| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Structure and Development |
2Harvard Forest, Harvard University, Petersham, Massachusetts 01366 USA and National Tropical Botanical Garden, 3530 Papalina Road, Kalaheo, Hawaii 96741 USA; 3Department of Ecology, Evolution and Behavior, University of Minnesota, 100 Ecology Building, 1987 Upper Buford Circle, St. Paul, Minnesota 55108 USA
Received for publication October 18, 2001. Accepted for publication February 5, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Arecaceae • Calamus • metaxylem • procambium • protoxylem • rattan • stem histology • vascular development • vessels
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). A recent investigation (Tomlinson et al., 2001
) has shown that the vascular system is distinctive among monocotyledons because there is limited axial continuity of phloem and especially xylem. In the present article we describe equally distinctive features of development and show how the vascular arrangement of the Calamus stem is modified by the absence of features that characterize stems of palms with a normal habit. The specialization in relation to habit includes an early blocking out of the vascular pattern combined with late and extended internodal elongation without concomitant increase in diameter. The information is relevant to an understanding of long-distance transport mechanisms in woody plants. It must be emphasized that monocotyledons with a pronounced climbing habit are peculiarly vulnerable to vascular dysfunction because they cannot produce secondary vascular tissues. Furthermore, recent observation suggests that in climbing monocotyledons a close correlation exists between axial metaxylem type and degree of xylem continuity (Tomlinson and Fisher, 2000
), with as yet unknown functional consequences.
The vascular system of Calamus is paradoxical in its apparent inefficiency for long distance transport, especially of water. Stems can reach lengths approaching 200 m (Burkill, 1966
) by producing large numbers of long (up to 1 m) but narrow internodes, these supplying a single crown of leaves. However, the axial transport system consists of vascular bundles that are connected distally only to leaves, but proximally over long distances the same bundles are tapered gradually and end blindly. In the measured example (Tomlinson et al., 2001
), a single bundle may extend through about 15 internodes (a total of approximately 3.5 m) with a single uninterrupted metaxylem vessel over most of this distance. Furthermore, although metaxylem vessels are wide (up to 350 µm diameter), as is to be expected in a woody liane, they are never in contact with the protoxylem tracheids that directly irrigate the leaves. The only direct connection among metaxylem vessels of different axial bundles is via short, narrow, irregular connections (transverse commissures) with narrow vessels and little phloem. This suggests that as far as water movement is concerned there is considerable internal resistance both laterally and from stem to leaf. The presence of long, wide vessels as efficient conduits seems obviated by the overall vascular architecture. There are equivalent major discontinuities in the phloem. The most likely, but still speculative, conclusion was that (a) vessels serve as much for water storage as for transport and (b) that the system minimizes xylem cavitation in an otherwise highly vulnerable system (Tomlinson et al., 2001
).
In order to extend these observations the present article adds a developmental dimension by describing how the vascular system is initiated and the limited vascular continuity is maintained during late internodal extension. Associated histological features also need to be understood in the same context. Quantitative descriptions of shoot development in rattans have been provided by Fisher (1978)
and Fisher and Dransfield (1977
, 1979
). Our observations fit well with these descriptions so our quantitative analysis of shoot form is minimal.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. The detached canes (up to 10 m long) were cut into shorter lengths for transport to the laboratory. Leaf dimensions from older to younger parts were measured as they were detached in order, up to the insertion of the youngest visible (but still incompletely expanded) leaf protruding from the crown, i.e., the "spear leaf," Ps, which was used as a reference level. Younger, enclosed leaves were progressively removed and measured to within a few primordia of the apex proper, but the leaves numbered basipetally, with P1 the youngest. To provide precise dimensions of younger stages, the shoot, together with attached early leaf primordia, was fixed in formalin-acetic-alcohol (FAA). A total of 12 apices were prepared, but with some loss of material and measurements because of breakage during dissection. For sectioning, material was embedded in "Paraplast" following routine procedures of dehydration. Serial transverse and longitudinal sections were cut on a rotary microtome at 10–15 µm, stained in safranin and alcian green, and mounted permanently in "Permount." It was found that with larger specimens there were some problems of wax penetration related to the surface texture of the stem, causing some sections to fall out of the ribbon, but less than 3% of the total number of sections were lost and axial continuity was not jeopardized. The most completely analyzed series included 570 sections and the insertion of nine developing leaves. To facilitate measurement the sections were numbered individually. Longitudinal sections produced in the same way were largely used to make measurements of early leaf stages and to complement histological observations.
For scanning electron microscopy (SEM) a number of dissected shoots were observed at different stages of leaf development (see Figs. 1–4). These were first dehydrated, critical point dried, and metal coated in the usual way and photographed with an AMR 1000 (Cambridge, Massachusetts, USA) at 10 kV. Some of these specimens were rehydrated, dissected further, and rephotographed, all were eventually rehydrated and embedded in wax for serial sectioning in order to provide as complete a series of measurements as possible for several shoots (e.g., Fig. 5B).
|
|
).
|
|
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). The flagellar axis includes several inconspicuous, reduced tubular leaves (bracts) and is provided distally with vicious recurved hooks. We have not studied adult canes at a later stage, when the flagellum branches into a flower-bearing panicle. We studied only post-juvenile vegetative stages. Each foliage leaf has a closed tubular, sheathing base, its mouth extended into a short ligule that is open on the inner (i.e., adaxial) side. The petiole is short (3–5 cm) and the rachis is provided with somewhat irregularly placed leaflets that project in a single plane. Leaves are spirally arranged with an average angular divergence of about 137.5°, corresponding to a phyllotactic fraction close to 3/8. Because the leaf itself has no consistently identifiable dorsal midpoint (e.g., Fig. 8) phyllotaxis was measured from the position of successive flagella (Fig. 7), since they are approximately in the ventral axillary position. Shoots may be described as right- or left-handed according to the direction of the ontogenetic helix, but chirality is most obvious from whether the flagellum lies to the right or left of the median ventral plane of the leaf (cf. Figs. 1 and 2 as left hand and right hand respectively).
|
for Calamus insignis and are not repeated here. Notably, the flagellum shows very late and considerable internodal extension (to an overall length of 1.75 m), even though its bracts are initiated early (as on P7 in Figs. 3 and 4).
Leaves are referred to in order of age, with P1 the youngest, whereas the flagella are referred to according to the number of the leaf to which they are attached (Figs. 1–3), with F2 the youngest. For example, F8 is attached to P8 (Fig. 7), although it originates in the axil of P9. There is no F1 because P1 lacks an attached flagellum as a consequence of its sheath still being incompletely encircling. However, meristematic tissue that can be identified as F1 is visible in the axil of P2. This precocious inception of the flagellum conforms to the description by Fisher and Dransfield (1977)
in other Calamus species. Differentiation of the various parts of the leaf are completed by P3 (Fig. 1) in which the first evidence of leaflet primordial can be recognized, although the short petiole is little extended. The flagellum becomes protuberant and clearly adnate by P4 (Fig. 1) and its apex is free distally by P5 (Fig. 2). Subsequently, the free portion initiates a series of distichously arranged tubular bracts (Figs. 3 and 4), but its extensive internodal elongation is one of the last features of leaf development, as quantified by Fisher (1978)
. The status of the flagellum as a lateral or branch axis is confirmed by the meristematic "shell zone" visible almost from inception and forming an arc of cambial-like cells separating the flagellum primordium from differentiated stem cortical parenchyma as described by Fisher and Dransfield (1977)
.
The length of the first 13 leaves on a single shoot changes through four orders of magnitude (Fig. 5A). The youngest externally visible leaf is the "spear" leaf (Ps); Pfe is the first fully extended leaf. Early leaf development with overall lengths for the first six leaves at the same plastochron stage but in several different shoots shows an appreciable range (Fig. 5B). There is limited overlap in leaf size from one plastochron to the next showing that leaf development is uniform in different shoots. This validates information from different samples as representing overall features of development; histological features are directly comparable from shoot to shoot. Leaf extension accelerates considerably from P5 to Ps, the latter corresponding to P9 in Fig. 5A. In this shoot P10 is the first fully extended leaf (Pfe in Fig. 5A) although the internode below is still incompletely extended. Consequently, a fully mature leaf–stem unit (mature in the sense of having completed extension growth), does not exist until P11 or P12 (= Ps+2), i.e., two leaves older than Ps. Therefore vascular development and maturation must proceed over a period of at least 11 internodes. Younger leaves, whose blade is exposed but whose sheath is still immature, are supported by encircling lignified sheaths of older leaves, as is usual in palms.
Origin of the vascular system
In the mature stem the vascular system includes a central core of about 200 large bundles, which occupy almost two-thirds of the total stem cross-sectional area, and a peripheral series of numerous narrow vascular and fibrous bundles. There is no discrete cortical system, as is found in palms of normal habit (Zimmermann and Tomlinson, 1965
; Tomlinson, 1990
). Our main focus is the origin of the central system, which becomes the main axial transport system. The vascular connection between stem and leaf in the developing shoot is made by progressive addition of leaf traces at successively older nodes of insertion as can be seen in transverse section (cf. Fig. 5C with Figs. 8–10). The youngest leaf (P1) always lacks such a connection so that the youngest stem–leaf connection is not made until P2 (Fig. 8). The five bundles of P2 can be seen in a different orientation at a lower level (Fig. 9, VB). Subsequent leaves show increasing numbers of leaf traces until about P7, which has about 60 major bundles connected to it. The insertion of P5 adds about 47 leaf traces to the axis (Fig. 10, cf. Fig. 5C). At this node the central system includes all traces from younger leaves (P2–P4), i.e., 5 + 17 + 33 = 55. The internode below P7 includes the total contribution to the central system (about 200 vascular bundles). This procambial system provides the template in which the main axial vascular tissue develops subsequently.
The peripheral system is formed as vascular bundles continue to differentiate in continuity with stem and leaf (dotted line in Fig. 5C). These are indistinctly circumscribed from ground tissue so that their number cannot be precisely counted. For the spear leaf (Ps) they number approximately 150, but their numbers are no longer additive so that there is an average of about 300 peripheral bundles in sectional view of the mature stem. Limited analysis suggests that they also end blindly below, i.e., without anastomoses, but over much shorter distances than central bundles. Except for a minority of bundles with few narrow late-differentiating xylem and phloem elements they are entirely made up of fibers and have a largely mechanical function. They are not considered further in our analysis.
Basal discontinuity
To facilitate discussion of overall shoot and vascular architecture an outline plot of the shoot is provided but with the leaves rotated into a single plane and with a single leaf trace drawn for each leaf (Fig. 6). The diagram is disarmingly simple since changes in the radial position of vascular bundles are ignored. To maintain the constant number of central bundles there must be progressive loss of vascular bundles. This is achieved by the progressive disappearance of bundles in a basal direction as demonstrated earlier (Tomlinson et al., 2001
). These tapering ends are seen as narrow strands, each with few narrow procambial elements, but only in older internodes. This is exactly to be expected from our knowledge of the vascular topography of the mature stem in which basipetally blind-ending bundles have been illustrated (Tomlinson et al., 2001
: fig. 43). However, traced distally these same bundles cannot connect with a leaf because we know that they may extend through as many as 15 internodes before this happens, i.e., to leaves that still have to be initiated. In fact, such bundles progressively disappear toward the younger internodes; the youngest level at which they can be recognized is about P5. A single bundle of this type is plotted in Fig. 6 (barred line). In this sense such bundles end blindly in each direction, but must continue to grow acropetally until they are "captured" by a developing leaf. Careful search of the ground tissue reveals these incipient "uncaptured bundles" as a series of narrow cells, which become more obvious as a procambial strand when followed basipetally. Differentiation within these strands is not necessarily continuous because one may fail to recognize a strand through a short series of sections. This is particularly clear even in the youngest primordia where the distal connection has been made since a procambial strand may fail to be recognized through several successive sections before it is rediscovered at a lower or higher level.
Apical discontinuity
Vacuolation of central ground tissue begins close beneath the shoot apical meristem so that newly initiated bundles are not visibly continuous with meristematic tissue (Fig. 8). However, the discreteness of leaf traces within the vacuolated ground tissue in lower internodes is clear from Figs. 9 and 10. Axial bundles thus appear by dedifferentiation of ground-tissue cells and subsequent internal longitudinal division. Early stages of this process are evident only in internodes older than P5. Each newly initiated leaf thus captures a progressive series of blind-ending vascular bundles. In order to maintain the system at progressively higher levels there must be initiated a corresponding new series of apically discontinuous axial bundles. These are not captured until a much later plastochron. This method of initiation and subsequent connection to a leaf is the most distinctive developmental feature of the rattan stem. The overall acropetal development of the vascular system thus depends on the contrasted process: on the one hand of "capture" of axial strands to form leaf traces and, on the other, generation of new apically discontinuous axial strands.
Differentiation of vascular tissue
Because of the long distance over which vascular bundles extend, a bundle connected as a procambial strand to any leaf primordium when traced basally shows a progressively differentiating complement of vascular tissues. The distance from the leaf connection for each bundle is reflected in the relative amount of differentiated and undifferentiated protoxylem. The stem center shortly below the insertion of P9, i.e., about 5 mm below the shoot apex shows the future configuration of the vascular tissue (Fig. 11), i.e., at the base of the plotted apex in Fig. 6. Although metaxylem vessels are conspicuous, they are incompletely differentiated, whereas mature protoxylem elements are present in most bundles. A section of a single mature vascular bundle is provided for comparison with earlier stages (Fig. 12).
|
, i.e., where nuclei of successive divisions remain at the same level to form a banded series. The resulting radial seriation of cells in the center of the vascular bundle may remain evident late in development, especially close to the departure of an axial bundle as a leaf trace (e.g., Fig. 15, ct). This configuration is lost in the bundle at maturity, as shown for comparison in Fig. 12. Late stages in bundle maturation, i.e., in internodes below P8 and P9, are shown in Figs. 13–16. An axial bundle close to bundle departure from the stem as a leaf trace (Fig. 13) indicates the abundant protoxylem (px), some of it mature; the bundle in Fig. 14 is at a level more remote from its departure as a leaf trace and so has little protoxylem. At this level the original location of the protophloem (pph in Fig. 14) is being usurped by fibers (f in Fig. 13). The future metaphloem in both bundles is blocked out as lateral arcs of tissue (e.g., ph in Fig. 13), but the original continuity of phloem tissue as a single strand is most evident in Fig. 14 in the early protophloem (pph) region. The contrast in number of protoxylem elements between distal (Fig. 13) and proximal (Fig. 14) positions of the same bundle is pronounced.
The level of vascular differentiation at opposite ends of a bundle, i.e., its distal and proximal extremities, is represented in Figs. 15–19, but not from a single bundle. In Fig. 15 the protoxylem (px) and protophloem (pph) are apparent at opposite poles of the bundle. This represents a leaf trace close to its departure into a leaf, which is evident from the numerous actual and potential protoxylem elements and the absence of differentiated metaxylem. At a lower level of a comparable bundle (Fig. 16) the metaxylem vessel is present (mx) and there is more limited mature protoxylem (px). In both these bundles the pph is a single strand and there is no differentiation of phloem in the lateral parts of the bundle. This stage should be contrasted with the later condition in Figs. 13 and 14 and the mature condition in Fig. 12. The metaxylem originates in the sinus between the differentiating metaphloem strands so that it eventually appears to be flanked by their mature end products and remains unconnected to the protoxylem (cf. Fig. 16 in contrast with Fig. 12). At the basal extremity of bundles (Figs. 17–18) there is little indication of vascular tissue, until finally there remains only a strand of narrow cells that never differentiate as vascular tissue (Fig. 19). The sequence represented by these figures in the order 15–16, 13–14, and 17–18 represents features that change as one follows an entering leaf trace in a basipetal direction from its insertion into the stem. These images may be compared with the sequence shown as Figs. 10–20 in Tomlinson et al. (2001)
. It is important to emphasize that these changes within the developing crown are progressive apically, since a bundle present in a mature internode, with fully differentiated xylem and phloem (e.g., Fig. 12), may be continuous distally with an immature bundle (e.g., Figs. 13 or 14). The basal end of a newly initiated axial bundle (as in Fig. 19) is likely to be continuous apically with a procambial strand that will not connect to a leaf until many plastochrons later.
Contrast between protoxlem and metaxylem elements is extreme, as seen in longitudinal section, because early developing protoxylem initials form an overlapping series of fusiform cells, in contrast to the linear series of late-developing barrel-shaped metaxylem elements with transverse end walls (cf-mx in Figs. 20 and 21 with px in Fig. 21). Protoxylem elements mature progressively first as annular and then as helical elements of wider diameter with subsequent extension as internodes elongate. This produces a wide range of diameters for protoxylem elements, with the most recently formed elements still incompletely differentiated (Fig. 11). Maturation, with pitted secondary walls, of the last-formed protoxylem elements and all the metaxylem elements does not begin until internodal elongation ceases. This does not occur until after the associated leaf is fully extended (i.e., Pfe in Fig. 5A). The collective first appearance of the metaxylem vessel initials is surprisingly late and in the internode below P8 (e.g., Fig. 20), with subsequent widening (Fig. 21) and then extension (Fig. 25). Metaxylem vessel differentiation refers to the extended middle part of the overall length of a vascular bundle: basally (at the blind ending of the bundle) there is a short series of overlapping narrow metaxylem elements; distally in the leaf trace the series of narrow metaxylem elements also ends blindly. The absence of mature metaxylem in the developing shoot until about P11 emphasizes the importance of protoxylem maturation in shoot development.
|
. It is equally clear that the attachment of these strands to a metaxylem vessel will depend on the simultaneous wall maturation of xylem in the two bundle types in order to preserve a continuous pathway for transport (Fig. 25).
Raphide canals
Linear series of cells that include mucilage and a cluster of raphides are a feature of the ground tissue of the Calamus stem. They appear as early as the internode below P4 as files of cells with dense cytoplasmic contents (Fig. 26) and so demonstrate histological differentiation well above the level of differentiation of vascular tissues. Surrounding cells are somewhat shorter than other ground tissue cells as a continuous layer without intercellular spaces, but do not constitute a distinct epithelium. Cell contents of the immature raphide sacs themselves stain intensely with safranin so that these initials are obvious in transverse view (Fig. 28). Once differentiated, the cells of the raphide canals do not divide so that they become considerably extended by subsequent internodal elongation. A cluster of raphide crystals appears early in each cell of the canal within one to two plastochrons of their initial differentiation (Fig. 27). Extension of the cells of the raphide canals is considerable, but the raphide crystals do not increase in number and are easily overlooked in mature tissue, although they are easily seen in mucilage exuding from cut surfaces.
Ground parenchyma
The central ground parenchyma of the axis shows vacuolation as early as the first plastochron (Figs. 8 and 9), but with continued transverse cell division so that extension growth is promoted. This results in extended files of cells within the wall of the original mother cell, as is very evident in longitudinal section (Fig. 25), but with persistence of the original intercellular space. A distinctive feature of this developmental process is the thickened, hydrated texture of the wall adjacent to the intercellular spaces, made conspicuous by the intense staining with alcian green (e.g., Figs. 13 and 24). The feature is retained in mature stem internodes even after wall thickening and lignification are completed. The significance of this structural feature is not clear. Tannin cells are not observed in these stems.
Flagellar vasculature
Even though the flagellum is recognizable as an axillary meristem as early as P2, and is post-genitally fused to stem and leaf sheath above its node of insertion, its vascular system develops much later than that of the main axis. For the first five plastochrons it remains uniformly meristematic. Vascular bundles become recognizable as 1–2 strands in transverse sections at the level of insertion of P5, with five bundles at a corresponding level in P6, the number increasing to the maximum (10–13) at the insertion of F8. The basal extent of flagellar bundles is limited to the internode below because they disappear relatively abruptly without making any connection to stem bundles. This restricted development shows that they are confined morphogenetically to the original axillary meristem. Their discreteness is further indicated by the limits of the shell zone, which in later stages becomes a zone of narrow parenchyma cells. Vascular differentiation in the flagellum is recognizable at about the level of insertion of P8 as protoxylem and protophloem in all but the most recently formed bundles. The only displacement of axial bundles in association with the insertion of the flagellum is a slight irregularity caused by the outgoing leaf traces. Axial bundles are here seen as obliquely sectioned strands in transverse view, below the flagellar attachment.
Summary
From our knowledge of the mature vascular architecture of Calamus (Tomlinson et al., 2001
) and the present information we can summarize our understanding of how the axial system of the shoot in Calamus is initiated (illustrated in Figs. 29 and 30). Figure 29 represents the described structure, based on the earlier analysis, but presented in a developmental context, with the shoot drawn in an extended form for clarity but with elongating and mature internodes at the same magnification. Within this shoot a complete leaf trace continuous into the node of insertion of P2 is drawn, with its basal end terminating in the internode below the insertion of P16, i.e., the axial bundle extends through 15 internodes, of which the upper eight are still elongating. We choose 15 internodes because this is the maximum distance estimated for the length of an axial strand (fig. 43 in Tomlinson et al., 2001
). Within the same stem diagram we represent an equivalent bundle that would enter the youngest leaf primordium (P1) by the next plastochron, this bundle having originated within the internode below the insertion of P15. Its distal terminus would not easily be recognized and is represented by a dotted line. The right-hand portion of the figure represents a single bundle continuous into each successively older leaf, plus an additional bundle that would be expected to become continuous with each of the successive future leaf primordia (I1 to I15) in order of their age. Their position is shown outside the left-hand diagram to simplify the illustration. These bundles are shown existing up to I10, but their distal extremities disappear (as dotted lines) in the internode below P4. Each of them, as expected, would end blindly below in order of their age. Further bundles in the "incipient" box might be expected to exist, but would not be recognizable in serial sections.
This model is based on a system with only a single bundle to each leaf and all bundles extending through 15 internodes. At any level the number of axial bundles remains constant at 15, since for each bundle lost by capture as a leaf trace, one is gained by differentiation of a new bundle from vacuolating ground parenchyma cells. The actual complexity of the stem itself can be appreciated from the fact that about 60 central bundles are lost at each leaf insertion as leaf traces and that there is no constant length for every one of them. Later-differentiating bundles are likely to be shorter.
The change in configuration of the leaf insertion and its associated internode during primary expansion of the stem that must exist in an actual shoot is illustrated (Fig. 30). The diagram is realistic because it is based on the plotted example of Fig. 6, but simplified in representing a constant distance for any axial bundle as ten internodes. The solid lines represent bundles and tissue at the initiation of the expansion of the crown, the dotted lines their further development once internodal extension and expansion is complete (i.e., ten plastochrons later). The diagram is simplified because we have not represented bundles that would be continuous into leaves younger than P1 (cf. Fig. 29).
The only illustrated strand that is discontinuous both apically and basally is the one leading to P1. The strands to P2 and P3 end blindly below within the figure, but are continuous into a leaf. All other bundles (from P4 to P12) extend basally out of the limits of the figure. The simplification of this figure is evident, but it represents a reasonable facsimile of our observations and their interpretation as a developmental process.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
, 1966
), in which the cane-like palm Rhapis excelsa was used as a model. In Rhapis vascular development is determined by a "meristematic cap," i.e., a dome-shaped meristem that generates axial bundles as branches of existing leaf traces; the axial bundles eventually make contact with a leaf to become, in turn, leaf traces. As in Calamus axial strands are initially uncommitted because leaf traces appear well before the leaf they will supply. Calamus differs from this system in that there is no meristematic cap. Instead, vascular bundles are generated by dedifferentiation of vacuolated ground parenchyma cells, such that their level of initiation is represented by a basally blind-ending vascular strand and their distal, also blind-ending, apical portion continues to differentiate until "captured" by a newly initiated leaf primordium. As the leaf primordium enlarges through about nine plastochrons vascular connections continue to be made up to a maximum of about 200 central bundles. These first-formed connections result in bundles of the central system, which provide the main stem-conducting system. Late-differentiating bundles are small, numerous, and mainly fibrous, but are not regarded as a cortical system, as develops in the Rhapis model. They have a primary mechanical function in the mature stem.
Vascular differentiation within the procambial strands follows in principle the same sequence as in Rhapis (Tomlinson and Vincent, 1984
), with the protoxylem providing the exclusive leaf–stem hydraulic system within the developing crown. In Calamus protoxylem is first evident, together with protophloem, in the bundles connecting the axis with P5. This represents the upper limit of continuously generated vascular tissue. Because such leaf traces represent the distal connection made by preexisting axial bundles, it is certain that vascular tissue already exists in older internodes. In contrast, metaxylem vessels do not begin to differentiate until the internode below P8. They do not mature until extension growth has ceased, i.e., at about the internode below P10, which is the first fully extended foliage leaf. The peculiarity of the Calamus system, in which axial bundles may extend unconnected for as much as 15 internodes, is that the basal end of a bundle may be fully mature, whereas its distal extremity can still be an axial strand, represented by a blind-ending procambial bundle. From our knowledge of the uniform construction of Calamus canes (Weiner, 1992
) we can assume that these developmental details are universally applicable, although canes of larger dimensions are likely to be more complex numerically (cf. Fisher, 1978
).
Although the method of vascular development is highly distinctive, some of its features may exist in other monocotyledons, albeit in examples where stems are of determinate growth. For example, in their study of Alstroemeria, Priestley, Scott, and Gillett (1935)
demonstrate that an axial bundle diverges from a parent strand within the apical region of the stem at a level that is eight plastochrons before the appearance of the leaf it will eventually supply. Consequently, as in Calamus, axial strands are differentiated well before they become committed to a specific leaf primordium. On the other hand, axial bundles are interconnected in a very regular way, unlike Calamus. The relatively well-investigated shoot of Zea mays is too highly specialized to afford comparison because it is precisely determinate in organization and has pronounced nodal plexi (Sharman, 1942
; Pizzolato and Sundberg, 1999
). In conifers and gymnosperms, in which the vascular system is largely continuous, there is evidence that axial strands can also appear well before the leaf they will supply (Esau, 1965
), but within the residual meristematic ring of the shoot apex. Here, vascular analysis is made easy by the precise association between phyllotaxis and vascular architecture, with a small number of bundles continuous over relatively short distances. The limited understanding of these developmental processes in monocotyledons makes comparative discussion difficult.
The uniqueness of the Calamus system is not an exclusive consequence of its pronounced late internodal elongation because a preliminary analysis of the stem of the American climbing palm Desmoncus shows that it is constructed according to the Rhapis principle. Further research should be directed to comparative analysis of the vascular architecture of climbing palms, but most obviously to those calamoid palms that show a stem anatomy contrasted with that of members of the tribe Calaminae (Uhl and Dransfield, 1987
; Weiner, 1992
). Furthermore, there is preliminary evidence that there are consistent correlations between vessel types and vascular anatomy of other climbing monocotyledons (Tomlinson and Fisher, 2000
). The general suggestion is that more detailed comparative study of these taxa may produce informative results about the hydraulic system of plants in general. Because climbing monocotyledons lack secondary growth, they are particularly vulnerable to xylem dysfunction. The present contribution is preliminary and is certainly incomplete in details.
|
FOOTNOTES |
|---|
4 Author for reprint requests (pbtomlin{at}fas.harvard.edu
)
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Esau K. 1965 Vascular differentiation in plants. Holt, Rinehart, and Winston, New York, New York, USA
Fisher J. B. 1978 A quantitative description of shoot development in three rattan palms. Malaysian Forester 41: 280-293
Fisher J. B. J. Dransfield 1977 Comparative morphology and development of inflorescence adnation in rattan palms. Botanical Journal of the Linnean Society 75: 119-140
Fisher J. B. J. Dransfield 1979 Development of axillary and leaf-opposed buds in rattan palms. Annals of Botany 44: 57-66
Pizzolato T. D. M. D. Sundberg 1999 Initiation of the vascular system and the transition to flowering in early tassels of Zea mays land race chapalote (Poaceae). International Journal of Plant Science 160: 1111-1125[CrossRef][ISI][Medline]
Priestley J. H. L. I. Scott E. C. Gillett 1935 The development of the shoot in Alstroemeria and the unit of shoot growth in Monocotyledons. Annals of Botany 49: 161-179
Sharman B. C. 1942 Developmental anatomy of the shoot of Zea mays L. Annals of Botany n.s. 22: 245-282
Tomlinson P. B. 1990 The structural biology of palms. Oxford University Press, Oxford, UK
Tomlinson P. B. J. B. Fisher 2000 Stem vasculature in climbing monocotyledons: a comparative approach. In K. L. Wilson and D. A. Morrison [eds.], Monocotyledons—systematics and evolution. Vol. 1. Proceedings of the Second International Conference on the Comparative Biology of the Monocotyledons, September 1998. C.S.I.R.O., Melbourne, Australia
Tomlinson P. B. J. B. Fisher R. E. Spangler R. A. Richer 2001 Stem vascular architecture in the rattan Calamus (Arecaceae—Calamoideae—Calaminae). American Journal of Botany 88: 797-809
Tomlinson P. B. J. F. Vincent 1984 Anatomy of the palm Rhapis excelsa X. Differentiation of stem conducting tissues. Journal of the Arnold Arboretum 65: 191-214[ISI]
Uhl N. W. J. Dransfield 1987 Genera palmarum. Allen Press, Lawrence, Kansas, USA
Weiner G. 1992 Zur Stammanatomie der Rattanpalmen. Ph.D. dissertation, University of Hamburg, Hamburg, Germany
Zimmermann M. H. P. B. Tomlinson 1965 Anatomy of the palm Rhapis excelsa. I. Mature vegetative axis. Journal of the Arnold Arboretum 46: 160-180
Zimmermann M. H. P. B. Tomlinson 1966 Anatomy of the palm Rhapis excelsa. IV. Vascular development in the apex of the vegetative aerial axis and rhizome. Journal of the Arnold Arboretum 48: 122-142[ISI]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
