| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
2 Department of Agronomy and Center for Aquatic and Invasive Plants, P.O. Box 110500, University of Florida, Gainesville, Florida 32611-0500 USA
Received for publication January 25, 1999. Accepted for publication September 23, 1999.
ABSTRACT
Giant cutgrass [Zizaniopsis miliacea], a tall emergent grass native to the southeastern United States, was studied in two Florida lakes. In Lake Seminole (15 176 ha) giant cutgrass forms large expanding stands, but in Lake Alice (9 ha) it is confined to a stable narrow fringe. By monitoring individual plants in Lake Seminole, it was found that an average decumbent flowering stem produced three flowers and ten nodes, 80% of which became rooted in the substrate. Such flowering stem development could potentially result in stand expansion of 2.2–2.7 m/yr, depending upon water levels and rates of node rooting. Once flowering stems became decumbent in Lake Alice, they typically broke, producing no more than two flowers with four nodes in a growing season. While still attached to the parent plant, few of these nodes were able to become rooted in the substrate, limiting the rate of stand expansion in Lake Alice. Sections of flowering stems bearing axillary shoots that were detached from the parent plant and free-floating could become rooted on reaching shallow water and produce robust, new, flowering plants. This interesting mode of population dispersal and spread has important implications for the distribution and management of giant cutgrass.
Key Words: adventitious roots • axillary shoots • dispersal • giant cutgrass • functional stolons • Poaceae • stand expansion • vegetative reproduction
Giant cutgrass [Zizaniopsis miliacea (Michx.) Doell. & Asch.] is a rhizomatous perennial grass, which grows in shallow fresh water. Leaves, which are 4 cm wide and may reach 1.5 m in length, have extremely rough margins that fully justify the "cut" aspect of this common name. Synonyms include water millet and southern wildrice, the latter because Zizaniopsis miliacea in its vegetative state may easily be confused with species of wildrice [Zizania spp.]. Giant cutgrass panicles, however, do not show the clear partitioning of male (below) and female (above) spikelets that is distinctive of wildrice (Cook, 1990
; Fox, 1993
).
A native of the southeastern United States of America, giant cutgrass is principally found in Arkansas and the coastal states from Maryland to Texas (Martin, 1953
). Giant cutgrass may be found both as a fringing emergent or in dense stands in marshes, ditches, creeks, and along the edges of lakes, rivers, and streams. Large stands of giant cutgrass are often associated with abandoned rice fields (e.g., Baden, Batson, and Stalter, 1975
; Latham, Pearlstein, and Kitchens, 1994
), freshwater tidal marshes (e.g., Birch and Cooley, 1982
; Odum, Birch, and Cooley, 1983
), or shallow lakes and reservoirs, where rates of colonization may be quite rapid. For example, coverage of giant cutgrass expanded in Reelfoot Lake, a shallow lake in northwest Tennessee, from 770 ha in 1942 to 1000 ha in 1960 (Burbank, 1963
). In Lake Seminole, a 15 176-ha reservoir impounded in 1957 on the Georgia-Florida- Alabama borders, giant cutgrass coverage increased from 1.2 ha in 1960 (Kight, 1980
) to 3240 ha by 1983 (Gholson, 1984
).
The mechanisms by which such rapid littoral zone colonization is possible have not been fully studied. Although large numbers of 2-mm long seeds are produced (~3000 seeds per inflorescence; A. M. Fox, University of Florida, unpublished data), only occasional observations of seedling growth, always on exposed mud flats, have been recorded (Steenis and Cottam, 1945
; Smart and Barko, 1982
). Creeping rhizomes, protected by water and soil, provide an important mechanism for regrowth of plants damaged by frost, fire, disease, or mechanical damage, but with growth rates under ideal conditions of 55 cm over 5 mo (A. M. Fox, University of Florida, unpublished data) these seem unlikely to account for some of the observed rates of stand expansion of 2–3 m/yr (Fox and Haller, 1990
).
Rapid colonization of littoral zones has been attributed to the production of "stolons" or "runners" up to 4 m long from which leafy buds and adventitious roots develop (Steenis and Cottam, 1945
; Martin, 1953
; Smart and Barko, 1982
). The morphological origins of these "stolons" were hinted at by Kight (1980)
who observed that stalks of giant cutgrass would fall over and produce roots and leaves from the nodes. The realization that these so-called "stolons" were actually decumbent flowering stems was first described by Cutshall, Glennon, and Biles (1989)
.
The mechanism by which flowering stems of giant cutgrass may behave as functional stolons and result in new plants becoming rooted up to 3–4 m away is outlined by Fox and Haller (1990)
, Fox (1993)
, and in Fig. 1. In northern Florida, most flowering stems are produced between March and May and initially have three or four nodes, with a single leaf arising from each. The sheath of each leaf surrounds the internode above it so that, superficially, the leaf blade appears to arise from the next node. As inflorescences on the periphery of a stand mature, their 1.5–2 m long stems lean away from the stand (Fig. 1B). By the time all the seeds are shed, the stem will be fully lodged at the water surface, and shoots and adventitious roots will have appeared at one or two of the nodes. A shoot arising from a node initially will be wrapped within the sheath of that node's original leaf, which will no longer be clasping the stem. As the shoot develops, this sheath splits or unfurls until it no longer restricts expansion of the new shoot. It is because new shoots formed at the nodes arise in the axils of these original stem leaves that they may be referred to as axillary shoots (Holmes and Stalling, 1990
), to distinguish between them and shoots arising from the rhizome.
|
provided quantitative data on the production of axillary shoots arising from nodes on the flowering stem (fertile culms) and on internode and bud dimensions for giant cutgrass. They also suggested that the mass of the axillary shoots causes fertile culms at the water's edge to fall to a horizontal position. These authors concluded that only "... up to three vegetative buds ..." would be produced per flowering stem.
Fox (1993)
noted, however, that the shoot at the distal node of a decumbent flowering stem usually produces another inflorescence, and further shoots may develop from the three or four nodes on such secondary flowering stems (Fig. 1D). This process may continue with the production of three or more flowers from a single plant base, and this functional stolon may be up to 4.9 m long (Fig. 2). The production of more than two flowering stems per plant is usually possible only if the horizontal stem is anchored at some of the nodes by adventitious roots that have reached the substrate. If none of the nodes is anchored, the horizontal stems eventually will break from the parent plant (Fox and Haller, 1990
; Holmes and Stalling, 1990
). It is the production and survivorship of such secondary flowering stems that the present study was designed to quantify.
|
MATERIALS AND METHODS
Sites
Lake Alice is a 9-ha lake on the University of Florida campus in Gainesville, Florida (29°38' N, 82°22' W), which receives secondary-treated wastewater that has passed through a 21-ha marsh on the lake's east side. Lake water is pumped into the Florida aquifer by two injection wells at the west edge of the lake, which are regulated to maintain a minimum lake level with an average depth of 1 m (Korhnak, 1996
). Giant cutgrass first appeared in Lake Alice in the 1970s (W. T. Haller, University of Florida, personal communication), and the 3–10 m wide fringe around the lake had not noticeably expanded in the 1980s.
Lake Seminole, a reservoir on the lower Chattahoochee and Flint Rivers (30°45' N, 84°50' W) contained 3440 ha of giant cutgrass in 1992 (J. Kight, U.S. Army Corps of Engineers, Lake Seminole, personal communication). Although the lake-wide coverage of giant cutgrass had not increased greatly in the preceding decade (since Gholson, 1984
), this was partly because areas of continuing localized expansion were offset by losses due to vegetation control programs in other areas. The study site in Lake Seminole was Fox Island, a 5.1-ha island located 1.5 km south of Sealy Point Landing, Georgia.
Tracking of individual plant development in the field
In March and April 1988, 40 plants with immature inflorescences were selected from around the littoral edges of both Lake Alice and Fox Island (Lake Seminole) and were marked with flagging tape. Every 6 wk the following variables were measured on each plant: water depth at the plant base, total number of nodes per plant, internode lengths, number of nodes with shoots and total number of axillary shoots, number of nodes with roots and number of nodes rooted in the substrate, and total number of inflorescences produced.
In April 1989, 20 plants were flagged and attached at their base to clearly visible PVC (polyvinyl chloride) poles on the south and west shorelines of Lake Alice and on the north and west edges of Fox Island. The variables listed above (except internode lengths) again were measured at 6-wk intervals. Even if the horizontal stem broke, data on node and axillary shoot numbers were still collected, provided that the origins of separated sections of stem and/or of rooted shoots could be clearly identified by the presence of flagging tape.
For each plant tracked from April to August 1988 in Lake Alice, the length of the longest adventitious root at any of the nonrooted flowering stem nodes was recorded. Similar measurements were made in August 1989 on Fox Island, prior to substantial node loss or rooting. The ten greatest lengths from each site were then compared with a Student's t test. Only the longest ten root lengths were compared because the objective was to compare maximum possible root lengths, not the average. Averages would have been influenced by water depth and the incidence of rooting, with shallow sites averaging more short roots remaining unrooted than at deep sites.
Daily water elevations at the Jim Woodruff dam (8 km from Fox Island) on Lake Seminole were provided by the U.S. Army Corps of Engineers. Average monthly elevations were calculated from these data. At each sampling date, comparisons of water elevation at the dam with water depths observed at a fixed scale on Fox Island indicated that changes in water elevations at the dam were proportional to water-level changes that occurred at Fox Island (A.M. Fox, University of Florida, unpublished data).
Data were analyzed per sampling time by Student's t test or by analysis of variance.
RESULTS
In 1988, when tagged plants were marked only with flagging tape, it was difficult to keep track of all plants for more than 2 or 3 mo. Less than half of the 40 plants tagged in each lake in April were found by the end of June. This was either because flagged stems already had been broken off and lost (especially in Lake Alice) or because plant growth and expansion had been so rapid that the original flagged stems were buried in other giant cutgrass foliage (particularly in Lake Seminole).
Depending upon which of these eventualities had occurred to most of the plants lost from a site, the remaining tracked plants would bias the estimated average for that site. The overall tendency of this bias would probably be to indicate larger plants than the true population average, since it was usually broken plants that were lost. This problem was overcome in 1989 by attaching the plant bases to clearly marked poles. For this reason most of the discussion of results from the tracked plants will concentrate on 1989 data.
Maximum number of nodes along the main flowering stem axes
To make 1988 and 1989 data comparable (Fig. 3), the average number of nodes along the main flowering stem were calculated per unbroken plant in 1989 rather than for the total number of all plants tracked (see "Total remaining nodes" below). Thus, as the season progressed and stems were broken, fewer plants were included in the survey so that by December 1988 only two plants were being tracked in Lake Alice and six in Lake Seminole. By November 1989, only two plants remained intact in Lake Alice and 15 in Lake Seminole.
|
In Lake Seminole flowering started in April and May, but maximum numbers of nodes per plant were not produced until November or December. The longest individual plant had as many as 22 longitudinally attached nodes by October 1988 (Fig. 2).
Numbers of nodes on the longitudinal axes of plants that were growing out perpendicular to the stand edge would relate to estimates of rates of stand expansion. Average internode length was estimated from 21 Lake Seminole plants tagged in 1988 as 24.0 cm (N = 195, SE = 0.6). Thus, when maximum numbers of longitudinal nodes were present, stands in Lake Seminole had potentially expanded by 2.7 and 2.2 m in 1988 and 1989, respectively. This contrasted with the maximum potential expansion in Lake Alice of 1.3 m in 1988 and 1.0 m in 1989.
A comparison of the basal and distal three internode lengths per plant (from Lake Seminole), showed that internodes were longer toward the base of the plant (32.2 cm) compared to the distal end of the stem (19.8 cm; P < 0.05, Student's t test, N = 63). This average basal internode length was shorter than the comparable values of 48–51 cm reported by Holmes and Stalling (1990)
.
Total remaining nodes
These data for 1989 consisted of the average number of all nodes (includes side branches, not just those on the main axis, e.g., see near plant base in Fig. 2) for all tracked plants (i.e., including rooted remains of broken stems). Although average values were lower for these data (not shown) than for maximum longitudinal node numbers (as above) the overall trends with time were similar. The maximum total number of remaining nodes in Lake Alice was achieved in late June (3.8 nodes per plant) with a linear decline to only 0.3 nodes by late November. The peak number of total remaining nodes in Lake Seminole occurred in late September (8.5 nodes per plant) with a steady decline until the following May, when the average was 3.2 nodes, but one plant had nine nodes remaining.
Production of axillary shoots
Throughout the growing season, and at both sites, the average number of nodes with axillary shoots increased in proportion to the total number of nodes. In Lake Alice in 1989, the maximum average of 1.4 nodes with shoots per plant, and of 1.9 axillary shoots per plant, occurred in late June and declined thereafter. The percentage of nodes with shoots varied around 40% during this period (Fig. 4A). Since the number of nodes with shoots declined faster than the total number of shoots, the number of shoots per node (averaged for only nodes with shoots) actually increased throughout the summer to a maximum of 2.1 shoots per node by late August (Fig. 4B).
|
The total number of axillary shoots per plant increased linearly in Lake Seminole from April to a peak in late September when 17.8 shoots per plant occurred. This peak in total numbers of axillary shoots coincided with a maximum of 4.3 shoots per node (Fig. 4B). The number of shoots per node then declined through the winter until January when this value increased again with the start of the new year's growth season (Fig. 4B). The average number of shoots per node in Lake Seminole reflected changes in growth pattern of undamaged plants. In a few instances in both sites, particularly where terminal nodes were damaged and stem expansion could not continue, many (up to 25) small shoots were produced at a single node. Often only a few of these shoots survived, or the resulting large and heavy node was broken off the parent plant.
Production and maximum length of adventitious roots and likelihood of rooting
The maximum number of nodes with roots in Lake Alice occurred in late June with 1.5 nodes per plant. Thus, about the same proportion of all nodes had roots (40%; Fig. 4C) as had shoots. Comparable values in Lake Seminole were 6.0 nodes with roots per plant (slightly higher than the number of shoots) and 64% of all nodes (Fig. 4C). Although in each site the percentages of nodes with shoots or roots were very similar, this did not necessarily mean that nodes with shoots always had roots; both roots and shoots could grow separately from each other. The typical percentage of nodes with shoots had been attained in each site by the end of May, but the proportion of nodes with roots continued to increase until August.
The average length of the ten longest adventitious roots at nonrooted nodes in Lake Alice of 29.6 cm (SE = 1.4) was shorter than the average length of 46.5 cm (SE = 1.1) in Lake Seminole (P < 0.05, Student's t test).
In 1988 only one node on a tagged plant in Lake Alice was found rooted into the substrate, and no rooted nodes were found there in 1989 (Table 1). The total number of rooted nodes in Lake Seminole reached a maximum for 1988 plants of 8.7 nodes per plant in January 1989 (but only three plants remained). This represented 81% of all nodes and 96% of nodes that had roots (Fig. 5). The latter percentage had steadily increased since late May 1988 when 51% of all nodes with roots were rooted (Fig. 5). Prior to that time, none of the nodes had roots.
|
|
Water depths at the bases of the tagged plants, node, and adventitious root production were compared for 1988 and 1989, within and between the lakes (Table 1). Data were used from late June/early July, prior to most 1988 plants being lost in Lake Alice. Water depths were deeper in Lake Seminole in 1989 compared with 1988, and in Lake Alice compared with Lake Seminole in 1988 (P < 0.05, Student's t tests). Significant numbers of nodes only became rooted when water depths in July were shallow, such as the 22 cm deep water in Lake Seminole in 1988, despite similar percentages of nodes having roots in each year.
Comparisons of water elevations at the Lake Seminole dam for 1988, 1989, and 1990 with year as the main effect showed that, averaged over the whole year, water levels in Lake Seminole were higher by 13 cm in 1989 than in 1988 (Fig. 6; P < 0.05, ANOVA). This difference increased to 23 cm if compared for the typical growing season from May to November (period when new nodes are produced). Average monthly water elevations for June differed by 53 cm between 1988 and 1989, with June 1988 showing the lowest water levels in the 3.5 yr of data analyzed (Fig. 6).
|
The maximum numbers of inflorescences produced by any individual plant were three in Lake Alice in 1988 and two in 1989, and in Lake Seminole, six in 1988 (Fig. 2) and five in 1989. The average numbers of inflorescences per plant in Lake Alice were 1.7 (SE = 0.2) and 1.2 (SE = 0.1) for 1988 and 1989, respectively, and 2.8 (SE = 0.3) and 2.4 (SE = 0.3), respectively, in Lake Seminole. No additional inflorescences were formed after June in Lake Alice or after September in Lake Seminole.
Initial inflorescences may be associated with stems composed of as few as two, or as many as five, nodes. Typically, subsequent inflorescences arise only from the distal node on the longitudinal axis. Occasionally inflorescences form at both the distal node and the adjacent one, but usually only the distal inflorescence continues to produce axillary shoots with further inflorescences and nodes.
DISCUSSION
The type of functional stolons described in this study, which develop axillary shoots, adventitious roots, and secondary inflorescences from the nodes of decumbent flowering stems, are produced by various members of the Poaceae (David W. Hall, D. W. Hall Consulting, Inc, personal communication). These data provide quantitative support for the field observations of Cutshall, Glennon, and Biles (1989)
that the vegetative growth from the nodes will "... produce a seedhead, whether the new plant is rooted to the soil or floating." It may be surmised that Holmes and Stalling (1990)
did not observe secondary inflorescences, or more than three vegetative buds per flowering stem, on the giant cutgrass that they observed for two reasons. Firstly, their plants were collected from a pasture drainage ditch in which stems toppled over and the axillary shoots made "... contact with the substrate." The continued growth of axillary shoots and the production of secondary inflorescences may not have occurred if the substrate was not sufficiently wet to stimulate further adventitious root production (A. M. Fox, University of Florida, unpublished data). All plants studied in Lakes Alice and Seminole were at the lakeside edge of stands, such that the decumbent stems were in open water.
The other important issue was that Holmes and Stalling collected their samples from central Louisiana between 2 May and 12 June 1987. Although secondary inflorescences were produced in Lakes Alice and Seminole prior to mid-June in 1988, it is possible that differences in annual climatic conditions, site, or plant population characteristics could be responsible for a later, or less common, development of these structures at the Louisiana site. The greater basal internode lengths noted by Holmes and Stalling (1990)
, compared to plants from Lake Seminole, could indicate population differences, which might coincidentally include retarded secondary inflorescence formation. Thus, Holmes and Stalling may have ceased collecting samples too early to detect this continued sexual function of the functional stolons.
If axillary shoots are able to take root in the substrate prior to breakage of the functional stolons from the parent plant, these shoots can survive to initiate the new lakeward edge of the stand. For plants in Lake Seminole, although only 3.2 nodes remained by May 1990 (when averaged over all tagged plants including those that had failed to produce functional stolons or that had been broken prior to nodes becoming rooted), this represented an average stand expansion of 0.8 m/yr. One in four plants had more than five nodes remaining by May 1990, representing over 1.2 m of stand expansion, and the functional stolon of the plant that retained nine nodes was 2.2 m long. With approximately four inflorescences produced per square metre along a stand edge (A. M. Fox, University of Florida, unpublished data), this indicates that on average one plant in every metre along the shoreline would produce shoots that survive into the next growing season and extend 1.2 m into the lake.
Such stand expansion would have been greater if there had been a better survival rate of nodes from 1989 into 1990. The peak distance of stand expansion, averaged for all plants in September 1989, would have been over 2 m (8.5 nodes) but with only 40–50% of nodes becoming rooted in the substrate in late 1989, it was inevitable that this potential for expansion would decline over winter. Stand expansion of 1–2 m/yr was found to be fairly typical on Fox Island when this parameter was specifically evaluated (A. M. Fox, University of Florida, unpublished data).
This contrasts with the situation in Lake Alice where no nodes became rooted in the substrate in 1989 and hence no stand expansion was likely from functional stolons. A single node became rooted and survived in 1988, representing ~1 m of lakeward stand expansion by one plant in 5 m of shoreline. This illustrates that differences in the ability of axillary shoots to become rooted in the substrate probably explain the contrasting rates of stand expansion observed at the two sites. Although the percentages of nodes with roots may be lower in Lake Alice than in Lake Seminole (Fig. 4C; 1989 in Table 1), this is not sufficient to explain the extreme differences in proportions of nodes that become rooted in the substrate. This will be influenced by four main factors: root length, water depth, likelihood that the functional stolon will be broken and float away, and substrate type.
Maximum root lengths were not only significantly shorter in Lake Alice than in Lake Seminole, but at 29.6 cm they were shorter than the average water depth of over 50 cm at the plant bases in late June 1988 (Table 1). This was the time at which the single node had become rooted in Lake Alice, and by which over 50% of nodes with roots had become rooted in Lake Seminole. Water levels do not vary widely in Lake Alice because they are regulated such that while higher water levels may occur after heavy rain, a minimum level is usually maintained even during prolonged dry periods. Thus, significant stand expansion from functional stolons might only be expected in Lake Alice if water levels were allowed to fall >10 cm below average minima or if the adventitious roots on giant cutgrass nodes were longer.
Why maximum possible adventitious root lengths should be different between the two lakes is not known but could be related to differences in water quality. Water concentrations of total nitrogen and phosphorus were typically 0.7 mg/L N and 0.03 mg/L P in Lake Seminole (Meadows, Martin, and Mixson, 1992
) and 2.5 mg/L N and 1.1 mg/L P in Lake Alice (Korhnak, 1996
). Roots were reported to be larger on water hyacinths grown in water with low nutrient concentrations compared to water hyacinths grown in nutrient-rich conditions (Knipling, West, and Haller, 1970
; Richards, 1982
). The same influence of nutrient availability may be responsible for the shorter length of adventitious roots on giant cutgrass growing in the nutrient-rich waters of Lake Alice compared to those found in Lake Seminole. The likelihood of this influence would depend upon whether the growth of adventitious roots is affected more by the quality of the surrounding water (as in floating plants) or by the nutrient conditions of the sediments that are affecting the main roots of the parent plant. It would be useful to study the relationships between these influences and between the nutrient contents of the soil and water in these lakes, to further explain the differences in adventitious root lengths between these sites.
Adventitious roots on plants in Lake Seminole were long enough at 46.5 cm to reach and take root in the substrate when water levels were relatively low, such as the average water depth of 21.9 cm in July 1988 (Table 1). With an even lower lake elevation in June 1988 (Fig. 6) it is not surprising that almost 70% of nodes with roots were rooted by early July. In contrast, the significantly higher lake elevations during these months in 1989, which resulted in average water depths at the plants of over 50 cm by late June (Table 1), indicate why none of the almost 60% of all nodes that had roots was rooted in the substrate. The ~20-cm decrease in lake elevation during August and September 1989 (Fig. 6) corresponded with the period when the maximum number of nodes became rooted in the substrate (Fig. 5).
Such delays in nodes becoming rooted would increase the likelihood that functional stolons might be broken prior to stabilization. Although the specific causes of functional stolon breakage were not determined nor measured in this study, it was evident by December 1989 when no functional stolons remained unbroken in Lake Alice, that disturbance was an influential factor in this lake. Disturbance along the stand edge in Lake Alice was most likely from wave action. Collisions from floating debris were more likely in Lake Seminole, where many logs or rafts of uprooted submersed plants, such as hydrilla [Hydrilla verticillata (L.f.) Royle], have been found along the shoreline of Fox Island. Disturbances by animals or humans are other possible mechanisms for functional stolon damage or mechanical stresses in flowing water.
Broken sections of functional stolons that have axillary shoots will float (Holmes and Stalling, 1990
) and, judging from broken sections which remained tagged to their parent plants, remain viable for many weeks. In 1988, a large raft of many broken functional stolons was observed floating in Lake Alice, eventually running aground and forming a new giant cutgrass island on the northeast shore. Other sections of shoreline in Lake Alice, which had been cleared by back-hoe to facilitate observation of the lake and its alligators by visitors to the university, were recolonized by floating sections of giant cutgrass within a few weeks. The potential for long-distance dispersal of giant cutgrass by such vegetative mechanisms was suggested by Holmes and Stalling (1990)
and Fox and Haller (1990)
. Possible consequences of these mechanisms of vegetative reproduction and dispersal on the distribution and management of giant cutgrass in sites with new, expanding, or established populations were noted by Fox (1993)
.
It is possible that substrate quality might influence the ability of axillary shoots to become, and remain, rooted, particularly for extremely flocculent or hard sediments. Soil samples were not collected nor compared for these sites although it was noted that there was a much greater proportion of sand in the Lake Seminole substrate. Observations from wading around in Lake Alice did not indicate that this substrate was exceptionally hard or flocculent, but this would be an interesting factor to investigate further.
This study has quantitatively documented the production and survivorship of secondary inflorescences and functional stolons by giant cutgrass. The potential for these reproductive mechanisms to facilitate stand expansion at rates of 1–2 m per year is evident from these data. Differences between the survival and growth of functional stolons in different sites could adequately explain contrasting rates of stand expansion. A relationship between functional stolon survival and the rooting and persistence of axillary shoots with factors such as water depth and disturbance has been proposed and should be investigated further.
FOOTNOTES
1 The authors thank Margaret Glenn and Jan Miller for valuable assistance in the field and Randall Stocker and David Sutton for reviews of the draft manuscript. Financial support was provided in part by the U.S. Department of Agriculture and the Center for Aquatic and Invasive Plants under cooperative agreement Number ARS 58-43YK-9-0001. Published as Florida Agricultural Experiment Station Journal Series Number R-06618. 
LITERATURE CITED
Baden, J., W. T. Batson, and R. Stalter. 1975 Factors affecting the distribution of vegetation of abandoned rice fields, Georgetown Co., South Carolina. Castanea 40: 171–184.
Birch, J. B., and J. L. Cooley. 1982 Production and standing crop patterns of giant cutgrass [Zizaniopsis miliacea] in a freshwater tidal marsh. Oecologia 52: 230–235.[CrossRef][ISI]
Burbank, J. H. 1963 An evaluation of the aquatic pest plant control program at Reelfoot Lake. Journal of the Tennessee Academy of Science 38: 42–48.
Cook, C. D. K. 1990 Aquatic plant book. SPB Academic Publishing, The Hague, The Netherlands.
Cutshall, J. R., R. Glennon, and L. T. Biles. 1989 Vegetative propagation of giant cutgrass for fresh marsh erosion control. In W. G. Duffy and D. Clark [eds.], Proceedings of a Symposium, Marsh management in coastal Louisiana: effects and issues, 239–241. U.S. Fish and Wildlife Service Biological Report 89(22), Washington, D.C., USA.
Fox, A. M. 1993 Giant cutgrass—an unfriendly native. Aquatics (Magazine of the Florida Aquatic Plant Management Society) 15 (4): 4–9.
———, and W. T. Haller. 1990 Reproductive strategies in giant cutgrass [Zizaniopsis miliacea]. Proceedings of the European Weed Research Society Eighth Symposium on Aquatic Weeds 8: 103–104.
Gholson, A. K. 1984 History of aquatic weeds in Lake Seminole. Aquatics (Magazine of the Florida Aquatic Plant Management Society) 6(4): 17.
Holmes, W. C., and D. T. Stalling. 1990 Studies on the reproductive strategy of Zizaniopsis miliacea (Michx.) Doell. & Asch. (Gramineae: Tribe Oryzeae). Castanea 55: 113–121.
Kight, J. 1980 USAE Division/district presentations, aquatic plant problems—operations activities. Proceedings of the 14th Annual Meeting Aquatic Plant Control Research Planning and Operations Review, 26–29 November 1979, 57–61. Miscellaneous Paper A-80–3. U.S. Army Engineer Waterways Experiment Station, Vicksburg, Mississippi, USA.
Knipling, E. B., S. H. West, and W. T. Haller. 1970 Growth characteristics, yield potential, and nutritive content of water hyacinths. Proceedings of the Soil and Crop Science Society of Florida 30: 51–63.
Korhnak, L. V. 1996 Water, phosphorus, nitrogen, and chloride budgets for Lake Alice, Florida, and documentation of the effects of wastewater effluent removal. Master's thesis, University of Florida, Gainesville, Florida, USA.
Latham, P. J., L. G. Pearlstein, and W. M. Kitchens. 1994 Species association changes across a gradient of freshwater, oligohaline, and mesohaline tidal marshes along the lower Savannah River. Wetlands 14: 174–183.[ISI]
Martin, A. C. 1953 Improving duck marshes by weed control. U.S. Department of the Interior, Fish and Wildlife Service Circular Number 19, U.S. Government Printing Office, Washington, D.C., USA.
Meadows, P. E., J. B. Martin, and P. R. Mixson. 1992 Water resources data Florida, Water year 1990, Vol. 4, Northwest Florida. U.S. Geological Survey Report USGS-WDR-FL-91-4.
Odum, E. P., J. B. Birch, and J. L. Cooley. 1983 Comparison of giant cutgrass productivity in tidal and impounded marshes with special reference to tidal subsidy and waste assimilation. Estuaries 6: 88–94.[CrossRef][ISI]
Richards, J. H. 1982 Developmental potential of axillary buds of water hyacinth, Eichhornia crassipes Solms. (Pontederiaceae). American Journal of Botany 69: 615–622.[CrossRef][ISI]
Smart, R. M., and J. W. Barko. 1982 Ecology of giant cutgrass [Zizaniopsis miliacea] in Lake Seminole. Proceedings of the 16th Annual Meeting of the Aquatic Plant Control Research Planning and Operations Review, 17–19 November 1981, 107–109. Miscellaneous Paper A-82–3. U.S. Army Engineer Waterways Experiment Station, Vicksburg, Mississippi, USA.
Steenis, J. H., and C. Cottam. 1945 A progress report on the marsh and aquatic plant problem: Reelfoot Lake. Journal of the Tennessee Academy of Science 20: 6–19.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
