Dynamics of leaf litter accumulation and its effects on riparian vegetation: a review.
The dynamics of plant litter is important in most plant communities (Facelli & Pickett, 1991a), and a process that is closely linked to climatic conditions and environmental management. Therefore, predictions of the effects of global and local changes in climate and land use should seriously consider litter cycling. Besides being the substrate for the detritus food web, a nutrient reservoir, and a contributor to mineral-nutrient cycling and energy flow within ecosystems (Odum, 1960), litter may deeply alter the microenvironment and the structure and dynamics of plant communities (Sydes & Grime, 1981a, 1981b; Beatty & Sholes, 1988; Carson & Peterson, 1990; Facelli & Pickett 1991a, 1991b, 1991c; Facelli & Facelli, 1993).
Litter may be especially important in riparian corridors because of a higher production and a faster decomposition than occur in uplands (for litter production, see Conner & Day, 1976; Shure & Gottschalk, 1985; Brinson, 1990; Malanson, 1993; for litter decomposition, see Merritt & Lawson, 1980; Malanson, 1993), and the redistribution of litter varies strongly both between years and between parches (Nilsson & Grelsson, 1990; Nilsson et al., 1993). The big litter source in riparian corridors implies a high potential for influences on riparian and aquatic communities. A high rate of litter decomposition may imply not only a high rate of mineral recycling which favors primary production but also an availability of bare soil for plant colonization during the later stages of decomposition. The high dynamics of litter redistribution may contribute to the heterogeneity of riparian habitats, including a variety of substrates, nutrient conditions, and vegetation mosaics.
Furthermore, during floods, riverborne litter enhances species flows within riparian corridors, both by providing floating devices for seed dispersal and by creating bare ground for plant colonization on erosion sites (Nilsson & Grelsson, 1990). Not surprisingly, being a carrier of materials and energy in these processes, litter is an essential element in the concepts of the river continuum (Vannote et al., 1980), the nutrient spiraling (Webster, 1975), the riparian filtering (Ward, 1989), and the flood pulse (Junk et al., 1989).
Recent studies on plant litter comprise many different ecosystems, including forests (Sydes & Grime, 1981a, 1981b; Abbott & Crossley, 1982; Vogt el al., 1986; Staaf, 1987; Beatty & Sholes, 1988), grasslands (Rice, 1979; Knapp & Seastedt, 1986; Fowler, 1988), oldfields (Pastor et al., 1987; Carson & Peterson, 1990; Facelli & Carson, 1991; Facelli & Pickett, 1991b, 1991c), arid ecosystems (de Jong & Klinkhamer, 1985), fresh waters (Polunin, 1984; van der Valk, 1986), and river corridors (Keller & Swanson, 1979; Conners & Naiman, 1984; Chauvet & Decamps, 1989; Nilsson & Grelsson, 1990). However, most studies have focused on inputs and outputs of litter (Whittaker & Woodwell, 1969; Conners & Naiman, 1984; Cuffney, 1988; Chauvet & Jean-Louis, 1988), nutrient dynamics of litter (Aber & Melillo, 1980; Peterson & Rolfe, 1982; Pastor et al., 1987), and rates of litter decomposition (Abbott & Crossley, 1982; Chauvet, 1988; Berg el al., 1993); only a few were conducted to reveal the direct effects of litter on vegetation (Werner, 1975; Cheplick & Quinn, 1987; Facelli & Pickett, 1991a). Not until recently has attention been paid to the effects of plant litter on the dynamics and structure of plant communities (Sydes & Grime, 1981a, 1981b; Beatty & Sholes, 1988; Carson & Peterson, 1990; Nilsson & Grelsson, 1990; Facelli & Pickett, 1991b, 1991c; Facelli & Facelli, 1993).
Much work on litter processes has been done in aquatic and adjacent riparian areas (Malanson, 1993, and references therein), mainly emphasizing the flows of matter and energy in terms of production (Brinson, 1990), decomposition (Chauvet, 1988), nutrient dynamics (Brinson et al., 1980), and transport (Conners & Naiman, 1984; Cuffney, 1988). Other papers deal with the importance of litter for terrestrial (Merritt & Lawson, 1980) and aquatic animal communities (Petersen et al., 1989). However, little is known about the importance of litter in structuring riparian vegetation (but see Nilsson & Grelsson, 1990).
In this paper, we review the present knowledge of the production, redistribution, and decomposition of plant litter within the riparian corridor. We compare the differences in physical, chemical, and biological impacts between riparian and other systems; and we discuss the effects of leaf litter on the structure and dynamics of riparian vegetation.
III. Litter Dynamics in the Riparian Zone
A. LITTER PRODUCTION
Almost all data available about litter fall in riparian areas are from the past two decades and consider aboveground litter production and riparian forests (Brinson, 1990; Malanson, 1993); examples from riparian meadows (Gurtz et al., 1988), riparian oldfields (Webster & Patten, 1979), and other riparian ecosystems are scarce. Most studies of litter production in riverine areas represent single cases, and comparisons with other ecosystems are rare. A few researchers have attempted to review geographical patterns of litter production (Conner & Day, 1976; Brinson, 1990), but their data are still too limited to reveal the global features. In this section, we compare litter production patterns between terrestrial and riparian forests on a global scale.
We analyzed the global relationship between riparian litter production and latitude and found a negative relationship ([[r.sup.2].sub.adj] = 0.35, P [less than] 0.0001, N = 87; [ILLUSTRATION FOR FIGURE 1 OMITTED], similar to the global pattern for upland forests (Bray & Gorham, 1964; Meentemeyer et al., 1982; Vogt el al., 1986). Most riparian forests [ILLUSTRATION FOR FIGURE 1 OMITTED], deciduous as well as coniferous, exceed the world averages of upland litter production (Vogt et al., 1986). Only a small fraction of the sites included fell below the upland means [ILLUSTRATION FOR FIGURE 1 OMITTED], and most of these were represented by still-water swamps (Conner & Day, 1976; Brown, 1981; Brown & Peterson, 1983) and CPOM (coarse particulate organic matter) inputs in headwaters (de la Cruz & Post, 1977; Cowan & Oswood, 1983; Conners & Naiman, 1984).
The estimation of litter production presents some sampling problems. For example, litter may lose weight through leaching if the litter trap protrudes above the water surface (Dawson, 1976; Blackburn & Petr, 1979). Small traps used for CPOM investigation may exclude branches and twigs (Schlesinger, 1978; Webster & Patten, 1979; Conners & Naiman, 1984) as well as big leaves, Thus, the data included in Figure 1 may represent underestimates. The result strongly indicates that, all over the world, litter production in riparian forests is higher than that in terrestrial systems, supporting the suggestions made by Brinson (1990) and Malanson (1993).
As pointed out by Facelli and Pickett (1991a) for upland ecosystems, climatic conditions are the major determinants of the latitudinal patterns for litter production. Light availability during the growing season (Jordan, 1971), minimum monthly mean temperature and precipitation (Vogt el al., 1986), and actual evapotranspiration (Meentemeyer el al., 1982) are all strongly related to the production of forest litter, as is net primary production.
Litter production should be positively related to the total net production of aboveground biomass. Brinson (1990) estimated that litter production accounted for ah average of 47% of the annual aboveground production in riparian forests. This level is higher than that of 39.4% measured by Whittaker and Woodwell (1969) in an oak-pine forest, and 40.6% by Hughes (1971) in an alder-birch forest, both in uplands, indicating that litter may account for a higher proportion of biomass gain in riparian zones than it does in uplands. It remains to be seen, however, whether litter production is consistently related to total net primary production, both in the uplands and in the riparian zone, all over the world (Meentemeyer el al., 1982).
Water-level fluctuation and soil conditions are of specific importance in the riparian zone. Many such areas receive, regularly or irregularly, fluvial inputs of water, oxygen, and nutrients that can increase nutrient cycling and productivity (Brinson el al., 1980; Brinson, 1990; Shure & Gottschalk, 1985). In a comparison of forests with changed flooding regimes in Louisiana, Conner el al. (1981) found that infrequent flooding resulted in higher litter production (549 g [m.sup.2]) and permanent flooding resulted in lower litter production (328 g [m.sup.2]) than in the naturally flooded area. Cuffney (1988) also found that litter fall in the habitats flooded [less than] 25% of the year was higher than that in habitats flooded 60-90% of the year. Shure and Gottschalk (1985) reported that leaf litter production was highest in the transition zone between the riverbank and the uplands, and lower toward the uplands and the frequently flooded riverbank. Although different studies face differences in local topography, community composition, and successional stage, they suggest that the magnitude and timing of flooding, soil water content, and soil fertility largely control litter production as well as net primary production in the riparian zone (Brinson, 1990).
Natural disturbances may also cause variation in litter production. Strong winds can make leaves fall prematurely (Conner & Day, 1976) and increase the litter amount further by adding twigs and branches. Ice storms also increase the proportion of wood litter (Peterson & Rolfe, 1982). Insect consumption can reduce litter production (Conner & Day, 1976), and beaver activity may be responsible for both increasing and decreasing litter production.
Leaf litter accounts for 72% and 80% of the total litter production in broad-leaved and needle-leaved riparian forests, respectively. Both exceed the expected worldwide average for uplands of 70% (Meentemeyer et al., 1982). Such a high proportion of leaf litter has been observed by several authors (Gosz et al., 1972; Shure & Gottschalk, 1985; Cuffney, 1988) and was attributed to the big leaf area in these alluvial habitats (Shure & Gottschalk, 1985). However, Cuffney (1988) argued that the relatively large fraction of leaves might be due also to an underestimation of the branch litter. On the other hand, our present study shows that the fraction of leaves in the litter of riparian forests has no significant relation to latitude ([[r.sup.2].sub.adj] = 0.05, P = 0.14, N = 44). It seems that local factors in the riparian zone may be responsible for the variations in the proportion of leaf litter. Muzika et al. (1987) reported that the leaf and stem litter of early successional sites exceeded that of late successional sites by 1.66 and 1.54 times, respectively, whereas the flower and fruit litter of the latter sites exceeded that of the former by 6 times.
Another piece of information still lacking is the magnitude of litter production in riparian types of vegetation other than forest. Among the few studies available so far, most data consider CPOM inputs to the water channel (Otto, 1975; Webster & Patten, 1979; Conners & Naiman, 1984; Gurtz et al., 1988; Chauvet & Decamps, 1989), ranging from 100 to 780 g [m.sup.2]. No clear latitudinal pattern can be seen.
B. LITTER REDISTRIBUTION
Litter redistribution along riparian corridors includes three major processes: erosion, transport, and deposition.
Erosion of litter is mainly related to water flood periods when water moves over vegetated ground. Litter erosion is strongest at sites with high current velocity during floods, such as along rapids and at the upstream end of islands (Nilsson et al., 1993). By placing tagged leaves and sticks on two differently elevated floodplains prior to flooding of the Ogeechee River, U.S.A., Cuffney (1988) found that a more frequently flooded site had larger displacements than did a less frequently flooded site. He also found, on both sites, that the degree of displacement decreased from the river channel toward the top of the riparian zone. This is consistent with our field observations along boreal rivers where most litter is deposited as a string at the top of the riparian zone (Xiong et al., pers. obs.). However, the structure of riparian vegetation may largely change the amounts and patterns of litter output to the rivers. Gurtz et al. (1988) reported that lateral transport of litter increased from 5.3 g m of the length of riverbank in a prairie section to 18.1 g m in a prairie/shrub section to 368.9 g m in a gallery forest section. When transformed, the data represent 10%, 19%, and 35% of the litter production for the prairie, prairie/shrub, and gallery forest sections, respectively. Conners and Naiman (1984) also reported that lateral transport of litter to rivers was more strongly influenced by vegetation structure and sediment obstacles than by stream size. Sediment accumulated on the ground has been found to prevent litter from being eroded by floods (Conners & Naiman, 1984; Shure et al., 1986). The slope of the riparian zone has also been suggested as a controlling factor (Fisher, 1977).
There are only few data on the amounts of direct litter fall from trees into the river. Focusing on organic inputs to the river, Conners and Naiman (1984) determined that direct litter fall per unit of stream surface declined exponentially with increasing river order, ranging from 421 g [m.sup.2] yr AFDM (ash free dry matter) in a first-order stream to 16 g [m.sup.2] in a sixth-order river. These masses correspond to about 52% and 90% of the total inputs of organic matter in the first- and sixth-order rivers, respectively (Conners & Naiman, 1984). In a small beech forest stream in New Zealand, Winterbourn (1976) found that about 81% of 567 g [m.sup.2] annual litter input to the river was directly fallen. Chauvet and Jean-Louis (1988) also showed that an average of 42 g [m.sup.2] yr litter (water area basis) from a Salix stand fell directly into the river, representing a large source of litter exported from the floodplain to the river. It has been suggested that the direct litter fall from riparian forests to rivers may be a major part of the total allochthonous input to rivers, whereas the lateral input of litter removed from the ground is far less (Chauvet & Dcamps, 1989). Flood events and weather conditions, especially wind, may largely influence the direct litter fall to the river.
The pattern of litter transport along the river depends on the hydrological regime. The transport distance is positively correlated with the discharge (Winterbourn, 1976; Cummins el al., 1983), and negatively correlated with the elevation of the riparian site (Cuffney, 1988).
The waterborne litter deposits on the banks of boreal rivers may range in size from a few hundred to several thousand grams per square meter (Nilsson & Grelsson, 1990; E. Nilsson el al., unpubl. data). This deposition occurs mainly along outer bends of rapids and in backwater reaches downstream from rapids (Johansson & Nilsson, 1993). Riparian trees and shrubs may effectively trap riverborne litter. For example, Hardin and Wistendahl (1983) reported that litter accumulation was greatest next to trees, especially on the upstream side. Both the abundance and the composition of litter are quite variable, from none to high and from fragments of leaves to big logs (Nilsson & Grelsson, 1990).
Litter packs may be of local origin or may come from upstream regions of the surrounding uplands. However, we know little about how the proportions of these three types of litter vary among and within rivers. Cuffney (1988) reported that between 2% and 100% of locally produced litter remained in situ after a major flood, and the degree of output from the riparian zone increased with the duration of the flood. In effect, much of the deposits of litter originating from upstream regions has been moved by the water not only in the upstream-downstream direction but also from lower to upper parts of the riparian zone.
C. LITTER DECOMPOSITION
Roughly, litter decomposition in wetlands comprises two major stages: an initial rapid weight loss, primarily through leaching, and a later breakdown phase (Polunin, 1984). The leaching stage is mainly an abiotic loss of soluble components, whereas the breakdown stage depends largely on the activities of decomposer organisms (Polunin, 1984). However, sometimes microbes and invertebrates can stimulate leaching (Merritt & Lawson, 1980; Day, 1983), and particle fragmentation is likely to facilitate breakdown.
Leaching is a prominent process during the first stages of litter inundation. Petersen and Cummins (1974) reported that the majority of soluble organic matter could leach from the leaves in the stream within the first 24 hours of initial wetting. Blackburn and Petr (1979) reported a weight loss of as much as 10-40%, depending on species and types of litter, within 2 weeks, with 7-18% net loss through leaching. By means of controlled leaching experiments in microcosms, Day (1983) found that 31% of litter net weight was leached into the water during the first 93 hours. Neiff and Poi de Neiff (1990) demonstrated a litter half-life of 20 days, once the litter was transferred to the stream. Not surprisingly, after seasonal flooding every year, stranded litter (i.e., riverborne litter) may have quite a low nutrient content because of leaching losses (Xiong & Nilsson, pers. obs.).
Litter breakdown is associated with the activity of invertebrates, which, besides moisture, is considered the most important factor governing litter decomposition in riparian areas (Malanson, 1993). It has been reported that soil animals, mainly macroinvertebrates, are more abundant and diverse in floodplain woodlands than in the adjacent uplands, and responsible for at least 29-32% loss of original litter weight in the whole season, although low oxygen in the soil reduces subterranean animal activity (Merritt & Lawson, 1980). Microbes may also play an important role in the litter breakdown in riparian areas. It has been found that the litter exposed to air is decomposed mostly by fungi (Holland & Coleman, 1987; Facelli & Pickett, 1991a), whereas submerged litter is predominantly processed by bacteria, especially at an early stage (Polunin, 1984). Probably, fungi and bacteria can shift their relative abundances in riparian areas following the fluctuations between wet and dry conditions. Decomposition by fungi is quite slow during dry conditions (Holland & Coleman, 1987), but bacteria may burst after wetting (Polunin, 1984) and result in an increase in litter decomposition. However, Gessner and Chauvet (1994) recently reported that aquatic fungi ate especially important for leaf decomposition in streams. They found that high species diversity of hyphomycetes and peak fungal biomass coincided with roughly a 50% loss in leaf mass in a French stream (Gessner el al., 1993). It has been estimated that microbial activities cause approximately 30% weight loss from the original leaf mass in the stream (Petersen & Cummins, 1974). However, there are few such data from riparian systems.
Malanson (1993) conclude that the rate of litter decomposition generally decreases from the riparian zone toward the uplands (see also Merritt & Lawson, 1980; Peterson & Rolfe, 1982; Shure el al., 1986). Since water availability is higher in the riparian than in the upland ecosystems, one could expect that it is a major governing factor for the decomposition process. More specifically, the alternating wet and dry periods could be important. For example, Polunin (1984) suggested that the higher the frequency of change between wetting and drying, the greater the potential release of nutrients from the litter. However, so far no one has found any significant relationship between the extent of decomposition and the frequency of floods.
The differences in types and qualities of litter may result in different rates of leaching and breakdown (Day, 1982; Berg el al., 1993). In general, leaves of gymnosperm species leach less and thus decompose slower than do angiosperm species (Blackburn & Petr, 1979; Monk & Gabrielson, 1985). Young leaves usually contain less lignin, fiber, and secondary chemicals, and consequently decay faster than old leaves (Facelli & Pickett, 1991a; Polunin, 1984). Day (1983) reported that senescent leaves leached nearly 10% more than green leaves of the same species over 93 hours. Studies also show that litter decomposition rates are positively correlated with nutrient content and negatively correlated with substrate quality of litter (Meentemeyer, 1978; Vogt et al., 1986). Even within the same species, high-quality litter which contains more nutrients may attract more decomposers and thus decay more rapidly (Berg el al., 1993). However, if alternative food supplies for decomposers are poor, low-quality litter may also be consumed and/or decomposed rapidly (Vogt el al., 1986).
Temperature has been considered one of the most important environmental determinants of litter decomposition; high temperatures can promote the litter decomposition, and vice versa (Vogt el al., 1986; Facelli & Pickett, 1991a). Meentemeyer (1978) and Berg et al. (1993) found that the actual evapotranspiration (AET) was positively correlated with litter decomposition on both a global and a continental scale. Riparian areas consist of relatively stable and warm microenvironments (Malanson, 1993), especially in late fall when airborne litter deposits begin to decompose. Although there is no study available on AET examination in riparian areas, the relatively well-saturated soil in the riparian zone reasonably implies a higher AET level (Dolph et al., 1992) than is present in the adjacent uplands.
Soil fertility of site nutrient conditions are positively related to litter decomposition (Davis & van der Valk, 1978; Polunin, 1982; Staaf, 1987). The mechanism is little known, but Polunin (1984) proposed that exogenous nutrients such as nitrate and phosphate may increase the microbial respiration and thus the actual decomposition. Therefore, a rapid litter decomposition may be enhanced by the fertile riparian soils (Brinson, 1990).
Low oxygen availability and low pH may slow down litter decomposition in riparian areas (Reice, 1974; Polunin, 1984). However, the reduction of the decomposition rate resulting from low oxygen availability and low pH is generally small (Polunin, 1984) in comparison to the high rate that is maintained by leaching (Day, 1983), invertebrate breakdown (Merritt & Lawson, 1980), and microbial respiration (Petersen & Cummins, 1974).
Silt accumulation may also influence the litter decomposition process. Chauvet (1988) reported that litter decomposition is slower on floodplains than it is in aquatic and dry upland sites when the litter is covered by sediments, However, Mayack et al. (1989) found that buried litter decomposed slowly in the winter, but faster in spring because of increasing invertebrate activity, and therefore concluded that the overall rate of decomposition was similar between uncovered and covered litter.
IV. The Effects of Leaf Litter on Riparian Vegetation
The dynamics of litter accumulation may strongly influence the patterns and dynamics of riparian vegetation. We will discuss this topic by examining the basic litter processes that have been studied in terrestrial systems. We concentrate on leaf litter effects, simply because leaves are the dominant litter component.
A. THE LITTER IMPACTS
Litter accumulation may affect the structure and dynamics of plant communities physically, chemically, and biologically (Nilsson et al., 1993). The physical and chemical impacts are reviewed by Facelli and Pickett (1991a), but the biological impact seems to be poorly known. None of these impacts has been studied in any detail in riparian vegetation.
1. Physical Effects
Physical impact of litter has a significant role in riparian areas. However, in boreal regions, litter rain occurs mainly in late fall and early winter, whereas riverborne litter accumulates during the spring flood at the onset of the growing season. Therefore, in riparian areas, physical impact of litter on vegetation is caused mainly by litter that is transported from upstream sites of at least redistributed locally by the river water.
Litter deposited on riverbanks may cover and suppress plants in situ by acting as a barrier that isolates seeds, seedlings, and shoots from the available resources (Facelli & Pickett, 1991a). Therefore, seeds retained in the litter layer where soil is not available may undergo either delayed germination (e.g., Fowler, 1986) of unsuccessful germination (Hamrick & Lee, 1987). The emergence of seedlings or sprouts can also be inhibited in some cases (e.g., Fowler, 1986; Tao et al., 1987), depending on the species and the structure of the litter mat (e.g., Knapp & Seastedt, 1986). Only species with big seeds (Tao et al., 1987; Molofsky & Augspurger, 1992), seedlings with a long hypocotyl, or individuals with vigorous vegetative growth (Facelli & Pickett, 1991a) may have the ability to penetrate the litter mat.
Benefits of burial by litter have also been documented. For instance, Shaw (1968) and Sydes and Grime (1981b) observed that a dense litter may protect seeds from predators. Heady (1956) reported that a litter mat prevented top soil, and probably seeds and young plants us well, from erosion. It has been suggested that the litter mat may protect some low-growing plants from the erosion of floods in riparian areas, although a surplus of litter may reverse this protective function (Nilsson & Grelsson, 1990).
Many studies have shown that litter reduces both radiation amount and light quality on the soil surface (Knapp & Seastedt, 1986; Facelli & Pickett, 1991b). Such a reduction of light may prevent seed germination (Grime, 1979; Sydes & Grime, 1981b) and inhibit the growth of seedlings and sprouts (Barrett, 1931; Hamrick & Lee, 1987; Leishman & Westoby, 1994). However, some shade-demanding species may instead be favored (Grime, 1979). The impact of litter on light conditions also differs with litter amount us well us litter types and quality, because of differences in light interception (Facelli & Pickett, 1991b).
In riparian areas, the effect of litter on soil temperature is probably less significant than it is in uplands, because of a high underground water table and high air moisture. In a two-year field experiment along a free-flowing river in northern Sweden, litter additions of [less than] 1200 g [m.sup.2] did not significantly change the mean daily temperature of the topsoil (Xiong & Nilsson, unpubl. data). In contrast, significant temperature reductions were observed by Sydes and Grime (1981b) under a litter layer of 516 g [m.sup.2] in a terrestrial forest, and by Facelli and Pickett (1991b) under a litter layer of 400 g [m.sup.2] in an oldfield experiment. Beatty and Sholes (1988) also reported a significantly higher temperature in litter removal pits in a deciduous forest in eastern United States. Soil temperature may be modified by litter because litter intercepts radiation and insulates the soil (Sydes & Grime, 1981b; Beatty & Sholes, 1988; Facelli & Pickett, 1991b). Cool soils may inhibit the emergence of seedlings, delay growth of grass shoots (Knapp & Seastedt, 1986), and reduce flowering (Facelli & Pickett, 1991a) in the spring. Changes in soil temperature produced by litter may also decrease the activity of plant roots and underground animals us well us microbes. This, in turn, has negative effects on the plants, because nutrient uptake is reduced (Knapp & Seastedt, 1986). On the other hand, litter may protect plants against lethal frosts (Watt, 1974) and may delay senescence of leaves in late fall (Facelli & Pickett, 1991a). The effect of litter on soil temperature changes not only with litter amounts but also with litter types (Sydes & Grime, 1981b).
There is little direct evidence about the litter effect on soil moisture in riparian areas, although we observed no soil humidity changes along a litter gradient from none (removal) to 6000 g [m.sup.2] along the Vindel River, northern Sweden (Xiong & Nilsson, unpubl. data). Probably, a high water table in the riparian zone may weaken the differences in soil moisture between litter-covered and uncovered areas. However, in mesic grasslands and crop fields, the cover of litter may directly change soil water availability by intercepting rainfall, decreasing runoff, and reducing evaporation from soil (Facelli & Pickett, 1991a, and references therein). In drier and less productive grasslands, a certain amount of litter may help to conserve soil moisture and thus increase growth and plant diversity (Heady, 1956; West, 1979; Fowler, 1986).
2. Chemical Effects
Nutrient dynamics induced by litter have been extensively studied in grasslands and oldfields (Pastor et al., 1987, and references therein), forests (Staaf, 1987), and riverine corridors (Conners & Naiman, 1984; Brinson et al., 1980). On the basis of experiments in oldfields, Tilman (1987) viewed nutrient dynamics as the major determinant of succession; however, the role of these dynamics in determining the vegetation structure along riverbanks has not been evaluated.
Leaf litter may greatly affect the mineral composition of soil by releasing nutrients during its decomposition (Facelli & Pickett, 1991a, and references therein). In terrestrial habitats, litter may indirectly reduce nutrient availability by lowering soil temperature, thus slowing down litter decomposition rate, and by intercepting nutrients from rainwater (Knapp & Seastedt, 1986). Another indirect effect of litter is to alter soil acidity (Szczeponska, 1977), which in turn affects nutrient availability. However, the differences in types, quality, and decomposition patterns of litter may greatly influence nutrient dynamics (Facelli & Pickett, 1991a). This is especially relevant in riparian areas. Generally, litter fallen directly on the ground is more nutrient rich than is riverborne litter (Xiong & Nilsson, unpubl. data). Therefore, the chemical effects produced by riverborne litter are likely to be small, at least in the short term.
Litter usually accumulates in the fall under the canopies where it has been produced. It begins to decompose after contacts with soil or water (Chauvet, 1988; Merritt & Lawson, 1980). Therefore, the topsoil is likely to be enriched during the fall and winter. Peterson and Rolfe (1982) reported that substantial amounts of nutrients in floodplains were transferred to the mineral soil prior to the removal of remaining litter by spring floods. However, we know little about how such enrichments affect the vegetation in the following growing season.
Phytotoxin effects produced by decomposition of litter have been well documented (see reviews by Rice, 1979; Facelli & Pickett, 1991a). Laboratory and greenhouse studies have shown that litter leachates reduce germination and growth of seedlings (Rice, 1979; Ward & McCormick, 1982; van der Valk, 1986). In contrast, phytotoxins seem to be of little importance in the field (Facelli & Pickett, 1991a). Van der Valk (1986) proposed that phytotoxins are unimportant in wetlands because water-level fluctuations can remove or dilute eventual effects. However, I. Ridge (pers. comm.) found that unidentified phytotoxin substances of oak-leaf litter remained inhibitory for phytoplankton for more than 18 months in aerated water, suggesting an unknown complexity of phytotoxin effects.
A few studies document the enrichment of litter by nutrients during its decomposition (Polunin, 1984; Jordan et al., 1989). Jordan et al. (1989), having found this to be true for experimentally manipulated litter in a brackish tidal marsh, attributed it to the physical structure of the litter layer. It may be that riverborne litter enhances nutrient uptake in the riparian zone by intercepting and trapping suspended substances from the water. It may also be that microorganisms immobilize nutrients during litter decomposition (Polunin, 1984; de Jong & Klinkhamer, 1985). For example, de Jong and Klinkhamer (1985) reported that the growth of Cirsium vulgare (Savi) Ten. was reduced by the nutrient deficiency in the soil resulting from immobilization of nutrients by microorganisms that decompose the litter. Day (1982, and references therein) proposed that this nutrient immobilization is more likely to occur if the initial nutrient content in the litter is low. As Brinson (1990) pointed out, the immobilization of nutrients is a pathway for nutrient conservation in wetland forests.
A time lag may exist between the nutrient release from litter and the plant response. For example, Tilman and Cowan (1989) found that the differences between species in their response to nitrogen additions were greater after two seasons than they were after one season. The reasons for such a lag ate to be found in a combination of soil fertility prior to litter addition and the time needed for plants to adapt to changed nutrient conditions.
3. Biological Effects
Riparian systems are important natural corridors for the flows of species (Nilsson el al., 1991a, 1991b; Malanson, 1993; Naiman et al., 1993). Riverborne litter may have an important role in these flows across different landscape units. The major biological effect of litter is to bring species to new places during flooding. Nilsson and Grelsson (1990) documented a maximum of 189,000 seeds per [m.sup.2] in stranded litter along the Vindel River, northern Sweden. This addition implies a high potential for plant colonization. Simultaneously, litter packs also work as floaters for diaspores during dispersal. Johansson and Nilsson (1993) reported that both the experimentally and naturally released vegetative rhizomes of Ranunculus lingua L. were mostly found in drift accumulations stranded around the high-water level. Nilsson et al. (1993) found that released mimics of diaspores were often trapped in litter packs and that the distance of dispersal could be 147.5 km during a single flood event. These findings imply that litter may increase the dispersal capacity of species and may be an important factor in species turnover in the riparian vegetation.
Several studies have examined the indirect biological effects of litter through the attraction of animals and microbes that may change plant communities (Facelli & Pickett, 1991a, and references therein). Obviously, litter provides food and shade for decomposers (Anderson & Sedell, 1979; Merritt & Lawson, 1980; Petersen et al., 1989). In addition to determining the amount of litter, animals and microbes attracted by litter may also affect living plants and species interactions (Facelli, 1994).
The creation of bare ground for plant colonization is another indirect biological effect. Litter may create bare ground by burying and suppressing established vegetation and by leaving bare soil after decomposition. Direct erosion of litter during litter movement may also produce parches of bare soil.
The biological implications of litter are important to consider for the restoration and management of riparian zones. Litter redistribution may play an important role in maintaining species diversity on regional as well as landscape scales. However, at this point little is known about litter biology - for example, it is not yet clear what quantities of litter are needed for a maximal establishment success of waterborne diaspores.
B. THE EFFECTS OF LITTER ON THE PLANT COMMUNITY
All abiotic and biotic effects induced by litter may ultimately affect community structure and dynamics. Table I summarizes studies of litter effects on germination and establishment, community production, and species richness. Most of them are made in terrestrial areas.
1. Germination and Establishment
Germination and establishment seem to be particularly sensitive to the presence of litter, and there exist both positive and negative responses to litter accumulation. Certain amounts of litter may enhance the germination and establishment of some species by improving water conditions in dry habitats (West, 1979; Fowler, 1986), by protecting seeds from predation (Myster & Pickett, 1993), or by reducing competition (Facelli & Pickett, 1991c; Facelli, 1994). Litter may also inhibit establishment of many species by shading (Goldberg & Werner, 1983) or obstructing plants (van der Valk, 1986), reducing temperature (Haslam, 1971a, 1971b), attracting predators (Facelli, 1994), and probably fostering pathogens (Fowler, [TABULAR DATA FOR TABLE I OMITTED] 1988). The differences between species in how their seedlings withstand litter accumulation may contribute to the differences in distribution limits (Wilson & Zammit, 1992) and regeneration patterns of forests (Collins & Good, 1987), and to the species replacements which may result in a community succession in oldfields (Facelli & Pickett, 1991b; Facelli & Facelli, 1993).
2. Community Productivity
In six available examples (Table I), aboveground biomass production was enhanced by litter in one case, unchanged in one case, and inhibited in four cases. Different reactions may be attributable to different substrate conditions, species compositions, and litter amounts, as well as litter types and qualities. Substrate may be one of the most important factors, because substrate variation may affect the rates of both litter production and litter disappearance (Staaf, 1987; Cuffney, 1988), and thus community responses. Different successional stages may also respond differently. Under nearly the same environmental conditions in Hutcheson Memorial Forest Centre, New Jersey, a 1-year-old oldfield was much more sensitive to litter than a 14-year-old oldfield (Carson & Peterson, 1990). The fact that different authors have used different amounts of litter could also have influenced results. However, a well-designed long-term experiment, including a long gradient of litter mass, still remains to be done. For example, Heady's (1956) experiment might have given different results had he used a longer gradient of litter including also large deposits, or had he examined a longer period instead of only the early successional stage.
The inhibition of community production by riverborne litter has been experimentally studied along a boreal riverbank (Nilsson et al., unpubl. data). It was found that the aboveground biomass did not change significantly during one growing season until the litter amount exceeded 2400 g[m.sup.2].
3. Species Richness
In most of the reviewed examples, litter has had significant effects on species richness (Table I). Both positive and negative effects on species richness have been found and these probably depend on the relative amounts of litter deposited. Sydes and Grime (1981b) proposed that a moderate input of litter under a discontinuous tree canopy may reduce dominance and increase diversity within the ground herb layer. In contrast, a thick mat of tree litter may reduce the diversity of ground vegetation. In a riparian area in northern Sweden, Nilsson and Grelsson (1990) reported a quadratic relationship between species richness and leaf litter mass, with the highest numbers of species at low to intermediate accumulations of litter, However, little is known about how such patterns change along gradients of climate and substrate.
4. Community Dynamics
Grime (1979) postulated that litter accumulation may produce rapid changes in the microenvironment and strongly affect successional communities. It has been reported that litter-induced changes in soil nutrients (Kellman, 1979), light (Facelli & Pickett, 1991b), and soil temperature and moisture (Beatty & Sholes, 1988) initiate community successions in different systems. A community succession may be associated with a change of interspecific relations induced by litter. Sydes and Grime (1981b) reported that litter accumulation changed species interactions of ground vegetation in a deciduous forest. In an oldfield community, Monk and Gabrielson (1985) reported that litter changed the relative importance of different species categories, although litter was less important than root competition and shading. Facelli and Pickett (1991c; see also Facelli & Facelli, 1993; Facelli, 1994) found that the succession of oldfield communities was indirectly controlled by changing interspecific interactions because of litter accumulation. On the basis of experimental measures in heathlands, Berendse (1994) developed a model suggesting that the species replacements that might initiate a succession were triggered by the decomposability of litter.
Litter accumulation may also affect seasonal changes of a community. For example, Carson and Peterson (1990) reported that litter addition strongly suppressed both plant density and species richness in the early season, but these differences disappeared near the end of the growing season, irrespective of whether litter was added in the previous fall or spring. Carson and Peterson (1990) also found that removal of accumulated litter on the ground in the previous fall increased plant density and species richness more than removal in the spring. Our observations (Xiong & Nilsson, unpubl. data) in the riparian zone also suggest a potential effect of previous litter on species richness. Probably, a thick litter layer during fall and winter may prevent many seeds from reaching the soil and germinating in early spring (Carson & Peterson, 1990). These facts indicate that the redistribution of litter in different years may have different outcomes for a specific reach of riparian vegetation, because the redistribution time varies between early spring and early summer, depending on the spring flood.
In addition, the accumulation of litter may also affect the dynamics of plant communities through an interaction between litter and the dynamics of animals and microbes (Facelli & Pickett, 1991a). A thick litter layer can provide shelter for invertebrates, whereas the removal of litter may favor herbivores instead because more biomass will be produced. In the riparian areas, drifting litter may also relocate invertebrates (Xiong & Nilsson, pers. obs.). Hence, there are dynamic changes in the animal community composition associated with litter deposition. On the other hand, however, such dynamics of animal communities may subsequently change plant communities, because animals consume the litter, and change its structure by trampling (Facelli & Canon, 1991).
V. Concluding Remarks
This review was stimulated by two separate kinds of field observations over the years. First, we have seen many free-flowing rivers with a high production of plant biomass in the riparian zone, a considerable downstream transport of plant litter and subsequent food supply to aquatic animals, and also a high accumulation of litter on some riverbanks. Second, we have seen numerous hydrologically disrupted rivers with a low-productive riparian vegetation and little riverborne litter. Apparently, humans have contributed to a deterioration of the role of organic-matter cycling in the riparian corridor and impaired the terrestrial-aquatic linkages. Considering the great number of rivers that have been fragmented and regulated by dams (Dynesius & Nilsson, 1994), the effects on litter dynamics, and consequent ecological changes, not least on the regional scale, might be immense. These observations highlighted the importance of leaf litter in naturally functioning riparian ecosystems.
This review clearly demonstrates that riparian corridors differ from upland systems by having a high production, un effective transport, and a fast decomposition of litter. This places riparian systems among the key ecosystems in landscape functioning (cf. Malanson, 1993). It also shows that litter impacts on riparian vegetation van/with the location and type of biota, and that the amounts, types, and qualities of litter, us well us the timing of the impacts, also vary. However, the relationships between variables are insufficiently known. For example, we do not know what effects litter may have on plant germination, establishment, species richness, and biomass production, nor what directions effects might have in riparian areas. Furthermore, we do not know how the magnitude of litter impact varies with latitude and ecosystem structure, as well as with amounts and types of litter. The relative importance among physical, chemical, and biological effects of litter on the vegetation also needs to be evaluated. Therefore, many questions remain to be answered before we can accurately predict the changes in leaf litter dynamics in the riparian corridor, and the consequent changes in riparian and aquatic ecosystems that will result from changes in climate and land use.
We thank H. Dcamps, R. Jansson, and M. E. Johansson for valuable comments on the manuscript, and G. Pinay for the French translation of the abstract. Funding was provided by the European Commission Environment Programme under Research Grant EV5V-CT92-0100 (ERMAS) (to CN), and the Swedish Institute (to SX). The authors are grateful to Dr H. Barth for his interest in the project.
VII. Literature Cited
Abbott, D. T. & D. A. Crossley. 1982. Woody litter decomposition following clear-cutting. Ecology 63: 35-42.
Aber, J. D. & J. M. Melillo. 1980. Litter decomposition: measuring relative contributions of organic matter and nitrogen to forest soils. Canad. J. Bot. 58: 416-421.
Adis, J., K. Furch & U. Irmler. 1979. Litter production of a CentraI-Amazonian black water inundation forest. Trop. Ecol. 20: 236-245.
Alvarez-Lopez, M. 1990. Ecology of Pterocarpus officinalis forested wetlands in Puerto Rico. Pages 251-265 in A. E. Lugo, M. M. Brinson & S. Brown (eds.), Forested wetlands. Ecosystems of the World 15. Elsevier, Amsterdam.
Anderson, N. H. & J. R. Sedell, 1979. Detritus processing by macroinvertebrates in stream ecosystems. Ann. Rev. Entomol. 24: 351-377.
Barrett, L. I. 1931. Influence of forest litter on the germination and early survival of chestnut oak, Quercus montana. Ecology 12: 476-487.
Beatty, S. W. & O. D. V. Sholes. 1988, Leaf litter effects on plant species composition of deciduous forest treefall pits. Canad. J. Forest Res. 18: 553-559.
Bell, D. T., F. L. Johnson & A. R. Gilmore. 1978. Dynamics of litterfall, decomposition, and incorporation in the streamside forest ecosystem, Oikos 30: 76-82.
Berendse, F. 1994. Litter decomposability - a neglected component of plant fitness. J. Ecol. 82: 187-190.
Berg, B., M. P. Berg, P. Bottner, E. Box, A. Breymeyer, R. Calvo de Anta, M. Couteaux, A. Escudero, A. Gallardo, W. Kratz, M. Madeira, E. Malkonen, C. McClaugherty, V. Meentemeyer, F. Munoz, P. Piussi, J. Remacle & A. Virzo de Santo. 1993. Litter mass loss rates in pine forests of Europe and eastern United States: some relationships with climate and litter quality. Biogeochemistry 20: 127-159.
Bird, G. A. & N. K. Kaushik. 1981. Coarse particulate organic matter in streams. Pages 41-68 in M. A. Lock & D. D. Williams (eds.), Perspectives in running water ecology, Plenum, New York.
Blackburn, W. M. & T. Petr. 1979. Forest litter decomposition and benthos in a mountain stream in Victoria Australia. Arch. Hydrobiol. 86: 453-498.
Boling, R. H., E. D. Goodman, J. A. Van Sickle, J. O. Zimmer, K. W. Cummins, R. C. Petersen & S. R. Reice. 1975. Towards a model of detritus processing in a woodland stream. Ecology 56: 141-151.
Bray, J. R. & E. Gorham. 1964. Litter production in forests of the world. Adv. Ecol. Res. 2: 101-158.
Brinson, M. M. 1990. Riverine forests. Pages 87-141 in A. E. Lugo, M. M. Brinson & S. Brown (eds.), Forested wetlands. Ecosystems of the World 15. Elsevier, Amsterdam.
-----, H. D. Bradshaw, R. N. Holmes & J. B. Elkins. 1980. Litterfall, stemflow, and throughfall nutrient fluxes in an alluvial swamp forest. Ecology 61: 827-835.
Brown, S. 1981. A comparison of structure, primary productivity, and transpiration of cypress ecosystems in Florida. Ecol. Monogr. 51: 403-427.
----- & D. L. Peterson. 1983. Structural characteristics and biomass production of two Illinois bottomland forests. Amer. Midl. Naturalist 110: 107-117.
Carson, W. P. & C. J. Peterson. 1990. The role of litter in an oldfield community: impact of litter quantity in different seasons on plant species richness and abundance. Oecologia 85: 8-13.
Chauvet, E. 1988. Influence of the environment on willow leaf litter decomposition in the alluvial corridor of the Garonne River. Arch. Hydrobiol. 112: 371-386.
----- & A. M. Jean-Louis. 1988. Production de litiere de la ripisylve de la Garonne et apport au fleuve. Oecol. Gener. 9: 265-279.
----- & H. Decamps. 1989. Lateral interactions in a fluvial landscape: the river Garonne, France. J. N. Amer. Benthol. Soc. 8: 9-17.
Cheplick, G. P. & J. A. Quinn. 1987. The role of seed depth, litter, and fire in the seedling establishment of amphicarpic peanut-grass (Amphicarpum purshii). Oecologia 73: 459-463.
Collins, S. L. & R. E. Good. 1987. The seedling regeneration niche: habitat structure of tree seedlings in an oak-pine forest. Oikos 48: 89-98.
Conner, W. H. & J. W. Day. 1976. Productivity and composition of a baldcypress-water tupelo site and a bottomland hardwood site in a Louisiana swamp. Amer. J. Bot. 63: 1354-1364.
-----, J. G. Gosselink & R. T. Parrondo. 1981. Comparison of the vegetation of three Louisiana swamp sites with different flooding regimes. Amer. J. Bot. 68: 320-331.
Conners, M. E. & R. J. Naiman. 1984. Particulate allochthonous input: relationship with stream size in an undisturbed watershed. Canad. J. Fish. Aquatic Sci. 41: 1473-1484.
Cowan, C. A. & M. W. Oswood. 1983. Input and storage of benthic detritus in an Alaskan subarctic stream. Polar Biol. 2: 35-40.
Cromack, K. & C. D. Monk. 1975. Litter production, decomposition and nutrient cycling in a mixed hardwood watershed and a white pine watershed. Pages 609-624 in F. G. Howell, J. B. Gentry & M. H. Smith (eds.), Mineral cycling in south-eastern ecosystems. Energy Research and Development Administration Symposium Series, CONF-470513, Washington, DC.
Cuffney, T. F. 1988. Input, movement and exchange of organic matter within a subtropical coastal blackwater river-floodplain system. Freshwat. Biol. 19: 305-320.
Cummins, K. W., J. R. Sedell, F. J. Swanson, G. W. Minshall, S. G. Fisher, C. E. Cushing, R. C. Petersen & R. L. Vannote. 1983. Organic matter budgets for stream ecosystems: problems in their evaluation. Pages 299-353 in J. R. Barnes & G. W. Minshall (eds.), Stream ecology: application and testing of general ecological theory. Plenum Press, New York.
Dance, K. W. 1981. Seasonal aspects of transport of organic and inorganic matter in streams. Pages 69-95 in M. A. Lock & D. D. Williams (eds.), Perspectives in running water ecology. Plenum Press, New York.
Davis, C. B. & A. G. van der Valk. 1978. Litter decomposition in prairie glacial marshes. Pages 99-113 in R. E. Good, D. F. Whighham & R. L. Simpson (eds.), Freshwater wetlands. Academic Press, New York.
Dawson, F. H. 1976. Organic contribution of stream edge forest litter fall to the chalk stream ecosystem. Oikos 27: 13-18.
Day, F. P. 1982. Litter decomposition rates in the seasonally flooded Great Dismal Swamp. Ecology 63: 670-678.
-----. 1983. Effects of flooding on litter decomposition in microcosms. Oecologia 56: 180-184.
de Jong, T. J. & P. G. L. Klinkhamer. 1985. The negative effect of litter of parent plants of Cirsium vulgare to their offsprings: autotoxicity or immobilization? Oecologia 65: 153-160.
de la Cruz, A. A. & H. A. Post. 1977. Production and transport of organic matter in a woodland stream. Arch. Hydrobiol. 80: 227-238.
Dolph, J., D. Marks & G. A. King. 1992. Sensitivity of the regional water balance in the Columbia River basin to climate variability: application of a spatially distributed water balance model. Pages 233-265 in R. J. Naiman (ed.), Watershed management. Springer-Verlag, New York.
Dynesius, M. & C. Nilsson. 1994. Fragmentation and flow regulation of river systems in the northern third of the world. Science 266: 753-762.
Elder, J. F. & D. J. Cairns. 1982. Production and decomposition of forest litter fall on the Apalachicola river flood plain, Florida. U.S. Geol. Surv. Water-Supply Paper 2196-B. Government Printing Office, Washington, DC.
Facelli, J. M. 1994. Multiple indirect effects of plant litter affect the establishment of woody seedlings in oldfields. Ecology 75: 1727-1735.
----- & W. P. Carson. 1991. Heterogeneity of litter accumulation in oldfields. Bull. Torrey Bot. Club 118: 62-66.
----- & E. Facelli. 1993. Interactions after death: plant litter controls priority effects in a successional plant community. Oecologia 95: 277-282.
----- & S. T. A. Pickett. 1991a. Plant litter: its dynamics and effects on plant community structure. Bot. Rev. (Lancaster) 57: 1-32.
----- & -----. 1991b. Plant litter: light interception and effects on an oldfield plant community. Ecology 72: 1024-1031.
----- & -----. 1991c. Indirect effects of litter on woody seedlings subject to herb competition. Oikos 62: 129-138.
-----, C. M. Montero & R. J. C. Leon. 1988. Effect of different disturbance regimen on seminatural grasslands from the subhumid Pampa. Flora 180: 241-249.
Fisher, S. G. 1977. Organic matter processing by a stream-segment ecosystem: Fort River, Massachusetts, U.S.A. Intl. Rev. Ges. Hydrobiol. 62: 701-727.
----- & G. W. Likens. 1973. Energy flow in Bear Brook, New Hampshire: an integrative approach to stream ecosystem metabolism. Ecol. Monogr. 43: 421-439.
Fowler, N. L. 1986. Microsite requirements for germination and establishment of three grass species. Amer. Midl. Naturalist 115: 131-145.
-----. 1988. What is a safe site?: neighbor, litter, germination date, and patch effects. Ecology 69: 947-961.
Frangi, J. L. & A. E. Lugo. 1985. Ecosystem dynamics of a subtropical floodplain forest. Ecol. Monogr. 55: 351-369.
Franken, M., V. Irmler & H. Klinge. 1979. Litterfall in inundation, riverine, and terra firma forests of central Amazonia. Trop. Ecol. 20: 225-235.
Gessner, M. O. & E. Chauvet. 1994. Importance of stream microfungi in controlling breakdown rates of leaf litter. Ecology 75: 1807-1817.
-----, M. Thomas, A. Jean-Louis & E. Chauvet. 1993. Stable successional patterns of aquatic hyphomycetes on leaves decaying in a summer cool stream. Mycol. Res. 97: 163-172.
Goldberg, D. E. & P. A. Werner. 1983. The effects of size of opening in vegetation and litter cover on seedling establishment of goldenrods (Solidago spp.). Oecologia 60: 149-155.
Gomez, M. M. & F. P. Day. 1982. Litter nutrient content and production in the Great Dismal Swamp. Amer. J. Bot. 69: 1314-1321.
Gosz, J. F., G. E. Likens & F. H. Bormann. 1972. Nutrient content of litter fall on the Hubbard Brook Experimental Forest, New Hampshire. Ecology 53: 769-784.
Grime, J. P. 1979. Plant strategies and vegetation processes. John Wiley & Sons, New York.
Gurtz, M. E., G. R. Marzolf, K. T. Killingbeck, D. L. Smith & J. V. McArthur. 1988. Hydrologic and riparian influences on the import and storage of coarse particulate organic matter in a prairie stream. Canad. J. Fish. Aquatic Sci. 45: 655-665.
Hamrick, J. L. & J. M. Lee. 1987. Effects of soil surface topography and litter cover on germination, survival and growth of musk thistle (Carduus nutans). Amer. J. Bot. 74: 451-457.
Hardin, E. D. & W. A. Wistendahl. 1983. The effects of floodplain trees on herbaceous vegetation patterns, microtopography and litter. Bull. Torrey Bot. Club 110: 23-30.
Haslam, S. M. 1971a. Community regulation in Phragmites communis Trin. I. Monodominant stands. J. Ecol. 59: 65-73.
-----. 1971b. Community regulation in Phragmites communis Trin. II. Mixed stands. J. Ecol. 59: 75-88.
Heady, H. F. 1956. Changes in the central California annual plant community induced by the manipulation of natural mulch. Ecology 37: 798-811.
Holland, E. A. & D. C. Coleman. 1987. Litter placement effects on microbial and organic matter dynamics in an agroecosystem. Ecology 68: 425-433.
Hughes, M. 1971. Tree biocontent, net production and litter fall in a deciduous woodland. Oikos 22: 62-73.
Iversen, T. M., J. Thorup & J. Skriver. 1982. Inputs and transformation of allochthonous particulate organic matter in a headwater stream. Holarct. Ecol. 5: 10-19.
Johansson, M. E. & C. Nilsson. 1993. Hydrochory, population dynamics and distribution of the clonal aquatic plant Ranunculus lingua. J. Ecol. 81: 81-91.
Jordan, C. F. 1971. A world pattern in plant energetics. Amer. Sci. 59: 426-433.
Jordan, T. E., D. F. Whigham & D. L. Correll. 1989. The role of litter in nutrient cycling in a brackish tidal marsh. Ecology 70: 1906-1915.
Junk, W. J., P. R. Bayley & R. E. Spark. 1989. The flood pulse concept in river floodplain systems. Canad. Spec. Publ. Fish. Aquatic Sci. 106: 110-127.
Keller, E. A & F. J. Swanson. 1979. Effects of large organic material on channel form and fluvial processes. Earth Surf. Proc. 4: 361-380.
Kellman, M. 1979. Soil enrichment by neotropical savanna trees. J. Ecol. 67: 565-577.
Killingbeck, K. T. 1986. Litterfall dynamics and element use efficiency in a Kansas gallery forest. Amer. Midl. Naturalist 116: 180-189.
----- & M. K. Wall. 1978. Analysis of a North Dakota gallery forest: nutrient, trace element and productivity relations. Oikos 30: 29-60.
Knapp, A. K. & T. R. Seastedt. 1986. Detritus accumulation limits productivity of tall grass prairie. BioScience 36: 622-668.
Leishman, M. R. & M. Westoby. 1994. The role of large seed size in shaded conditions: experimental evidence. Funct. Ecol. 8: 205-214.
Malanson, G. P. 1993. Riparian landscapes. Cambridge University Press, Cambridge.
Mayack, D. T., J. H. Thorp & M. Cothran. 1989. Effects of burial and floodplain retention on stream processing of allochthonous litter.
Oikos 54: 378-388. Meentemeyer, V. 1978. Macroclimate and lignin control of litter decomposition rates. Ecology 59: 465-472.
-----, E. O. Box & R. Thompson. 1982. World patterns and amounts of terrestrial plant litter production. BioScience 32: 125-128.
Merritt, R. W. & D. L. Lawson. 1980. Leaf litter processing in floodplain and stream communities. U.S.D.A. Forest Serv. Gen. Techn. Rep. WO-12: 93-105.
Molofsky, J. & C. K. Augspurger. 1992. The effects of leaf litter on early seedling establishment in a tropical forest. Ecology 73: 68-77.
Monk, C. D. & F. C. Gabrielson. 1985. Effects of shade, litter and root competition on old-field vegetation in South Carolina. Bull. Torrey Bot. Club 112: 383-392.
Mulholland, P. J. 1981. Organic carbon flow in a swamp-stream ecosystem. Ecol. Monogr. 51: 307-322.
Muzika, R. M., J. B. Gladden & J. D. Haddock. 1987. Structural and functional aspects of succession in south-eastern floodplain forests following a major disturbance. Amer. Midl. Naturalist 117: 1-9.
Myster, R. W. & S. T. A. Pickett. 1993. Effects of litter, distance, density and vegetation patch type on postdispersal tree seed predation in old fields. Oikos 66: 381-388.
Naiman, R. J., H. Decamps & M. Pollock. 1993. The role of riparian corridors in maintaining regional biodiversity. Ecol. Appl. 3: 209-212.
Neiff, J. J. & A. Poi de Neiff. 1990. Litterfall, leaf decomposition and litter colonisation of Tessaria integrifolia (Compositae) in the Parana river floodplain. Hydrobiologia 203: 45-52.
Nilsson, C. & G. Grelsson. 1990. The effects of litter displacement on riverbank vegetation. Canad. J. Bot. 68: 735-741.
-----, G. Grelsson, M. Dynesius, M. E. Johansson & U. Sperens. 1991a. Small rivers behave like large rivers: effects of postglacial history on plant species richness along riverbanks. J. Biogeogr. 18: 533-541.
-----, M. Gardfjell & G. Grelsson. 1991b. Importance of hydrochory in structuring plant communities along rivers. Canad. J. Bot. 69: 2631-2633.
-----, E. Nilsson, M. E. Johansson, M. Dynesius, G. Grelsson, S. Xiong, R. Jansson & M. Danvind. 1993. Processes structuring riparian vegetation. Pages 419-431 in J. Menon (ed.), Current topics in botanical research. Council of Scientific Research Integration, Trivandrum.
Odum, E. P. 1960. Organic production and turnover in old field succession. Ecology 41: 34-49.
Otto, C. 1975. Energetic relationships of the larva population of Potamophylax cingulatus (Trichoptera) in a south Swedish stream. Oikos 26: 159-169.
Pastor, J., M. A. Stillwell & D. Tilman. 1987. Little bluestem litter dynamics in Minnesota oldfields. Oecologia 72: 327-330.
Petersen, R. C. & K. W. Cummins. 1974. Leaf processing in a woodland stream ecosystem. Freshwat. Biol. 4: 343-368.
-----, ----- & G. M. Ward. 1989. Microbia and animal processing of detritus in a woodland stream. Ecol. Monogr. 59: 21-39.
Peterson, D. L. & G. L. Rolfe. 1982. Nutrient dynamics and decomposition of litterfall in floodplain and upland forests of central Illinois. Forest Sci. 28: 667-681.
Polunin, N. V. C. 1982. Processes contributing to the decay of reed (Phragmites australis) litter in fresh water. Arch. Hydrobiol. 94: 182-209.
-----. 1984. The decomposition of emergent macrophytes in fresh water. Adv. Ecol. Res. 14: 115-166.
Reice, S. R. 1974. Environmental patchiness and the breakdown of leaf litter in a woodland stream. Ecology 55: 1271-1282.
Reiners, W. A. 1972. Structure and energetics of three Minnesota forests. Ecol. Monogr. 42: 71-94.
Rice, E. L. 1979. Allelopathy: an update. Bot. Rev. (Lancaster) 45: 15-109.
Schlesinger, W. H. 1978. Community structure, dynamics and nutrient cycling in the Okefenokee cypress swamp-forest. Ecol. Monogr. 48: 43-65.
Shaw, M. W. 1968. Factors affecting the regeneration of sessile oak (Quercus petraea) in North Wales. II. Acorn losses and germination under field condition. J. Ecol. 56: 647-660.
Shure, D. J. & M. R. Gottschalk. 1985. Litter-fall patterns within a floodplain forest. Amer. Midl. Naturalist 114: 98-111.
-----, ----- & K. A. Parsons. 1986. Litter decomposition processes in a floodplain forest. Amer. Midl. Naturalist 115: 314-327.
Staaf, H. 1987. Foliage litter turnover and earthworm populations in three beech forests of contrasting soil and vegetation types. Oecologia 72: 58-64.
Sydes, C. L. & J. P. Grime. 1981a. Effects of tree leaf litter on herbaceous vegetation in the deciduous woodlands. I. Field investigations. J. Ecol. 69: 237-248.
----- & ----- . 1981b. Effects of tree leaf litter on herbaceous vegetation in the deciduous woodlands. II. Ah experimental investigation. J. Ecol. 69: 249-262.
Szczeponska, W. 1977. The effect of remains of helophytes on the growth of Phragmites communis Trin. and Typha latifolia L. Ekol. Polska 25: 437-445.
Tao, D. L., Z. B. Xu & X. Li. 1987. Effect of litter layer on natural regeneration of companion tree species in the Korean pine forest. Environm. Exp. Bot. 27: 53-65.
Tilman, D. 1987. Secondary succession and the pattern of plant dominance along experimental nitrogen gradients. Ecol. Monogr. 57: 189-214.
----- & M. L. Cowan. 1989. Growth of old field herbs on a nitrogen gradient. Funct. Ecol. 3: 425-438.
van der Valk, A. G. 1986. The impact of litter and annual plants on recruitment from the seed bank of a lacustrine wetland. Aquatic Bot. 24: 13-26.
Vannote, R. L., G. W. Minshall, K. W. Cummins, J. R. Shedell & C. E. Cushing. 1980. The river continuum concept. Canad. J. Fish. Aquatic Sci. 37: 130-137.
Vasicek, F. 1985. Natural conditions of floodplain forests. Pages 13-29 in M. Penka et al. (eds.), Flood-plain forest ecosystem. I. Before water management measures. Elsevier, Amsterdam.
Vitousek, P. M. 1984. Litterfall, nutrient cycling, and nutrient limitation in tropical forests. Ecology 65: 285-298.
Vogt, K. A., C. C. Grier & D. J. Vogt. 1986. Production, turnover, and nutrient dynamics of above- and below-ground detritus of world forests. Adv. Ecol. Res. 15: 303-377.
Ward, J. V. 1989. Riverine-wetland interactions. Pp. 385-400 in R. R. Sharitz & J. W. Gibbons (eds.), Freshwater wetlands and wildlife. Office of Scientific and Technical Information, U.S. Dept. of Energy, Oak Ridge.
Ward, H. A. & L. H. McCormick. 1982. Eastern hemlock allelopathy. Forest Sci. 28: 681-686.
Watt, A. S. 1974. Senescence and rejuvenation in ungrazed chalk grassland (grassland B) in Breckland: the significance of litter and moles. J. Appl. Ecol. 11: 1157-1171.
Weaver, J. E. & N. W. Rowland. 1952. Effect of excessive natural mulch on the development, yield, and structure of native grassland. Bot. Gaz. 114: 1-19.
Webster, J. R. 1975. Analysis of potassium and calcium dynamics in stream ecosystems on three southern Appalachian watersheds of contrasting vegetation. Ph.D. dissertation, University of Georgia, Athens.
----- & B. C. Patten. 1979. Effects of watershed perturbation on stream potassium and calcium dynamics. Ecol. Monogr. 49: 51-72.
Werner, P. A. 1975. The effect of plant litter on germination in teasel, Dipsacus sylvestris Huds. Amer. Midl. Naturalist 94: 470-476.
West, N. E. 1979. Formation, distribution, and function of plant litter in desert ecosystems. Pp. 647-659 in J. A. Perry & D. W. Goodall (eds.), Arid land ecosystems: structure, function, and management. Cambridge University Press, Cambridge.
Whittaker, R. H. & G. M. Woodwell. 1969. Structure, production, and diversity of the oak-pine forest at Brookhaven, New York. J. Ecol. 57: 155-174.
Williams, R. J. & D. H. Ashton. 1987. Effect of disturbance and grazing by cattle on the dynamics of heathland and grassland communities on the Bogong High Plains, Victoria. Austral. J. Bot. 35: 413-431.
Wilson, S. D. & C. A. Zammit. 1992. Tree litter and the lower limits of subalpine herbs and grasses in the Brindabella Range, ACT. Austral. J. Ecol. 17: 321-327.
Winterbourn, M. J. 1976. Fluxes of litter falling into a small beech forest stream. New Zealand J. Marsh Freshwat. Res. 10: 399-416.
[TABULAR DATA FOR APPENDIX VIII OMITTED]