Morphological Traffic between the Inflorescence and the Vegetative Shoot in Helobial Monocotyledons.
Previous studies of reproductive structures in the helobial monocotyledons (Alismatidae) indicate that partitioning between flower and inflorescence is not always clear (e.g., Lilaea, Scheuchzeria) and that this may be the result of ancestral, unisexual modules coming together to form flowers and/or inflorescences. Later evolutionary changes may have included the inflorescence becoming involved or mixed in with vegetative growth. Substitution of vegetative buds for flowers is the simplest version, and there can be additional modifications to the growth behavior of the inflorescence, such as horizontal growth and dorsiventrality. In the Alismataceae and Limnocharitaceae the derivation of stolonlike structures from inflorescences is obvious: vegetative features have been incorporated into structures that are recognizably inflorescences. In the Hydrocharitaceae the interrelationships between the inflorescence and the vegetative body are much less well defined. We previously suggested for Hydrocharis, where a si ngle axillary complex can contain both inflorescence and stolons, that the stolon is basically a sterilized inflorescence and that features of the inflorescence have become incorporated into the vegetative body. Here we will explore this theme further for the Hydrocharitaceae, using information from within and outside the family.
In an earlier review contribution (Posluszny & Charlton, 1993), we attempted to bring together and integrate old and new suggestions on floral evolution in helobial monocotyledons. We concluded that the floral structures in helobial monocotyledons might have evolved from a primitive multiaxial reproductive structure that became modified in different ways. We also suggested that, in the lines that led to the tepaloid families (i.e., Potamogetonaceae, Juncaginaceae, etc.), the lateral axes became more flowerlike and the main axis became more inforescencelike. However, the "flowers" still show a mixture of flowerlike and inflorescencelike features: the common stamen-tepal superposition resembles an axillary relationship and can be considered more akin to "inflorescence" than to "flower"; the gynoecium is made up simply of carpels inserted on the floral axis and is consequently just like that of a conventional flower; the phyllotactic sequence of the "flower" is nevertheless continuous through the initiation of the stamen-tepal associations and the gynoecium. The "inflorescence" also has some characters that are more usual in flowers; it often bears lateral structures (the "flowers") directly without subtending foliar organs; the phyllotaxis is often paired or whorled; and there is frequently a residual terminal meristem. In the "petaloid" families (Alismataceae, Hydrocharitaceae, etc.) the flower itself is quite clearly defined but is always carried on a distinctive inflorescence; and we suggested for these groups that the hypothetical original multiaxial reproductive structure had become differentiated into "flower" in its distal regions and "inflorescence" in its proximal region. We concluded that the archetypal inflorescence was one with two subopposite bracts and a terminal flower, a pattern discernible in many cases in Hydrocharitaceae, in Limnocharitaceae, and in Ranalisma in the Alismataceae. Inflorescences of most of the Alismataceae have bracts in sets of three forming pseudowhorls (i.e., a set of leaves t hat appear to form a whorl but are initiated sequentially). Generally there are several pseudowhorls; this appears to be a further derived condition in which the inflorescence has extended growth. Within the "flower" residua of the original multiaxial system are detectable in early floral development as CA primordia from which a petal and a number of stamens arise in many Alismataceae s/1., or as primary androecial primordia in Limnocharis. We concluded that, in the helobial monocotyledons as a whole, inflorescence bracts, sepals, and tepals are homologous structures derived from the phyllomic appendages of the original multiaxial reproductive structure and, therefore, that the divergence between the petaloid and tepaloid groups involved divergence of the relationships between phyllome and subtended structure, at the same time as the axes of the original multiaxial structure became differentiated into "flower" and "inflorescence." Finally, in a number of helobial families (e.g., Najadaceae, Cymodoceaceae, and Zannichelliaceae) reproductive structures are so reduced that morphological approaches to patterns of evolution founder for lack of information.
Although these comments on structural evolution will probably remain no more than speculations, since we wrote our review in 1991 advances have been made in understanding the interrelationships and phylogeny of the helobial monocotyledons, and it has become evident that the morphological speculations conform to the new data. One advance was made on the basis of classical anatomical studies: seed-coat structures (Shaffer-Fehre, 1991a, 1991b) showed that Naias is closely related to the Hydrocharitaceae, which supports the idea that its reproductive structures are extremely reduced. A synthesis of cladistic analyses (Les & Haynes, 1995), including data from molecular studies (rbcL gene sequence data) indicates that the helobial monocotyledons are a monophyletic group, derived from within the monocotyledon lineage, and that the "tepaloid" and "petaloid" groups represent a fundamental and early divergence. Most of the families with extremely reduced floral structures are associated with the tepaloid groups, but t he overall assessment confirms the association of Naias with the Hydrocharitaceae.
Sculthorpe (1967) gives numerous examples among aquatic plants of what he calls "pseudo-vivipary," in which sexual structures are replaced by vegetative propagules. A number of clear instances of this exist among the helobial monocotyledons, particularly in the Alismataceae and Limnocharitaceae. In these families it is quite common to find that the development of a cymose cluster of flowers in the axil of a bract is terminated by the formation of a vegetative bud in place of the last flower, or in some cases a vegetative bud occurs directly in the bract axil. However, there are instances among the helobial monocotyledons, in particular in the Hydrocharitaceae, in which the boundary between vegetative growth and reproductive growth has become quite unclear, so that features of the reproductive shoot may have been incorporated into the vegetative developmental pattern of the plant. These are the cases we shall address here. In itself this is not a new topic: Miki (1937) speculated on the equivalence of the pla nt body in Najas to the stolon in Hydrocharis, and this type of consideration appeared again later, in Wilder (1975). We acknowledge also that a number of studies have been published on the organization of specific members of the Hydrocharitaceae (e.g., Brunaud, 1976, 1977; Bugnon & Joffrin, 1962, 1963; Wilder, 1974b, 1974c, 1975), and we ourselves have contributed to the field (Posluszny & Tomlinson, 1991; Posluszny & Charlton, 1999). However, we believe that further studies of these interesting plants can add to understanding of the possibilities for morphological evolution of inflorescences. Of course, once one begins to visualize intermixing of inflorescence and vegetative characters it may become difficult to keep track of what is meant by any particular term, particularly "inflorescence." As far as the Hydrocharitaceae are concerned, when we use the term directly we are using it in the sense of Kaul (1970), and we mean a structure that has a long first internode surmounted by one or more bracts and that bears one or more flowers. Other structures may be considered to be derived or possibly derived from inflorescences, but we assume that the distinction between these and what are at present defined as inflorescences will be kept in mind.
III. Materials and Methods
Plants of Limnobium spongia (Bosc) Richard (Hydrocharitaceae) were obtained from a commercial source and grown in a pond in the second author's garden. Material of Luronium natans (Alismataceae) was obtained from plants grown in about 15 cm of water in a container in the first author's garden. This material was originally derived from a local canal (Charlton, 1973). Plants of Stratiotes aloides (Hydrocharitaceae) were obtained from the University of Liverpool Botanic Garden. Vallisneria americana Michx. (Hydrocharitaceae) was collected by Gordon Lemon (currently a graduate student with the second author) from the old Welland Canal (now known as the Recreational Waterway), in the town of Welland, Ontario, Canada.
All material was fixed in FAA fixative and processed for epi-illumination microscopy, and in the case of Stratiotes for wax sectioning, by standard methods (Charlton & others, 1989; Johansen, 1940). Buds of Stratiotes were sectioned transversely at 15 [micro]m.
A. LURONIUM NATANS
This member of the Alismataceae provides an interesting illustration of the way in which vegetative features can be incorporated into an inflorescence as a regular feature of development. The plant is usually rooted beneath the water, with ovate floating or emergent leaves in summer. The inflorescence itself generally floats below the water surface, or it may grow over the surface of the substrate and root, and the flowers emerge into the air above. As in most Alismataceae it bears bracts in pseudowhorls of three, and all bracts (except the very first) subtend lateral buds. Arber noted in 1920 that, in each set of three lateral buds, two were flowers and one was a vegetative bud. This patterned distribution was confirmed by Charlton (1973), who noted that the first two bracts in each pseudowhorl subtend flowers while the third subtends a vegetative bud and that under certain conditions the flower buds abort so that the inforescence is effectively solely vegetative. Details of development have not previously been studied. As in other alismads the vegetative shoot apex bifurcates to form the inflorescence and a large vegetative bud, the latter continuing the growth of the vegetative shoot (Fig. 1). The three bracts of each pseudowhorl are conjoined, and this gives the inflorescence bud a rather flowerlike appearance, particularly when the lateral buds have become prominent (Fig. 2). The first two bracts of each pseudowhorl arise close together in time, but the inflorescence apex appears to bifurcate unequally after each bract has been initiated to form a large bud in the axil of the bract (Figs. 3-5). Each set of bracts overgrows the lateral buds and the inflorescence apex (Fig. 6). Although the buds that give rise to them appear to be similar in their early stages (Figs. 3-5), the two floral buds and single vegetative bud in each set become detectably different very early on (Fig. 7); in fact, even the prophylls of floral and vegetative buds are rather different. Floral prophylls tend to have a single major upgro wth at one side of the flower (Fig. 7), whereas prophylls of vegetative buds are more symmetrical and tend to have two main points of upgrowth (Figs. 7, 8).
B. STRATIOTES ALOIDES
This is a floating stoloniferous aquatic with a rosette of leaves that partly emerge from the water during the summer. In the United Kingdom all plants appear to be female, and male flowers only occur as an exception. The female inflorescence has two subopposite bracts and a single terminal flower, which is aerial and rather showy.
Unlike other Hydrocharitaceae that have been investigated in detail, Stratiotes has at first sight a rather irregular distribution of axillary buds in the vegetative rosette (Fig. 9). When the spacing between buds was examined in sectioned material and dissected material it was found that almost invariably (68 times out of 69) the buds occurred an even number of leaves apart along the genetic spiral, and distances of 4-34 nodes were found, with 6 and 8 the most frequent spacing. The phyllotaxis can be seen as 1+2 or 2+3; consequently, when the buds are an even number of nodes apart they occur in the same "2" parastichy in the phyllotaxis.
In Stratiotes the lateral buds give rise to axillary complexes that always form more than one stolon and may also form an inflorescence. Early development of the axillary complex seems to be the same in flowering and nonflowering plants. The buds arise high on the apical meristem as quite large outgrowths and appear to be subtended quite symmetrically by a leaf (Fig. 10). The bud seems to grow asymmetrically upward at first (Fig. 11), and then the lower part of the bud also begins to bulge (Fig. 12), so the bud develops into two components (Fig. 13). The taller component is always at the anodic side of the complex (i.e., the side facing up the genetic spiral of the parent shoot). Subsequent development differs in flowering and nonflowering plants.
In flowering plants the taller component develops into an inflorescence, and the shorter one develops into what we call a "stolon complex," which gives rise to a succession of stolons. The taller component initiates a first bract on the side away from the stolon complex (Fig. 13) and then a second bract opposite the first (Fig. 14); following this, floral organs are formed (Fig. 15). The bracts develop a somewhat hooded appearance and are strongly keeled at maturity. The stolon complex initiates a foliar structure at the side opposite the inflorescence and then bifurcates unequally, so that a small bud appears in the axil of the foliar structure and the larger bud develops into a stolon (Fig. 16). The small bud repeats the pattern of its parent, bifurcating again into a larger part, which becomes another stolon, and a smaller part, which repeats the process at least once more (Fig. 16). The first stolon in a stolon complex normally forms two subopposite scale leaves (Fig. 17), which often rather resemble the bracts of the inflorescence (Fig. 18). Subsequent stolons in a complex usually do not initiate their first leaves in anything approaching a paired arrangement (Figs. 16, 19).
In sterile plants, the shorter component of the axillary complex invariably develops into a stolon complex (Figs. 20-23). In only one case the stolon complex developed with its first foliar structure inserted at the side adjacent to the taller complex (Fig. 24). The taller component may develop into a single stolon (Fig. 21) or a stolon complex (Figs. 22,23). The symmetry of the whole axillary complex is always the same as in fertile plants, with the taller component at the anodic side of the bud.
C. LIMNOBIUM SPONGLA
Known as American frogbit, this is a floating stoloniferous aquatic that, unlike its European counterpart, Hydrocharis norsus-ranae, does not form overwintering turions and therefore cannot survive in lakes that freeze during winter. As far as we know, L. spongia is monoecious, forming either male or female inflorescences on fertile ramets. Branching patterns and the development of stolons and stolon complexes are very similar to H. morsusranae (Posluszny & Charlton, 1999), although the variability of handedness (clockwise versus counterclockwise) of leaf formation in the rosettelike ramets seems greater in L. spongia. In sterile ramets the axillary complex develops into a stolon and two stolon complexes, on either side (Fig. 25). The formation of three subunits in the axil seems consistent in both Limnobium and Hydrocharis and involves an initial bifurcation of an axillary primordium. The uppermost or anodic portion of the bifurcated primordium becomes a stolon complex (similar to that described for Stratio tes), while the other, cathodic portion of the primordium bifurcates again, forming the stolon in the middle and another stolon complex in the position farthest from the next youngest leaf primordium. When the ramet becomes fertile, the stolon complex on the anodic side of the axillary complex is replaced by either a male or a female inflorescence (Figs. 26, 27). In the inflorescence the sheathing scale leaves in the stolon complexes are replaced by two bracts; again, very similar to what was described above for Stratiotes. The male inflorescence forms a small cyme of 3-6 flowers and usually produces a stolon complex at the last bifurcation (Fig. 28). The female inflorescence usually comprises a single flower within two sheathing bracts and has also been observed to produce a stolon complex at the base. Similar observations were made for H. morsus-ranae (Posluszny & Charlton, 1999).
D. VALLISNERIA AMERICANA
This is a rooted member of the Hydrocharitaceae found in rivers, lakes, and ponds, usually in deep water. It has even been reported as the dominant macrophyte in freshwater tidal estuaries such as the Chesapeake Bay (Wigand & Stevenson, 1997). The unusual and highly specialized mode of pollination in this dioecious plant has been much studied and extensively reported in the literature (see Sculthorpe, 1967: 306-307). In its vegetative state stolons are rarely produced. The plant grows like a rosette, producing long, straplike leaves that reach lengths of 50 cm or more. Toward midsummer, stolons are produced that grow out horizontally under the substrate and replace the axillary buds usually found in leaf axils (Fig. 29). The rosettes that form from these stolons tend to be reproductive, producing either male or female inflorescences. In the axil of every second leaf an axillary complex develops. The initial primordium of the axillary complex elongates and bifurcates, with the anodic portion of the primordium (closest to the next youngest leaf in the spiral) becoming an inflorescence (Figs. 30-32). The cathodic portion of the initial axillary complex primordium elongates and bifurcates again; and once again the anodic portion of the primordium (closest to the initial inflorescence) develops into an inflorescence (Figs. 31-33). This process of elongation and bifurcation can produce up to four inflorescences, though usually only two are produced (Figs. 33, 34). In all of these bifurcations, no subtending appendage has been observed. The final product of this series of bifurcations is a stolon, usually one (Fig. 33), though in one instance there seemed to be one more bifurcation after the stolon (Fig. 34). Both the male and female inflorescences form two bracts, which develop at a relatively even rate and tend to meet at the apex of the inflorescence (Fig. 30). The female inflorescence is composed of a solitary flower, with no evidence of further bifurcations or ancillary stolons associated with it. The male inflore scence is made up of hundreds of flowers that will eventually break off and float to the surface. Once again, we found no evidence of stolon or inflorescence remnants within the male inflorescence. The stolons, like the stolons described for the other species above, form two subopposite, sheathing scale leaves that protect the developing rosette of leaf primordia. We noted that the stolons produced in the reproductive rosettes turn down and grow into the substrate, acting as overwintering turions.
Our primary interest in this article is the Hydrocharitaceae. However, we included the alismataceous Luronium because, taken together with some other processes that have been found in the family, it offers a complementary view of possible evolutionary occurrences in the Hydrocharitaceae.
While Luronium is only one example among many in the closely related families Alismataceae and Limnocharitaceae, in which vegetative buds occur within the inflorescence, it differs from almost all the others that have been described (Charlton, 1973; Charlton & Ahmed, 1973; Wilder, 1974a) in two ways. The mixed nature of the inflorescence is constant in Luronium, while in most other instances vegetative buds occur only in submerged inflorescences. The vegetative buds occur in a strictly patterned distribution, inserted directly in bract axils, and in this they differ from all other cases except the pseudostolons (sterilized horizontal inflorescences) of Echinodorus tenellus and its allies. Together these features put Luronium into a situation in which the inflorescence, a strongly patterned shoot, could actually be the main framework of the plant. Seen in this light, it would give rise to flowers and to leaf-bearing lateral shoots, which would further form inflorescences in the branching system. We present Lu ronium in this light to show that it is not at all far-fetched to suggest that in other taxa, such as some Hydrocharitaceae, the main framework of the plant body is derived from an inflorescence.
Also in the Alismataceae and the related Limnocharitaceae it is common to find cymose clusters of flowers in which the cymose development is terminated by the formation of a vegetative bud (Charlton, 1973; Wilder, 1974a). In Hydrocleis nymphoides this vegetative bud was thought to form one foliar structure and then immediately bifurcate to form another inflorescence, which continued the stolonlike growth of the plant, and a vegetative bud that formed a new plant (Charlton & Ahmed, 1973). In Ranalisma humile one of the two bracts of the inflorescence subtends an axillary complex consisting of an inflorescence branch and a vegetative bud; and Charlton and Ahmed (1973) considered this situation to have arisen from the replacement of a flower or floral cluster in the bract axil by a vegetative bud that behaved in the same way as the vegetative bud in Hydrocleis. These cases then illustrate how a situation can easily be developed in which an inflorescence can appear in an unusual situation, either within a cymose cluster of flowers or even in place of one; and again this can reflect on some of the arrangements that may be found in the Hydrocharitaceae.
Coming back to the Hydrocharitaceae, a number of detailed analyses have been conducted of the architecture of the shoot system, particularly of freshwater representatives such as Limnobium, Hydrocharis, Elodea, Vallisneria, and Stratiotes (e.g., Brunaud, 1976, 1977; Bugnon & Joffrin, 1962, 1963; Posluszny & Charlton, 1999; Wilder, 1974b, 1974c, 1975). With the exception of Wilder and ourselves, these authors based their conclusions on serial sections of shoots, in which it is very difficult to be sure precisely how structures are related at initiation. For example, Bugnon and Joffrin, and Brunaud, saw the first foliar structure initiated on what we call a "stolon complex" as a prophyll of the whole axillary complex: but in material studied by dissection it is perfectly clear that the first foliar structure is initiated on the stolon complex itself. It could still be argued that the foliar structure is a prophyll of the whole axillary complex displaced onto the stolon complex, which these workers in fact also consider to be the first bract of the inflorescence. However, the curious case in Stratiotes (Fig. 16), where the first foliar structure occurred between the stolon complex and the rest of the axillary complex, probably militates against this argument. Wilder used dissection techniques but worked with fresh material, which is very difficult to stain and visualize adequately. Elodea has relatively simple shoot architecture but is probably reduced: in the other genera, all of which are stoloniferous, the most difficult problems are the interpretation of the complex axillary buds found in the vegetative rosette and in some cases in the stolons, in particular the interpretation of the relationship among the stolon, stolon-containing structures, and the inflorescence.
In Hydrocharis, discounting events in the inflorescence proper, we concluded (Posluszny & Charlton, 1999) that there are basically two forms of axillary complex. In one form, during the early development of the axillary complex, the initial outgrowth in a leaf axil undergoes one bifurcation before any foliar structures are formed: the product at the anodic side of the leaf axil develops into a stolon complex, and that toward the cathodic side develops into a single stolon. In the other form there are two successive bifurcations, each before any foliar structures are formed. The first bifurcation produces one component toward the anodic side of the axil, which develops into an inflorescence, and another component, which bifurcates again. The second bifurcation, in the same plane, produces a large component, which occupies a central position in the whole axillary complex and develops into a stolon, and a smaller component toward the cathodic side of the leaf axil, which develops into a stolon complex. We concl uded that the inflorescence was positionally homologous to the stolon complex. Putting earlier information from other stoloniferous Hydrocharitaceae (e.g., Brunaud, 1976, 1977; Bugnon & Joffrin, 1962, 1963; Wilder, 1974b, 1974c) together with that for Hydrocharis, we concluded that the axillary complex, in its present form in the stoloniferous Hydrocharitaceae, developed as a meristem that was destined to develop as a stolon, but before it did so would bifurcate once or twice, occasionally more, and that the lateral components of these bifurcations could develop as either of the positional homologues stolon complex or inflorescence. We noted that, if this interpretation was correct, if the axillary complex meristem failed to bifurcate it would be expected to develop directly into a stolon. This had been reported in Vallisneria (Bugnon & Joffrin, 1962; Wilder, 1974c).
Our observations on Limnobium and Vallisneria fit in with the same general interpretation of the relationships within the axillary complex, although the absence of stolon complexes in Vallisneria is notable. It is possible that the numerous inflorescences that form in Vallisneria are replacing what might have been stolon complexes. Stratiotes, however, does not seem to fit the framework exactly, because the main component of the axillary complex does not seem to be destined to become a stolon. In contrast, in Stratiotes what seems to be the main component of the axillary complex develops directly into an inflorescence rather than a stolon in flowering plants of Stratiotes. In nonflowering plants it develops into a stolon complex or, sometimes, directly into a stolon. Nevertheless, it demonstrates positional homology of inflorescence and stolon complex, since the main component of the axillary complex can develop into either. It also demonstrates something that the other cases do not: direct positional homolo gy of stolon and inflorescence.
Based on developmental information, there are therefore two rather different aspects of homology involving the inflorescence and vegetative structures in stoloniferous Hydrocharitaceae. First is the suspicion, voiced most clearly by Wilder (1975) and ourselves (1999), that the stolon itself represents a sterilized inflorescence. The general resemblance between the two structures is certainly strong. The archetypal inflorescence in Hydrocharitaceae seems to be a long internode surmounted by a terminal flower above two bracts, which probably subtend lateral structures (Kaul, 1970), while the stolon of Hydrocharis (Posluszny & Charlton, 1999), Limnobium (Wilder, I 974b), and Stratiotes. tends to have a long internode surmounted by a vegetative shoot above two scale leaves, at least one of which normally subtends a lateral structure. It is strange that direct positional equivalence of inflorescence and stolon has not been found, but we have now shown that this does occur in Stratiotes. The general positional ho mology of inflorescence and stolon complex indicates that these structures are also equivalents. When allowance is made for differences in approach, it is evident that Wilder (1974b, 1 974c, 1975) came to the same conclusion, even though he did not strictly define any part of the Hydrocharitaceae he studied as "inflorescence," and even though he saw the other components of axillary complexes in a different light.
As an intermediate summation, then, in these stoloniferous Hydrocharitaceae a bifurcation product of an axillary complex is capable of giving rise either to an inflorescence, or to a single stolon that is almost certainly equivalent to a sterilized inflorescence, or to a stolon complex that is also apparently equivalent to a sterilized inflorescence but itself produces a number of stolons. But what is the basis for the difference between the alternatives of a single stolon or a stolon complex, if they are both sterilized inflorescences? Even if we cannot see directly how the difference arises, from within and without the Hydrocharitaceae, we can find the means of identifying the sterilized inflorescences that could have been involved.
Derivation of the stolon itself could occur easily enough by substitution of a vegetative bud for the terminal flower of the archetypal inflorescence. As pointed out above, in numerous instances in the helobial monocotyledons a flower is replaced by a vegetative bud, though usually lateral flowers rather than terminal ones are involved.
The conversion of an inflorescence into a stolon complex is more difficult to visualize. We showed that, in the male inflorescences of Hydrocharis, the sequence of bifurcations that forms a cluster of flowers in the bract axils can be terminated by the formation of a stolon complex instead of a floral meristem and that a stolon complex can occur in the axil of the single bract of the female inflorescence. A stolon complex can therefore replace a flower. In the analogous situation in the Alismataceae and Limnocharitaceae, as we mentioned above, it is common for a vegetative bud to replace a flower, but in two instances the bud appears to bifurcate immediately to form an inflorescence and a vegetative bud. The insertion of a stolon complex, a structure homologous with an inflorescence, into an inflorescence instead of a flower could presumably occur by a modification of this process. An inflorescence could therefore be converted into an axis bearing bracts or scale leaves and "terminating" in a stolon complex. Relatively small heterochronic and heterotopic shifts could eliminate the long internode of the original inflorescence and bract development per se, resulting in the form of stolon complex that is actually found. Earlier accounts describe the insertion of "vegetative buds" in male inflorescences of Stratiotes (Rohrbach, 1871) and in female inflorescences of Limnobium (Wilder, 1974b); we suspect that these were also stolon complexes rather than straightforward vegetative buds with foliage leaves.
The bracts or scale leaves might also subtend stolon complexes, which opens up other interesting possibilities. If the elongation of the lowest internode of the original inflorescence were displaced to the internode above the bracts/scale leaves, the result could be a structure remarkably like the occasional form of axillary complex found in Hydrocharis (Posluszny & Charlton, 1999) and Limnobium (Wilder, 1974b), in which there is a central stolon with a stolon complex at either side. This is actually very similar to the interpretation of axillary complexes made by Bugnon and Joffrin (1963, 1964) and Brunaud (1976, 1977). The axillary complexes in the Hydrocharitaceae are themselves very unusual structures. Formation of a complex bud by successive bifurcations occurs in other helobial monocotyledons, but it is normally restricted to the buds subtended by the bracts of inflorescences. Only in the Hydrocharitaceae does it occur outside the inflorescence, and it is possible that this is a result of transference o f the specialized branching process into the vegetative shoot.
Stolons of Vallisneria seem to have completely lost the paired bracts, and in Stratiotes it appears that successively formed stolons in a stolon complex are progressively less inflorescencelike in this respect. Presumably the loss of bractlike appearance here is a symptom of progressive conversion of "inflorescence" into the part of the vegetative architecture of the plant. Complete loss of inflorescencelike character would make these cases more difficult to interpret without the range of related forms.
While we have concentrated on the stoloniferous freshwater Hydrocharitaceae, in at least one marine genus there seems to be introgression of inflorescence and vegetative features. The main framework of the plant body of the genus Halophila consists of a horizontal shoot system (rhizome) that bears scale leaves in subopposite pairs, and one scale of each pair subtends a shoot bearing foliage leaves and also gives rise to inflorescences and new horizontal shoots. The horizontal system could easily be seen as derived from the archetypal two-bracted inflorescence of the family: it has retained the pairs of bracts, but instead of forming a terminal flower it has gained unlimited growth, forming further pairs of bracts; the buds in the axils of the bracts are sterilized and grow out as vegetative shoots at first. If this interpretation is correct, the inflorescence in Halophila shows a rather different form of sterilization than do the stoloniferous types and reinforces Wilder's (1975) conclusion that sterilizatio n of inflorescences is probably polyphyletic in the Alismatidae.
The authors would like to thank Gordon Lemon for his help in collecting material of Vallisneria americana. We are also thankful for the financial support of the Natural Sciences and Engineering Research Council of Canada operating grant (A6260) held by the second author.
VII. Literature Cited
Brunaud, A. 1976. Ramification chez les Hydrocharitaceae I--Ontogenie du systeme des pousses. Rev. Gen. Bot. 83: 397-413.
-----. 1977. Ramification chez les Hydrocharitaceae II--Organisation des rameaux lateraux. Rev. Gen. Bot. 84: 137-157.
Bugnon, F. & G. Joffrin. 1962. Recherches sur la ramification de la pousse chez la Vallisneria spiralis. Mem. Soc. Bat. France (1962): 61-72.
----- & -----. 1963. Ramification de la pousse chez l'Hydrocharis morsus-ranae L.: Comparaison avec le cas du Vallisneria spiralis L. Bull. Soc. Bot. France 110: 34-42.
Charlton, W. A. 1973. Studies in the Alismataceae. II. Inflorescences of the Alismataceac. Canad. J. Bot. 51: 775-789.
----- & A. Ahmed. 1973. Studies in the Alismataceae. IV. Developmental morphology of Ranalisma humile and comparisons with two members of the Butomaceae, Hydrocleis nymphoides and Butomus umbellatus. Canad. J. Bot. 51: 899-910.
-----, A. D. Macdonald, U. Posluszny & C. P. Wilkins. 1989. Additions to the technique of epillumination light microscopy for the study of floral and vegetative apices. Canad. J. Bot. 67: 1739-1743.
Johansen, D. A. 1940. Plant Microtechnique. McGraw-Hill, New York.
Kaul, R. B. 1970. Evolution and adaptation of inflorescences in the Hydrocharitaceae. Amer. J. Bot. 57: 708-715.
Les, D. H. & R. R. Haynes. 1995. Systematics of subclass Alismatidae: A synthesis of approaches. Pp. 353-377 in P.J. Rudall, P.J.Cribb, D.F. Cutler and C.J. Humphries (eds.), Monocotyledons: Systematics and evolution. Royal Botanic Gardens, Kew.
Miki, S. 1937. The origin of Najas and Potamogeton. Bot. Mag. (Tokyo) 51: 472-480.
Posluszny, U. & W. A. Charlton. 1993. Evolution of the helobial flower. Aquatic Bot. 44: 303-324.
----- & -----. 1999. Multiple redundancy in Hydrocharis morsus-ranae. Pp. 135-146 in M. H. Kurmann and A.R. Hemsley (eds.), The evolution of plant architecture. Royal Botanic Gardens, Kew.
----- & P.B. Tomlinson. 1991. Shoot organisation in the seagrass Halophila (Hydrocharitaceae). Canad. J. Bot. 69: 1600-1615.
Rohrbach, P. 1871. Beitrage zur Kenntniss einiger Hydrocharideen. Abh. Naturf. Ges. Halle 12 (75):53-114.
Sculthorpe, C. D. 1967. The biology of aquatic vascular plants. Edward Arnold, London.
Shaffer-Fehre, M. 1991a. The endotegmen tuberculse. An account of little-known structures from the seed coat of the Hydrocharitoideae (Hydrocharitacese) and of Najas (Najadaceac). J. Linn. Soc., Bot. 107: 169-188.
-----. 1991b. The position of Najas within the subclass Alismatidae (Monocotyledones) in the light of new evidence from seed coat structures in the Hydrocharitoideae (Hydrocharitales). J. Linn. Soc., Bot. 107: 189-209.
Wigand, C. & J. C. Stevenson. 1997. Facilitation of phosphate assimilation by aquatic mycorrhizae of Vallisneria americana Michx. Hydrobiologia 342/343: 35-41.
Wilder, G. J. 1974a. Symmetry and development of Butomus umbellatus (Butomaceae) and Limnocharis flava (Limnocharitaceae). Amer. J. Bot. 61: 379-394.
-----. 1974b. Symmetry and development of Limnobium spongia (Hydrocharitaceae). Amer. J. Bot. 61: 624-642.
-----. 1974c. Symmetry and development of pistillate Vallisneria americana (Hydrocharitaceac). Amer. J. Bot. 6l: 846-866.
-----. 1975. Phylogenetic trends in the Alismatidae (Monocotyledoneae) Bot. Gaz. 136: 159-170.
|Printer friendly Cite/link Email Feedback|
|Author:||CHARLTON, W. ALAN; POSLUSZNY, USHER|
|Publication:||The Botanical Review|
|Date:||Oct 1, 1999|
|Previous Article:||Capitula in the Asteridae: A Widespread and Varied Phenomenon.|
|Next Article:||Inflorescence Architecture: A Developmental Genetics Approach [*].|