PITFALLS of TRADITIONAL TECHNIQUES WHEN STUDYING DECOMPOSITION of VASCULAR PLANT REMAINS in AQUATIC HABITATS E Barlocher Dept

Home , Plant litter

PITFALLS OF TRADITIONAL TECHNIQUES WHEN STUDYING DECOMPOSITION OF VASCULAR PLANT REMAINS IN AQUATIC HABITATS E Barlocher Dept. of Biology, Mount Allison University, Sackville, N.B., E4L 1G7. Canada

SUMMARY

The bulk of vascular plarit production in terrestrial or aquatic habitats enters the detrilal carbon pool. Micrcmrganisms initiate incor- poration of the detritus into food webs. Foliowing the icad by soil ecologists, liinnologicts have investigated decomposition by exposing dried plant parts in litter bage. Under these conditions. leaching (rapid, abiotic Ioss of coluble compounds such as phenolics, carbohydrates, amino acids) is common. Leaching is largeiy absent when fresh, rather than predried, aider aiid willow leaves are exposed in water. Reduced leaching WÍIC subsequently shown to deiay colonization by aquatic hyphoinycetes and invertebrate feeding. In a series of experiments with 27 leaf speciec (some species tected more than once), drying sigriificantly increased leaching in 18 effect in 3 caces, and decreased leaching in 8 cases. Many aquaiic inacrophytes that contribute to the organic budget of rivers and streams do not abscise their leaves or stems. This includes Bryophyta (e.g., Foiiririulis sp.), subriierged (e.g., Rnizunculus sp.) and emergent ular planta. Almost without exception, decomposition of such plant material has beeii studied by removing and drying plant parts followed by their cxposure in litter bags. Based oii comparable studies from marches, it is likely that this introduces several sources of error: the course and magnitude of leaching may change. and there may be shifts between inicrobial groupc (fungi vs. bacteria). To avoid some of these pitfallc. it ie essential to closely observe the natural introduction of detritus into streams and conditions during its decay, and attempt to reproduce these conditions during experiments.

INTRODUCTION lead to erroneous conclusions concerning the courae of decomposition arid the relative coritributions of fungi and bacteria. In Production iii many terrestrial and aquatic ecosycteins is thic review, 1 shall discuss two potential sources of error. The dominated by vascular piants (WESTLAKE, 1963; WETZEL. first concerns the use of pre-dried plant materials. and the 1990). Consumption «f living vascular plant material is often second thc confinement in litterbagc »f detritus that normaily minirnai, and the bulk of the primary production enters the remains attached to the plant beyond senescence and death. detrital, or non-iiving, carbon pool. It is generally assurned that Decay in the “standing-dead” phace predominatee in grasses microorganisms, primarily fungi and bacteria, initinte incorpo- and grass-like piants. ration of this detritus into food webs. Microbial decomposition is therefore a key procese. and has been studied extensively. AUTUMN-SHED LEAVES Most inveatigations have been based on a technique pioneered by terrestrial ecoiogists come 60 years ago (FALCONER et al.. Leaching in dried and fresh leaves 1933: LUNT, 1933): plant material is coilected, dried (conieti- mes at room temperature, more often in an oven). and preweig- Leaching refers to the rapid, abiotic removal of soluble com- hed portions are exposed in containerc such as boxes, open- pounds from plant litter. One of the first thorough investigations ended tubes, or, most commonly, in litterbags with variable of this process was conductcd by NYKVIST (1963). He worked mesh sizes. Periodically, some containers are recovered, the with several cpecies of air-dried leaves, and found that they lost remaining mass ot’ the detritus is determined and chemicai between 7. I (Quercus rohur) and 16.5 5% (Fruxinus oxr.ulsior) analysea are performed. Thic approach obviously deviates in of their total macs within one day of being submerged in disti- several ways from the natural decomposition process and rnay lled water. Lcaching losses were much lower from conifer

Liiritirtit.

n E\piiiiola dc Llninologia. Madrid, Spain. 2 needles (for example, 0.9 % in Pinus silvestris). NYKVIST nes do deteriorate during senescence, resulting in increased lea- (1963) also demonstrated that leaching is accelerated when the kage; THOMPSON et al., 1997). Drying, however, is known to leaves or needles are first ground into smaller particles. Amino disrupt interna1 membrane structures and to increase leakage of acids, sugars and fatty acids together accounted for 10-25 % of solutes through the plasmalemma (BEWLEY, 1979). Drought the total organic leachate. Among other potential components, intolerant plants are unable to reverse these changes, which which he did not analyze, NYKVIST (1 963) mentioned water- lcads to increasing fragmentation of organelles and membrane soluble phenolics. When leachates were stored under aerobic structures. conditions, “clots” of bacteria, fungal hyphae and protozoa The observation that leaching in fresh (non-dried) leaves is appeared within a few days, apparently sustained by the dissol- much reduced has potentially far-reaching consequences ved organic compounds. Under anaerobic conditions, only bac- (GESSNER, 1991). The cubstances retained by the leaf (amino teria could be observed. Obviously, leached substances make up acids, carbohydrates, phenolics, etc.) may influence its coloni- a substantial portion of total leaf mass, and their retention or zation by aquatic microbes and palatability to stream inverte- removal is likely to affect organisms that feed on the leaf, as brates. Conversely, a reduced supply of leaf leachate to the ctre- well as organisms that depend on nutrients dissolved in water. am water would presumably have negative effects on thc acti- NYKVIST was interested in how leaching would affect soil vity of planktonic bacteria (CUMMINS et ul., 1972) and of bio- formation, but autumn-shed leaves also make an important con- film communities (associations of bacteria, fungi and protozoa tribution to food webs in streams (BIRD & KAUSHIK, 1981; embedded in polysaccharide matrices, covering most BARLOCHER, 1992 a). In a pioneering study of leaf break- solid/water interfaces; LOCK, 1993). 1 am not aware of any down in streams, KAUSHIK & HYNES (1971) differentiated study comparing the effect of dried and non-dried leaves on between an early, rapid loss due to leaching, followed by a such “extemal” microbes, but there are some data on microbial more gradual loss due to microbial and invertebrate attack. This colonization and invertebrate consumption of fresh and dried distinction became generally accepted, and leaf decomposition leaves. in streams has been suhdivided into three more or less distinct In a field study started on 19 September, fresh A. glutinosu phases: leaching, microbial colonization and invertebrate fee- leaves lost less mass than dried leaves during the first four ding (CUMMINS, 1974). Leaching in streams is generally weeks of exposure in the River Teign in Devon, England. Over complete within 24 - 48 h and results in a loss of up to 30 % of the entire course of the study (1 1 weeks), the decay rates did the original mass (WEBSTER & BENFIELD, 1986). SUBER- not differ significantly (BARLOCHER, 199 1). Similarly, colo- KROPP et al. (1976) identified phenolics and soluble carbohy- nization by aquatic hyphomycetes proceeded more rapidly on drates as two classes of organic compounds that are particularly dried leaves than on fresh leaves. After two weeks in the stre- susceptible to leaching. am, recovered dried leaves released close to 5000 conidia per In order to achieve more uniform initial conditions, leaves mg leaf mass during two days of aeration; only 40 were recove- were almost always dried at room temperature or in an oven red from fresh leaves. There was some indication that oomyce- before their decomposition was studied. GESSNER & SCH- tes were more common on fresh than on dried leaves during the WOERBEL (1 989) demonstrated that this pretreatment greatly first two weeks. increased mass losses of alder (Alnus glutinosa (L.) Gaertner) When the experiment was repeated on 14 November, there and willow (Su1ixJrugili.s L.) leaves during the first few days. were no longer significant differences between fungal coloniza- Fresh, ¡.e. non-dried, leaves did not lose any appreciable m tion and mass losses during the first few weeks during the first few days of immersion. (BARLOCHER, 1991). One possihle explanation was based on GESSNER & SCHWOERBEL (1989) attributed their obser- the fact that between the two experiments, the air temperature vations to the fact that the shedding of leaves is preceded by had repeatedly dropped below freezing. Like desiccation, free- senescence, which is an orderly process, requiring maintenance zing can disrupt the integrity of leaves and damage their mem- of the cell’s compartments and functioning biochemical proces- branes, which may result in increased leakage (BURKE et ul., ses (MATILE, 1986). Death of the leaf generally occurs when 1976). A third experiment was therefore initiated in the follo- phenolic compounds are released from vacuoles and make con- wing spring: young, green leaves were collected, exposed in a tact with phenoloxidases, resulting in the browning of the leaf. stream without further treatment or first dried or frozen. Autumn-shed leaves of alder and many other trees are not Colonization by aquatic hyphomycetea was delayed on fresh necrotic upon abscission and largely maintain their integrity leaves, but not on frozen or dried leaves. The conclusion was (there is recent evidence, however, that to some extent membra- that drying or freezing accelerates leaching of suhstances that inhibit fungal colonization. This was tested by extracting mean that microbial colonization was not greatly delayed in variously treated leaves with distilled water, and exposing aqua- fresh leaves. Combined with the observation that cellulose tic hyphomycetes to these leachates (BARLOCHER, 1990). As decay was initially slower in fresh leaves, this suggests that expected, leachates from whole, fresh leaves did not inhibit early microbial colonizers of fresh leaves used the labile orga- fungal growth, while leachates from whole, dried leaves did. nic compounds that were leached from dried leaves. In addi- Leachates contain amino acids, sugars, phenolics and alipha- tion, phenolics and tannins in fresh leaves probably contributed tic acids. Some of these compounds are valuable nutrients, whi- to the formation of artifact lignins (complexes of phenolics, le others are known to deter potential consumers. It is therefore polysaccharides and proteins); due to rapid leaching, no such conceivable that fresh, unleached leaves are more attractive to effect was observed in dried leaves. The stability of such arti- invertebratec. There is little evidence that this is the case fact lignins depends, among other factors, on pH; they may the- (BARLOCHER, 1990; CHERGUI & PATTÉE, 1993; GESS- refore be a source of food for invertebrates with highly alkaline NER & DOBSON, 1993). Neither Gammurus pulex gut fluids such as Tipula larvae (BARLOCHER et al., 1989). (Amphipoda) nor Asellus aquaticus (Isopoda) discriminated It may be argued that the distinction between dried and fresh between leached dried and unleached fresh A. glutinosa leaves, leaves becomes irrelevant in hot climates. Thus, in Alabama, but both strongly preferred conditioned (colonized by aquatic leaves of Liriodendron tulipifera L. often become dry and brit- hyphomycetes) over unconditioned leaves (BARLOCHER, tle while still attached to the tree (K. SUBERKROPP, pers. 1990). In a related study, STOUT et al. (1985) found no eviden- comm.). In Base], Switzerland, newly shed leaves of the same ce for invertebrate feeding on fresh summer leaves (Alnus rug(i- species generally have retained much of their moisture and su) until 26 days after immersion in two hardwater streams in remain rubberlike (pers. obs.). Neveriheless, the only published Michigan. Nevertheless, a diverse macroinvertebrate commu- study from a subtropical region, Morocco, also showed clear nity invaded fresh leaf packs during this initial period. STOUT differences between fresh and dried willow leaves (Salix sp., et al. (1985) suggest that fresh leaves might support a rich later identified as. S. pedicellatu; CHERGUI & PATTÉE, 1992; epiphyllic community of fungi, bacteria, algae, protozoa and 1993). The rapid initial mas los, assumed to be due to lea- micrometazoa. This biofilm, in turn, could attract invertebrate ching, wac absent in fresh leaves, and during the first two “browsers”. ROUNICK & WINTERBOURN (1983) demons- months spore production and fungal species numbers were hig- trated that organic layers of slime, fine particles and microbes her on dried leaves. The gastropod shredder Melanopsis prue- are potential food sourcec for stream invertebrates, and that leaf morsa initially preferred dried leaves, and its mortality was hig- leachates enhance the thickness of this layer. Since fresh leaves her on fresh leaves. These observations again suggest that the retain soluble substances, a biofilm may form on the leaves soluble compounds that are leached from dried but not from themselves. STOUT et al. (1985) commented that fresh leaves fresh leaves have an overall inhibitory effect on aquatic hyp- were “more slimy” to the touch than were autumn leaves during homycetes and leaf-eating invertebrates. As in other studies, the first 26 days. This indicates a superficial microbial film, decay rates over the entire study period (1 7 wks) did not differ possibly sustained by slow leakage of organic molecules from significantly between fresh and dried leaves. Average monthly the leaf. STOUT et al. (1 985) used summer leaves, and they air temperature during leaf fa11 at the Moroccan sites was suggest that cellular activity, including photosynthesis, may approx. 20 ”C, with extremes between 12-28 “C (E. PATTÉE, have continued in the stream. GESSNER & DOBSON (1993) pers. comm.), which is considerably higher than temperatures found no significant differences in invertebrate colonization of in Central Europe. fresh and dried A. glutinosa litter. However, invertebrate num- When fungal colonization is delayed, changing externa1 con- bers peaked four weeks after immersion of the leaves. By then, ditions may affect fungal succession. In temperate deciduous chemical differences between fresh and dried leaves may have forests, the shedding of leaves coincides with falling temperatu- become negligible (GESSNER, 1991). res (in temperate evergreen forests, dominant in the southern GESSNER (1991) compared the decomposition of fresh and hemisphere, leaf fa11 occurs in summer, when water temperatu- dried A. glutinosa leaves during the first eight weeks in a stre- res are high; CAMPBELL & FUCHSHUBER, 1994). As a con- am. AS expected, he observed an early sharp decline in the m sequence, species preferring higher water temperatures would of dried but not of fresh leaves. In addition to soluble organic benefit if leaves were predried, and be less common if leaves compounds, phosphorus and potassium also leached rapidly were introduced in a natural, fresh state. This was shown to from dried leaves. Dynamics of nitrogen and protein were simi- occur in the River Teign: Lunulospora curvulu, a species gene- lar in the leaf types, which GESSNER (1991) interpreted to rally more common at warmer temperatures (WEBSTER et al., 1976), doininated on dried leaves introduced on I9 Septernher: seems to be supported by the observation that drying of Hetuli on frech leaves, where fungal colonization was delayed by pcipvr(fera aiid Acer saccharum leaves collected very early in several weeks, it never reached the same dominance the season had no effect on fungal colonization (SRIDHAR & (BARLOCHER, 199 1). During the experiment, the water tem- BARLOCHER, 1993). perature dropped from 1 1 .5 to 3.1 “C. In evergreen temperate forests, leaf fall is less seasonal and In a similar study in the French Pyrenees, fungal coloniza- dominated by abscission. There ceems to be little overland tion of fresh A. glutinosa leaves w gain delayed (GESSNER transport of fallen leaves from dry areas into streams (CAMP- rt al., 1993). In one year, correspondence analysis demonstra- BELL et al., 1992); leaves decaying in streams are therefore ted a slight difference in the fungal communitics of fresh and unlikely to have experienced a lengthy phase of aerial decom- dried leaves. In the second year, no difference was found. This pocition. may be due to the fact that the study was done in a “sumnier Do most leaves remain “fresh’ until abscission, ¡.e.,do they cool” stream, and Luriulnspora curvultr and other species typi- maintain thcir structural integrity? One would assume that in cal of warm streams were absent (GESSNER et d.,1993). conifers and deciduous species that retain senescent or dead In the Moroccan streams, a total of 16 species were found on leaves for extended periods of time (for exaniple, oak and pre-dried leaves, vs. 12 on fresh leaves (CHERGUI & PATTÉE, beech), additional drying before decomposition experirrients 1993). has little effect. Species of Euculyptus and Acacia are common Thus, several studies have shown that in fresh (non-dried) in arid cliniates arid presumably have adapted to retain moisture leaves of A. glutinosa and Sulix sp., leaching will be reduced, under these conditions. Whether these mechanisms reniain and fungal colonization and invertebrate consumption delayed. effective during senescence and abscission is unknown. One factor that might simulate drying is exposure to freezing In a Canadian study with three species (Betula pupj‘ri f’ era, temperatures. Two questions remain: how much of the annual Ultnus aniericann and Awr sa<.charum),hoth drying and free- leaf production enters streams in a fresh state. and do al1 lcaf zing increased leaching (BARLOCHER, 1992 b). However, species react in a similar manner to drying? leaching wiis not completely ahcent in fresh leaves of the same From the limited information availahle, it appears that in species, and exceeded 10 % in U. amrric.crnci. This very pro- temperate deciduous forests, a majority of leaves will enter the nounced leaching may have been due to the fact that a subctan- stream immediately after being separated from the tree (pnnia- tial proportion of the elm leaves were visibly damaged by insect rily through natural abscission during fall, but see below; FIS- feeding. This raises another important point: it is often standard HER, 1977; CONNERS & NAIMAN, 1984). Conditions are practice to use unblemished leaves in decomposition studies. As likely to be more variable in subtropical and tropical regions: pointed out by BOUI>TON& BOON ( IY9 I ), this may give mis- some species lose a cmall proportion of their leaves throuphout leading information in systems where heavy feeding by insects the year, others Iose them as the dry season approaches, a third or other herbivores is common (CHOUDHOURY, 19x8). group drops them after the oncet of the rainy season (COVICH. In the most extensive ctudy puhlished to date, leaching of 1988; SRIDHAR rt al., 1992). As ii consequence, some leaves fresh and pre-dried leaves from 27 species, collected across may accumulate on temporarily dry ground. In a Puerto Rico Canada. wac compared (TAYLOR & BARLOCHER. 1996). rainforest, terrestrial basidiomycetes reduced downhill move- Some leaves were collected from more than one location or in ments of leaves on steeply sloped stream hanks by 40 % (LOD- two successive years, giving a total of 35 measurementc. No GE & ASBURY. 1988). Aerial rhizomorphc oE Mt~ia.siniu.s, attempt was made to select healthy or undamaged leaves, but Psaihurella and others function much like spider webs and most leaves, with the exception of Alnus crispu (Ait.) Pursh and entrap leaves before they reach the ground (HEDGER. 1990; Sorhus arnrricuricr Marsh., were largely free o1 conspicuous COVICH. 1988). This may again introduce an aerial phaae of insect damage or necrosis (some leaves, however, were sticky. decomposition before the leaves reach a stream. indicating that some “leaching” rnay have been due to washing In temperate deciduous forests, a variable proportion of the off of externa1 compounds). In slightly more than half of thc leavec becomes detached while still green due to storms oI due case^ ( 18 out of 35), air-drying significantly increaced leaching to insect damage (BRAY & GORHAM, 1964). Often, yourig losses; in another 7 cases, it decreased leaching, and in 10 cases leaves are less well defended against herbivores than older lea- it had no significant effect. The most credible explanation for veh (CHOUDHOURY, 1988). It is therefore conceivable that decreased leaching is that complexation and precipitation of the effects OP increased leaching (which may remove inhibitory cell components occurred during drying. Neither moisture con- compounds) on fungal colonization are age-specific. This tent. nor leaf or cuticle thickness proved useful as predictors of 5

leaching losses or the effect of air-drying. It seems that both iinportant to realize, however, that this approach can be enor- magnitude and direction of changes in leaching due to drying mously time-consuming, especially if one wants to compare may be highly variable, not just between species, but also wit- several streams with species-rich riparian vegetation. Variability hin species collected in different years, or at different sites. will be high, which ¡neVitdbly lowers the power to detect pat- Factors that intluence leaching patterns may include the extent terns and identify iinportant factors. In many cases, a more suc- of insect damage, temperature and availability of nutrients and cessful strategy involves focusing on relatively narrow, well- water during growth and senescence. The nature of the leacha- defined questions, which can generally be investigated adequa- ble material may be as important as its quantity in affecting the tely under lesc realistic, but better controlled conditions. course of decomposition, colonization by aquatic hyphomycetes and invertebrates, and those microbes not directly associated EMERGENT FRESHWATER AND ESTUARI- with the leaves. NE MACROPHYTES

Exposure techniques Leaching of fresh and dried detritus

Once the leaves have been collected, they have to be expo- Freshwater and estuarine marshes have long been recognized sed in a stream. There are essentially three techniques: 1. as being among the most productive ecocystems (WESTLAKE, unconstrained leaves; 2. stacking leaves on top of each other 1963). Much of their productivity is due to emergent macrophy- and loosely sowing them together into packs; 3. placing leaves res of the littoral zone, dominated by genera such as Typhri, in bags with variable mesh sizes. A thorough discussion of the Phragmites, Scirpus and Spartina (WETZEL, 1990; NEWELL. relative merits of these approaches is beyond the scope of this 1993). In freshwater and estuarine streams, they are generally paper; a useful review can be found in BOULTON & BOON restricted to banks and shoals (HYNES, 1970; ALLAN, 1995). (199 1). Measuring decay in unconstrained leaves most closely but additional material may be swept in from upstream or by approximates the natural process, but ic difficult in practice tides. There are literally hundreds of studies investigating the (CUMMINS et al., 1980; BENFIELD ef al., 1991; D’ANGELO decomposition of emergent macrophytes (for reviews, see & WEBSTER, 1992; GRUBBS & CUMMINS, 1994). It gene- POLUNIN, 1984; NEWELL, 1993; GESSNER et al., 1997). rally results in higher estimated rates of mas loss than studies The vast majority is based on the use of pre-dried material (e.g., with leaf packs or bags, precumably because direct exposure to MASON & BRYANT, 1975; ANDERSON. 1978; GODS- the current maximizes mechanical fragmentation. Packs of lea- HALK & WETZEL, 1978 a, b, c; MORRIS & LAJTHA, 1986; ves held together with nylon filament or staples and tethered to CHERGUI & PATTÉE, 1990). Some studies reported faster objects within the stream allow free access to invertebrates and leaching of organic carbon due to drying, but as with deciduous are considered a reasonable compromise between realism and leaves, this effect is not universal (GODSHALK & WETZEL, reproducibility. However, decay rates depend on the size of leaf 1978 a; LARSEN, 1982; ROGERS & BREEN, 1982). packs (REICE, 1974; CAMPBELL et u/.. 1994); in streams, Generally, leaching became more pronounced when dried mate- leaf accumulations continuously form and reform, and it is the- rial was cut or ground into small particles (BRUQETAS DE refore difficult to define a representative pack size. In addition, ZOZAYA & NEIFF, 1 99 I ; OLÁH, 1 972). In two recent ctudies re often attached to bricks; if large numbers are introdu- with Spurtina altern~flora(SAMIAJI & BARLOCHER, 1996), ced in a stream reach, current patterns, and microbial and invei-- Tvphu lat@lia and Lythrum salicrrria (BARLOCHER & BID- tebrate colonizaiion inay be profoundly influenced. Finally, lit- DISCOMBE, 1996), total niass loss and loss ofsugars and phe- ter bags may drastically lower the impact of the current on the nolics were generally higher during the first I - 2 wks in pre- enclosed leaves, and depending on mesh sire, restrict the access dried leaves, indicating that leaching was less pronounced iii of invertebrates. Nutrient and gas exchange, and therefore frech leaves. Dissolved organic carbon is the dominant form of microbial metabolism, may also be inhibited. organic carbon in most aquatic ecosystems (WETZEL, 1983; When deciding on which exposure metliod to use, it is essen- THURMAN. I985), and accurate knowledge of how and when tial to clearly define the objectives of the study. If the goal ih it is introduced into the water is clearly important. broadly descriptive, ¡.e., to determine how leaf material is trans- ferred to various compartineiits such as dissolved organic rnat- Exposure techniques ter, fungal and bacteria1 biomass, etc.. an investigation of unronstrained leaves should give the most accurate results. It is The earliest studies of decomposition iri inarshes were based on cut and dried plant parts, which were permanently submer- European salt marsh macrophytes will be rewarding. ged in boxes or litter bags, or on ground up leaves incubated in NEWELL (1993) wrote: “Researchers interested in accura- water with marsh sediment (for reviews, see POLUNIN, 1984; tely describing natural microbial participation in the decay of NEWELL, 1993). Under these conditions, bacteria often predo- portions of vascular plants must try to avoid altering genuine minate and the fungal contribution to decay was assumed to be conditions of decay via their methods”. It seems obvious that negligible (TEAL, 1962; BENNER et al., 1986; MORAN et al., this reasoning also applies to freshwater marshes. This was sta- 1988; for notable exceptions, see MASON, 1976; MASON & ted explicitly by DAVIS & VAN DER VALK (1978): “Any BRYANT, 1975). Mycologists, on the other hand, were well study of emergent macrophyte decomposition ...must recognize aware of the diverse mycoflora that can be found on Spurtina, the fact that the proceses involved begin in an aerial environ- Typha, Phragmites and other marsh plants (INGOLD, 1955; ment and conclude in an aquatic environment.” These authors APINIS & CHESTERS, 1964; PUGH & MULDER, 1971; loss of macrophyte detritus in litterbags suspen- TALIGOOLA et al., 1972; APlNIS et al., 1972 a, b; GESSNER ded in the air and in the water, but did not measure bacteria1 & GOOS, 1973; KOHLMEYER & KOHLMEYER, 1979; and fungal biomass or activity. The first published attempt to ELLIS & ELLIS, 1985). Surprisingly, DESJARDIN et al. simulate the natural process by studying Cara leaves decaying (1995) even found a psychrophilic agaric on culms of Scirpus in situ again demonstrated that fungal standing crop and pro- californicus submerged under a thin layer of ice. The real ductivity greatly exceeded those of bacteria during the initial, breakthrough, however, came with the recognition that most aerial stage of decay (NEWELL et al., 1995). Naturally deca- emergent macrophytes do not abscise leaves or stems. ying T‘ypha leaves accumulated considerably more ergosterol NEWELL & FALLON (1989) and NEWELL et al. (1989) were than pre-dried leaves submerged in litterbags (BARLOCHER & the first to systematically apply this insight by comparing the BIDDISCOMBE, 1996). The same trend was found with leaves decomposition of dried Spartina alternzfloru leaves in litterbags of the purple loosestrife (Lythrum salicaria L.), which, howe- with that of standing leaves marked with electric cable ties. ver, started to shed leaves at an exponential rate after senescen- They concluded that ascomycetous fungi dominate the micro- ce. For this particular plant, therefore, placing leaves (prefe- bial biomass that accumulates on naturally decaying leaves. rably undried) in a litterbag is unlikely to misrepresent the natu- Fungi captured up to 90 % of the nitrogen present in decaying ral process to the same extent as it would in Typha or Spartina. S. altern@oru leaves within 8 - 10 wks. KUEHN (1997) made an important addition to our unders- In North America, Spartina salt marshes extend from Texas tanding of decay in freshwater macrophytes by measuring daily

(latitude 27 O) al1 the way to the Gulf of St. Lawrence (latitude variation of microbial respiration. In Alabama, day temperature

46 O; MANN, 1982). Conceivably, the natural action of ice and during summer can reach 36 “C. Under these conditions, CO, snow during late fall and winter at higher latitudes might simu- release is close to O. After nightfall, temperature drops to values late to some extent decay of detached leaves, and the study of in the low 2Os, and air humidity increases. This in turn allows detritus in litter bags might be a close approximation of the considerable microbial respiration and release of fungal spores. natural process. A recent study in a New Brunswick marsh (lati- Most previous studies were conducted during the day, and the-

tude 45 O) showed that this is not the case (SAMIAJI & refore missed this very significan1 contribution of primarily BARLOCHER, 1996). The decay of over SO % of leaves for- fungal respiration to the overall carbon budget. med during a growing season is initiated while they are still attached and upright. On such leaves, fungal biomass (as esti- OTHER PLANTS mated by ergosterol) is again considerably higher than that on dried leaves placed in litterbags. In addition to emergent plants, there is a variety of other On the European side of the North Atlantic and in other parts macrophytes that can make important contributions to the detri- of the world, the marsh flora is much more diverse and hetero- tus food webs in aquatic habitats. They are generally grouped geneous (MANN, 1982). There are several studies of the decay into three broad categories, namely floating-leaved taxa, free- of Spartina anglica, S. townsendii and S. maritima (for exam- floating plants, and submerged taxa (HYNES, 1970; HASLAM, ple, POZO & COLINO, 1992), but to my knowledge they are 1978; ALLAN, 199.5). NEWMAN (1991) compared evidence al1 based on dried material exposed in litterbags. BARATA et for herbivory vs. detritivory for some of these plants, and con- ul. (1 997) recently described a new fungal species from cluded that many seem to be protected by feeding deterrents. Spartina maritima in the Mira River estuary (Portugal), sugges- Such deterrents often persist beyond senescence and death, and ting that a closer look at the role and diversity of fungi on may have antimicrobial properties. Treatments that influence leaching dynamics, and therefore the removal of such com- Reasonably consistent results have been found with Alnus gluti- pounds, may profoundly alter the course of decomposition and nosa and Salix sp. decaying in streams: pre-dried leaves lose the relative contributions of bacteria and fungi. Since for the soluble organic compounds much more rapidly than fresh lea- most part these plants remain covered by water throughout ves, and are more readily coionized by aquatic hyphomycetes, growth and decomposition, pre-drying is very likely to introdu- which in turn makes them more quickly acceptable to leaf- ce artifacts, and ideally should be avoided. However, it is often eating invertebrates. In a comprehensive study with other leaf difficult with partiy or wholly submersed plants to distinguish species, the results were often contradictory: in roughly haif of between living, senescent and dead sections. Instead of deca- the reported cases, drying also increased leaching; in the remai- ying, collected and exposed material may actually grow. To pre- ning cases, there was either no change or the opposite effect vent this, severa1 authors used frozen plants (BARTODZIEJ & was observed. More progress probably depends on a closer look PERRY, 1990; NEWMAN, 1990), but cell death upon thawing at how different ciasses of compounds are affected by drying. is likely to have the same effect as drying. KORNIJOW et al. With emergent macrophytes in freshwater or estuarine mars- (1995) worked with material that had been incubated in the hes, it seems equally clear that early studies. based on dried dark for 7 days at 35 "C. material in litterbags, underestimated fungal participation. Most studies were again done with pre-dried leaves or leaf There is an entire fungal community specifically adapted to the sections in litter bags. For example, GAUR et al. (1992) com- cyclical changes of temperature, humidity and salinity. Even in pared fungal and bacterial contributions to water hyacinth hot climates, daily fluctuations are sufficient to allow temporary (Eichhornia crassipes) decomposition by using air-dried tissue. resumption of fungal activity. IKUSIMA & GENTIL (1996) dried water lily leaves to investi- For submersed plants, there is insufficient iiiformation of gate leaching. KOK et al. (1990) used frozen disks for decom- how decomposition might proceed under natural conditions. position experiments, while KOK & VAN DER VELDE (1994) Since they are generally delicate and often smail, it is difficult prepared disks from leaves stored at 4 "C. 1 am not aware of to recognize and tag senescent parts, and a simuiation of natural any study that tried to imitate the natural breakdown of water decay presents formidable challenges. lilies by tagging leaves and estimating mass loss, and fungal Ciose adherence to natural conditions of decay is most and bacterial biomass. Fungi do occur on water lily leaves, both important when the goal is some kind of description and quanti- as saprophytes (KOK et al., 1992) and as pathogens (JOHN- fication of microbial groups responsible for natural decay, and SON et al., 1997). and the large size and obvious signs of to follow the fate of the various detritus fractions. In these senescence in this plant should make a more realistic approach cases, there is simpiy no substitute for carefully observing the to studying decomposition feasible. natural process, and trying to imitate it as faithfully as possible. HANLON (1 982) noticed that two macrophytes, Isoetes In many other cases, the question of interest may be much more lacustris L. and Potamogeton pefcdiatus L. became extremely circumscribed, and deviations from the "normal' conditions fragile when oven dried. He therefore worked with air-dried may be acceptable, or even desirable. material, while CHERGUI & PATTÉE (1990) used Potamogeton leaves dried at 40 "C. ACKNOWLEDGEMENTS GODSHALK & WETZEL (1978 a, c) worked with a variety of floating-leaved (Nuphar vctriegatum) and submersed plants 1 acknowiedge with gratitude advice from E. Pattée and K. (Myriophyllum keterophyllum, Najas jlexilis, Zosteru marina). Suberkropp during the writing of this review. I'm thankful to They measured release of dissolved components from air-dried the organizers of this meeting for financia1 support. material. REFERENCES CONCLUSIONS ALLAN, J.D., 1995. Streum ecology. Chapman & Hall. The current review discusses two widespread practiccs that London. may give a misleading picture of how deconiposition procecds ANDERSON, F.O.. 1978. Effects of nutrient leve1 on the in iiature: the use of pre-dried material, and exposure of thic decomposition of Phrqrnites conzmunis Trin. Arch. Hydrobiol., material under conditions that deviate significantly from reality. 84: 42-54. The severity and direction of artifacts that may be introduced APINIS, A.E. & C.G.C. CHESTERS, 1964. Ascomycetes of vary hetween systems, and generalizations are difficult. some salt marshec and sand dunes. Trurz.s. BI: Myol. Soc.. 47: 419-435. environments: time to turn over an old leaf? Austr. J. Mar. APINIS, A.E., C.G.C. CHESTERS & H.K. TALIGOOLA, Freshw Res., 42: 1-43. 1972 a. Colonization of Phragmites cornmunis Trin. leaves by BRAY, J.R. & E. GORHAM, 1964. Litter production in fungi. Nova Hedwigia, 23: 113-124. forests of the worlds. Adv. Ecol. Res., 2: 101-157. APINIS, A.E., C.G.C. CHESTERS & H.K. TALIGOOLA, BRUQUETAS DE ZOZAYA, 1. Y. & J. J. NEIFF, 1991. 1972 b. Microfungi colonizing submerged standing culms of Decomposition and colonization by invertebrates of Typha lati- Phragmites communis Trin. Nova Hedwigia, 23: 473-480. ,folia L. litter in Chaco cattail swamp (Argentina). Aquat. Bof., BARATA, M., M.C. BASILIO & J.L. BAPTISTA-FERREI- 40: 185-193. RA, 1997. Nia globospora, a new marine gasteromycete on BURKE, M.J., L.V. GUSTA, H.A. QUAMME, C.M. WEIS- baits of Spartina maritima in Portugal. Mycol. Res., 101: 687- TER & P.H. LI, 1976. Freezing and injury in plants. Ann. Rev. 690. P1. Physiol., 27: 507-528. BARLOCHER, E, 1990. Factors that delay colonization of CAMPBELL, I.C. & L. FUCHSHUBER, 1994. Amount, fresh alder leaves by aquatic hyphomycetes. Arch. Hydrobiol., composition and seasonality of terrestrial litter accession to an I 19: 249-235. Australian cool temperate rainforest stream. Arch. Hydrobiol., BARLOCHER, F., 1991. Fungal colonization of fresh and 130: 499-5 12. dried alder leaves in the River Teign (Devon, England). Nova CAMPBELL, I.C., K.R. JAMES, B.T. HART & A. DEVE- Hedwigia, 52: 349-357. REAUX, 1992. Allochthonous coarse particulate organic mate- BARLOCHER, F., 1992 a. Research on aquatic hyphomyce- rial in forest and pasture reaches of two southeastern Australian tes: historical background and overview. In: The ecology of streams. 1. Litter accession. Freshwat. Biol.,27: 341-352. uquatic hyphomycetes. (E BARLOCHER, ed.): 1- 15. Springer- CAMPBELL, I.C., G.M. ENIERGA & L. FUCHSHUBER, Verlag, Berlin. 1994. The influence of pack size and position, leaf type, and BARLOCHER, F., 1992 b. Effects of drying and freezing shredder access on the processing rate of Afherosperma mos- autumn leaves on leaching and colonization by aquatic hyp- chatum leaves in an Australian cool temperate rainforest homycetes. Freshwat. Biol., 28: 1-7. stream. Int. Rev. ges. Hydrobiol., 19: 557-568. BARLOCHER, F. & N.R. BIDDISCOMBE, 1996. CHERGUI, H. & E. PATTÉE, 1990. The processing of lea- Geratology and decomposition of Tvpha latifolia and Lythrum ves of trees and aquatic macrophytes in the network of the salicaria in a freshwater marsh. Arch. Hydrobiol., 136: 309-325. River Rhone. Int. Rev. ges. Hydrobiol., 75: 281-302. BARLOCHER, F., P.G. TIBBO & S. CHRISTIE, 1989. CHERGUI, H. & E. PATTÉE, 1992. Processing of fresh and Formation of phenol-protein complexes and their use by two dry Salix leaves in a Moroccan river system. Acta Oecol., 13: stream invertebrates. Hydrobiologia, 173: 243-249. 29 1-298. BARTODZIEJ, W. & J.A. PERRY, 1990. Litter processing CHERGUI, H. & E. PATTÉE, 1993. Fungal and invertebrate in diffuse and conduit springs. Hydrobiologia, 206: 87-97. colonization of Salix fresh and dry leaves in a Moroccan river BENFIELD, E.F., J.R. WEBSTER, S.W. GOLLADAY, G.T. system. Arch. Hydrobiol., 127: 57-72. PETERS & B.M. STOUT, 1991. Effects of forest disturbance CHOUDHOURY, D., 1988. Herbivore induced changes in on leaf breakdown in southern Appalachian streams. Verh. Int. leaf-litter resource quality: a neglected aspect of herbivory in Ver: Limnol., 24: 1687-1690. ecosystem nutrient dynamics. Oikos, 5 1 : 389-393. BENNER, R., M.A. MORAN & R.E. HODSON, 1986. CONNERS, M.E. & R. NAIMAN, 1984. Particulate allocht- Biogeochemical cycling of lignocellulosic carbon in marine and honous inputs: relationships with stream-size in an undisturbed freshwater ecosystems: relative contributions of procaryotes watershed. Can. J. Fish. aquat. Sci., 4 I : 1473- 1484. and eucaryotes. Limnol. Oceunogr., 3 1: 89- 1OO. COVICH, A.P., 1988. Geographical and historical compari- BEWLEY, J.D., 1979. Physiological aspects of desiccation sons of neotropical streams: biotic diversity and detrital proces- tolerance. Ann. Rev. Pl. Physiol., 30: 195-238. sing in highly variable habitats. J. N. Am. Benthol. Soc., 7: 361- BIRD, C.A. & N.K. KAUSHIK, 1981. Coarse particulate 386. organic matter in streams. In: Perspectives in running water CUMMINS, K.W., 1974. Structure and function of stream ecology. (M.A LOCK & D.D. WILLIAMS, eds.): 41-68. ecosystems. BioScience, 24: 61-64. Plenum Press, New York. CUMMINS, K.W., J.J. KLUG, R.G WETZEL, R.C. PETER- BOULTON, A.J. & P.I. BOON, 1991. A review of the met- SEN, K.F. SUBERKROPP, B.A. MANNY, J.C. WUYCHECK hodology used to measure leaf litter decomposition in lotic & F.O. HOWARD, 1972. Organic enrichment with leaf leachate 9 in experimental lotic ecosystems. BioScience, 22: 7 19-722. GODSHALK, G.L. & R.G. WETZEL, 1978 b. CUMMINS, K.W., G.L. SPENGLER, G.M. WARD, R.M. Decomposition of aquatic angiosperms. 11. Particulate compo- SPEAKER, R.W. OVINK, D.C. MAHAN & R.L. MAT- nents. Aquaf. Bot.. 5: 301-327. TINGLY, 1980. Processing of confined and naturally entrained GODSHALK, G.L. & R.G. WETZEL, 1978 c. leaf litter in a woodland stream ecosystem. Limnol. Oceanogr., Decomposition of aquatic angiosperms. 111. Zostera marina L. 25: 952-957. and a conceptual model of decomposition. Aquat. Bot., 5: 329- D’ANGELO, D.J. & J.R. WEBSTER, 1992. Natural and 354. containment-induced factorc influencing the breakdown of dog- GRUBBS. S.A. & K.W. CUMMINS, 1994. A leaf-toughness wood and oak leaves. Hydrohiologia, 237: 39-46. method for directly measuring the processing of naturally DAVIS, C.B. & A.G. VAN DER VALK, 1978. The decom- entrained leaf detritus in streams. J. N. Am. Benthol. Soc., 13: position of standing and fallen litter of Typha glauca and 68-73. Scirpus,fluviatalis. Can. J. Bot., 56: 662-675. HANLON. D.G., 1982. The breakdown and decomposition DESJARDIN, D.E., L. MARTlNEZ-PECK & M. RAJ- of ailochthonous and aiitochthonous piant iitter in an oligotrop- CHENBERG, 1995. An unusual psychrophiiic aquatic agaric hic lake (Llyn Frongoch). Hydrohiologia, 88: 281 -288. from Argentina. Mycologia, 87: 547-550. HASLAM, S.M., 1978. River plarits. Cambridge University ELLIS, M.B. & J.P. ELLIS, 1985. Microfungi on Iand Presc, Cambridge. plants. Macmillan Publishing Company, New York. HEDGER, J., 1990. Fungi in the tropical forest canopy. FALCONER, G. J., J.W. WRIGHT & H. W. BEALL, 1933. Mycologi.rt, 4: 200-202. The decomposition of certain types of fresh litter under field HYNES, H.B.N., 1970. The ecology of running wuters. conditions. Am. J. Bot., 20: 196-203. University of Toronto Press, Toronto. FISHER, S.G., 1977. Organic rnatter processing by a stream IKUSIMA, 1. & J.G. GENTIL, 1996. Evaluation of the faster segment ecosystem: Fort River, Massachusetts, USA. hter: Rei: initial decomposition of tropical floating ieaves of Nymphaea ges. Hydrobiol.. 62: 701-727. eleguris Hook. Ecol. Res., 1 I : 201 -206. GAUR, S., P.K. SINGHAL & S.K. HASIJA, 1992. Relative INGOLD, C.T., 1955. Aquatic ascomycetes: further species contributionc of bacteria and fungi to water hyacinth decompo- from the Engiish Lake District. Trans. Br. Mycol. SOL.,38: 157- cition. Ayuat. Bot., 43: 1-15. 168. GESSNER, M.O., 1991. Differences in processing dynamics JOHNSON. D.A., L.M. CARRIS & J.D. ROGERS, 1997. of fresh and dried ieaf iitter in a stream ecosytem. Freshwat. Morphoiogical and molecular characterization of Biol., 26: 387-398. Colletotrichum nymphaeae and C. nuplzaricoln sp. nov. on GESSNER, M.O. & M. DOBSON, 1993. Colonisation of water-liiies (Nymphaea and Nuphar). Mycol. Res., 101: 64 1- fresh and dried leaf litter by lotic invertebrates. Arch. 649. Hydrobiol., 127: 14 1- 149. KAUSHIK, N.K. & H.B.N. HYNES, 1971. The fate of the GESSNER, M.O. & J. SCHWOERBEL, 1989. Leaching dead leaves that fa11 into streams. Arch. Hydrohiol., 68: 465- kinetics of fresh ieaf-iitter with implications for the current con- 515. cept of leaf-processing in streams. Arch. Hydrobiol., 1 15: 8 1-90. KOHLMEYER, J. & E. KOHLMEYER, 1979. Marine GESSNER, M.O., M. THOMAS, A.-M. JEAN-LOUIS & E. mjicology. Academic Press, New York. CHAUVET, 1993. Stable successionai patterns of aquatic hyp- KOK, C.J. & G. VAN DER VELDE, 1994. Decomposition homycetes on leaves decaying in a summer cool stream. Mycol. and macroinvertebrate coionization of aquatic and terrestriai Res., 97: 163.172. leaf material in alkaline and acid still water. Freslzwat. Biol., GESSNER, M.O., K. SUBERKROPP & E. CHAUVET, 3 1 : 65-75. 1997. Decomposition of piant litter in marine and freshwater KOK, C.J., W. HAVERKAMP & H.A. VAN DER AA, 1992. ecosystems. In: The Mycota (G.C. WICKLOW & B.E. Influence of pH on the growth and leaf-macerating ability of SODERSTROM, eds.): 303-322. Springer-Verlag, Beriin. fungi involved in the decomposition of floating leaves of GESSNER, R.V. & R.D. GOOS, 1973. Fungi from decoin- Nymphaeu ulhu in an acid water. J. Gen. Microbiol.. 138: 103- posing Spurtina ulterniflom. Can. J. Bor., 5 1 : 51 -55. 108. GODSHALK, G.L. & R.G. WETZEL. 1978 a. KOK, C.J., H.W.G. MEESTERS & A.J. KEMPERS, 1990. Decomposition of aquatic angiosperms. 1. Disolved compo- Decoinposition rate, chemicai composition and nutrient recy- nents. Aqituf. Bot., 5: 281-300. cling of Nynzphara ctlha L. floating leaf biade detritus as 10

influenced by pH, alkalinity and aluminium in laboratory expe- SON, 1995. Productivities of microbial consurners during early riments. Aquat. Bot., 37: 215-227. stages of decomposition of leaves of a freshwater sedge. KORNIJÓW, R., R.D. GULATI & T. OZIMEK, 1995. Food Freshwat. Biol., 34: 135-148. preference of freshwater invertebrates: comparing fresh and NEWMAN, R.M. 1990. Effects of shredding amphipod den- decomposed angiosperm and a filamentous alga. Freshwat. sity on watercress Nasturtium officinule breakdown. Holarct. Biol., 33: 205-212. Ecol., 13: 293-299. KUEHN, K.A., 1997. Standing dead decomposition of the NEWMAN, R.M., 1991. Herbivory and detritivory on frech- freshwater emergent macrophyte Juncus effusus L. Ph. D. water macrophytes by invertebrates: a review. J. N. Am. Thesis, University of Alabama, Tuscaloosa. Benfhol. Soc., 10: 89-114. LARSEN, V. J., 1982. The effects of pre-drying and frag- NYKVIST, N., 1963. Leaching and decomposition of water- mentation on the leaching of nutrient elements and organic mat- soluble organic substances from different types of leaf and ter from Phragmires australis (Cav.) Trin. litter. Aquut. Bof., 14: needle litter. Stud. Foresf. Suec., 3: 3-3 l. 29-39. OLÁH, J., 1972. Leaching, colonization and stabilization LOCK, M.A., 1993. Attached microbial communities in during detritus formation. Mem. Ist. Ifal. ldrobiol.. 29: 105-127. rivers. In: Aquatic microbiology. (T.A. FORD, ed.): 113-138. POLUNIN, N.V.C., 1984. The decomposition of emergent Blackwell, Oxford. macrophytes in fresh water. Adv. Ecol. Res., 14: 115-166. LODGE, D.J. & C.E. ASBURY, 1988. Basidiomycetes redu- POZO, J. & R. COLINO, 1992. Decomposition processes of ce export of organic matter from forest slopes. Mycologia, 80: Spartina maritima in a calt marsh of the Basque Country. 888-890. Hydrohiologia, 231: 165-175. LUNT, H.A., 1933. Effects of weathering upon composition PUGH, G.J.F. & J.L. MULDER, 1971. Mycoflora of hardwood leaves. J. Forest., 3 1 : 43-45. with Tvpha latijolia. Trans. BI: Mycol. Soc., 57: 273-282. MANN, K.H., 1982. Ecology qf coastal water. University of REICE, S.R. 1974. Environmental patchiness and the break- California Press, Berkeley & Los Angeles. down of leaf litter in a woodland stream. Ecology, 55: 1271- MASON, C.F., 1976. Relative importance of fungi and bac- 1282. teria in the decomposition of Phragmites leaves. ROGERS, K.H. & C.M. BREEN, 1982. Decomposition of Hydrobiologiu, 5 1: 65-69. Potamogeton crispus L.: the effects of drying on the pattern of MASON, C.F. & R.J. BRYANT, 1975. Production, nutrient mass and nutrient loss. Aquat. Bot., 12: 1 - 12. content and decomposition of Phragmites communis Trin. and ROUNICK, J.S. & M.J. WINTERBOURN, 1983. The for- Typha angustijolia L. J. Ecol., 63:71-95. mation, structure and utilization of stone surface layers in two MATILE, P. 1986. Blattseneszenz. Biol. Rundsch., 24: 349- New Zealand streams. Freshwat. Biol., 13, 57-72. 65. SAMIAJI, J. & F. BARLOCHER, 1996. Geratology and MORAN, M.A., T. LEGOVIC, R. BENNER & R.E. HOD- decomposition of Spartina ulterrzifloru Loisel in a New SON, 1988. Carbon flow from lignocellulose: a simulation Brunswick saltmarsh. J. Exp. Mal: Bid. Ecol., 201 : 233-252. analysis of a detritus-based ecosystem. Eccology, 69: 1525-1536. SRIDHAR, K.R. & E BARLOCHER, 1993. Seasonal chan- MORRIS, J.T. & K. LAJTHA, 1986. Decomposition and ges in microbial colonization of fresh and dried leaves. Arch. nutrient dynamics of litter from four species of freshwater Hydrobiol., 128: 1-12. emergent macrophytes. Hydrobiologiu, 13 1: 2 15-223. SRIDHAR, K.R., K.R. CHANDRASHEKHAR & K.M. NEWELL, S.Y., 1993. Decomposition of shoots of a salt- KAVERIAPPA, 1992. Research on the Indian subcontinent. In: marsh grass. Methodology and dynamics of microbial assem- The ecology of aquatir hyphomycetes. (F. BARLOCHER, ed.). blages. Adv. Microb. Eco/., 13: 301-326. 182-21 1. Springer-Verlag, Berlin. NEWELL, S.Y. & R.D. FALLON, 1989. Litterbags, leaf STOUT, R.J., W.H. TAFT & R.W. MERRIT, 1985. Patterns tags, and decay of nonabscised intertidal leaves. Can. J. Bot., of macroinvertebrate colonization on fresh and senescent alder 67: 2324-2327. leaves in two Michigan streams. Freshwat. Bid., 15: 573-580. NEWELL, S.Y., J.D. MILLER & R.D. FALLON, 1989. SUBERKROPP, K., G.L. GODSHALK & M.J. KLUG, Decomposition and microbial dynamics for standing, naturally 1976. Changes in the chemical composition of leaves during positioned leaves of the salt-marsh grass Spartina alterrzijlora. processing in a woodland stream. Ecology, 57: 720-727. Mar: Biol., 101: 471-481. TALIGOOLA, T.K., A.E. APINIS & C.G.C. CHESTERS, NEWELL, S.Y., M.A. MORAN, R. WICKS & R.E. HOD- 1972. Microfungi colonizing collapsed aerial parts of 11

Phrugniitrs conimurii.r Trin. in water. Now Hedivigirr, 23: 465- WEBSTER. J., S.T. MORAN 6i R.A.DAVEY, 1976. Growth 472. and qm-ulation oí Tridadiurn c,h