Author for correspondence: Roger T. Koide Tel: +1 814 8630710 Fax: +1 814 8636139 Email: email@example.com
• Ectomycorrhizas may interact with saprotrophic microorganisms to influence decomposition. These interactions are poorly understood.
• We buried mesh envelopes filled with either litter or F-layer material in the forest floor of a Pinus resinosa plantation. After 16 months we characterized decomposition, C : N ratio, water content, and the density of four ectomycorrhiza morphotypes.
• For both substrates there were negative correlations between ectomycorrhiza density and decomposition, and between ectomycorrhiza density and water content. The relative proportion of morphotypes was significantly different in litter and F-layer material buried at the same depth in the forest floor.
• We conclude that the negative effect of ectomycorrhizas on decomposition could be mediated by extraction of water, particularly in relatively dry years. The ‘Gadgil effect’ may be explained by this phenomenon. In wetter years, water extraction may be less consequential. Also, variation in substrate quality may influence the structure of ectomycorrhizal fungal communities.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
The decomposition of biological residues is an essential process in all ecosystems. Decomposition counterbalances primary production as it returns the carbon in organic detritus (litter) to the atmosphere whence it came. Decomposition also results in the transformation of organic forms of nitrogen, phosphorus and sulfur, which are found in litter, into mineral forms that are readily absorbed by plants (Swift et al., 1979; Attiwill & Adams, 1993). Several kinds of organisms participate in the decomposition process. The most important of these may be the saprotrophic bacteria and fungi because they produce the enzymes that hydrolyze the major litter components including cellulose, lignin and chitin. Temperature, moisture and litter chemistry are repeatedly shown to influence the activity of the saprotrophic microorganisms, and thus the rate of decomposition of litter (Swift et al., 1979; Gosz, 1984; Fog, 1988; Hunt et al., 1988; Vestgarden, 2001; Zimmer, 2002).
Gadgil & Gadgil (1971, 1975) suggested that ectomycorrhizal fungi might actually retard decomposition by competing with saprotrophic microorganisms for nutrients. Abuzinadah et al. (1986) more specifically suggested that ectomycorrhizal fungi could compete with saprotrophic microorganisms for organic nitrogen. Saprotrophic microorganisms certainly may decrease nutrient uptake by mycorrhizal fungi and their host plants (Koide & Kabir, 2001; Lindahl et al., 2001). If two-sided competition between ectomycorrhizal fungi and saprotrophic microorganisms occurs, which is at least theoretically possible, then ectomycorrhizal fungi may retard decomposition by limiting the activities of the saprotrophs.
Competition may occur for resources other than nutrients. Water may be such a resource. Therefore, another hypothesis to explain the negative effect of ectomycorrhizal roots on decomposition is that the removal of water from forest floor substrates by mycorrhizal fungi and colonized roots limits the activity of saprotrophic microorganisms. Moisture, of course, is one factor that is well known to influence decomposition (Swift et al., 1979). Moreover, some species of ectomycorrhizal fungi can absorb significant quantities of water from the surrounding medium (Brownlee et al., 1983). We therefore examine this hypothesis in the forest floor of a red pine (Pinus resinosa Ait.) plantation in central Pennsylvania, USA.
Materials and Methods
This research was conducted in an c. 65-yr-old-red pine (Pinus resinosa Ait) plantation located in State College, Centre County, PA, USA. The plantation has a well-developed organic horizon comprising litter, fermentation (F) and humified (H) layers overlying a thin, sandy, eluviated A-horizon and a thicker, B-horizon (Morrison sandy loam, Ultic Hapludalfs). There is little understory vegetation, which consists of small patches of Vaccinium spp., Acer rubrum and Pinus resinosa saplings, and individual saplings of Quercus alba. These were avoided in the current study.
Forty-eight envelopes, c. 13 × 10 cm, were fabricated from fiberglass window screening (1 mm mesh) and were filled with pine needle litter. This was collected from the forest floor at several locations in the plantation, pooled, mixed, dried at room temperature, and cut into 1–2 cm pieces. Litter is defined in this study as organic material on the forest floor (typically 90% needles) that has undergone only preliminary stages of decomposition. Needles in the litter, for example, are whole or nearly whole. Each of the litter envelopes contained untreated (not sterilized, not heated) litter equivalent to 14.0 g after 70°C oven-drying. Another 48 envelopes were filled with F-layer material, also collected from the forest floor at several locations in the plantation, pooled, mixed, sieved (4.0 cm) to remove most roots and dried at room temperature. Each F-layer envelope contained untreated F-layer material equivalent to 17.8 g after 70°C oven drying. F-layer material in this study is defined as partially decomposed organic material below the litter layer, brown in color, the components of which are still clearly recognizable. On 17 July 2000 one litter and one F-layer envelope were buried adjacent to each other at 48 random locations (but at least 1 m from any tree) in the pine plantation at the interface between the organic horizon (comprising the litter, F and H layers) and the uppermost mineral horizon. The organic horizon was, on average, 5.1 cm thick.
On 16 August, 15 September, 13 November 2000 and on 15 January 2001, six envelopes of each substrate were collected from the forest floor. We measured the moist weight of the remaining material in each envelope. Moist subsamples from each envelope were also weighed, oven-dried (70°C) and weighed again to determine the ratio of moist to d. wt. The d. wt of the total remaining material in each envelope was then calculated.
On 27 November 2001, c. 16 months after initiation, all remaining envelopes were harvested (a total of 24 litter and 24 F-layer envelopes). Pine roots that had grown into each envelope were removed carefully and placed in Petri dishes containing wet filter paper. The substrates from each envelope were weighed as before to calculate d. wts and water contents. Carbon and nitrogen concentrations of subsamples from each envelope were determined on an elemental analyzer (Fisons Instruments, model 1108, Beverly, MA, USA). For each envelope, the numbers of individual root tips of four ectomycorrhizal morphotypes were counted. The morphotypes were distinguished by visual means under a dissecting microscope using reflected light. These four morphotypes included Cenococcum, Tylopilus felleus, and two other morphotypes we refer to as E (arbitrary designation), and mixed yellow/brown (Fig. 1). The Cenococcum, E and the Tylopilus felleus morphotypes are distinctive and rather easy to distinguish visually. The yellow/brown morphotype, however, is rather variable and probably comprises multiple species. For Cenococcum, only nonbifurcated tips were found. For the E morphotype, each was considered as a single root tip. For the Tylopilus felleus and mixed yellow/brown morphotypes each bifurcated root was considered as two tips. The density of the mycorrhizas (tips cm−3) was calculated from the number of root tips and the volume of the material within each envelope, which was calculated from its oven d. wt and its bulk density. The bulk density was calculated from the weight and volume (graduated cylinder method) of subsamples.
Comparisons between litter and F-layer mesh envelopes were performed using the one-way analysis of variance module of the Statgraphics Programs (STSC, 1991). The correlations between decomposition and mycorrhiza density, water content or C : N were determined using the linear regression analysis module of the same programs to obtain the best-fit equations.
Both litter and F-layer material exhibited significant decomposition during the first 4 months of residence in the forest floor. During that period there were no significant differences between litter and F-layer material in the proportion of substrate remaining (Fig. 2). After 4 months decomposition of the litter continued, but no further decomposition of F-layer material occurred. For the samples harvested at the end of the experiment (27 November 2001) litter had decomposed significantly (P < 0.0001) more than F-layer material. The d. wt loss was 33.8%, s.e.m. 1.4 and 15.6%, s.e.m. 1.3, for litter and F-layer material, respectively. Litter also contained significantly (P = 0.0007) more moisture than F-layer material at the final harvest (149.9%, s.e.m. 10.6 vs 101.9%, s.e.m. 7.3, respectively), and the litter C : N was significantly (P < 0.0001) lower than for F-layer material (20.2, s.e.m. 0.4 vs 35.1, s.e.m. 0.3, respectively).
For the samples harvested at the end of the experiment (27 November 2001), decomposition of both litter and F-layer material was negatively related to the total density of ectomycorrhizal roots (Fig. 3). Thirty-six percent of the variation in litter decomposition was associated with variation in ectomycorrhiza density (P = 0.003). For the F-layer material, 34% of the variation in decomposition was associated with variation in ectomycorrhiza density (P = 0.004). No significant relationships were found between decomposition and the density of individual morphotypes for either litter or F-layer material.
When considering both substrate types together (litter needles and F-layer material) there was an obvious overall negative correlation between C : N and decomposition (Fig. 4). However, there were no significant relationships between C : N and decomposition for either litter or F-layer material when considered separately (P = 0.17 and 0.74, respectively). Litter and F-layer material C concentrations were not significantly affected by the 16 months incubation in the forest floor (Table 1). However, for both litter and F-layer material there were significant increases in the N concentration over time. Thus, the C : N ratios decreased with time for both substrates.
Table 1. Means (s.e.m) of substrate C and N concentrations and C : N ratios at the beginning of the experiment and after 16 months in the forest floor. n = 3 for initial values and n = 24 for 16 month values. Within a substrate type a different letter indicates means are significantly different (P ≤ 0.05) according to the analysis of variance
C : N ratio
1.22 (0.05) b
40.7 (1.69) a
After 16 months
2.56 (0.04) a
20.2 (0.40) b
0.98 (0.02) y
43.6 (1.16) x
After 16 months
1.25 (0.01) x
35.1 (0.29) y
There were significant positive correlations between decomposition and water content for both litter and F-layer material (Fig. 5). For litter, 23% of the variation in decomposition was associated with variation in water content (P = 0.021). For F-layer material, 31% of the variation in decomposition was associated with variation in water content (P = 0.006).
There were significant negative relationships between substrate water content and ectomycorrhiza density for both litter and F-layer material (Fig. 6). For litter, 21% of the variation in water content (P = 0.028), and for F-layer material, 40% of the variation in water content (P = 0.001), were associated with variation in ectomycorrhiza density.
The total density of ectomycorrhizas was significantly higher in litter (4.4 tips cm−3, s.e.m. 0.6) than in F-layer (1.6 tips cm−3, s.e.m. 0.2). Litter and F-layer material possessed significantly different ectomycorrhiza morphotype compositions (Table 2). There was a significantly higher density for each of the morphotypes in the litter than in the F-layer. Proportionally, there was not a significant effect of substrate on the Cenococcum morphotype. However, the Tylopilus felleus morphotype and the E morphotype were both proportionally more abundant in litter than in F-layer material, while the yellow-brown morphotype was proportionally less abundant.
Table 2. The total number of root tips of each morphotype [mean (s.e.m)] and the proportion of root tips [mean (s.e.m)] of each morphotype found within litter and F-layer material after 16 months. A single asterisk indicates that the mean for the litter differs significantly (P ≤ 0.05) from the mean for the F-layer
In this study we found a significant negative correlation between ectomycorrhiza density and decomposition for both litter and F-layer material, consistent with the results of Gadgil & Gadgil (1971, 1975). Gadgil & Gadgil (1975) suggested that the retardation of litter decomposition by ectomycorrhizas was caused by competition for nutrients. Abuzinadah et al. (1986) found that ectomycorrhizal fungi could directly obtain N from organic sources under laboratory conditions, which could theoretically result in reduced activity by saprotrophic microorganisms if they were limited by N. However, some research indicates that ectomycorrhizal fungi generally have a poor ability to obtain N from naturally occurring organic sources in the forest floor (Bending & Read, 1996; Colpaert & van Tichelen, 1996), primarily because the major forms of organic N there are complexed by polyphenolic compounds (Qualls et al., 1991; Northup et al., 1995), and many ectomycorrhizal fungi have little capacity to hydrolyze polyphenolics (Bending & Read, 1996). Therefore, ectomycorrhizal fungi may not compete directly against saprotrophic microorganisms for organic N. Instead, ectomycorrhizal fungi may actually depend on the saprotrophic fungi to mobilize N from the recalcitrant organic forms (Colpaert & van Laere, 1996). Therefore, competition for organic N may not always be the sole reason for the ‘Gadgil effect’.
In the current research, increased density of ectomycorrhizas was associated with decreased substrate water content, which is to be expected. Others have previously shown that trenching of the soil to cut off roots increases soil moisture (Fisher & Gosz, 1986a; Staaf, 1988), and that naturally occurring lower densities of roots are associated with higher soil water contents (Parmelee et al., 1993). Roots alone can certainly withdraw water from the forest floor, but this could be compounded by ectomycorrhization because some rhizomorph-forming ectomycorrhizal fungi are themselves capable of absorbing water and transporting it to their hosts (Brownlee et al., 1983). If, as a consequence of this water withdrawal, the substrate water content fell below some critical level, saprotroph activity could be depressed (Swift et al., 1979). Forest litter decomposition in other ecosystems can be strongly affected by water availability within the range of water contents described in this study (Fioretto et al., 1998; Reichstein et al., 2002). Therefore, our observations suggest another possible explanation for the ‘Gadgil effect’, a simple withdrawal of water from the forest floor by ectomycorrhizal roots.
The effect of water removal on litter decomposition should depend on the weather. In drier weather water removal by roots may have a significant negative effect on decomposition. For the period July 2000 to November 2001 (this study), the central mountains division of Pennsylvania received 118 cm of precipitation, approximately 27 cm less than the average between 1961 and 1990 (Pennsylvania Agricultural Statistics, 2000–01; Pennsylvania Agricultural Statistics, 2001–02). In generally wet weather water removal by ectomycorrhizas may have little effect on forest floor water content. In fact, it is possible for decomposition to be limited by too much moisture (Progar et al., 2000). Expression of the ‘Gadgil effect’, if caused by water withdrawal, may thus depend on weather. This may explain some of the discrepancies in the literature.
There is often a negative relationship between C : N ratio of forest floor substrates and decomposition rates (Swift et al., 1979). Indeed, there was such a relationship for litter and F-layer material examined together. However, for a given substrate (litter or F-layer material separately), no such relationship existed. Thus, variation in C : N ratio did not seem to be a large factor in determining variation in the rate of decomposition of a given substrate in this study.
The ectomycorrhizal fungal community was affected by substrate type (litter vs F-layer material), even though the two substrate types were buried at the same depth in the forest floor. This may have been caused by differences in substrate physical or chemical properties. We have previously shown that different ectomycorrhizal fungi exhibit unique responses to pine needle litter extracts (Koide et al., 1998). In conjunction with spatial variability in substrate quality that surely exists in the forest floor, differential preferences and tolerances to the physical and chemical properties of various substrates exhibited by the fungi could result in spatial partitioning, which has been observed (Malajczuk & Hingston, 1981; Sagara, 1995; Griffiths et al., 1996; Goodman & Trofymow, 1998; Conn & Dighton, 2000; Dickie et al., 2002). This kind of partitioning may be important in the maintenance of species diversity.
We gratefully acknowledge financial support from the A. W. Mellon Foundation.