The Drilosphere As A Selforganizing System

Earthworms and other major soil ecosystem engineers create physical domains in soils that have all the characteristics of self-organized systems as defined by Perry (1995): Based on strong and rather specific interactions within physical boundaries, these systems change the constraints of their environment with positive feedbacks on their own living conditions (Lavelle et al. 2006; see Figure 5.1).

Soil constraints indeed have pushed soil organisms to develop intense interactions along evolutionary time, mostly of a mutualistic type (Lavelle and Spain 2006). These interactions operate within the boundaries of the rhizosphere of roots, drilosphere of earthworms, and termitosphere of termites, as well as a few other such domains (Lavelle 2002) that have

Communities

FIGURE 5.1 A general model of the drilosphere system. Within the boundaries (large dotted line) of their functional domain, the drilosphere, earthworms accumulate macro aggregates, galleries, and other pores that constitute the habitat for specific communities of microorganisms and invertebrates. Earthworms interact with these organisms. These interactions affect the external environment, especially plant growth and the composition of their communities, hydraulic conditions, and organic matter inputs and storage. Note that + signs indicate positive feedbacks.

Communities

FIGURE 5.1 A general model of the drilosphere system. Within the boundaries (large dotted line) of their functional domain, the drilosphere, earthworms accumulate macro aggregates, galleries, and other pores that constitute the habitat for specific communities of microorganisms and invertebrates. Earthworms interact with these organisms. These interactions affect the external environment, especially plant growth and the composition of their communities, hydraulic conditions, and organic matter inputs and storage. Note that + signs indicate positive feedbacks.

more or less recognizable limits. These systems in turn have feedback effects on external constraints. Roots and earthworms, for example, significantly affect soil structure with known impacts on water availability and their own ability to further penetrate this environment.

These systems, finally, are in a metastable state of equilibrium: The engineer makes the environment on which it and other species depend, and this situation lasts as long as engineers produce structures that replace the ones that have been degraded or destroyed. The large amount of energy channeled into the drilosphere actually is invested in the building of solid aggregates and creation of voids that allow soils to provide ecosystem services at a high rate.

We shall first focus on the description of these individual structures before considering their assemblage in soils and their emerging effects on soil properties.

BIOGENIC STRUCTURES

Earthworm casts may have diverse shapes and sizes. A first classification separates granular casts formed from an accumulation of small, fragile, and fine-textured pellets from globular casts comprised of coa-lescent round or flattened units (Lee 1985). While soil texture has a great influence over the final shape and structure of casts, some anatomic features of the posterior part of the earthworm gut also influence the process. Some earthworms produce a continuous flow of small independent pellets that rarely stick together to form a globular mass. Others expel at discrete intervals rather large amounts of wet and plastic digested soil material that tends to form units of up to 1 cm depending on the species. These units when wet easily stick to others, forming sometimes large and solid structures after they have been dried at least once (Shipitalo and Protz 1988, Blanchart et al. 1993). When deposited and regularly accumulated at the soil surface, globular casts may form spectacular tower-shaped structures up to 10-15 cm high and several hundred g dry mass (Figure 5.2).

Granular casts are fragile structures easily whipped off by rain when deposited at the soil surface. Globular casts may persist for very long periods of time, especially when they have been deposited in the soil and stabilized by one or two drying-rewetting cycles (Shipitalo and Protz 1988, Marinissen and Dexter 1990). In the African savannah of Lamto (Ivory Coast), Blanchart et al. (1997) showed that the large casts that comprise the macroaggregated structure of these soils in the upper 15 cm can still be found almost intact 32 months after removal of earthworms by a 48 hr artificial flooding. Dry globular casts deposited at the

Figure 5.2 Tower-shaped earthworm cast in fallows in Vietnam. This structure is formed after weeks of daily deposition of casts at the top edge of the structure. Note dense colonization by fine roots. Photo P. Jouquet. (See color plate.)

soil surface can persist for periods of a few days to several weeks or months, depending on their own mineral constitution, the degree of protection by plant cover, and the intensity of rain and other climatic events (alternance of dry and moist periods; freezing/thawing) (Decaens 2000, Le Bayon and Binet 1999). They may also be crushed by large mammals or broken by invertebrates that use them as shelter and/or food (Decaens 2000).

When fresh, casts are the seat of intense microbial activities, and ammonium and other nutrients are found at relatively high concentrations (Lavelle et al. 1992, Blair et al. 1995). In fresh casts of the pantropi-cal endogeic species Pontoscolex corethrurus, for example, ammonium concentrations in fresh casts vary from 67 to 1052 |g g-1 soil depending on clay mineralogy and N (nitrogen) content of the soil they had ingested. This represents on average 4 to 10% of the ingested organic N (Lavelle and Spain 2006). Assimilable P (phosphorus) concentration is also mul tiplied by a factor of 2 to 8 in the same casts as compared to a nonin-gested control soil (Lopez-Hernandez et al. 1993, Chapuis-Lardy et al. 1996). Once dried, casts become a harsh environment for microorganisms. Porosity is often extremely reduced. Casts of R. omodeoi, for example, have a bulk density of 2.3 as compared to 1.4 on average in soil. A superficial 15 |im pellicle rich in clay minerals and polysaccharides seems to isolate the cast environment from the outside and limit water and air penetration (Blanchart et al. 1993). Laboratory incubations have shown that organic matter mineralization was reduced to almost zero in these structures after 30 days, while a control nonaggregated soil continued to lose C (carbon) (Martin 1991). The quality of organic matter contained in earthworm casts is significantly different from the one in the nondigested control soil. Spectral signatures (Near Infrared Reflectance Spectrometry) allow separating them from aggregates produced by other biological or physical processes (Velasquez et al. 2007).

In soils favorable to earthworm activities, subterranean casts tend to accumulate as stable macroaggregates forming >40% of the total soil volume (Blanchart et al. 1999). Persistence and dynamics in time of these biogenic structures are still poorly documented. Highly unstable fresh globular casts can be easily dispersed or included into larger structures made by the addition of a number of similar structures.

The continuous deposition of casts at the soil surface is a response of earthworms to the general trend of soil to compact and a contribution to soil-forming processes. The proportion of the ingested soil that they deposit at the soil surface may vary from less than 5% to over 80% depending on species and soil conditions. Surface cast deposition is therefore a very poor indicator of earthworm activity; in the Lamto savannah, overall soil ingestion by endogeic earthworms estimated by a simulation model and surface cast depositions actually had opposite temporal patterns (Lavelle and Spain 2006). Surface cast deposition was maximum at the onset of the rainy season, while actual maximum soil ingestion by populations occurred several months later when surface deposition was very low.

Both categories of surface casts participate in the soil-creeping process, a general mechanism that transfers small-sized organic and mineral soil particles from the most elevated parts of the landscapes to low-lying areas where they accumulate (Nooren et al. 1995). Surface cast deposition also contributes to the progressive burial of gravels and stones by covering each year the soil surface with a continuous layer of 0.250.50 mm (Darwin 1881) to 1-2 mm (Lavelle 1978).

Gallery networks and burrows made by anecic and a few endogeic species have been studied independently from aggregate assemblages.

In a 12-year-old pasture in France, Bastardie et al. (2005) made a thorough quantitative description of earthworm burrow systems by applying X-ray tomography to 12 soil cores 25 cm in diameter and 60 cm in depth. Earthworm mean density was 101+-3 S.D. individuals m-2 distributed among 8 species. Three were anecic, four endogeic, and one was epigeic. Total burrow length ranged from 687 to 1212 mm-3. Volume represented 13.3 to 24.4 l m-3, which is less than 2.5% of soil volume. Total area of internal burrow walls represented 1069 to 7237 cm2 m-2. Only 9-43% of the volume was connected to the soil surface, and large seasonal variations did occur.

Burrow systems seem to have species-specific shapes and organizations. The diameter of galleries, their branching, orientation, and the continuity of the burrow system significantly vary among species (Kretzschmar 1990, Lamparski et al. 1987, Lightart et al. 1993, Bastardie et al. 2005).

Earthworm burrowing activities are highly sensitive to soil compaction (Kretzschmar 1991) and such soil pollutants as heavy metals (Nahmani et al. 2005) or pesticides (imidacloprid also known as "gaucho"; Capowiez et al. 2005). Galleries may act as preferential ways of circulation for gases and water. Their walls are regularly recoated with cutaneous mucus and sometimes with cast deposits each time the worm passes through. Cast deposition occurs more frequently in deep soil strata than closer to the surface; as a result, continuity between gallery and porosity of the rest of the soil is much better achieved in upper soil horizons than in the deeper soil.

THE TOPOLOGY AND DYNAMICS OF DRILOSPHERIC ASSEMBLAGES

Recent studies have demonstrated a significant relationship between soil macroaggregation, especially the abundance and size of biogenic aggregates, and the presence of earthworms and other soil ecosystem engineers (Blanchart et al. 1999, Bossuyt et al. 2006, Velasquez et al. in press). In the Brazilian Amazonian region of Parà, pastures derived from a primary forest cut 6 years ago were planted to four different plant species and all possible combinations of them, in a complete randomly designed experiment replicated in three blocks. There were two shrub species, the local weed Solanum nigrum and the legume Leucaena leu-cocephala, and two herbaceous species, the legume Arachis pintoi and the African grass Brachiaria bryzantha, the same grass that had been planted 6 years ago when the pasture had been created. Soil macro-invertebrate communities significantly responded to the change that

Figure 5.3 Significant co-inertia among soil fauna and soil morphology parameters in Amazonian pasture soils submitted to all possible combinations of four different plants. Projection on factorial plane 1-2 of fauna and soil morphology variables shows close location, in the right half of the figure, of earthworm species and biogenic aggregates, suggesting that they are formed by these earthworms. Key: pm, pl, ps: medium, large, and small physical aggregates; bs, bm, bl: biogenic aggregates of small, medium, and large size; rs, rm, rb: root aggregates of different sizes; inv: invertebrates found in small soil blocks.

Figure 5.3 Significant co-inertia among soil fauna and soil morphology parameters in Amazonian pasture soils submitted to all possible combinations of four different plants. Projection on factorial plane 1-2 of fauna and soil morphology variables shows close location, in the right half of the figure, of earthworm species and biogenic aggregates, suggesting that they are formed by these earthworms. Key: pm, pl, ps: medium, large, and small physical aggregates; bs, bm, bl: biogenic aggregates of small, medium, and large size; rs, rm, rb: root aggregates of different sizes; inv: invertebrates found in small soil blocks.

occurred in vegetation. Soil macroaggregation also changed, and co-inertia analysis showed a significant relationship of this soil attribute with macrofauna communities (p < 0.01). Earthworms, especially endo-geic species, were responsible for a great part of this aggregation as shown by location of their projections close to respective projections of biogenic aggregates in the factorial plan (Lavelle et al. unpublished data, Figure 5.3).

Drilosphere boundaries have never yet been directly described. They can be seen when examining thin sections of soil showing a discrete array of aggregates of different sizes and shapes; they also are felt when manually separating soil blocks into different classes of aggregates that further exhibit distinct spectral signatures (Velasquez et al. 2007). It will be particularly interesting to observe the frontiers of rhizospheres and drilospheres, two systems that otherwise develop very intense interactions.

Topology and dynamics of earthworm structure assemblages inside drilosphere boundaries are largely ignored. Laboratory and field experiments indicate only a few basic features that determine the spatial distribution of earthworms and the biogenic structures that they produce at spatial scales of a few centimeters to decameters.

First, endogeic earthworms do not seem to re-ingest casts of their own species unless they have been totally disintegrated (Lavelle 1978). This observation made on endogeic earthworms that produce globular casts has profound implications on the spatial distribution of their populations (Rossi 2003). In African moist savannahs at Lamto (Côte d'Ivoire), earthworms that produce globular casts (as the "compacting" species) have opposite patterns of horizontal distribution to species that produce granular casts ("decompacting" species). Patches with dominant decom-pacting populations actually had significantly lower bulk density (hence higher porosity) and a larger density of fine roots than patches predominantly occupied by "compacting" species. Statistical tests (partial Mantel test) showed that the nature of earthworm communities was responsible for these differences, not the opposite. The hypothesis that patches of opposite functional groups should move in time when transformation of soil has been completed has not been tested so far in the field. A modeling exercise predicts a shift in population distribution after 2-3 years of activity (Barot, in press).

Second, anecic earthworms seem to have rather sedentary and territorial ways of life (Edwards and Bohlen 1996). This allows soils that host dense populations to have rather regularly distributed vertical (and sometimes horizontal) drainage networks. A very interesting case was observed in rainforests of Madagascar that led to formulate a hypothesis on the role of anecic earthworms on soil conservation in these environments (Lavelle et al. unpublished data). The observed forest grows on highly unstable oxisols. Below a Ao 5 cm thick holorganic horizon, a 30 cm thick A1 horizon tops a 60 cm deep clayey B horizon. This B horizon has a special prismatic vertical structure that tends to disaggregate in case of physical disruptions like the one created by cutting a slope to create a road (Figure 5.4). Any excessive water infiltration in this soil layer is likely to generate horizontal disruptions leading to massive erosion events. Such events are prevented by absorption and drainage of the water in the upper 30 cm of soil, maintained by biological activities. The surface humic horizon, an accumulation of invertebrate fecal pellets

Water circulation Erosion

Water circulation Erosion

M 0l & A horizon fflyï'1 B horizon C horizon

Figure 5.4 Hypothesized role of giant anecic earthworms in soil conservation of tropical rainforest in Madagascar (Ambohilero forest). Rainfall is first absorbed by a 10 cm thick surface organic layer and then enters soil of the A horizon, where a subhorizontal network of earthworm galleries stores and channels water toward low-lying areas. This prevents water from penetrating too much in unstable low-lying horizons. Cutting a road across the slope exposes the B horizon (with a prismatic fragile structure) and C horizon made of highly dispersable alterites, eliminates the natural drainage system and accelerates massive soil erosion and the occurrence of major landslides ("lavakas"). Drawing by R.L. Andriamarisoa.

M 0l & A horizon fflyï'1 B horizon C horizon

Figure 5.4 Hypothesized role of giant anecic earthworms in soil conservation of tropical rainforest in Madagascar (Ambohilero forest). Rainfall is first absorbed by a 10 cm thick surface organic layer and then enters soil of the A horizon, where a subhorizontal network of earthworm galleries stores and channels water toward low-lying areas. This prevents water from penetrating too much in unstable low-lying horizons. Cutting a road across the slope exposes the B horizon (with a prismatic fragile structure) and C horizon made of highly dispersable alterites, eliminates the natural drainage system and accelerates massive soil erosion and the occurrence of major landslides ("lavakas"). Drawing by R.L. Andriamarisoa.

(mainly Diptera larvae, in that case), acts like a sponge able to absorb the equivalent of approximately 100 mm rainfall. Below this spongelike structure, giant anecic earthworms create a dense network of regularly distributed horizontal galleries that seem to act as a pipe network allowing water to store and convey by a horizontal transfer to low-lying areas and natural effluents (Figure 5.5). If confirmed, this hypothesis would explain how the destruction of these self-organized systems may lead to such spectacular landslides at the landscape scale.

Another consequence of the relative sedentarity of anecic and some endogeic earthworms is the accumulation over time of surface casts at the same place, which end up making rather large, sometimes tower-like, structures at the soil surface (Edwards and Bohlen 1996). In predominantly herbaceous fallows in Vietnam, individual cast accumulations

Figure 5.5 Section of the A horizon in soil of the Ambohilero Forest (Madagascar) showing anecic earthworm horizontal galleries (arrows and smaller photo) forming a pipe network allowing subhorizontal drainage. (Photos by P. Lavelle).

may amount to several hundred g dry weight, and total mass deposited at the soil surface may be approximately 10 kg dry mass m2 (Jouquet unpublished data). Anecics also may collect litter deposited around the mouth of their burrow, creating "middens" colonized by a dense and rather specific fauna and microflora (Hamilton and Sillman 1989, Subler and Kirsch 1998, Bohlen et al. 2002). This community is thought to achieve a preliminary decomposition of litter before earthworm ingestion. This process achieves an "external rumen" type of digestion as defined by Swift et al. (1979).

All these observations still are rather isolated and need to be supported by more fieldwork and modeling exercises and extended to a larger number of species and situations to be considered as general features. They also need to be interpreted in terms of the interactions that earthworms and other organisms develop inside the drilosphere.

BIOLOGICAL INTERACTIONS INSIDE THE DRILOSPHERE

With a few exceptions, studies show increased microbial activities in gut contents, fresh casts, and burrow walls produced by earthworm activities (Parle 1963, Barois and Lavelle 1986, Scheu 1987, Daniel and Anderson 1992, Fischer et al. 1995, Karsten and Drake 1997, Lattaud et al. 1997, Winding et al. 1997, Zhang 2000, Tiunov et al. 2001, Kersante et al. 2006). This activity—largely associated with earthworm digestion processes—is mostly that of soil-dwelling microorganisms. They used to be in resting stages in the soil and took advantage of optimal conditions created by the earthworm in its foregut to resume their activity. This process has been described as the "Sleeping Beauty" paradigm, which states that most microbial activity in soils occurs in specific microsites created by the activities of macroorganisms. Microbial communities in soil are mostly in resting stages, waiting for these "Prince Charmings" to reactivate them (Lavelle et al. 1995). Earthworm guts, gallery walls, and fresh casts are the drilospheric microsites where such activations do occur. There is growing evidence that only part of microorganisms are stimulated in this process, and more research is required to know how specific this interaction is and whether activated microbial communities differ among earthworm species and among the different soil functional domains (Lavelle et al. 2005).

Drilosphere structures are the habitat of very diverse communities of invertebrates of all sizes, while fine roots often concentrate in this specific environment. Decaens et al. (1999a) showed a rather fast colonization of casts of the anecic neotropical earthworm Martiodrilus carimaguensis by fine roots and a diverse community of invertebrates of the macro- and mesofauna. Drilospheres are also highly favorable habitats for Collembola and Acari (Marinissen and Bok 1988, Loranger et al. 1998). In a pasture of Martinique (French West Indies), patches densely colonized by the earthworm Polypheretima elongata had 28 instead of 23 species of Collembola; population density was 13,000 m-2 instead of 9000 outside the patches, and the Shannon index of diversity for their communities was 3.53 instead of 2.74 outside the patches.

FEEDBACK EFFECTS OF THE DRILOSPHERE ON SOIL CONDITIONS AND OTHER ORGANISMS

Feedback effects of biological interaction on environment constraints are expected to occur in drilosphere as a result of self-organization (Perry 1995). The accumulation of earthworm biogenic structures in soils has significant effects on soil physical properties that may, or may not, have positive feedback effects on earthworms through changes in moisture regime in soils, a fundamental determinant of earthworm activities (Edwards and Bohlen 1996).

Many experiments in laboratory and observations in field conditions have indicated such effects (Edwards and Bohlen 1996, Chauvel et al. 1999, Decaens et al. 1999b, Hallaire et al. 2000). Clay mineralogy seems to be one clue to the occurrence of significant influence of earthworms with more pronounced and lasting effects in soils with kaolinitic 1 : 1 type clays than with smectitic 2 : 1 clay materials (Blanchart et al. 2004).

Pontoscolex corethrurus, a very active endogeic invasive earthworm that produces globular casts, has been claimed to be responsible for soil compaction in sweet potato cultures (Rose and Wood 1980), maize crops (Hallaire et al. 2000), and recently installed Amazonian pastures following conversion of primary forest to pasture (Chauvel et al. 1999). P. core-thrurus is actually a clear example of invasive engineer as defined by Cuddington and Hastings (2004), as it is able to survive in conditions that many native species cannot withstand and modify the habitat in ways that make other species' return more difficult (Lavelle et al. 1987, Lapied and Lavelle 2003).

Unlike compacting species, decompacting filiform endogeics significantly decrease soil bulk density when kept alone in experimental soils (Blanchart et al. 1999). Although more data clearly are needed to reach a conclusion, it seems that single earthworm species are not able to maintain alone suitable physical conditions in most cases. They probably need to interact with other earthworm and other invertebrate engineer species, or natural physical processes, in order to achieve this feature. These findings support the view of Jouquet et al. (2006) that endogeic earthworms may be accidental rather than extended pheno-type ecosystem engineers. However, regulations obviously occur at the scale of communities, and positive feedback of soil structure maintained by a community may further affect each of the species in the community. Earthworms seem to re-ingest casts of other species, thus converting certain types of structures (e.g., casts of anecics rich in organic residues, or compact casts of large endogeic species) into other types (epigeic or other litter invertebrates, Scheu and Wolters 1991; loose granular casts of epigeics, Mariani et al. 2001), and thus exerting regulatory effects on the proportions of each type and preventing the accumulation in excess of a single category of casts.

Compacting species seem to exert negative feedbacks on their own survival by reducing the porous space and eating out the small aggregates that are their feeding resource; however, decompacting species develop at the same time opposite effects in adjacent patches. The resulting effect of the two functional groups likely has positive feedbacks on both groups at the scale of the ecosystem.

The experiment conducted in natural field conditions in Brazilian Amazonia (Figure 5.3) by Lavelle et al. (unpublished data) and observations of Velasquez et al. (2006) showed that earthworms actually may be responsible for a significant part of aggregation in the upper 10-20 cm of many soils. Soils that have improved biogenic aggregation are less compact and likely present improved hydraulic properties in the upper few cm below surface. This result, however, largely depends on the diversity and composition of earthworm and other soil engineer communities and the nature of their respective biogenic structures.

Feedback effects on direct or indirect competitors comprise the decrease in litter dwelling arthropods when anecic populations increase and adverse effects on communities of plant parasitic nema-todes (Yeates 1981, Lavelle et al. 2004). As regards microbial communities, drilospheres tend to be colonized by bacteria rather than fungi (Hendrix et al. 1986). Mutualist digestion systems developed in earthworm guts seem to involve only bacteria, and there is slight evidence that earthworm cutaneous mucus sprayed over litter accumulated in "middens" or in burrow walls might have some fungistatic effects (Tiunov et al. 2001). Positive effects on plant growth also are likely to increase the amount of food available to earthworms (Brown et al. 1999, Scheu 2003).

ALLELOCHEMICAL AND BIOLOGICAL EFFECTS ON PLANT HEALTH AND COMMUNITIES

Interactions among earthworms and plants are intense and involve a rather diverse range of mechanisms. Plants' growth and resistance to parasites are improved in the presence of earthworms. Their communities also may be affected by the selective effect of earthworms on the germination of the soil seed bank. Several hundreds of laboratory and field experiments have shown significant increases in plant production in over 70% of cases (Brown et al. 1999, Scheu 2003). The sense and intensity of this effect vary with plant and earthworm species. Shoot and grain productions are generally significantly enhanced, while root production remains unaffected or decreases (Brown et al. 1999). Effects generally are greater in poor than in fertile soils. This supports the hypothesis that earthworm effects are constant and proportional to their overall activity; their contribution is less visible when plant production is not limited by soil constraints. Five mechanisms likely explain earthworm effects:

the release of nutrients in fresh casts and their uptake by fine roots; favorable effects on soil physical properties;

enhanced activities of mutualist microorganisms, mycorhizae, and N-fixing bacteria;

direct protection from belowground parasites; indirect protection from aboveground parasites; hormone-like effects on plant growth.

Allelochemical effects are clearly involved in the last mechanism and likely operating to some extent in mechanisms 3 and 4.

Recently, Blouin et al. (2006) have shown that the enhancement of rice growth in the presence of Reginaldia omodeoi was due neither to an enhanced nutrient availability, nor to any change in soil physical properties. Plants received different amounts of mineral-N fertilizer, from 0 to 1600 |imol l-1. In the presence of earthworms, a rather constant increase was observed, whatever the mineral-N concentration. Since the experiment did not allow parasites or specific root mutualists to act, and because no limitation in water availability or other nutrients was present, they concluded that a "hormone-like effect" probably was responsible for the observed effects (Figure 5.6). This effect, first mentioned by Tomati et al. (1988), has been found in Eisenia fetida lombricompost extracts (Atiyeh et al. 2002, Arancon et al. 2003).

Earthworm effect therefore is more than a simple indirect effect of their physical engineering activities on plants. This was shown again in an experiment where rice plants (Oryza sativa) had been infested with a cyst-forming nematode with or without earthworms (R. omodeoi) in the soil (Blouin et al. 2005). Earthworm activities changed the expression of stress-responsive genes in the leaves of rice plants and allowed them to become tolerant instead of drying out, as was observed when earthworms were absent (Blouin et al. 2005). This systemic response of plants to earthworm activities recently has been confirmed with Arabidopsis thaliana interacting with the Lumbricidae Aporrectodea caliginosa (U. Jana, A. Reppelin, Y. Zuily-Fodil unpublished data). It is an indication that highly sophisticated communication and interactions among earthworms and plants have been selected by evolution. The exact nature of the interaction—the signal molecules likely involved and their origin (produced by the earthworm or by specific microbes activated by the earthworm)—is not known.

Another example of a systemic response of plants to earthworm activities is observed in tea plants restored with the FBO (Fertilisation Bio-Organique) patented method in South China (Senapati et al. 1999, P. Lavelle, J. Dai, E. Velasquez, and N. Ruiz-Camacho unpublished data).

Figure 5.6 Response of rice (Oriza sativa) to increasing inputs of mineral nitrogen, in the presence (dotted line) and absence (solid line) of earthworms. The effect of earthworms on plant growth (distance between the two curves) is constant whatever the nutrient status of soil, which allows rejecting the hypothesis of enhanced mineralization to explain the gain in plant growth observed in their presence.

Figure 5.6 Response of rice (Oriza sativa) to increasing inputs of mineral nitrogen, in the presence (dotted line) and absence (solid line) of earthworms. The effect of earthworms on plant growth (distance between the two curves) is constant whatever the nutrient status of soil, which allows rejecting the hypothesis of enhanced mineralization to explain the gain in plant growth observed in their presence.

Following the inoculation of earthworms and stimulation of their activities by organic amendments in soil, tea quality evaluated by systematic tasting assessment was significantly improved.

At the larger scale of a pasture plot, several studies have shown that earthworms have significant effects on the germination of seed banks (Decaens et al. 2003, Milcu et al. 2006). Other examples show how earthworms and other soil organisms may influence the composition of plant communities and their natural successions through different effects (Bernier and Ponge 1994, De Deyn et al. 2003).

0 0

Post a comment