorphic (morphologically identical or very similar) gametophytes and sporophytes, such as Ulva, Cladophora suhriana, Chaetomorpha (cf. Bold, 1957; Bold & Wynne, 1985; Lee, 1999). The sporophyte is so obviously related to the gametophyte in these algae that it has been regarded merely as a "modified gametophyte" (Bold, 1957); gametophyte and spornphyte appear to be homologous structures. From this has come the extrapolation that it must be so with all sporophytes (e.g., Bold, 1957), including those of land plants. Although green algae are related to green plants (in combination, the "Viridiplantae"; cf. Sluiman, 1985), evidence reviewed in some of the lettered headings in this section (VI), and in other sections of this article (e.g., section VIII), indicates that algae with isomorphic life cycles (isomorphic alternating generations) are definitely not among those green algae most closely related to land plants. Rather, it is charophytes (e.g., Coleochaete, Chara, Nitella) that are closely related to land-plant (embryophyte) ancestors (cf. Graham, 1993). In none of the closest living (or fossil, as far as we know) algal relatives of land plants does alternation of generations (i.e., a sporophyte) actually occur; advanced charophytes are gametophytic plants. Thus, it is a moot point (to the origin of land plants) whether other kinds of algae--Ulva or members of the Cladophorales, tbr example--have alternating generations (or whether these generations are isomorphic).
B. APOGAMY AND APOSPORY
Apogamy ("without gametes") and apospory ("without spores") are, to an extent, misnomers, because in some instances gametes and spores are still produced in the life cycle when these phenomena occur; the chromosome complement of the apogamously tbrmed sporophyte and aposporously produced gametophyte is, however, sometimes other than the usual for these generations (for example, both gametophyte and sporophyte in the life cycle may be found to be diploid). Apogamy and apospory are rather well known in bryophytes (s.l.) and ferns, with apogamy perhaps being the more common phenomenon in nature (Bold et al., 1987); apomixis in angiosperms constitutes forms of apogamy (cf. Stuessy, 1990, re "apogameon").
Apogamy, by definition, is the formation of a sporophyte by (cells of) the gametophyte, without fertilization (without sexual union of gametes). Apospory is the formation of a gametophyte by the sporophyte, without the function of actual melospores. Apogamy and apospory have been alleged as evidence for the homologous theory of alternating generations (e.g., Bold, 1957; Bold et al., 1987), since in some cases gametophyte and sporophyte seem rather readily interconvertible. However, because the gametophyte and sporophyte of a plant necessarily share a large percentage (or all) of a genome, it is probably more accurate to state, simply, that the gametophyte and the sporophyte may be expected on occasion to exhibit a fundamental (genetic) relationship (Niklas, 1997) and even the capacity for interconversion. Because of significant genomic overlap, this genetic relationship would be true regardless of whether the sporophyte originated in a homologous manner or an antithetic manner. There is, thus, no special reason to prefer either the homologous theory or the antithetic theory based on observed phenomena of apogamy/apospory.
C. REVERSION OF VEGETATIVE TO SPOROGENOUS TISSUE
The land sporophyte generation is thought of by some as a neutral generation, not concerned with sex and genetic diversity but "introduced" into the life cycle for the primary "purpose" of producing (often large numbers of) spores for asexual propagation and (perhaps rapid) spread in a terrestrial environment; its function is thus quite different from that of the gametophyte (cf. Niklas, 1997). In an "antithetic view" of sporophyte origin, the only view well suited to this particular "asexual scenario," the sporophyte is seen as a generally upgrade development, progressively coping with distribution in a dynamic, often harsh, land-based environment; it is seen as derived from original sporophytes that were probably little more than small masses of spores (sporophytes represented in appearance perhaps by those of such liverworts as Riccia and Ricciocarpus). This antithetic view holds that more elaborate sporophytes developed by progressive sterilization, and further vegetative development, of originally almost totally sporogenous tissues. One support for this (development by sterilization) assumption is the observation of occasional "reversions," in Porella and some other liverworts, of vegetative tissues of the sporophyte (e.g., of the seta and foot) to sporogenous tissue (Smith, 1938)--an indication of an original, more completely sporogenous state. A more general interpretation is that cells of these rather simple plants retain a similar genetic potential, regardless of the usual course of cell differentiation.
D. PRESENCE OF LIMITED CHLOROPHYLL IN SIMPLE (REDUCED OR PRIMITIVE?) SPOROPHYTES
Bold (1948, 1957) interpreted the finding of a modicum of chlorophyll in the small, simple, ball-like sporophyte of the liverwort Ricciocarpus as evidence of a relic remaining from the reduction of a formerly more elaborate chlorophyll-producing sporophyte derived according to the homologous theory of sporophyte origin. This represents a possibly spurious, and contradictory, interpretation of Rieciocarpus as having a relatively primitive (i.e., thallose) gametophyte but a highly reduced (derived!) sporophyte (Bold, 1957). There is no substantial basis for interpreting either the gametophyte or the sporophyte of Ricciocarpus as anything other than relatively primitive, although the gametuphyte in this case is not the simplest among Hepatics (cf. Schofield, 1985; Bold et al., 1987). The fact is that the vast majority of land-plant sporophytes, of all groups of land plants (excepting certain bryophytes), are heavily chlorophyll bearing, a major autotrophic adaptation evident in a virtually unbroken sequence. The presence of limited chlorophyll in Ricciocarpus, rather than being regarded as relictual, could just as well be regarded as part of a relatively primitive (inchoate) condition of a protected (more or less enclosed), antithetically developed sporophyte--representative, perhaps, of a relatively early stage in an elaboration sequence leading to the much more abundantly chlorophyllose sporophytes of some bryophytes and most tracheophytes.
E. DEPENDENCY OF SPOROPHYTES OF EMBRYOPHYTES
Sporophytes of bryophytes (liverworts, hornworts, mosses) are at least partially dependent upon gametophytes for nutrition in the life cycle; that is, total sporophyte independency is not attained in the Bryophyta (s.l.). In vascular cryptogams (psilophytes, lycopods, sphenophytes, ferns), the sporophyte, though usually becoming independent, nonetheless has early stages that are heavily dependent on the gametophyte. Only a very limited dependency of the sporophyte is retained in gymnosperms and angiosperms, and only during the embryonic stages of development; by contrast, the microscopic gametophytes of seed plants have, in reversal of fortune, established complete dependency on the sporophyte. In any case, overwhelmingly, the sporophytes of embryophytes do not closely resemble the gametophytes on which they depend. Although one could conjure various scenarios for these facts, it is actually difficult to explain them logically with any other than the antithetic theory of sporophyte origin (Wardlaw, 1955); if the homologous theory (deriving the generations from isomorphic precursors) were true, we would expect a greater similarity (and independence) than is observed, at least in some lower groups of embryophytes.
F. OCCURRENCE OF MOTILE (FLAGELLATED) CELLS IN PLANTS
Flagellated sperm occur in bryophytes, vascular cryptogams, and some groups of gymnosperms (cf. Sporne, 1965; Bold et al., 1987), but not in angiosperms. The apical biflagellation of bryophytes and most lycopods is perhaps traceable to a similar pattern in green algae (Bold & Wynne, 1985; Gifford & Foster, 1989; Van den Hoek et al., 1995), including members of the Charophyceae. Flagellation of sperm in land plants is a trait retained to achieve fertilization in aquatic (or past aquatic) environments. Many algae, such as Ulva, in addition to biflagellate gametes, exhibit flagellated zoospores (quadriflagellate in the case of Ulva) produced by the sporophyte. If the homologous theory of origin of the land-plant sporophyte were true, we would expect to find some evidence (in lower land plants) of either motile meiospores or of spores that exhibit (cytologically) a remnant of former flagellation. In fact, no flagellation, or remnant of flagellation, is found in spores of land plants. On the other hand, if the land-plant sporophyte developed antithetically, in response to (initiated in) a terrestrial environment, one would not necessarily expect to find spore flagellation (or even residual evidence of same). It is quite possible that flagella were present on spores of earliest land plants but were lost relatively soon after the incursion onto land--with little or no evidence of former flagellation remaining. The loss of flagella does not represent a special morphogenetic hurdle, since flagella are readily retracted in a number of kinds of flagellated algae, as a natural part of their cell cycles (cf. Beech et al., 1991). This question (of why we do not observe residual evidence of flagellation in land-plant spores), however, should cause us to consider carefully which charophytes might be most representative of algae that were land-plant ancestors. Was it those with flagellated spores (e.g., forms such as Coleochaete) or those forms, such as Nitella and Chara, that produce no spores at all (flagellated or otherwise)? Could not only the sporophyte but also spores of land plants have been an innovation associated with land invasion? It is quite possible that the land-plant spore (even in primitive land plants) represents a structure which is rather different from the spores of any algal group; at a minimum, there is a significant difference in appearance, in both internal and external features (cf. Gray, 1985; Bold et al., 1987; Taylor & Taylor, 1993).
G. PALEOBOTANICAL EVIDENCE
Fossil material provides useful evidence for understanding plant evolution. However, compared with sporophytes, gametophytes are less commonly preserved. Some ancient fossil gametophytes are quite different from their respective sporophytes, as is the case in land plants today (Taylor & Taylor, 1993)--such information tentatively supports the antithetic theory. Other fossil gametophytes, however, are more similar to sporophytes in structure (Remy & Remy, 1980; Remy & Hass, 1986), a fact that has been claimed in support of the homologous theory (see the discussion in Graham, 1993). The Devonian garnetophytes discussed by Remy and Remy and Remy and Hass are, however, some 40 50 million years too recent in the geologic record to provided clues as to the appearance of the earliest land-plant gametophytes. As Niklas (1997) suggested, a repertory of ancient forms (sporophytes and gametophytes) probably occurred (but it is unlikely that we will ever know of all of them). In speculation, gametophytes and sporophytes, which were either similar or dissimilar to each other, could be projected from either the homologous theory (i.e., isomorphic vs. heteromorphic life cycles) or the antithetic theory (especially as reinterpreted by Bower, 1935). Thus, Ordovician, Silurian, and Devonian gametophytes have not yet offered special help in choosing between the homologous theory and the antithetic theory. A possible exception, the Silurian genus Parka, is parenchymatous and padlike (Taylor & Taylor, 1993) and, if eventually demonstrated to be a gametophyte, may be concluded to resemble Coleochaete (Charophyceae). If Parka is eventually shown to have a relationship to Coleochaete. this finding would lend support to the antithetic theory (since charophytes are thought to be most representative of the algal progenitors of land plants; cf. Graham, 1984, 1993). Fossil representatives of the Charales (though not entirely the equivalent of modern genera such as Chara and Nitella) have great antiquity in the fossil record, as exemplified by finds of "gyrogonites" (fossil oogonia); such structures are known from the early Devonian (cf. Croft, 1952; Taylor & Taylor, 1993).
H. CYTOLOGICAL EVIDENCE
Since 1970 a great deal of evidence on cell structure and ultrastructure has accumulated, shedding light on ancestry of land plants. The starlike, flagellar transition zone (as seen in cross-section; cf. Dodge, 1973) possessed by green algae (Chlorophyta) was clearly carried over to motile cells (sperm) of land plants (cf. Mishler & Churchill, 1985). Pinning the (green algal) ancestry of land plants down further, vegetative cells of advanced charophytes (e.g., Chara, Nitella, and Coleochaete) have plasmodesmata, as do land plants. Similarly, these advanced charophytes exhibit a distinctly land-plant ("embryophytic") pattern of mitotic and cytokinetic events, including, an open mitosis with a persistent, interzonal spindle (the daughter mitotic nuclei remaining separated at a distance) and a phragmoplast (spindle fiber remnants perpendicular to the cytokinetic plane and associated golgi-derived vesicles; el. Wolfe, 1983). A cell plate is formed, which often begins centrally and progresses centrifugally. Advanced charophytes thus exhibit a land-plant type of vegetative cell division or "desmoschisis" in which the "parental wall forms part of the wall of the cellular progeny" (Bold & Wynne, 1985: 649). While this definition of desmoschisis (of vegetative cell division, pro parte) is correct insofar as it goes, Groover and Bold (1969) and Bold and Wynne (1985) actually employed "desmoschisis" in the limited context of packets of cells formed by chlorosarcinalean algae (which are chlorophycean, not charophycean, algae; cytokinetic details of the two groups are typically different). The context of Groover and Bold is a concept of cell division actually traceable back to Fritsch (1935). The more demanding and inclusive concept of the land-plant type of vegetative cell division discussed above (presence of phragmoplast, persistent spindle, cell plate, etc.) is in Pickett-Heaps (1975, 1976), Stewart and Mattox (1975), and Mattox and Stewart (1984). Smith (1950) actually used "vegetative cell division" in the "loosest" sense, implying simply cell division that resulted in vegetative (i.e., other than reproductive) cells. Thus, confusingly perhaps, vegetative cell division is used differently by different authors and also is not entirely the equivalent of "desmoschisis."
Returning to motile cells, the flagellated cells of charophytes have a unilateral (asymmetrical) flagellar rootlet system. There are usually two distinct roots, a larger root and a smaller root (and these are sometimes rather closely spaced; cf. Mattox & Stewart, 1984). The larger flagellar root is a "band" with many microtubules (perhaps 60 or so) and is associated (toward the base) with a distinctive multilayered structure (MLS) composed of microtubules and laminate plates; this composite structure is similar to that found in sperm of embryophytes (cf. Van den Hock et al., 1995; Lee, 1999). The ultrastructure of the motile cell of charophytes, particularly the rootlet structure, is in fact quite suggestive of land-plant motile-cell ultrastructure (Melkonian, 1982; Bold & Wynne, 1985; Mishler & Churchill, 1985). Motile cells of" green algae other than charophytes (e.g., Ulvophyceae and Chlorophyceae), by contrast, tend to have a "cruciate" (symmetrical) flagellar root system (i.e., four equally spaced rootlets) and lack the larger root with the MLS (Mattox & Stewart, 1984; Bold & Wynne, 1985).
I. BIOCHEMICAL AND MOLECULAR--GENETIC EVIDENCE
Green algae s.l. (Chrorophyta and Charophyta) share the same types of chlorophylls (a and b) and carotenoids (e.g., lutein, beta-carotene) with embryophytes and, associatedly, similar chloroplast structure and thylakoid arrangement (Van den Hock et al., 1995). The storage product is the same (starch) as higher plants and is similarly stored within plastids; most algal groups other than chlorophytes and charophytes (e.g., chrysophytes, xanthophytes, phaeophytes), exhibit extraplastidal storage of photosynthate (cf. Lee, 1999). The cell-wall composition of green algae and higher plants is similar as well (Green, 1962). Comparing chlorophytes and charophytes, biochemically, Frederick et al. (1973), Al-Houty and Syrett (1984), and Syrett and Al-Houty (1984) determined that charophytes and embryophytes (bryophytes and tracheophytes) utilize the enzyme glycolate oxidase in their photorespiratory pathway; chlorophytes (meaning green algae other than charophytes) utilize glycolate dehydrogenase instead. A further distinction was found in the systematic distribution of the specific urea-utilizing enzyme (Al-Houty & Syrett, 1984; Syrett & Al-Houty, 1984); Charophyceae and land plants possess Urease; Chlorophyceae and Ulvophyceae have urea amidolyase (UAL-ase). Al-Houty and Syrett (1984) found such biochemical (enzymatic) information useful in confirming taxonomic alignment of certain genera; for example, placement of Klebsormidium Silva et al. (1972) with the Charophyceae was affirmed with enzymatic evidence.
Small subunit ribosomal RNA and DNA sequences (e.g., Nakayama et al., 1998; Katana et al., 2001) support the general lineages of algae outlined by Mattox and Stewart (1984)--chlorophytes, ulvophytes, charophytes, etc. and basal groups ofprasinophytes leading to some of these lineages (Nakayama et al., 1998). Mattox and Stewart's (1984) revised classification of green algae (s.l.) has received support (though perhaps limited) from some authors of phycology texts (e.g., Bold & Wynne, 1985). However, a detraction from the Mattox and Stewart (1984) proposals was their consideration (in a "phylogenetic tree") of Charophyceae as a "primitive" group among green algae, based on structure of "swarmers" (motile cells), a mistaken conclusion that could have been avoided through application of sound cladistic methodology to ultrastructural and biochemical data (as discussed by Bremer, 1986). Work on tRNA introns (Manhart & Palmer, 1990), on SSU rRNA gene sequences (Kranz et al., 1995), and rbcl sequences (cf. Graham & Wilcox, 2000) all point to charophytes (of one type or another) as the group of extant "algae" most closely related to land plants.
J. CLADISTIC INTERPRETATIONS
A number of cladistic analyses, based on multiple sorts of information, have examined the question of the lineage order of Viridiplantae (green algae and green plants): e.g., Bremer and Wanntorp (1981a, 1981b), Mishler and Churchill (1984, 1985), Bremer (1985), Sluiman (1985), Bremer et al. (1987), and Theriot (1988). Many characters, of various kinds, have been included in these analyses, viz., morphological, ultrastructural and biochemical (cf. Mishler & Churchill, 1985). Some minor disagreements have occurred, as in Theriot's (1988) criticism of Sluiman's (1985) failure to resolve grouping relationships of Coleochaete, other charophytes, bryophytes and tracheophyes (all arising in Sluiman's work as a sort of polychotomy). Generally, however, the filiation sequence has been relatively clear: chlorophytes, then charophytes, then either liverworts or hornworts, then mosses, and finally, tracheophytes (cf. Theriot, 1988; Niklas, 1997). Some authors (e.g., Niklas, 1997; Purves et al., 1998) considered liverworts to have preceded hornworts in evolution; others (e.g., Theriot, 1988; Renzaglia & Vaughn, 2000) thought hornworts to have the more primitive assemblage of traits and/or to have arisen first. Solomon et al. (2002) considered the matter equivocal. Regardless, the Viridiplantae are considered by virtually all authors to represent generally related groups of organisms, some details not withstanding; in most cladistic analyses it seems clear that, among algae, charophytes place the closet to lower embryophytes.
K. OTHER EVIDENCE?
Is there further evidence (characters), so far largely overlooked, amid the many characters used in cladistic analysis, and recently in molecular genetic analysis, that might be brought to bear on the question of relative filiation "position" of hornworts (anthocerophytes) and liverworts (marchantiophyes)? One such piece of evidence, not found to be a part of cladistic data sets reviewed, might be found in something as seemingly relatively simple and straightforward as chromosome number. As pointed out by Schofield (1985: 274), "there is considerable uniformity in the chromosome number of hepatics (n = 8, 9, or 10) and hornworts (n = 4, 5, or 6) with secondary polyploidy occurring relatively infrequently." Bold et al. (1987) noted that the usual number for hornworts is, in fact, n = 5. In any event, I assume here that, with such a low number, n is the equivalent of x (one chromosome set). Schofield (1985) further indicated that, among hepatics, fully three-fourths (with known chromosome numbers) have n = 9. This last fact is potentially enlightening, because (for example) of Mandal and Ray's (2001) cytotaxonomic study of the genus Nitella (Charophyceae), in which a uniform chromosome number of n = 18 was found. If this (n = 18) in Nitella were actually considered a "diploid" number (the zygote being in effect tetraploid at 36), it would be very difficult to derive a haploid number to match the usual hornwort n (i.e., n = 5); "polyploids" of the hornwort number would probably be 10 or 20, for example, not 18. However, a "haploid" of the 18 number would, of course, be n = 9, matching perfectly the most prevalent n found in the Hepatophyta. This idea is put forward not as any sort of definitive answer to the charophyte--liverwort-hornwort filiation sequence but, rather, as another source of evidence of a possible charophyte--hepatophyte connection--which could be carried further by detailed, karyotype analysis (as initiated by Mandal & Ray, 2001). In short, the limited chromosome number evidence available so far suggests, given scenarios available, the filiation sequence of charophytes, then liverworts, and then hornworts (i.e., prior to reaching the moss clade).
VII. Plant Life Cycles, Morphology, and Habitats: Keys to the Kingdom (and to the Land-Plant Sporophyte)
Considerable evidence exists, some presented herein, that the plant kingdom, Plantae or Viridiplantae (cf. Cavalier-Smith, 1981; Blackwell & Powell 1995, 1999; viz., green algae and green plants) is a sizable but relatively coherent phylogenetic lineage (i.e., a large clade). There is, as discussed, strong evidence from cytology and biochemistry that links green algae, especially charophytes, to higher plants. Perhaps surprisingly, some of the most compelling evidence comes from study of morphology and life cycles (and, in association, habitat); this is, in fact, evidence that has been available for a long time. The selection of the flattened-spheroidal, sometimes parenchymatous Coleochaetae as possibly representative of the putative ancestor of embryophytes (particularly thallose liverworts) is not a new idea and was clearly suggested by Bower (1908) and later supported by Campbell (1940). Wardlaw (1955) and Jeffrey (1962) mentioned charophytes as an algal group possibly involved in land-plant origins. Among more recent students of green algae-green plants, some authors (e.g., Graham, 1984; Scagel et al., 1984) have more apparently credited these past insights and suggestions than have other authors (e.g., Mattox & Stewart, 1984). Regardless, summaries of recent evidence (Graham, 1993; Niklas, 1997), and information presented herein, generally support the ideas of Bower, Campbell, Wardlaw, and Jeffrey.
What can be said, morphologically, is that algae belonging to the Charophyceae, particularly forms like Coleochaete. are now thought (based on considerable evidence) to resemble (i.e., be representative of) putative land-plant ancestors. But if we allow that the land-plant (embryophyte) lineage is traceable to charophycean forms (i.e., to these sorts of algae), does this help us decide which theory of alternating generations, and sporophyte origin, to adopt? The answer is, yes, most definitely it does. First then, a brief, but necessary, digression to consider algal (and plant) life cycles.
Anyone who has studied either Bold's morphology texts or his phycology texts (in their various editions) has become familiar with the three basic life cycles in algae. To simplify, these are (as per Bold's editions of Morphology of Plants (and Fungi), up through the fourth edition, 1980, with or without coauthors):
1. Haplobiontic, Haploid: one type of generation (i.e., the gametophyte) only, in the life cycle, the only "diploid" cell usually being the zygote; meiosis is therefore zygotic; many simple green algae such as Chlamydomonas and Ulothrix exhibit this type of life cycle.
2. Haplobiontic, Diploid: gametophyte only, but the gametophyte is diploid and meiosis is gametic (e.g., the genus Codium, a member of the green algae). As an aside it may be noted that animals generally are diploid organisms, with gametic meiosis.
3. Diplobiontic: alternating generations present, gametophyte (haploid) and sporophyte (diploid); the sporophyte and gametophyte may be similar (isomorphic), as in Ulva and some Cladophora species (both genera being members of the Chlorophyta), or dissimilar (heteromorphic), as in a species of Bryopsis (Chlorophyta) and in Laminaria (belonging to the Phaeophyta); meiosis is sporic in diplobiontic life cycles. A diplobionitic type of life cycle (sporic meiosis) is typically found in land plants, but this land-plant life cycle cannot necessarily be extrapolated to have been derived from algae which are/were diplobiontic (a point discussed a number of times in this article).
Since the terms haplobiontic and diplobiontic actually refer to the number of generations in the life cycle and not chromosome number per se, these terms came to be viewed as potentially confusing. Thus, in the fifth edition of his morphology text (1987), Bold and his coauthors switched to the terms "monobiontic" (haploid), "monobiontic" (diploid), and "dibiontic," respectively, for the three types of life cycles. Other authors have used still different terminology for these three life cycles: "haplontic,'" "diplontic," and "diplohaplontic" (Pritchard & Bradt, 1984). However, it is simpler to refer to these life cycles by the timing or "position" of meiosis in the life cycle, respectively, "zygotic," "'gametic," and "'sporic," as did South and Whittick (1987). South and Whittick (p. 246) presented an easy-to-follow "flow diagram" of the evolution of meiosis, life cycles, alternation of generations, and so forth; their tracings led first to eukaryotes (after the development of mitosis and meiosis), and then to several major eukaryotic lineages and sublineages, including land plants based on the timing of meiosis in the life cycle and the supposed development of alternating generations.
South and Whittick (1987) rightly saw land plants as possessing sporic meiosis (since they do!) but envisioned this lineage as arising from ancestors with preexisting alternation of generations, for which there is no substantive evidence (only speculation), as I have discussed and will continue to discuss. South and Whittick speculated that this alternation of generations in algae, connecting to land-plant origins, would be heteromorphic (they did not name a specific group of algae). This thesis of heteromorphic precedence (of land plants) is a more plausible proposition, but less common, than an isomorphic hypothesis (since extant land-plant lifecycle generations are indeed overwhelmingly heteromorphic); but even a heteromorphic hypothesis is not truly plausible, because there is no evidence for algae that were already in possession of alternating generations, as being among those leading directly to land plants. The less plausible, but seemingly more common, suggestion has been that green algae with isomorphic alternating generations, such as Ulva (Bold, 1957; Tippo & Stern, 1977), were land-plant progenitors. However, in Ulvophyceae the furrowing type of somatic cytokinesis, the cruciate, flagellar roots of the motile cell, and the fact that not only male but also female gametes of Ulva are motile (the gametes being isogamous or anisogamous) would all but preclude the direct involvement of ulvophytes in land-plant origins (cf. Mattox & Stewart, 1984). Also, Ulva and relatives are marine organisms.
It is likely that land plants originated in amphibious environments involving fresh water (and land; cf. Niklas, 1997), regardless of the past counterspeculations of a few (cf. Church, 1919). Fritschiella, a terrestrial green alga (sometimes found on tree bark), with a growth form varying from filamentous to parenchymatous, has also been postulated as a land-plant progenitor (Cronquist, 1961). However, Fritschiella (a chaetophoracean or ulotrichacean form) does not develop a phragamoplast during cytokinesis but, rather, has a more primitive, "phycoplast" configuration (cf. Lee, 1999)--again virtually eliminating it as a possible direct land-plant ancestor. Concerning other types of algae, there is little point in considering algae not related to the green-plant lineage (cf. Wardlaw, 1952; Cavalier-Smith, 1981 ; Blackwell & Powell, 1995, 1999). For example, Ectocarpus and Laminaria (Phaeophyceae, i.e., brown algae) exhibit, respectively, fundamentally isomorphic and heteromorphic alternations of generations. Regardless of the appealing and instructive nature of their life cycles, these brown algae are not only marine organisms but Stramenopiles (cf. Blackwell & Powell, 2000), relatively unrelated to Viridiplantae (cf. Blackwell & Powell, 1995). The alternating life cycles of brown algae are thus best interpreted as parallel (not related) developments to any such life cycles in green algae (e.g., Ulva, Derbesia).
VIII. The Coleochaete Connection (or Is the Connection with Nitella, or Klebsormidium, or Something Else?)
Given greatly improved knowledge of the relationships of particular algal groups to land plants (i.e., to embryophytes), as discussed throughout this article, the focus need now be primarily on those groups of algae clearly related to embryophytes (in attempting to account for land-plant origins). It is specifically from the morphology and life cycle of these algae, with a close and certain connection to land plants, that one should hope to solve the riddle of the homologous versus antithetic origin of the land-plant sporophyte. A review and development of a considerable body of evidence has indicated the phylogenetic significance of Charophyceae to land-plant origins. The discussion, thus, will necessarily proceed to an analysis of the life cycle of this particular group of algae.
As discussed, along several lines of evidence, advanced charophytes provide the best insight into land-plant origins. Singled out among charophytes as possibly morphologically representative of very primitive land plants has often been the small, freshwater, sometimes filamentous, sometimes flattened and padlike (depending on the species) genus Coleochaete. Coleochaete indeed appears to offer developmental clues to embryophyte origins (Graham, 1984). The parenchymatous thallus and more or less circular shape of certain species of Coleochaete (e.g., C. scutata, C orbicularis; cf. Graham, 1982) are consistent with forms postulated for primitive land plants (Niklas, 1992). Such a "dorsiventral bifacial" thalloid structure is considered probable in the ancestry of land plants, according to some theories of land-plant origin (cf. Sattler, 1998: 784). While establishing an essential bipolarity in development (Haberlandt, 1914; Smith, 1938; Wardlaw, 1952), portions of the gametophytes of liverworts and some vascular plants (Equisetum. lycopods, certain terns) retain a similar pattern of parenchymatous growth (cf. Bold et al., 1987). Graham (1982) mentioned that the branching pattern of one species of Coleochaete in particular, C. soluta, suggests a possible "pathway" to land-plant patterns. Features of morphology, life cycle, cytology, ultrastructure, and biochemistry have combined to provide a focused viewpoint on the possible significance of Coleochaete. and other advanced charophytes, in phylogenetic interpretation.
In understanding Coleochaete as possibly representative of land-plant progenitors, illuminating is the work of Marchant and Pickett-Heaps (1973) and Pickett-Heaps (1975, 1976) on cell division (including cell-plate formation), indicating a land-plant pattern of cell division in Coleochaete. The parenchymatous (tissue-like) thallus of some species of Coleochaete has struck certain workers (e.g., Bower, 1908) as being similar to thalli of simple embryophytes (e.g., thallose liverworts). Comparing species of Coleochaete, a transition from filamentous to parenchymatous may be outlined (Graham, 1982, 1984), suggestive of a conversion of algal thallus to a thallus type possibly suitable for land existence. The thallus of Coleochaete is gamete producing (i.e., is a gametophyte). The apically biflagellate sperm of Coleochaete, formed in small but distinct antheridial cells, are consistent with sperm of land plants. A single-celled oogonium (not a multicellular archegonium, as found in embryophytes) surrounds the single egg cell; however, the nonmotile female gamete, and oogamous reproduction, are similar to embryophytes. Unlike most algae, the zygote of Coleochaete remains attached to the gametophyte, becoming surrounded by a layer of protective cells (of gametophytic origin). These cells surrounding the enlarging zygote (see the discussion of "spermocarp" in Smith, 1950) may develop invaginations indicative of nutrient transfer, as in the archegonial venter cells of lower embryophytes (cf. Graham & Wilcox, 1983; Niklas, 1997); such invaginated, "placental" cells (cf. Graham & Wilcox, 1983) may provide evidence of archegonial origins. The zygote divides initially by meiosis, and a small mass of biflagellate zoospores is produced (Bold & Wynne, 1978, 1985).
In the original terminology of Bold (1957), the life cycle of Coleochaete is "haplobiontichaploid," the zygote being the only diploid stage; meiosis is, thus, zygotic. As discussed, we may refer to the life cycle of Coleochaete as, simply, "zygotic" (South & Whittick, 1987). If it is convincing that Coleoehaete is similar to the algal ancestor of primitive embryophytes (cf. Graham, 1984, 1993), and if Coleoehaete and other advanced charophytes (e.g., Nitella, Chara) have only a zygotic type of life cycle (are only gametophytes), then consideration should be given to how this rather simple type of life cycle could have been modified into that of a land plant with a diplobiontic life cycle and sporic meiosis. In this consideration, it is the putative beginning stages of sporophyte development that require initial and primary attention, not all the subsequent (and also important, of course) land-plant adaptations, such as rhizoids, stomata, cuticles, nonflagellated, aerial spores, distinct organs, and, eventually, vascular tissue. Bower (1908) provided an answer (in his antithetic theory) for such sporophytic beginnings when he suggested that a delay in zygotic meiosis, with mitotic divisions of the zygote added to the life cycle (to produce a mass of cells), just prior to the occurrence (or potential occurrence) of meiosis, is primarily what would be necessary to add or intercalate a sporophyte into the life cycle. Without a great deal of morphological modification, thus, a multicellular structure (a sporophyte produced by zygotic, mitotic divisions) could be "interpolated" (Bower, 1935) in the life cycle and would presumably be retained on the gametophyte (cf. Scagel et al., 1984) as, for example, the zygote of Coleochaete is observed to be (no sporophyte, however, is developed in Coleochaete itself).
There is no supposition, or really even viable possibility, of a preexisting sporophyte in Bower's (1908, 1935) antithetic (interpolation) theory of origin of alternating generations; nor, for that matter, would such be expected based on careful scrutiny of the life cycle of Coleoehaete and other advanced charophytes (Chara, Nitella, etc.). As has been emphasized, advanced charophytes do not exhibit alternating generations; they possess only the gametophyte stage (no sporophyte is present). Thus, as Bower suggested, the sporophyte stage must have been something subsequently added to the life cycle, as an adaptation to a land-based existence. If this is so, the land-plant sporophyte is to be viewed as a structure "different" from the sporophytes of any algae. Graham (1984) made a case for the antithetic theory but, as discussed, later vacillated to an extent (1993) in the direction of the homologous theory--perhaps in response to paleobotanical evidence of a similarity of some early vascular plant gametophytes to sporophytes (see the discussion in section VI.G). The total paleobotancial evidence, though, as previously discussed, does not support either theory of alternation of generations conclusively; put another way, the existing evidence could be argued to support both theories more or less equally.
The soundest approach, perhaps, is to appreciate and understand the extensive evidence available for advanced charophyte--land-plant relationships and then to examine carefully the life cycles of these charophytes for the potential evolutionary insights they may bring (i.e., into the mechanism of origin of the land-plant sporophyte). From the understanding of morphological possibilities afforded by such life cycles, we may pursue the most logical and parsimonious path to the probable origin of a land-plant sporophyte. Only the antithetic theory (of sporophyte origin), not the homologous theory, is actually tenable given the possibilities offered by the life cycles of Nitella, Chara, and Coleochaete (as representative of forms ancestral to land plants). We also note that not only living but also fossil charophytes (of. Taylor & Taylor, 1993) show no evidence of alternating generations (do not exhibit, or left no evidence of, a sporophyte stage).
It is, thus, likely that algal ancestors of primitive embryophytes possessed only the gametophyte stage in the life cycle. The first (probably antithetically generated) sporophytes were, in all likelihood, "attempted" in amphibious habitats. Over time, intermittent periods of drying probably selected for the gene mutations and recombinations involved in incipient sporophyte development (Campbell et al., 1999), including the adaptation of desiccation-resistant, nonmotile spores. In the subsequent colonization of land by plants, a generally upgrade development (elaboration) of the sporophyte (by progressive sterilization of sporogenous tissues and subsequent diversification of vegetative tissue produced) occurred, in what were, often perhaps, increasingly drier environments. The sporophyte and gametophyte of land plants were thus cast upon different, though intimately related, courses (Niklas, 1997). Considering a possible ancestral form, perhaps represented by Coleochaete, loss of spore motility (flagellar reduction and loss) and development of a protective spore wall (with sporopollenin) probably occurred quite early in the process of land-plant adaptation (cf. Campbell et al., 1999). In the case of Chara and Nitella, these charophytic algae produce only gametes (do not produce spores); hence, spore flagellar loss need not even be contemplated in the case of these two advanced charophytes; in others words, it is possible that earliest land-plant spores were never flagellated.
B. OTHER CHAROPHYTES
Questions remain as to which charophycean algae (e.g., Coleochaete, Klebormidium, Nitella. Chara. etc.) are actually the closest, genetically, to bryophytes and other land plants (e.g., lycopods, ferns). Some mildly conflicting results are apparent in recent literature, in some cases related, perhaps, to how well the gene trees are "resolved." Regardless of the explanation, small differences in alignment of charophycean taxa are to be found. In Katana et al. (2001), based on nuclear SSU rDNA sequences, the placement of Coleochaete and Klebsormidium is somewhat closer to primitive land plants than is that of Nitella (and Chara); some of their results from chloroplast SSU rRNA genes, in fact, place Klebsormidium closer to land plants than even Coleochaete (echoed in the maximum likelihood tree based on combined data). On the other hand, the work of Nakayama et al. (1998), utilizing nuclear-encoded SSU rRNA sequences, illustrates a nice "bootstrap resolution," with Chara and Nitella more closely connected to land plants than are Coleochaete or Klebsormidium. However, there seems to be little doubt in these types of studies that charophytes (generally) are the group of algae representative of those leading to land plants (cf. Kranz et al., 1995).
Although it may eventually be determined which charophyte is "closest" to primitive land plants, in terms of molecular-genetic data, we must bear in mind that we are looking, in all these forms, at extant plants (as representatives of ancestral forms), not at the actual ancestors of land plants. As Campbell et al. (1999) pointed out, modern land plants and modern charophytes both probably evolved from a common ancestor. It is possible that this ancestor had a combination of traits that we see in modern charophyte genera; or, it may have differed, to an extent, from all of them. Given the morphology of marchantioid liverworts, Coleochaete may be the most plausible morphological counterpart of this ancestor. However, Coleochaete has flagellated spores. Nitella and Chara, as discussed, do not produce spores and are compelling to consider in a putative antithetic development of the land-plant sporophyte; spores of primitive land plants show no evidence of flagellation, or even residual flagellation. Also, the multicellular sex organs of Chara and Nitella are suggestive of land-plant gametangia (cf. Bold & Wynne, 1985; Bold et al., 1987).
An interesting phylogenetic diagram is in Lee (1999: 186), which infers that Coleochaete may be more along the path to liverworts and hornworts, than to mosses--Nitella and Chara being more aligned with mosses and vascular land plants. This diagram is consistent with Niklas's (1997) diagram and view of mosses and tracheophytes as sharing a "last common ancestor." Lee's (1999) conception is based in part on the work of Okuda and Brown (1992), showing a close relationship between the cellulose-synthesizing complex of Coleochaete scutata and that of hornworts, for example, but not those of mosses and tracheophytes--the cellulose-synthesizing complexes of mosses and tracheophytes being more similar to such complexes in Nitella and Chara.
Lee (1999) also cited McCourt's (1995) analysis of a compilation of molecular-genetic results (e.g., small and large subunit rRNA sequences and the corresponding, encoding rDNA) that indicated a more direct relationship of embryophytes (which, not specified) with the Charales (e.g., Chara, Nitella) than with Coleochaete, Klebsormidiales, or the Zygnematales. This close relationship of members of the Charales (several genera considered) and land plants was supported by further molecular-phylogenetic analyses (Karol et al., 2001). Thus, it is questionable whether, among charophytes s.1., Coleochaete is genetically as close to some groups of land plants as are characeans such as Chara, Nitella, and Tolypella. Graham and Wilcox (2000), however, maintained the position of a close relationship of the Coleochaetales with Embryophyta. Whatever the precise answer, the life cycles, morphology, and biochemistry of charophytes (s.1.) should be examined for clues to land origins. It seems reasonably certain that the origin of land plants occurred at fresh (at most, brackish) water--land interfaces, since it is very doubtful that even fossil charophytes were marine organisms (cf. Taylor & Taylor, 1993).
A review of historical and recent opinions concerning alternation of generations in land plants has led to the conclusion that there still has been no clear decision (i.e., approaching unanimity) as to which theory of land-plant sporophyte origin to accept, the homologous theory or the antithetic theory. The crux of the issue has centered on the exact mode of origin of the structure that was to eventually become the terrestrial sporophyte. The conclusions of the present article, as to land-plant sporophyte beginnings, are based on philosophical-analytic (logistic) considerations and on a corpus of evidence that has grown considerably over the past several decades; of particular interest is evidence bearing directly upon which group of algae is the most closely related to embryophytes.
Logistic (essentially a priori) assessments concern, specifically, questions of the feasibility of sporophyte origin by two (or more) different alleged processes. The homologous theory of alternation of generations presents certain problems of logical content. By definition, the homologous theory is based on the idea of homology and the belief in a fundamental similarity, or essential "equality," of gametophyte and sporophyte generations (the sporophyte being considered in this theory as basically a "transformed" gametophyte). This "similarity," in turn, is based, again almost by definition, on the observation of types of presumed precursor organisms (i.e., types of green algae, represented possibly by Ulva) with isomorphic (morphologically almost identical) gametophytes and sporophytes. The homologous theory (of land-plant sporophyte origin) thus predicts the preexistence of independent sporophytes and gametophytes in the ancestral algal life cycle and accounts for the existence of these alternating generations in embryophytes by a carryover of both generations to a primitive land-plant cycle. There is the further improbable suggestion (at least implied) of the subsequent establishment of dependency (or additional dependency) of this sporophyte (on the gametophyte); how this was supposed to have occurred has not been made clear in the literature. If, in fact, one is going to attempt to invoke preexisting sporophytes (and gametophytes) in an algal lineage leading to embryophytes, a better guess (since land-plant sporophytes are virtually always different in appearance from gametophytes) would, perhaps, be that some type of green alga with distinctly heteromorphic alternation was ancestral. However, no such heteromorphic algal candidates have been clearly suggested, and such "heteromorphism" in ancestry seems at least somewhat contrary to the concept of the fundamental similarity of generations so strongly emphasized in the homologous (transformation) theory. In any case, no specific mechanism has been proposed for the establishment of sporophyte dependency on the gametophyte, in accounts of the homologous theory, other than the following: An intercalative ontogenetic mechanism would seem to be the only plausible explanation (i.e., retention of the zygote on the gametophyte, the zygote thereupon developing in situ into an attached sporophyte) for the origin of a sporophyte that is completely or largely dependent on the gametophyte and dissimilar in form from the gametophyte. If so, then, by theory reduction, is it not the case that the homologous theory defaults at least in part to the theory with which it has been for so long in competition; that is, the antithetic (interpolation) theory?
A further problem with the homologous theory includes the presumption of (at least the likelihood of) separate origins of the sporophyte in bryophytes (s.1.) and vascular plants, both presumably taking place in the same or essentially the same environments. These environments were generally marshy, at least with abundant moisture for migration of the flagellated sperm produced by gametophytes; but intervals of drying probably occurred commonly, no doubt promoting adaptations leading to land-plant development. In any event, the point is that these early "amphibious" environments were similar! So, if logic is invoked, why expect two different patterns of sporophyte development (bryophytic and tracheophytic) to develop, in concurrence, in virtually identical environments? Also problematical in the homologous theory is that both upgrade and downgrade series of sporophyte development were "required" according to the tenets of this theory, even within one group of plants (such as the liverworts).
The antithetic theory does not presuppose the existence of a sporophyte in the initial land-plant ancestral life cycle; only the gametophyte of an alga (like Chara, for example, which is only a gametophyte) need be preexistent (i.e., be present in the land-plant ancestor). This antithetic (interpolation) theory suggests that the development of a land-plant sporophyte occurred in situ, specifically in response to an increasingly terrestrial habitat, as a "novel" innovation in land-plant existence. Suited to sounder logic than is the homologous theory, a definite ontogenetic mechanism for the antithetic theory is proposed; that is, a delay in meiosis by the zygote, during which a mass of sporogenous (or potentially sporogenous) tissue is developed (by zygotic mitosis). In this way, a small sporophyte, attached to the gametophyte, may be rather readily intercalated into the life cycle (of what was formerly primarily a gametophytic life cycle). Furthermore, independent origins of sporophytes in bryophytes and vascular plants were not necessary (i.e., are not necessary to postulate); and a primarily upgrade evolutionary development of the sporophyte ("progressing" from bryophytes through major groups of vascular plants) is envisioned, leading to eventual sporophyte dominance and independence. Arguments of parsimony lie clearly on the side of the antithetic theory; reversible phenomena on any large scale, at least, are not required. The antithetic theory thus is rather clearly the one considered more plausible given the preponderance of all of the evidence, and logic, discussed in this article. The homologous theory, by contrast, falls short on basic questions, "Was it really feasible?" and "How did it happen?" The homologous theory does not measure up well against the antithetic theory as to the precise mechanism of occurrence; if the ontogenetic mechanism of the homologous theory is in fact the same as the antithetic theory (interpolation by zygotic mitosis), then we are talking basically about the same theory (the only difference in the two theories then being whether the sporophyte originated before or after land occupancy).
In addition to questions of logical content of the above theories of land-plant origin, a range of factual information has been reviewed herein, relating (directly or indirectly) to which theory to accept. Much of the previous so-called evidence for the homologous theory does not point clearly to either the homologous or the antithetic theory as correct. This is particularly true, as discussed, of such often hailed points (allegedly supporting the homologous theory) as apogamy/ apospory and the presence of limited chlorophyll in allegedly reduced hepatophyte sporophytes (such as those of Ricciocarpus). On the other hand, such points as sporophyte dependency on the gametophyte in bryophytes and (to a lesser extent) in vascular cryptogams, the occasional reversion of vegetative to sporogenous tissue, and the usually striking dissimilarity of the gametophyte and the sporophyte in the life cycle of a given plant are seemingly more cogent arguments--and these arguments rather plainly suggest an antithetic origin of sporophytes of land plants. Paleobotanical evidence, though interesting and still promising, has not yet offered a definitive verdict on sporophyte origins (i.e., on theory preference). Continued paleobotanical investigation of the very earliest land plants could perhaps eventually provide valuable data and insights.
The strongest evidence at present (and the beginnings of resolution of the question of the initiation of alternating generations in land plants) comes to light when we understand which algal group, generally, is most probable as representative of forms involved in embryophyte origin. As has been reviewed rather thoroughly in this manuscript, strong evidence from morphology, cytology (particularly mitosis/cytokinesis and ultrastructure of motile cells), biochemical evidence (e.g., glycolate oxidase), and a variety of molecular-genetic data (often based on rRNA) supports Charophytes, particularly "advanced" forms (Coleochaete. Chara, Nitella, etc.), as the group of algae most related to land plants. Since these advanced charophyceans are fundamentally gametophytic organisms, with simple, haplontic life cycles (and zygotic meiosis)--no sporophyte being present in their life cycles--we should strongly suspect that the sporophyte was developed (in very early land plants originating from charophyte-like predecessors) antithetically, as an addition to the life cycle (by mitosis of a zygote retained on the gametophyte) and in response to newly available terrestrial (at least amphibious) habitats. This being true, the "interpolated" land-plant sporophyte has no palpable connection to the sporophyte per se of any known alga; however, the sporophytes of bryophytes and tracheophytes are evolutionarily related to each other, an idea recently endorsed by a number of authors. An understanding of the life cycle of advanced members of the Charophyta is thus key to selection of the antithetic (interpolation) theory of land sporophyte origin over the homologous (transformation) theory. Although Charophytes have clearly been shown to be the algal group most closely related to land plants, opinions still differ as to exactly which Charophytes are most representative of embryophyte ancestors.
I wish to thank Dr. Robert R. Haynes for reading this manuscript prior to review, and for his helpful suggestions.
XI. Literature Cited
Al-Houty, 17. A. A. & P. J. Syrett. 1984. The occurrence of urease/urea amidolyase and glycolate oxidase/ dehydrogenase in Klebsormidium spp. and members of the Ulotrichales. Brit. Phycol. J. 19: 1-10.
Arber, A. R. 1950. The natural philosophy of plant form. Cambridge Univ. Press, Cambridge.
Beech, P. L., K. Heimann & M. Melkonian. 1991. Development of the flagellar apparatus during the cell cycle in unicellular algae. Protoplasma 164: 23-37.
Blackwell, W. H. & M. J. Powell. 1995. Where have all the algae gone, or, how many kingdoms are there? Amer. Biol. Teacher 57: 160-167.
--. 1999. Reconciling kingdoms with codes of nomenclature: Is it necessary? Syst. Biol. 48: 406-412.
--. 2000. A review of group fililation of stramenopiles, additional approaches to the question. Evol. Theory & Rev. 12(3): 49-88.
Bold, H. C. 1948. The occurrence of chlorophyll in the sporophyte of Ricciocarpus natans. Amer. J. Bot. 35: 440-443.
--. 1957. Morphology of plants. Harper & Brothers, New York.
-- & M. J. Wynne. 1978. Introduction to the algae: Structure and reproduction. Prentice-Hall, Englewood Cliffs, NJ.
-- & --. 1985. Introduction to the algae: Structure and reproduction. Ed. 2. Prentice-Hall, Upper Saddle River, NJ.
--, C. J. Alexopoulos & T. Delevoryas. 1980. Morphology of plants and fungi. Ed. 4. Harper & Row, New York.
--, -- & --. 1987. Morphology of plants and fungi. Ed. 5. Harper & Row, New York.
Bower, F. O. 1908. The origin of a ]and flora. Macmillan, London.
--. 1935. Primitive land plants. Macmillan, London.
Bremer, K. 1985. Summary of green plant phylogeny and classification. Cladistics 1: 369-385.
--. 1986. Systematics of the green algae (Irvine, D.E.G. & John, D.M. (Eds.), a review. Cladistics 2: 378-381.
-- & H. E. Wanntorp. 1981a. A cladistic classification of green plants. Nor& J. Bot. 1: 1-3.
-- & --. 1981b. The cladistic approach to plant classification. Pp. 87-94 in V. Funk & D. R. Brooks (eds.), Advances in cladistics: Proceedings of the first meeting of the Willi Hennig Society. New York Botanical Garden, Bronx.
--, C. J. Humphries, B. D. Mishler & S. P. Churchill. 1987. On cladistic relationships in green plants. Taxon 36: 339-349.
Brown, W. H. 1935. The plant kingdom: A textbook of general botany. Ginn & Company, Boston.
Burns, G. W. 1974. The plant kingdom. Macmillan, New York.
Campbell, D. H. 1940. The evolution of land plants (Embryophyta). Stanford Univ. Press, Palo Alto, CA.
Campbell, N. A., J. B. Reece & L. G. Mitchell. 1999. Biology. Ed. 5. Benjamin Cummings, CA.
Cavalier-Smith, T. 1981. Eukaryote kingdoms, seven or nine? BioSystems 14: 461-481.
Church, A. H. 1919. Thalassiophyta and the subaerial transmigration. Botanical Memoirs, 3. Oxford Univ. Press, Oxford.
Croft, A. 1952. A new Trochiliscus (Charophyta) from the Downtownian of Podolia. Bull. Brit. Mus. (Nat. Hist.), Geol. 1: 187-220.
Cronquist, A. 1961. Introductory botany. Harper & Row, New York.
Delevoryas, T. 1977. Plant diversification. Ed. 2. Holt, Rinehart & Winston, New York.
Ditmer, H. J. 1964. Phylogeny and form in the plant kingdom. Van Nostrand, Toronto.
Dodge, J. D. 1973. The fine structure of algal cells. Academic Press, London.
Eames, A. J. 1936. Morphology of vascular plants: Lower groups (Psilophytales to Filicales). McGraw-Hill, New York.
Foster, A. S. & E. M. Gifford. 1959. Comparative morphology of vascular plants. W. H. Freeman, San Francisco.
Frederick, S. E., P. J. Gruber & N. E. Tolbert. 1973. The occurrence of glycolate dehydrogenase and glycolate oxidase in green plants: An evolutionary survey. Pl Physiol. (Lancaster) 52:318-323.
Fritsch, F. E. 1916. The algal ancestry of higher plants. New Phytol. 15: 233-250.
--. 1935. The structure and reproduction of the algae. Vol. 1. Cambridge Univ. Press, Cambridge.
Gifford, E. M. & A. S. Foster. 1989. Morphology and evolution of vascular plants. Ed. 3. W. H. Freeman, New York.
Goebel, K. 1930. Organographie der Pflanzen. Ed. 3. G. Fischer, Jena, Germany.
Graham, L. E. 1982. The occurrence, evolution, and phylogenetic significance of parenchyma in Coleochaete Breb. (Chlorophyta). Amer. J. Bot. 69: 447-454.
--. 1984. Coleochaete and the origin of land plants. Amer. J. Bot. 71: 603-608.
--. 1985. The origin of the life cycle of land plants. Amer. Scientist 73: 178-186.
--. 1993. Origin of land plants. Wiley & Sons, New York.
-- & L. W. Wilcox. 1983. The occurrence and phylogenetic significance of putative placental transfer cells in the green alga Coleochaete. Amer. J. Bot. 70, 113-120.
-- & --. 2000. Algae. Prentice Hall, Upper Saddle River, NJ.
Gray, J. 1985. The microfossil record of early land plants: Advances in understanding of early terrestrialization. Pp. 167 195 in W. G. Chaloner & J. D. Lawson (eds.), Evolution and environment in the late Silurian and early Devonian. Phil. Trans. Roy. Soc. London, B, 309.
Green, P. B. 1962. Cell expansion. Pp. 625-632 in R. A. Lewin (ed.), Physiology and biochemistry of algae. Academic Press, New York & London.
Groover, R. D. & H. C. Bold. 1969. The taxonomy and comparative physiology of the Chlorosarcinales, and certain other edaphic algae. Phycological Studies, 8. Univ. of Texas, Austin.
Haberlandt, G. 1914. Physiological plant anatomy. M. Drummond, trans. Macmillan, London.
Haupt, A. W. 1953. Plant morphology. McGraw-Hill, New York.
Hofmeister, W. 1851. Vergleichende Untersuchgungen. Friedrich Hofmeister, Leipzig, Germany.
Jeffrey, C. 1962. The origin and differentiation of the archegoniate land-plants. Bot. Not. 115: 446-454.
Kaplan, D. R. 2001. The science of plant morphology: Definition, history, and role in modern biology. Amer. J. Bot. 88: 1711-1741.
Karol, K. G., R. M. McCourt, M. T. Cimino & C. F. Delwiche. 2001. The closest living relatives of land plants. Science 294: 2351-2353.
Katana, A., J. Kwiatowski, K. Spalik & B. Zakrys. 2001. Phylogenetic position of Koliella (Chlorophyta) as interred from nuclear and chloroplast small subunit rDNA. I. Phycol. 37: 443-451.
Kranz, H. D., D. Miks, M. L. Siegler, I. Capesius, C. W. Sensen & V. A. R. Huss. 1995. The origin of land plants: Phylogenetic relationship among charophytes, bryophytes, and vascular plants inferred from complete small-subunit ribosomal RNA sequences. J. Molec. Evol. 41: 74-84.
Lee, R. E. 1999. Phycology. Ed. 3. Cambridge Univ. Press, Cambridge.
Mandal, D. K. & S. Ray. 2001. Karyotype analysis and cytotaxonomic study in the genus Nitella (Charophyceae). Phytomorphology 51: 27-32.
Manhart, J. R. & J. D. Palmer. 1990. The gain of two chloroplast tRNA introns marks the green algal ancestors of land plants. Nature 345: 268-270.
Marchant, H. J. & J. D. Pickett-Heaps. 1973. Mitosis and cytokinesis in Coleochaete scutata. J. Phycol. 9: 461-471.
Mattox, K. R. & K. D. Stewart. 1984. Classification of the green algae: A concept based on comparative cytology. Pp. 29-72 in D. E. G. Irvine & D. M. John (eds.), Systematics of the green algae. Academic Press, London & Orlando.
McCourt, R. M. 1995. Green algal phylogeny. Trends Ecol. Evol. 10:159-163.
Melkonian, M. 1982. Structural and evolutionary aspects of the flagellar apparatus in green algae and land plants. Taxon 31: 255-265.
Minkoff, E. C. 1983. Evolutionary biology. Addison Wesley, Reading, MA.
Mishler, B. D. & S. P. Churchill. 1984. A cladistic approach to the phylogeny of the "bryophytes." Brittonia 36: 406-424.
-- & --. 1985. Transition to a land flora: Phylogenetic relationships of the green algae and bryophytes. Cladistics 1: 305-328.
Nakayama, T., B. Marin, H. D. Kranz, B. Surek, I. Inouye & M. Melkonian. 1998. The basal position of scaly green flagellates among the green algae (Chlorophyta) is revealed by analyses of nuclear-encoded SSU rRNA sequences. Protist 149: 367-380.
Niklas, K. J. 1992. Plant biomechanics: An engineering approach to plant form and function. Univ. of Chicago Press, Chicago & London.
--. 1994. Plant allometry: The scaling of form and process. Univ. of Chicago Press, Chicago & London.
--. 1997. The evolutionary biology of plants. Univ. of Chicago Press, Chicago & London.
Okuda, K. & R. M. Brown Jr. 1992. A new putative cellulose-synthesizing complex of Coleochaete scutata. Protoplasma 168: 51-53.
Pickett-Heaps, J. D. 1975. Green algae: Structure, reproduction and evolution in selected genera. Sinauer Associates, Sunderland, MA.
--. 1976. Cell division in eukaryotic algae. BioScience 26: 445-450.
Prescott, G. W. 1968. The algae: A review. Houghton-Mifflin, Boston.
Pringsheim, N. 1878. Ueber Sprossung der Moosfruchte und den Generationswechsel der Thallophyten. Jahrb. Wiss. Bot. 11: 1-46.
Pritchard, 14. N. & P. T. Bradt. 1984. Biology of nonvascular plants. Times Mirror/Mosby, Saint Louis.
Purves, W. K., G. H. Orians, H. C. Heller & D. Savada. 1998. Life: The science of biology. Ed. 5. Sinauer, Sunderland, MA.
Remy, W. & H. Hass. 1986. Das Ur--Landpflanzen-Konzep--unter besonderer Berucksichtingung der Organization Altdevonisher Gametophyten. Argum. Palaeobot. 7: 173-214.
-- & R. Remy. 1980. Devonian gametophytes with anatomically preserved gametangia. Science 208: 295-296.
Renzaglia, K. S. & K. C. Vaughn. 2000. Anatomy, development and classification of hornworts. Pp. 1-20 in A. H. Shaw & B. Goffinet (eds.), Bryophyte biology. Cambridge Univ. Press, Cambridge.
Ruse, M. 1988. Philosophy of biology today. State Univ. Press of New York, Albany.
Sattler, R. 1986. Biophilosophy: Analytic and holistic perspectives. Springer-Verlag, Berlin.
--. 1998. On the origin of symmetry, branching and phyllotaxis in land plants. Pp. 775 793 in R. V. Jean & D. Barabe (eds.), Symmetry in plants. World Scientific, Singapore.
Scagel, R. E, R. J. Bandoni, J. R. Maze, G. E. Rouse, W. B. Schofield & J. R. Stein. 1984. Plants: An evolutionary survey. Wadsworth, Belmont, CA.
Schofield, W. B. 1985. Introduction to bryology. Macmillan, New York.
Silva, P. C., K. R. Mattox & W. H. Blackwell. 1972. The generic name Hormidium as applied to green algae. Taxon 21: 639-345.
Sluiman, H. J. 1985. A cladistic evaluation of the lower and higher green plants. Plant Syst. Evol. 149: 217-232.
Smith, G. M. 1938. Cryptogamic botany. Vol. 2. McGraw-Hill, New York & London.
--. 1950. The fresh-water algae of the United States. Ed. 2. McGraw-Hill, New York.
Solomon, E. P., L. R. Berg & D. W. Martin. 2002. Biology. Ed. 6. Brooks/Cole, South Melbourne, Australia.
South, G. R. & A. Whittick. 1987. Introduction to phycology. Blackwell Scientific, Oxford.
Sporne, K. R. 1965. The morphology of gymnosperms: The structure and evolution of primitive seedplants. Hutchinson, London.
Stewart, K. D. & K. R. Mattox. 1975. Comparative cytology, evolution and classification of the green algae with some consideration of the origin of other organisms with chlorophylls a & b. Bot. Rev. (Lancaster) 41: 104-135.
Stuessy, T. F. 1990. Plant taxonomy: The systematic evaluation of comparative data. Columbia Univ. Press, New York.
Syrett, P. J. & E A. A. Al-Houty. 1984. The phylogenetic significance of the occurrence of urease/urea amidolyase and glycolate oxidase/glycolate dehydrogenase in green algae. Brit. Phycol. J. 19: 11-21.
Taylor, T, N. & E. L. Taylor. 1993. The biology and evolution of fossil plants. Prentice Hail, Englewood Cliffs, NJ.
Theriot, E. 1988. A review of Sluiman's cladistic classification of green plants with particular reference to flagella data and to land plant origins. Taxon 37:913-919.
Tippo, O. & W. L. Stern. 1977. Humanistic botany. W. W. Norton, New York.
Van den Hoek, C., D. G. Mann & H. M. Jahns. 1995. Algae: An introduction to phycology. Cambridge Univ. Press, Cambridge.
Wardlaw, C. W. 1952. Phylogeny and morphogenesis: Contemporary aspects of botanical science. Macmillan, London.
--. 1955. Embryogenesis in plants. Methuen, London.
--. 1968. Essays on form in plants. Manchester Univ. Press, Manchester.
Wolfe, S. L. 1983. Introduction to cell biology. Wadsworth, Belmont, CA.
Zimmerman, W. 1930. Phylogenie der Pflanzen. G. Fischer, Jena, Germany.
--. 1952. Main results of the telome theory. Palaeobotanist 1: 456-470.
WILL H. BLACKWELL
Department of Biological Sciences
University of Alabama
Tuscaloosa, Alabama 35487, U.S.A.