Ecosystem Engineers in a Self-Organized Soil: a Review of Concepts and Future Research Questions

TECHNICAL ARTICLE

Ecosystem Engineers in a Self-organized Soil: A Review of Concepts and Future Research Questions

Patrick Lavelle,1 Alister Spain,2 Manuel Blouin,3 George Brown,4 Thibaud Decaëns,5 Michel Grimaldi,6 Juan José Jiménez,7 Doyle McKey,8 Jérôme Mathieu,1 Elena Velasquez,8 and Anne Zangerlé9

Abstract: their interactions with other organisms. With advances in research, Soils are self-organized ecological systems within which or- their participation in diverse soil processes has been progressively ganisms interact within a nested suite of discrete scales. Microorganisms detailed, and the scientific literature now offers a wide range of form communities and physical structures at the smallest scale (microns), publications that describe the many roles that these organisms followed by the community of their predators organized in microfoodwebs perform in soils (Lavelle and Spain, 2001; Lavelle et al., 2006; (tens of microns), the functional domains built by ecosystem engineers Blouin et al., 2013, and 103 articles in Web of Science with key (centimeters to meters), ecosystems, and landscapes. Ecosystem engineers, words “soil ecosystem engineers” and “ecosystem services,” principally plant roots, earthworms, termites, and ants, play key roles in March 11, 2016). Soil ecosystem engineers also comprise plant creating habitats for other organisms and controlling their activities through roots that play remarkable and multifaceted roles as engineers physical and biochemical processes. The biogenic, organic, and organomineral (Barea et al., 2005; Hinsinger et al., 2009, Berg and Smalla, structures that they produce accumulate in the soil space to form three- 2009). Vertebrates may also play important roles as physical dimensional mosaics of functional domains, inhabited by specific commu- engineers at specific locations and where populous (Butler, nities of smaller organisms (microfauna and mesofauna, microorganisms) 1995; Bancroft et al. 2005; Villarreal et al., 2008; James et al., that drive soil processes through specific pathways. Ecosystem engineers 2009; Dunham, 2011). also produce signaling and energy-rich molecules that act as ecological Although the term ecosystem engineers was defined rela- mediators of biological engineering processes. Energy-rich ecological me- tively recently by Jones et al. (1994), both their occurrence in soil diators may selectively activate microbial populations and trigger priming and their physical and other effects were originally recognized by effects, resulting in the degradation, synthesis, and sequestration of spe- Aristotle, long before modern scientific thought existed. White cific organic substrates. Signaling molecules inform soil organisms of their (1789) and Darwin (1881) provided much of the impetus for the ’ producers respective presences and change physiologies by modifying gene modern era of research that commenced in the second half of expression and through eliciting hormonal responses. Protection of plants the 20th century (Lee and Wood, 1971; Bouché, 1977; Lobry de against pests and diseases is largely achieved via these processes. At the Bruyn and Conacher, 1990); few of the general effects studied to- highest scales, the delivery of ecosystem services emerges through the func- day escaped their observation. Perhaps because of the complexity tioning of self-organized systems nested within each other. The integrity of of its study medium, soil science has become fragmented into a the different subsystems at each scale and the quality of their interconnec- number of subdisciplines. Reductionist approaches to the science tions are a precondition for an optimum and sustainable delivery of ecosys- have led to the situation in which soil science in general has tem services. Lastly, we present seven general research questions whose largely ignored the interactions that occur between soil ecosystem resolution will provide a firmer base for the proposed conceptual frame- engineers and other soil organisms and their participation in such work while offering new insights for sustainable use of the soil resource. important soil processes as carbon processing (van Breemen, Key Words: Ecological mediators, ecosystem engineering, 1993; Johnson and Hole, 1994; Lavelle, 2000; Schmidt et al., roles of the soil biota, self-organized systems, signaling molecules, 2011; Bottinelli et al., 2015). soil functioning, soil structures This article first proposes an updated conceptual version of – the soil model proposed 129 years ago by Dokuchaev (1889) (Soil Sci 2016;181: 91 109) and further adapted by Jenny (1941) and Duchaufour (1977) to illustrate the processes and dynamics of soil formation and function- he soil macroinvertebrates, now classified as “ecosystem en- ing. We acknowledge self-organization as a basic mechanism that gineers,” have long been studied for their spectacular effects T explains the complex interactions among organisms, structures, on soil functioning and development and for the wide variety of and processes in soils and that makes soil dynamics so difficult to describe and predict. By self-organization, we mean a process 1IEES, Université, Pierre et Marie Curie Paris 6, Paris, France. where some form of overall order or coordination arises out of 2The University of Western Australia, Perth, Australia. the local interactions between the components of an initially 3 Université Paris Est Créteil, Créteil, France. disordered system. We identify different types of soil ecosys- 4EMBRAPA Floresta, Curitiba, Brazil. 5CEFE-CNRS, Montpellier, France. tem engineers and detail the different roles they play as key ac- 6Institut de Recherche pour le Développement, Bondy, France. tors in self-organized patterns and processes within soils. 7Instituto Pirenaico de Ecologia, Jaca, Spain. Specific mechanisms that allow them to influence the provision 8 Universidad Nacional de Colombia, Palmira, Colombia. of ecosystem services are described (Fig. 1). We then discuss 9Technische Universität Braunschweig (TUB), Braunschweig, Germany. Address for correspondence: Dr. Patrick Lavelle, Centre IRD Ile de Fance, 32 rue the further research necessary to develop this model of soil Henri Varagnat, 93143 Bondy Cedex, France. E-mail: [email protected] functioning and meet the needs of future fundamental and Financial Disclosures/Conflicts of Interest: This project has received funding applied research. from NERC-IOF (project NE/M017656/1) and CNPq-CSF (project 401824/ 2013-6) and from the ALTERBIO project (Programme ECOPHYTO 2009, Ministere de l’Ecologie, France). Received October 2, 2015. ECOSYSTEM ENGINEERS: ANOTHER NAME FOR A Accepted for publication March 21, 2016. LONG ACKNOWLEDGED CONCEPT Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0038-075X Ecosystem engineers were defined by Jones et al. (1994) as DOI: 10.1097/SS.0000000000000155 “organisms that directly or indirectly modulate the availability of

Soil Science • Volume 181, Number 3/4, March/April 2016 www.soilsci.com 91

FIG. 1. Soil ecosystem engineers and the delivery of ecosystem services: three different processes (in red) 1. The organization of soil physical structure and associated ecosystem services. 2. Selection and activation of plant, microbial and smaller invertebrate communities that determine decomposition and nutrient cycling processes. 3. The release of hormones that regulate primary production. A color version of this figure is available in the online version of this article. resources to other species, by causing physical state changes in (Lee and Wood, 1971), earthworms and their drilospheres (Bouché, biotic and abiotic materials. In so doing they modify, maintain, 1977; Lavelle, 1984), and ants and their myrmecospheres (Folgarait, and create habitats.” Ants, termites, earthworms, and a number 1998), together with less widespread Coleoptera, Isopoda, or ver- of other soil-dwelling invertebrates and vertebrates closely con- tebrates and a host of others. form to this definition. Specific examples include the ants and ter- This research has clearly demonstrated that ecosystem engi- mites that build large earthen mounds and create extensive neering includes the capacity of organisms to drive ecosystem networks of subterranean galleries and chambers. They concen- processes and upgrade ecosystem performance: it involves sub- trate organic matter within their nests and deposit finer materials stantially more than simple physical effects on soils. While a nec- in the upper parts of the soil profile and are examples of central essary complement to theories that purport to explain ecosystem place foragers. The changes they mediate to soil-water relation- function and development solely through simple trophic relationships lead to different drainage characteristics while creating habitats ships, physical engineering forms only a part, possibly small, of for smaller organisms (Jones et al., 1994; Lavelle and Spain, 2001). the variety of nontrophic interactions that occur in natural systems. The soil-engineering concept is part of a general approach In the compact environment of the soil, with its limited and highly developed by ecologists who have long observed that some organ- fragmented pore space, the need to improve the efficiency of the isms play roles that are disproportionately large relative to their use of naturally low-quality resources has led organisms to coop- abundance. Paine (1969) defined “keystone species” as those that erate within self-organized systems (Lavelle, 1997). In this article, maintain the structure of biological communities by various pro- we consider three different types of ecosystem engineering: phys- cesses including predation (e.g., sea otters, Carnivora) and mutu- ical, biological, and biochemical. alism (e.g., plants that provide feeding resources to herbivores and therefore the entire foodweb), thereby sustaining the ecosystem- engineering activities of, for example, prairie dogs (Rodentia). SOIL AS A SELF-ORGANIZED SYSTEM: A GENERAL Soil ecologists have recognized these effects for many years: MODEL OF SOIL GENESIS AND FUNCTION earthworms, long acknowledged for their outstanding effects on Self-organization is a general property of natural systems in plant growth and soil physical properties, have been divided into which interactions occur among species at the discrete scales where separate functional groups that have quite different effects on soils these species coexist. Haken (2008) defines self-organization as the (Bouché, 1977; Lee, 1985). Biological systems of regulation were “spontaneous often seemingly purposeful formation of spatial, tem- then defined as associations among a macroorganism (a root, an poral, spatiotemporal structures or functions in systems composed invertebrate ecosystem engineer), a community of microorganisms of few or many components.” In physics, chemistry, and biology, that is selected and activated by the macroorganism, and an organic self-organization occurs in open systems driven far from equilib- resource that they use as a food resource (Lavelle, 1984). The later rium by the interactive processes thus created. The process of concept of functional domain (Lavelle, 2002) recognized the fact self-organization can also be found in such diverse contexts as eco- that soil ecosystem engineers organize the soil environment and nomics, sociology, medicine, and technology. create microenvironments with their own specific microbiologi- Highly diverse sets of species may interact within soils be- cal, chemical, and physical characteristics. These include roots cause of the great fragmentation of the void space in which species and their rhizospheres (Hiltner, 1904), termites and their termitospheres live. Nonetheless, elements potentially able to interact may not

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. Soil Science • Volume 181, Number 3/4, March/April 2016 Ecosystem Engineers in a Self-organized Soil necessarily be in contact, not all possible chemical or physical re- (1,000–10,000 years) and long-term (100,000 to 1,000,000 years) actions will occur, nor do all the predators present encounter their soil processes. Four of these processes are directly related to bio- possible prey species. As a result, potential prey may avoid their logical activity: litter formation, peat formation, pedoturbation, predators through spatial isolation, and anaerobic and aerobic pro- and the formation of soil structure. cesses may occur at the same time within a soil horizon. However, soil formation may also involve some self-limiting Self-organization may also be considered a consequence of irreversible processes (weathering, leaching, salt accumulation, or evolution. As systems evolve, interactions among organisms tend formation of hard pans) that are mostly unfavorable to unspecial- to increase with time, shifting progressively from negative (com- ized biological activity. According to Targulian and Krasilnikov petition, predation) to positive (mutualist) (Allee et al., 1949; (2007), only 12 of the 39 diagnostic horizons (mollic, umbric, Margulis, 1981; Lewis, 1985). Organisms tend to share their capa- chernic, melanic, histic-eutrophic, hortic, terric, andic, cambic, bilities, developing symbioses and mutualisms at different scales, calcic, nitic, and vitric horizons) of the World Reference Base from merging their bodies within a single integrated organism are considered favorable to biological activities. The other 27 (symbiogenesis, Margulis, 1981), to symbiotic relationships are considered unfavorable, in a productive sense “impaired” (Lewis, 1985) among organisms of similar (isosymbiotic) or by their own changes, and the biota adapts to such environ- different (anisosymbiotic) sizes. The relationships are said to ments rather than improving them. be inhabitational or exhabitational, according to the locations Contrastingly, soil maturation processes that lead to the for- of one of the mutualists, within or outside the body of the other mation of unfavorable weathered horizons also result in a gen- individual. At the scale of ecosystems, connectivity, or the inten- eral improvement of basic soil physical properties associated sity of interactions among organisms, is usually very low in incip- with deeper soil profiles. Therefore, during their formation, soils ient systems and increases with ecosystem maturation (Holling, evolve through a continuum of developmental stages from incipi- 2001). As evolution proceeds, organisms (and probably groups ent to mature and weathered. In the earlier stages, incipient and of organisms) are selected that use resources with progressively young, trophic conditions are optimal, while weathering and greater efficiency; this increases the resource base by creating leaching processes have not yet reduced the initially favorable bi- positive feedback effects between plants and soil organisms. ological nutrient and clay mineral contents of the profile. In the The “first link” hypothesis (Lavelle, 1986) explains how the later highly weathered stages, concentrations of the biologically mutualistic interactions that develop between roots and other available fractions of the nutrient elements (notably P) available soil organisms are the first link in a suite of processes that in- to organisms have become severely diminished. In contrast, the crease biodiversity: a larger amount of energy is captured by improved physical conditions associated with soil profile deepen- plants growing with other soil organisms. The large number ing offer more stable microclimatic conditions, including a greater of studies demonstrating positive interactions among roots, mi- water storage capacity and a more stable thermal environment, croorganisms, and invertebrates in plant rhizospheres supports particularly at depth. Invertebrates and roots can penetrate deeper into this hypothesis (Whipps, 2001; Bonkowski, 2004; Barea et al., these soils to find suitable moisture regimes during dry conditions. 2005; Bais et al., 2006; Berendsen et al., 2012; Andriuzzi et al., Whereas earthworm and ant abundances occur in two peaks, 2015). This is expected to allow an increased number of primary at opposite ends of the soil maturation spectrum, termites are and secondary consumers to coexist, thereby increasing biodiver- clearly associated with the generally more weathered soils of trop- sity over evolutionary time. Evolution may therefore have favored ical areas (Orgiazzi et al., 2016; Fig. 2). Earthworms exhibit their cooperation, a basic component of self-organized systems, by greatest densities in young fertile soils where productivity is high. selecting for increased energy capture efficiency and ecosystem Their communities are usually dominated by some combination stability. Within this conceptual framework, competition and of litter-feeding epigeic and anecic populations, the latter of which predation may act as secondary regulators of soil functioning, live in vertical galleries and feed on surface leaf litter and soil. organizing community structure, and driving nutrient cycling In mature soils, trophic conditions are less favorable, although processes at small spatial and temporal scales but shaped by the oldest and deepest soils support different types of species, the larger scale activity of ecosystem engineers (van der Putten the so-called endogeics. These species live deep within the soil, et al., 2004; Mulder et al., 2011; de Vries et al. 2013). Selection thereby taking advantage of the more stable moisture condi- may operate at the group level, probably on groups selected by tions associated with the larger water store that may be held ecoevolutionary processes rather than through the addition of in- within such soils; they feed in situ on soil organic matter. Ants dividual processes. Alternatively, it may be considered as a reg- are much more populous in younger soils where greater re- ulation process that promotes homeostasis within the more sources are available. The slight increase that occurs in old general context of self-organization (Hoelzer et al., 2006). The soils may simply reflect the fact that these soils mostly occur self-organized nature of soils has been acknowledged at the broad in tropical areas where climatic conditions favor high produc- scales of soil formation (Targulian and Krasilnikov, 2007) and tivity, at least seasonally. functioning (Lavelle et al., 2006). Soil Functioning Soil Formation According to the definitions proposed by Perry (1995), soils Targulian and Krasilnikov (2007) described soil formation as may be considered as self-organized systems, with clear implica- a synergetic process of self-organization in time, leading toward tions for their properties and functioning (Lavelle et al., 2006). the formation of a mature pedomatrix that itself becomes a power- Self-organized ecological systems involve specific living commu- ful regulator of further soil system functioning. They also state nities that create structures, which in turn influence processes with that “soil formation should be regarded as the transformation of positive feedbacks on their fitness. These communities interact at parent material by biota with a consequent increase in its fertility discrete scales of time and space, constrained within boundaries. and ecological suitability, in accordance with the Gaia hypothesis” Theyproducestructuresthathaveapositivefeedbackthatallevi- (Lovelock, 1988; van Breemen, 1993). They recognized nine ates environmental constraints and operates through the influence different categories of soil-forming processes and three major of organisms on processes (Jouquet et al., 2006). For example, in time scales, defining short-term (10–100 years), medium-term the case of roots and their associated organisms in the rhizosphere,

− FIG. 2. Changes in mean population densities (individuals m 2) (SE in yellow) of the main invertebrate soil ecosystem engineers across a chronosequence of soil maturation sites from incipient (1), to young (2), mature (3) and old weathered (4) (Orgiazzi et al., 2016). A color version of this figure is available in the online version of this article. a functional domain with well-delimited boundaries extends a few complexes or separated from microorganisms by physical millimeters away from the root surface. A specific community of barriers (Lavelle et al., 1995; Kuzyakov and Blagodatskaya, microorganisms exploits carbon resources released as root exudates 2015). Earthworms in their drilospheres, termites in their and interacts with predatory microfauna at the root surface (rhizo- termitospheres and ants in their myrmecospheres are fur- plane). This community captures nutrients from the surrounding ther major ecosystem engineers that create self-organized sys- soil and releases them in mineral forms at the root surface tems, or functional domains, operating at characteristic scales (Bonkowski, 2004; Bais et al., 2006), thereby improving plant (Lavelle, 2002). nutrition. Physical structures are also created from the mineral Self-organized systems are organized hierarchically across soil around the root: stable aggregates and pores that improve scales of time and space (Fig. 3). They occur as structures located the circulation of air and water to and within deeper soil hori- within a suite of embedded scales extending from a few micro- zons, facilitating greater root penetration. meters where self-organized microbial communities operate Another important property of self-organized systems is (Young and Crawford, 2004, Crawford et al., 2012), to larger their occurrence in metastable equilibrium states, “on the edge units (100–250 μm) within which their predators function of chaos,” far from the equilibrium maintained by the activity (Lavelle and Spain, 2001; Briar et al., 2011). These, in turn, of the organisms that form the system. Clearly, as soon as a root form part of the functional domains of ecosystem engineers, dies, the rhizosphere changes its state from a living rhizosphere which operate across scales of centimeters to meters where they system to a decomposition system fueled by the dead organic form mosaics of functional units driven by a diverse set of eco- matter of the root. It later becomes a biologically inactive sys- system engineers (Ettema and Wardle, 2002; Rossi 2003). Such tem once the fresh organic matter of the root has been trans- mosaics are ecosystems that combine to form landscape mo- formed into decomposition-resistant compounds that become saics (Lavelle et al., 2006). Paradoxically, this perfect succes- strongly associated with clay minerals to form organomineral sion of units across time and spatial scales results in the

FIG. 3. Soil as a self-organized system of biological entities. Organisms of increasing sizes, from microorganisms to micro invertebrates (microfoodwebs), ecosystem engineers and ecosystems (horizontal axis) create structures of different sizes (vertical axis) and organize soil at different nested scales from 1 (microbial aggregates inhabited by microbial communities) to 5 (landscape viewed as a mosaic of different ecosystems). Most soil-based ecosystem services are perceived and delivered at the landscape (5) scale. A color version of this figure is available in the online version of this article. general scale-free pattern of self-organized systems indicated earthworms (Blanchart et al., 1999, Shuster et al., 2004), and other by Gisiger (2001). organisms. A range of methodological approaches have revealed the linkages that occur between soil macroaggregation and the presence of ecosystem engineers; in some cases, the organisms di- ENGINEERS AS KEY ACTORS IN rectly responsible for aggregate formation have also been identi- SELF-ORGANIZED SYSTEMS fied (Velasquez et al., 2012). Ecosystem engineers operate principally at Scales 3 and 4 of the five scales in the continuum of scales shown in Fig.3 that links the smallest Scale 1, within which microorganisms medi- Biogenic Structures ate most of the chemical transformations in soil, and Scales 4 (ecosystem) and 5 (landscape), where most ecosystem services Ecosystem engineers produce a wide variety of aggregated are delivered. They select and activate microbial activities through and massive structures that may accumulate at the surface and a combination of physical and trophic mechanisms, constrained within the soil, leading to the formation of surface horizons with within the limits imposed by climatic (moisture and temperature) a high proportion of biogenic aggregates and pores (Fig. 5). Mi- and soil (clay minerals, nutrients, organic resources) conditions cromorphological analysis of soil thin sections (Hallaire, 2000; (Fig. 4). The mutualistic relationship with microbes is diversely Topoliantz et al., 2000; Davidson et al., 2004; Pulleman and organized: it may be mediated through internal or external diges- Marinissen, 2004) and manual separation of rounded biogenic ag- tive systems and allows ecosystem engineers to capture energy gregates (Topoliantz et al., 2000; Velasquez et al., 2007a) have and to develop engineering activities across three different scales. permitted identification and characterization of different structures on the basis of their shapes, resistances to mechanical pres- sure, and colors. As revealed by chemical and near-infrared Physical Engineering: Construction, Spatial spectral analysis, these structures differ greatly in their organic Organization, and Maintenance of Stable matter contents and compositions (Decaëns et al., 2001; Hedde Aggregates and Pores of Different Shapes and et al., 2005). Their microbial communities are rather specific Degrees of Connectivity (Lavelle et al., 2005), as are the enzymatic activities that they ex- Numerous studies have detailed the physical impacts of soil hibit (Mora et al., 2005). Detailed analyses of such structures indi- ecosystem engineers. Their constructs range from the minute struc- cate many unique materials, including the 50-μm-thick exterior tures created by microorganisms, either acting alone (Crawford coating of fine materials that impart hydrophobic properties to et al., 2012) or in combination with roots (Feeney et al., 2006), to earthworm casts (Blanchart et al., 1993). Termites often build roots, alone (Angers and Caron, 1998) or in combination with mounds comprising different kinds of materials between the out- earthworms (Zangerle et al., 2011). Other engineering groups in- side walls and the internal nesting areas (Blanchart et al., 1993; clude the Enchytraeidae (Davidson and Grieve, 2006), ants Korb, 2011). The sophisticated drainage systems built around the (Hölldobler and Wilson, 1990, Lobry de Bruyn and Conacher, nests of the mound-building ant species Camponotus punctulatus 1990; Folgarait, 1998; Gorosito et al., 2006; Jouquet et al. 2011), (Gorosito et al., 2006) and the spectacular tubular structures made

FIG. 4. Interactions among self-organized units across scales and the delivery of ecosystem services. Note that additional scales such as biome and biosphere may be added when ecosystem services are considered at the global scale. Ecosystem engineers select and activate microbial communities in their functional domains (1). A mutualist digestion system (2) releases energy (3) used by ecosystem engineers to build biogenic structures and organize soil physical structure (4) with resulting effects on water services (infiltration and storage) (5). Combination of mineralization processes during the digestion process and sequestration of organic compounds in biogenic structures affects soil organic matter dynamics (6), and hence, nutrient cycling and climate regulation (7). A color version of this figure is available in the online version of this article. by cicada nymphs in neotropical rainforests (Fig. 5) are two of stabilities, which allow them to either stabilize the soil struc- many possible examples. ture (compact globular casts) or induce local erosion and soil Termite mounds are a widespread feature of most of the creep (small granular casts) (Blanchart et al., 1999; Decaëns, savannas of Africa, Latin America, and Australia. Depending 2000, Jouquet et al., 2013). The compositions, structural forms, on the feeding habits of their constructors, their biogenic struc- and the time periods during which these structures persist deter- tures exhibit a range of nutrient compositions with higher nutri- mine their local impacts on the processes that sustain soil ecosys- ent contents in the structures created by litter feeders (Decaëns tem services. While compact structures with high concentrations et al., 2001; Jimenez et al., 2006) than those built by wood- of microaggregates may conserve carbon in soils and, when stabi- feeding species (Lee and Wood, 1971). In addition, interactions lized over a drying cycle, allow high rates of water infiltration among termite saliva, ingested organic matter, and clay min- and storage in soils, loose and physically unstable structures erals can modify clay mineralogy (Jouquet et al., 2007). Earth- may accelerate carbon mineralization and soil erosion. Cases of worm casts have very diverse compositions and structural soil creep triggered by the surface deposition of earthworm casts (Nooren et al., 1995; Le Bayon and Binet, 2001, Jouquet et al., 2013) exemplify the dual functions of biogenic structures on carbon cycling; whether protection or accelerated mineralization occurs depends on the types of structure, the depths at which they are deposited, and the times over which they persist. Any imbalance between the opposing “compacting” and “decompacting” functions may cause severe degradation of soil physical properties and plant death (Chauvel et al., 1999; Bartz et al., 2009).

Communities of Engineers, Soil Physical Properties, and the Delivery of Environmental Services Various groups of soil biota, especially mycorrhizal fungi, bacteria, earthworms, and termites, have been shown to significantly influence soil physical properties (e.g., Tisdall, 1994; Bossuyt et al., 2005; Pulleman et al., 2004; Rillig and Mummey, 2006; Velasquez et al., 2007b). It is known that the excretion FIG. 5. Tube made by a cicada nymph in an Amazonian rainforest of, for example, polysaccharides by bacteria and the physical en- (photo: P. Lavelle). A color version of this figure is available in the meshment of soil particles by fungal mycelia has a positive effect online version of this article. on soil aggregation (Tisdall and Oades, 1982; Degens et al.,

1996; Chenu, 1995; Rillig and Mummey, 2006). Soil physical Similar relationships have been observed in a number of engineers progressively accumulate biogenic structures in soils, studies in agroecosystems that illustrate how macroinvertebrates, and their spatial arrangements determine the physical properties where favored by appropriate management options, produce of the soil and, ultimately, the delivery of ecosystem services. The abundant biogenic macroaggregates and macropores, with amount of bioturbation carried out by soil ecosystem engineers subsequent positive effects on ecosystem hydrological services may often be very significant pedologically (Wilkinson et al., (Blanchart et al., 1999; Hallaire et al., 2000; Velasquez et al., 2009). In 1 m2 of a favorable temperate soil environment, approxi- 2007b; Velasquez et al., 2012; Marichal et al., 2014). In Neotrop- mately 1 L of soil is contained within the gut of the earthworm pop- ical savannas, modifying the plant cover or artificially exclud- ulation. Four percent to 10% of the total soil passes annually ing an important ecosystem engineer may rapidly affect physical through earthworm guts, that is, the equivalent of several hundreds soil properties (Decaëns et al., 1999a; Velasquez et al., 2012). of Mg of dry soil (Drake and Horn, 2007). In the humid tropics, an- Changes in chemical fertility occur more slowly and are un- nual rates of soil transit through earthworms may be much greater likely to be noted in the limited time spans of the experiments. and reach 1,250 Mg ha−1, which is equivalent to a layer of dry Termite and ant mounds create heterogeneity in physico- soil 10 cm thick (Lavelle, 1975, 1978). Recent information, gained chemical soil properties (Lee and Wood, 1971; Folgarait, 1998; using novel methodologies (Zangerlé et al., 2016) in temperate cli- Decaëns et al., 2001) and play a critical role in influencing both mate forests, suggests that the annual transit through earthworm ecosystem CO2 exchange and biological production (Otieno guts might be much greater than indicated by Drake and Horn et al., 2011). Large macropores made by Odontotermes species (2007). Although less spectacular, and possibly less well known, and Macrotermes subhyalinus in the Sahel (Léonard and Perrier, termite and ant depositions and constructions frequently reach 2001) have been reported to have a noticeable impact on soil hy- values of tens to hundreds of Mg per hectare (Lavelle and Spain, drological properties. Conversely, termite sheetings (protective 2001). Termites may also consume as much as 55% of surface plant constructions at the soil surface that cover termite food materials litter (Wood and Sands, 1978) and account for up to 20% of carbon during feeding) created by Ruptitermes species in the Neotropical mineralized (Holt, 1987). Such high rates of transit through engi- savannas (Decaëns et al., 2001) can erode rapidly during the in- neers’ guts have significant effects and Velasquez et al. (2007b) tense rainfall typical of the tropics. Termites are also considered demonstrated a clear relationship between the abundances of bio- to be net emitters of methane (Sanderson, 1996), although they genic macroaggregates in different Nicaraguan production systems probably do not emit more than 2% to 4% of global emissions. and the populations of the macroinvertebrates that inhabit these In addition, local emissions depend on the relative proportions soils (Fig. 6). Accumulation of stable macroaggregates in turn influ- of the different trophic groups present in a particular area: fungus ences the general physical properties of the soil and its overall qual- growers and soil feeders release more methane than wood feeders ity as measured using the General Indicator of Soil Quality . (Brauman et al., 1992). Soil engineers may also have medium- to

FIG. 6. Correlation circle of indicator variables (upper left), and projection of sample sites in the factorial plan F1F2 of the PCA analysis of macrofauna and soil quality indicators in a Nicaraguan multiple use (cropping, pasture and forested) landscape (right). Close projection of variables “Morphology” that measures soil macroaggregation and “Macrofauna” indicates correlation between these two indicators. GISQ: Global Indicator of Soil Quality combines the morphology, macrofauna, SOM (soil organic matter), physical and chemical sub indicators. Coffee plantation under tree cover had the best soil quality, “eroded” garden, mixed, and maize culture have the worst ones (Velasquez et al., 2007b).

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. Lavelle et al. Soil Science • Volume 181, Number 3/4, March/April 2016 long-term effects on soil dynamics. While termites are known to colonization by associated communities of smaller organisms oc- prevent desertification processes in arid systems (Bonachela curs, and the soil processes that determine ecosystem service de- et al., 2015), Isopoda may concentrate nutrient resources in livery follow specific pathways. deserts, promoting the maintenance of a diverse vegetation (Yair and Rutin, 1981). More widely, earthworms, termites, and ants progressively bury stones and artifacts through ongoing surface Earthworms deposition of earth materials and trigger soil creep by promoting Analysis of the spatial distributions of earthworm popula- downhill translocation of their surface deposits (Jouquet et al., tions has shown that they occur in patches, usually some 10 to 2011; Bonachela et al., 2015; Traore et al., 2015). The presence 30 m in diameter (Fig. 8). This pattern has been observed in a wide of abandoned termite mounds in some savanna plots may be due range of ecosystems, including natural temperate and tropical to predation during raids by army ants, which are the main grasslands, forests, introduced pastures, and in cropping systems predators of Macrotermitine termites (Lepage and Darlington, (Rossi et al., 1996; Jimenez et al., 2001; Rossi, 2003). 1984). These ants roam over large areas causing high mortality In a gallery forest in the Eastern Llanos region of Colombia, rates and short turnover times among termite colonies of Jimenez et al. (2014) found that soil intrinsic variability explains Macrotermes bellicosus, irrespective of their size. However, 48% of the population spatial variation for litter feeding epigeics even after depletion of their populations, their large biogenic but has virtually no effect on soil feeding endogeics (1% variance structures persist in the area for long periods, up to hundreds explained). Although relationships have often been established of years. between earthworm patches and soil characteristics, the exact sense of the relationship seems to be rather variable. While some species, generally endogeics, produce large quantities Interactions Among Ecosystem Engineers in of stable macroaggregates and create soil pores that may di- Building Biogenic Structures and rectly influence soil physical parameters, other smaller species Functional Domains associated with the litter layer are more dependent on local con- Although theory predicts that self-organized organisms cre- ditions (Jimenez et al., 2012).The formation of these patches ate functional domains with clear boundaries, there is evidence through self-organization has been demonstrated in a modeling in nature that interactions among ecosystem engineers do exist exercise based on initially homogeneous soil conditions (Barot and may result in the building of mixed domains with sometimes et al., 2007). In this case, two earthworm populations with alter- rather indeterminate boundaries (Decaëns et al., 1999b). Fine nate spatial distributions may coexist and have either compacting roots often colonize fresh earthworm casts (Decaëns et al., 2002), or decompacting effects on the soil structure, depending on the termite, and ant mounds (Ohashi et al., 2007; Dauber et al., types of casts they produce: hard and compact or loose and fragile. 2008), whereas root foraging is known to be facilitated by earth- Patches exhibit different soil conditions: both bulk density and worm and termite galleries (Lu et al., 2013). There is evidence that root density are significantly lower in the patches inhabited by roots and earthworms can coconstruct stable macroaggregates, decompacting earthworms than those inhabited by compacting although some specific affinities occur among plant and earth- species (Rossi, 2003). worm species in this process (Zangerle et al., 2011). In season- The negative correlations that occur between the spatial ally flooded savannas in South America, ecosystem engineers distributions of soil ecosystem engineers of different species may combine their activities to create the rather spectacular struc- producing separate cast types have been often interpreted as a tures, several meters in diameter, known as “surrales.” In French result of competition (Jimenez et al., 2001; Decaëns et al., Guyana, roots, termites, earthworms, and ants cooperate to main- 2009; Jimenez et al., 2012). Alternatively, this distribution patterns tain the raised fields initially created by pre-Colombian farmers in may be considered as the consequence of a microsuccessional periodically flooded savanna areas (McKey et al., 2010, Renard process. Compacting species cannot reingest their compact et al., 2012) (Fig. 7). Publications on this topic are extremely few, cast material and feed rather on small soil aggregates, whose and much remains to be learned. maximum size is limited by their mouth size. These organic matter–rich microaggregates are casts of the decompacting species that feed on large compact casts and decomposing root Spatial Design: Functional Domains and materials (Blanchart et al., 1999; Rossi, 2003). If demonstrated Their Boundaries to be true, this process would lead to the replacement of patches Communities of ecosystem engineers are usually organized of a given group by its complementary type. Observations in mosaics of patches of similar spatial extent. Within each patch, made at 1- and 2-year intervals have shown significant changes

FIG. 7. Surrales mounds made by ecosystem engineers in flooded savannas of the Orinoco region in South America (left: areal view; right: close view). A color version of this figure is available in the online version of this article.

− FIG. 8. Contour maps of the distributions and abundances (individuals m 2) of six earthworm species in a grass-legume pasture plot in the Oriental Llanos of Colombia (Jimenez et al., 2001).

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. Lavelle et al. Soil Science • Volume 181, Number 3/4, March/April 2016 in spatial distributions as well as persisting patterns (Rossi, Biological Engineers 2003, Jimenez et al., 2006). A further hypothesis, known as A large part of the influences that ecosystem engineers the coexistence-aggregation model, states that spatial aggregation have on nutrient cycling and plant growth is due to the activi- of competitors on patchily distributed resources can facili- ties of microorganisms and other smaller organisms (fine roots tate species coexistence (Price, 1984). In this case, the diversity and invertebrates) selectively activated within their functional of soil conditions created during soil formation by parent mate- domains. While part of this activation results from bioturba- rial variability and topography may allow species to coexist. tion and other physical processes, chemical compounds that The patterns observed in the field seem likely to be a combina- we propose to call ecological mediators also facilitate a great tion of both processes. part of it.

Termites and Ants Ecological Mediators Termite mounds of a given species may be either regular Ecological mediators are defined as any substance produced to randomly distributed, indicating at the extremes interactions by an organism within its environment that has the capacity to di- among colonies in partitioning colony foraging space or, alter- rectly affect the activities of other organisms. We distinguish two natively, the absence of such a process. Regular distributions major classes of such mediators. Energy-rich mediators are mostly are attributed to intraspecific competition (Spain et al., 1986; involved in the activation and selection of microbial communities Grohmann et al., 2010, Bourguignon et al., 2011), whereas ran- through priming effects (Jenkinson, 1966), whereas signaling dom distributions are considered to indicate a lack of competition. molecules have hormone-like effects and can influence the ex- However, there is no clear indication that in the absence of sur- pression of plant genes. rounding colonies a given colony putatively unconstrained spa- tially by competition would have expanded more than it did in the observed situation. Similarly, no relationship has been estab- Energy-Rich Mediators lished between the population density of colonies per hectare Root exudates and other components of rhizodeposited ma- and distribution patterns to show a threshold value in density be- terials, together with earthworm mucus, are the best known medi- yond which the distribution pattern would shift from random to ators within this category. regular as a result of competition constraints. Interestingly, it has been observed that young colonies of M. bellicosus (Smeathman) Root Exudates have a clumped distribution, whereas mature colonies are regu- Roots deposit in their rhizosphere exudates secreted by larly distributed (Korb and Linsenmair 2001). There are also some different types of cells (Uren, 1993; Kuzyakov and Domanski, reported cases where the nature of the parent material has no ap- 2000; Lavelle and Spain, 2001; Bertin et al., 2003; Bais et al., parent influence on mound distributions (Mujinya et al. 2014). 2006). This includes mucigel that comprises some bacterial prod- Ants build a wide variety of mounds, of very diverse sizes. As ucts and sloughed cells from the root cap region. These represent a for termites, these mounds may exhibit regular to clumped distri- massive annual input of several Mg per hectare of highly energy- butions and may or may not be affected by local soil variation rich materials predominantly released close to growing root tips. (Cushman et al., 1988; Cerdá et al., 2002). There is evidence that the compositions of these exudates may be species specific (Marschner et al., 2004), possibly allowing Conservation of Carbon plants to use nutrients from different soil pools, as proposed by Soil engineers affect carbon cycling at three different scales, Lavelle (1975). These compounds are involved in priming effects: with contrasting effects (Lavelle and Spain, 2001). Earthworm di- provision of a limited amount of this readily assimilable resource gestion results in mineralization of 5% to 20% of organic matter allows those microorganisms capable of using them to further de- on average during the very short period (from <1 h to 1 day) of compose complex substrates (Fontaine et al., 2003; Bird et al., a gut transit. Fresh casts and arthropod fecal pellets retain a high 2011; Dijkstra et al., 2013; Haichar et al., 2014; Shahzad et al., mineralization activity, although it soon decreases as the microbial 2015). The microbial loop model states that bacteria are selected biomass and plant roots quickly absorb and reorganize the nutri- and stimulated when the tip of a growing fine root comes in con- ents released in available forms. Dry or aged biogenic structures tact with them; these bacteria use the energy of the exudates to ex- have much-decreased and sometimes no detectable remaining tract nutrients from decomposing substrates in the vicinity of the mineralization activities, because organic matter is trapped within root. Finally, micropredators (nematodes or protists) prey upon compact structures and generally has a relatively high proportion bacteria, thereby releasing plant-available nutrients (Coleman of resistant compounds. Some biogenic structures, such as the et al., 1984, Bonkowski, 2004). hard external walls of termite mounds, may persist in the environment for very long periods. The age of carbon comprised in the Earthworm Mucus lower, oldest part of abandoned mounds of Macrotermes falciger Earthworms produce polysaccharide compounds termed in Congo varied from 600 to 2,300 years depending on the size “mucus” that is secreted at the body surface and in the forepart of the mounds (Erens et al., 2015). In the latter instance, the rapid of the gut. Daily cutaneous mucus production by a common tem- release of CO2 during their formation is largely compensated for perate endogeic species was estimated at 0.2% to 0.5% of the car- by carbon sequestration during their stable phase. bon content of the worm (Scheu, 1991). Direct observation of Invertebrates may thus influence the course of decomposi- living earthworms of different species studied under different soil tion of a large proportion of total soil organic matter in the up- conditions suggests that this value may vary substantially, but data per horizons. In the Lamto savanna of the Ivory Coast, 20% to are still lacking. 33% of the carbon stock transits annually through earthworm Intestinal mucus is mainly composed of glycoprotein approx- guts (Lavelle, 1978). Soil carbon dynamics studies generally ig- imately 60 kd in molecular size mixed with amino acids and simple nore these effects and are therefore unable to evaluate the conse- sugars (Martin et al., 1987). Another important secondary compo- quences of the sometimes drastic changes in their communities nent is called “drilodefensin” (Liebeke et al., 2015) and is furan that occur under different soil management options. sulfonic acid that counteracts the inhibitory effects of polyphenols

FIG. 9. Proportions of intestinal mucus mixed with the soil ingested by different earthworm species: M l: Millsonia lamtoiana;Dtn:Dichogaster terrae-nigrae,Mg:Millsonia ghanensis,Ma:Millsonia anomala (now Reginaldia omodeoi), A c: Amynthas corticis;Ag:Amynthas gracilis;Pc: Pontoscolex corethrurus;Am:Allolobophora molleri;Ot:Octolasion tyrtaeum (Trigo et al., 1999). on earthworm gut enzymes. Mucus may comprise from 4% to cells. Microorganisms can produce the same molecules and may more than 30% of the dry matter content of the anterior gut also produce phytotoxins or molecules that are considered to acti- (Trigo et al., 1999) (Fig. 9). Interestingly, mucus contents are vate plant defense mechanisms. Recent research has shown that higher in geophagous rather than litter-feeding species and in plants and soil organisms produce the same set of signaling mole- the species that occur in temperate rather than tropical areas. cules (Ausubel, 2005). This trend sustains the hypothesis that the general priming ef- However, a great proportion of biogenic molecules (such fect triggered in the earthworm gut requires less labile substrate as humic acids, or chitin) may elicit adaptive responses. The bio- when ambient temperatures are high and the food substrate is chemical dialog between organisms and plants involves emission of relatively high quality. and receptor devices and has a high energy cost, which suggests that Intestinal mucus does not occur in the posterior half of the they have been selected for their positive effects on organism fitness. earthworm gut. Given the huge energy investment that its production represents, this observation seems logical. However, we do not yet know what the fate of this mucus is: whether it Community Engineering is reabsorbed by the worm and metabolized or recycled by A universal property of the functional domains created by some internal transport process. In any case, considering this soil ecosystem engineers is the selection of communities of mucus only as a lubricant is erroneous. In addition, no measurable dependent organisms. amounts of mucus remain in the casts because the very tight energy budget of earthworms, especially that of the endogeic species, Microbial Communities would not permit such an energy loss (Lavelle and Spain, 2001). Microbial communities are highly modified during the formation of biogenic structures. A general process observed in all Signaling Molecules and Hormones invertebrate gut contents and other fresh biogenic structures is Signaling molecules are characterized by their marked ef- the selection and activation of a limited part of the soil community fects on organism physiology, despite their occurrence at very and the progressive evolution toward a common basic “bulk soil” low concentrations in the environment (Zhuang et al., 2013). Mi- community as structures age (Blackwood and Paul, 2003). croorganisms and plants produce many signaling molecules that Marschner et al. (2004) state that amounts and compositions of play a key role in the establishment of mutualist or parasitic rela- root exudates are the key drivers for differences in community tionships (Bais et al., 2004; Bertin et al., 2003). Well-known ex- structure. The secretion of root exudates into soil activates micro- amples are the Nod factors produced by Rhizobia that allow the bial communities, with a greater effect on bacteria than on fungi establishment of the Rhizobium-Fabaceae nodules and the plant (Buyer et al., 2002). Plant rhizospheres may have species-specific hormones released by free-living soil microorganisms. microbial communities, although they generally contain a rather Signaling molecules emitted into the environment are used limited number of fast-growing species, and soil effects may some- by organisms to acquire information on the spatiotemporal dy- times be more important than that of the plant (Marschner et al., namics of their resources. In this regard, signaling molecules alter 2004; Edwards et al., 2015; Lange et al., 2015). Rather similar ob- resource bioavailability by increasing organisms’ resource use ef- servations have been made on microbial communities within the ficiency (Puga-Freitas and Blouin, 2015) via changes in gene ex- earthworm gut content and within casts. While earthworms may pression or through RNA modifications. directly digest part of the microbial biomass they ingest, their The main signaling molecules in plants are hormones, overall effect is to activate a selected community that is involved such as auxins, cytokinins, gibberellins, abscisic acid, ethyl- in the digestion of ingested organic matter (Barois and Lavelle, ene, jasmonic acid, and salicylic acid synthesized within plant 1986; Furlong et al., 2002; Amador and Gorres, 2007; Byzov

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. Lavelle et al. Soil Science • Volume 181, Number 3/4, March/April 2016 et al., 2007; Milleret et al., 2009; Aira et al., 2010). Aging casts metabolites (allelochemicals) may intervene in interspecific plant generally have much lower microbial activities, and their micro- competition processes (Rice, 1979; Bais et al., 2006). bial communities are closer to that of the bulk soil. Termites also Growth stimulation is effected via improvements in nutrient maintain rather specific microbial communities within their guts and availability, soil physical conditions, and the production of growth biogenic structures (Brune and Friedrich, 2000; Aanen et al., 2002; hormones (Brown et al., 1999; Castellanos Suarez et al. 2014; Jouquet et al., 2005). Puga-Freitas and Blouin, 2015). This may occur directly through bacterial stimulation or indirectly, as when earthworms stimulate Plant Communities the bacteria that secrete these products. Recent research shows that Plant communities are affected through selective effects the hormonal effects triggered by earthworms and other soil- on seed germination and modification of soil fertility mediated dwelling invertebrates might be more important than purely tro- by soil ecosystem engineers. Plant communities associated phic influences (Puga-Freitas et al., 2012). with termite and ant mounds may differ from the general communities not so associated (Spain and McIvor, 1988; Folgarait, Plant Protection 1998; Folgarait et al., 2002). Such mounds may host a specialized Signaling molecules emitted by all soil organisms can trigger vegetation community on their surfaces (Schutz et al., 2008; defense mechanisms. Rhizosphere barriers (Chave et al., 2014) Grohmann et al., 2010) or around their bases that contrasts with and the modification of gene expression by earthworms are exam- communities more distant (Spain and McIvor, 1988). The ant ples of the rather sophisticated mechanisms that have been se- C. punctulatus, for example, builds large mounds in abandoned lected for during the course of evolution. rice fields in Argentina and Southern Brazil that have a highly diverse community of plants growing on their surfaces (Folgarait et al., 2002). Earthworms selectively affect seed germination Rhizosphere Barriers thereby favoring the establishment of specific plant communities Roots have developed protection systems against pathogens (Decaëns et al., 2003; Holdsworth et al., 2007; Mitchell et al., in their rhizospheres using signaling to identify the nature 2007; Forey et al., 2011). In some cases, invasive earthworms and importance of threats (Fig. 10). They may then produce toxic may locally decrease plant diversity at the time of their arrival compounds in their rhizospheres (Berg et al., 2005) and/or en- (Holdsworth et al., 2007). There is also some indication that gineer the compositions of rhizosphere populations to favor the specific affinities might exist between earthworms and plants enemies of microbial pests (Ryan et al., 2009). Roots also mod- at the species level (Zangerle et al., 2011), which leads us to hypoth- ify physical and chemical conditions within their rhizospheres, esize that self-organization in this case may favor specific soil, and this modification can last beyond the normal root life span plant, invertebrate, and microorganism associations. (Hastings et al., 2007). Termite mounds may positively influence both biotic and This ensemble of biotic and abiotic processes has the poten- abiotic processes in ecosystems and create foraging hotspots for tial to control pathogens in properly engineered systems. Soil herbivores. However, the magnitudes and spatial extents of these supressiveness, the capacity of soils to naturally suppress the ac- influences are poorly known. Levick et al. (2010) estimated that tivities of specific pathogens, may result from the presence of a 20% of the total landscape of an African savanna was influenced specific antipathogenic species or be a general effect of microbial by the presence of termite mounds and that termite influences on herbivore browsing operated at scales much larger than the spatial extents of their mound-building activities. Species building mounds with a higher nutrient status may host specialized vegetation communities on their surfaces (Grohman et al., 2010) or on the erosional pediments formed around their bases (Spain and McIvor, 1988) that differ from those of the general landscape.

Invertebrate Communities Small invertebrates are strongly influenced by the profound modifications of the physical and trophic environments that occur within functional domains. While the mesofauna may find suitable below-ground habitats within earthworm galleries (Marinissen and Bok, 1988; Martin and Marinissen, 1993; Loranger et al., 1998), a wide diversity of inquiline invertebrates may colonize ant and termite mounds (Lavelle and Spain, 2001) or live below large earthworm casts (Decaëns et al., 1999b). Collembola are strongly and specifically attracted by earthworm cutaneous mucus (Salmon and Ponge, 2001; Salmon et al., 2005). Many invertebrates find suitable feeding resources and shelter from their predators within these biogenic structures. FIG. 10. Disease-suppression strategies used to harness rhizosphere Stimulation of Plant Growth interactions (Chave et al., 2014). Plant breeding, soil Self-organization in plant-soil-organism systems has se- amendments, plant-microbiome associations (i.e., arbuscular lected rather sophisticated processes through which mutually mycorrhizal fungi), crop rotations, and mixed cropping systems favor rhizosphere processes: biocontrol of soil borne pathogens beneficial relationships allow plants to grow better and their as- (in red) via indigenous (in green and blue) or introduced (in sociated organisms to obtain increased benefit. These arise yellow) root microorganisms and disease-suppressive exudation from the different products of plant photosynthesis, root exudates, (T in red). A color version of this figure is available in the online leaf, and root litter. Conversely, growth-inhibiting secondary plant version of this article.

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. Soil Science • Volume 181, Number 3/4, March/April 2016 Ecosystem Engineers in a Self-organized Soil diversity (Weller et al., 2002; Postma et al., 2008). Mycorrhizae cooperate (or not) in the construction of aggregates? Can are known to provide a measure of plant protection, and the we identify the functional traits that determine compatibility use of companion plants that have a high potential for growth or complementarity among species of ecosystem engineers stimulation may also be used to control pathogens (Pozo and and thus allow a more effective mimicking of natural sys- Azcon-Aguilar 2007; Chave et al., 2014). tems in agroecosystems? Practical interventions that would enhance rhizosphere barriers (iii) What are the roles of the different microbial communities in agricultural practice involve organic amendments to enhance within functional domains? A large number of studies have biodiversity, particularly the activities of pathogen antagonists. confirmed that microbial communities are activated and Other interventions include plant breeding to enhance the capacity selected by ecosystem engineers within their functional of plants to select for a specific microbial genotype, the direct ma- domains. Although they seem to differ from one group nipulation of plant-microbiome associations, crop rotation, and to another, their exact origins and functions are still not the use of mixed cropping systems. clearly understood. Do they represent a similar component of a bulk soil microbial community whose activity Earthworm Effects is temporarily enhanced within functional domains, as Rice plants may become tolerant to such nematodes as suggested by Blackwood and Paul (2003)? What is the Heterodera sacchari in the presence of earthworms (Blouin functional meaning of the differences observed among et al., 2005). The decreased expression of stress genes PLD and functional domains and what are the mechanisms that SOD observed and the activation of genes involved in immuno- allow selection of dissimilar communities within the logical defense probably indicate that the plant no longer transfers individual types of functional domains? Do ecosystem proteins from the leaves to repair the roots but instead increases engineers produce specific signaling molecules that se- photosynthesis and the production of new roots (Jana et al., 2010). lect for specific communities and metabolic functions? AthoroughanalysisoftheArabidopsis thaliana (Brassica- Studies show that only the small part of the broad mi- ceae) transcriptome in the presence of Aporrectodea caliginosa crobial community that responds to high energy inputs (Puga-Freitas et al., 2012) has confirmed the ability of earthworms is enhanced. Is the slowly responding community also to modify plant gene expression. Genes known to be stress induc- enhanced, or does it respond to different stimuli, oper- ible and dependent on ethylene or jasmonic acid signaling are ac- ating over longer time scales that current experiments tually modulated in the presence of earthworms. Several of these and techniques have permitted us to observe? genes, such as PDF1.2 and PR-4, are known to be modulated dur- (iv) What interactions occur between self-organized systems ing the elicitation of induced systemic resistance, a vaccination- operating at disparate scales? What happens when an like process, by bacterial elicitors (van Loon et al., 2008). ecosystem engineer, for example, an earthworm (Scale 3), ingests the diverse components of the soil foodweb when feeding on soil (Scale 2)? How much does this CONCLUDING REMARKS: DEVELOPING happen in nature? Are such components as protists and THE CONCEPT nematodes just digested, or are specific groups promoted, Despite the important evidence that self-organization is a such as plant pathogens with further consequences for mi- key factor in explaining the organization and biological func- crobial control in the plant rhizosphere? Are the arbuscular tioning of soils, many questions need to be answered to validate mycorrhizal spores and other propagules that retain viabil- this approach. Nevertheless, acknowledging the self-organized ity when passed through the earthworm gut (Reddell and nature of soils allows a large number of novel questions to be Spain, 1991) solely transported to new locations in the soil, raised. The following seven general questions are of great im- or is spore viability stimulated or inhibited, with further portance in developing the model and obtaining a more holistic implications for rhizosphere dynamics? Several studies understanding of soil systems. of earthworm-nematode relationships have indicated a (i) What physical evidence exists for self-organization in soil? diversity of outcomes: while earthworms may either di- The self-organization theory predicts that self-organized gest nematodes or leave them apparently unaffected, subsystems concentrate their activities at a selected set gut transit may have adverse effects on the future sur- of discrete scales (Perry, 1995), which supposes a discon- vival of nematodes or of their progeny (Boyer et al., tinuous distribution of pore sizes, as suggested in Fig. 3. 2013). The fate of protists following transit through an Can we see in the distribution of the void space inhabited earthworm gut has seldom been studied. Cai et al. by organisms a track of the discrete distribution type pre- (2002) observed that cysts transited unharmed through dicted by the self-organization concept? There are cur- the earthworm gut, but do not form vegetative cells rently no measurements of pore-size distribution from within casts. In contrast, Rouelle (1983) maintained that the smallest ones accessible to bacteria (> 0.3 μm) to the protists were an important food source for earthworms. largest ones many centimeters in diameter and created (v) What are ecological mediators and how do they act? The im- by vertebrates. A few incomplete data sets currently exist portance of these chemical compounds has been revealed in (Grimaldi et al., 1993; Menendez et al., 2005), but they a number of studies. It is now important to analyze and com- cover only limited size ranges, compatible with the methods pare their structures, chemical compositions, and functional used. Lu et al. (2014) analyzed pore size distributions be- roles. What levels of specificity do they achieve, and what tween 0.001 and 100 μm and found three different peaks are the respective roles of energy-rich and signaling mole- of pores abundance in the ranges: 0.01 to 0.05, 0.1 to 2, cules in plant and other soil organism responses? The recent and70to100μm, respectively. discovery of “drilodefensins’ in earthworms suggests that (ii) What are the assembly rules for creating the mosaics of eco- similar molecules may act as ecological mediators and that system functional domains by engineers? Can we identify their discovery is likely to better determine some as yet un- patterns in their spatial distributions that favor such soil func- explained effects of the ecosystem engineers. tional processes as water infiltration or retention? What are (vi) How is carbon sequestered within the functional domains the mechanisms that lead certain plants and invertebrates to of ecosystem engineers? The issue of carbon balance in

the functional domains of ecosystem engineers may be of articles since the study of Bardgett et al. (1998). Part of this overemphasized, as when invasive species of ecosystem approach acknowledges the role of below-ground processes engineers enter natural ecosystems (Costello and on plant successional processes. Another important question Lamberti, 2009; Resner et al., 2015). Contrastingly, is the respective responses of below- and above-ground com- they may be totally ignored, as in present and next- ponents to changes occurring in the other half (van der Putten generation simulation models (Schmidt et al., 2011). Al- et al., 2009). though microorganisms are responsible for 95% to 99% of decomposition processes, ecosystem engineers are REFERENCES key promoters of soil larger-scale aggregation. A great part of the aggregates that they produce become microsites Aira M., C. Lazcano, M. Gomez-Brandon, and J. Dominguez. 2010. Ageing where carbon is conserved because microbial activity rap- effects of casts of Aporrectodea caliginosa on soil microbial commu- – idly diminishes as aggregates age (Martin, 1991). A clue nity structure and activity. App. Soil Ecol. 46:143 146. to understanding the capacity of the aggregated soil frac- AlleeW.C.,A.E.Emerson,O.Park,andK.P.Schmidt.1949.Principlesof tion to conserve carbon lies in determining the origin of animal ecology. W. B. Saunders, Philadelphia, PA. the individual aggregates (i.e., which ecosystem engineers Amador J. A., and J. H. Gorres. 2007. Microbiological characterization of the or physical process produced them) and their residence times structures built by earthworms and ants in an agricultural field. Soil Biol. in soil. The recently developed technique of near-infrared Biochem. 39:2070–2077. spectroscopy allows determination of the origins of indi- Andriuzzi W. S., M. M. Pulleman, O. Schmidt, J. H. Faber, and L. Brussaard. vidual aggregates located within a particular soil matrix. 2015. Anecic earthworms (Lumbricus terrestris) alleviate negative effects Aggregates produced by roots and earthworms can be of extreme rainfall events on soil and plants in field mesocosms. PLSO. identified and even separated by their ages and by the 397:103–113. species that produced them (Zangerle et al., 2011; Huerta Angers D. A., and J. Caron. 1998. Plant-induced changes in soil structure: et al., 2013; Zangerle et al., 2014). This powerful technol- Processes and feedbacks. Biogeochemistry. 42:55–72. ogy combined with specific microbial signatures should permit accurate evaluation of the contributions of ecosys- Aanen D. K., P. Eggleton, C. Rouland-Lefevre, T. Guldberg-Froslev, S. Rosendahl, and J. J. Boomsma. 2002. The evolution of fungus- tem engineers to carbon sequestration via the macroaggre- growing termites and their mutualistic fungal symbionts. Proc. Natl. gated structures that they form. Can we understand and Acad. Sci. U. S. A. 99: 14887–14892. model such processes as their turnover within a mosaic of functional domains of diverse ecosystem engineers? Ausubel F. M. 2005. Are innate immune signaling pathways in plants and ani- – (vii) What roles do ecosystem engineers play in plant protec- mals conserved? Nat. Immunol. 6:973 979. tion? What are the respective roles of the signaling mol- Bais H. P., S. W.Park, T. L. Weir, R. M. Callaway, and J. M. Vivanco. 2004. How ecules produced by different ecosystem engineers and plants communicate using the underground information superhighway. the mechanisms that allow plant protection or pest and Trends Plant Sci. 9:26–32. disease development? To what degree could we manipu- Bais H. P., T. L. Weir, L. G. Perry, S. Gilroy, and J. M. Vivanco. 2006. The role of late plant gene expression in managing the activities of root exudates in rhizosphere interations with plants and other organisms. ecosystem engineers? Annu.Rev.PlantBiol.57:233–266. Bancroft W.J., M. J. Garkaklis, and J. D. Roberts. 2005. Burrow building in sea- These are only a few of the research questions that have bird colonies: A soil-forming process in island ecosystems. Pedobiologia. arisen when considering the conceptual model proposed in this 49:149–165. review. The use of new approaches and benefits from novel mo- Bardgett R. D., D. A. Wardle, and G. W. Yeates. 1998. Linking above-ground lecular and spectral technologies will likely help in progressing and below-ground interactions: How plant responses to foliar herbivory development of a more refined and holistic view of soil function- influence soil organisms. Soil Biol. Biochem. 30:1867–1878. ing. Considering ecosystem functioning at different scales of time Barea J. M., M. J. Pozo, R. Azcon, and C. Azcon-Aguilar. 2005. Microbial co- and space, successional processes across all scales and the interac- operation in the rhizosphere. J. Exp. Bot. 56:1761–1778. tions among social, economic, and environmental parameters will provide the necessary template for developing truly sustainable Barois I., and P. Lavelle. 1986. Changes in respiration rate and some phys- soil management practices. icochemical properties of a tropical soil during transit through Another two, already thoroughly explored questions, would Pontoscolex-Corethrurus (Glossoscolecidae, Oligochaeta). Soil Biol. Biochem. 18:539–541. also contribute to developing the soil self-organization model. Stoichiometric and allometric constraints and above ground- Barot S., J. P. Rossi, and P. Lavelle. 2007. Self-organization in a simple below ground relationships are claimed to be of great importance consumer-resource system, the example of earthworms. Soil Biol. Biochem. – in determining both biodiversity and biomass densities of soil or- 39:2230 2240. ganisms (Mulder et al., 2006; Ott et al., 2014). The nutrient compo- Bartz M. L., G. G. Brown, A. Pasini, A. C. R. Lima, and D. N. Gasen. 2009. As sitions of soil organic resources are generally highly imbalanced minhocas e o manejo do solo: O caso do plantio direto e do arroz irrigado. with extremely high carbon-nutrient ratios compared with those Revist. Plant. Dir, Passo Fundo, pp. 4–8. of organisms. This leads to an important elimination of carbon, Berendsen R. L., C. M. J. Pieterse, and P. A. Bakker. 2012. The rhizosphere through respiration, secretion, and the excretion of labile carbon microbiome and plant health. Trends Plant Sci. 17:478–486. substrates in form of earthworm mucus, for example, and slow re- Berg G., C. Zachow, J. Lottmann, M. Gotz, R. Costa, and K. Smalla. 2005. lease of nutrients during the microbial immobilization process Impact of plant species and site on rhizosphere-associated fungi antagonis- (Lavelle and Spain, 2001). How do ecosystem engineers manage tic to Verticillium dahliae kleb. Appl. Environ. Microbiol. 71:4203–4213. to overcome this limitation and release nutrients in an available Berg G., and K. Smalla. 2009. Plant species and soil type cooperatively shape mineral form in an environment that is globally deficient in the structure and function of microbial communities in the rhizosphere. this nutrient? FEMS Microbiol. Ecol. 68:1–13. The relationships among below-ground and above- Bertin C., X. H. Yang, and L. A. Weston. 2003. The role of root exudates and ground processes have been emphasized in a large number allelochemicals in the rhizosphere. PLSO. 256:67–83.

Bird J. A., D. J. Herman, and M. K. Firestone. 2011. Rhizosphere priming of soil Castellanos Suarez D. E., A. Gigon, R. Puga-Freitas, P. Lavelle, E. Velasquez, organic matter by bacterial groups in a grassland soil. Soil Biol. Biochem. and M. Blouin. 2014. Combined effects of earthworms and IAA- 43:718–725. producing rhizobacteria on plant growth and development. App. Soil Ecol. Blackwood C. B., and E. A. Paul. 2003. Eubacterial community structure and 80:100–107. population size within the soil light fraction, rhizosphere and heavy fraction Cerdá X., A. Dahbi, and J. Retana. 2002. Spatial patterns, temporal variability, of several agricultural systems. Soil Biol. Biochem. 35:1245–1255. and the role of multi-nest colonies in a monogynous Spanish desert ant. Blanchart E., A. Bruand, and P.Lavelle. 1993. The physical structure of casts of Ecol. Entom. 27:7–15. Millsonia anomala (Oligochaeta, Megascolecidae) in shrub Savanna soils Chave M., M. Tchamitchian, and H. Ozier-Lafontaine. 2014. Agroecological En- (Cote-Divoire). Geoderma. 56:119–132. gineering to Biocontrol Soil Pests for Crop Health. In: Ozier-Lafontaine H., Blanchart E., A. Albrecht, J. Alegre, A. Duboisset, C. Gilot, B. Pashanasi, P. and M. Lesueur-Jannoyer (eds.). Sust. Agri. Rev. Springer Int. 269–297. Pub. Lavelle, and L. Brussaard. 1999. Effects of earthworms on soil structure Chauvel A., M. Grimaldi, E. Barros, E. Blanchart, T. Desjardins, M. Sarrazin, and physical properties in P. Lavelle L., P. Brussaard, and F. Hendrix and P.Lavelle. 1999. Pasture damage by an Amazonian earthworm. Nature. (eds). Earthworms Management in Tropical Agro-Ecosystems. CAB- 398:32–33. I, Wallingford. 149–172. Chenu C. 1995. Extracellular polysaccharides: An interface between microorgan- Blouin M., Y.Zuily-Fodil, A. T. Pham-Thi, D. Laffray, G. Reversat, A. Pando, J. isms and soil constituents. Env. Imp. Soil Comp. Int. 1:217–233. Tondoh, and P. Lavelle. 2005. Belowground organism activities affect plant aboveground phenotype, inducing plant tolerance to parasites. Coleman D. C., R. E. Ingham, J. F. McClellan, and J. A. Trofymow 1984. Ecol. Lett. 8:202–208. Soil nutrient transformations in the rhizosphere via animal-microbial interactions. In: Invertebrates-Microbial Interactions. Anderson J. M., Blouin M., M. E. Hodson, E. A. Delgado, G. Baker, L. Brussaard, K. R. Butt, J. A. D. M. Rayner, and D. W. H. Walton (eds.). Cambridge University Dai, L. Dendooven, G. Peres, J. E. Tondoh, D. Cluzeau, and J. J. Brun. 2013. Press, Cambridge, pp. 35–58. A review of earthworm impact on soil function and ecosystem services. Eur. J. Soil Sci. 64:161–182. Costello D. M., and G. A. Lamberti. 2009. Biological and physical effects of non-native earthworms on nitrogen cycling in riparian soils. Soil Biol. Bonachela J. A., R. M. Pringle, E. Sheffer, T. C. Coverdale, J. A. Guyton, K. K. Biochem. 41:2230–2235. Caylor, S. A. Levin, and C. E. Tarnita. 2015. Termite mounds can increase the robustness of dryland ecosystems to climatic change. Sci- CrawfordJ.W.,L.Deacon,D.Grinev,J.Harris,K.Ritz,B.Singh,andI. ence. 347:651–655. Young. 2012. Microbial diversity affects self-organization of the soil- microbe system with consequences for function. J. R. Soc. Interface. Bonkowski M. 2004. Protozoa and plant growth: The microbial loop in soil 9:1302–1310. revisited. New Phytol. 162:617–631. Cushman J. H., G. D. Martinsen, and A. I. Mazeroll. 1988. Density dependent Bossuyt H., J. Six, and P. F. Hendrix. 2005. Protection of soil carbon by and size dependent spacing of ant nests. Evidence for intraspecific compe- microaggregates within earthworm casts. Soil Biol. Biochem. 37:251–258. tition. Oecologia. 77:522–525. Bottinelli N., P. Jouquet, Y. Capowiez, P. Podwojewski, M. Grimaldi, and X. Darwin C. 1881. The Formation of Vegetable Mould Through the Action of Peng. 2015. Why is the influence of soil macrofauna on soil structure only Worms With Observations on Their Habits. Murray, London. considered by soil ecologists? Soil Till Res. 146:118–124. Dauber J., R. Niechoj, H. Baltruschat, and V. Wolters. 2008. Soil engineering Bouché M. B. 1977. Stratégies lombriciennes, soil organism as components of ants increase grass root arbuscular mycorrhizal colonization. Biol. Fertil. ecosystems. Ecol. Bull. (Stockolm), pp. 122–132. Soils. 44:791–796. Bourguignon T., M. Leponce, and Y. Roisin. 2011. Are the spatio-temporal dy- Davidson D. A., P.M. C. Bruneau, I. C. Grieve, and C. A. Wilson. 2004. Micro- namics of soil-feeding termite colonies shaped by intra-specific competi- morphological assessment of the effect of liming on faunal excrement in an – tion? Ecol Entom. 36:776 785. upland grassland soil. App. Soil Ecol. 26:169–177. Boyer J., G. Reversat, P. Lavelle, and A. Chabanne. 2013. Interactions between Davidson D. A., and I. C. Grieve. 2006. The influence of soil fauna on soil struc- – earthworms, and plant-parasitic nematodes. Eur. J. Soil Biol. 59:43 47. tural attributes under a limed and untreated upland grassland. Land Degr. Brauman A., M. D. Kane, M. Labat, and J. A. Breznak. 1992. Genesis of acetate Dev. 17:393–400. and methane by gut bacteria of nutritionally diverse termites. Science. Decaëns T., J. J. Jimenez, and P.Lavelle. 1999a. Effect of exclusion of the anecic – 257:1384 1387. earthworm Martiodrilus carimaguensis Jimenez and Moreno on soil prop- Brune A., and M. Friedrich. 2000. Microecology of the termite gut: Structure erties and plant growth in grasslands of the eastern plains of Colombia. and function on a microscale. Curr. Opin. Microbiol. 3:263–269. Pedobiologia. 43:835–841. Butler D. R. 1995. Zoogeomorphology. Cambridge University Press, New York. Decaëns T., L. Mariani, and P. Lavelle. 1999b. Soil surface macrofaunal com- Brown G. G., B. Pashanasi, C. Villenave, J. C. Patron, B. K. Senapati, S. Giri, I. munities associated with earthworm casts in grasslands of the Eastern Plains – Barois,P.Lavelle,E.Blanchart,R.J.Blakemore,A.V.Spain,andJ.Boyer. of Colombia. App. Soil Ecol. 13:87 100. 1999. Effects of earthworms on plana production in the tropics. In: Earth- Decaëns T. 2000. Degradation dynamics of surface earthworm casts in grass- worm Management in Tropical Agroecosystems. Lavelle P., L. Brussaard, lands of the eastern plains of Colombia. Biol. Fertil. Soils. 32:149–156. – and P. Hendrix (eds.). CAB International Edition, Wallingford, UK. 87 147. Decaëns T., J. H. Galvis, and E. Amezquita. 2001. Properties of the structures Briar S. S., S. J. Fonte, I. Park, J. Six, K. Scow, and H. Ferris. 2011. The distri- created by ecological engineers at the soil surface of a Colombian savanna. bution of nematodes and soil microbial communities across soil aggregate C. R. Acad. Sci. III. 324:465–478. fractions and farm management systems. Soil Biol. Biochem. 43:905–914. Decaëns T., N. Asakawa, J. H. Galvis, R. J. Thomas, and E. Amezquita. 2002. Buyer J. S., D. P.Roberts, and E. Russek-Cohen. 2002. Soil and plant effects on Surface activity of soil ecosystem engineers and soil structure in contrasted microbial community structure. Can. J. Microbiol. 48:955–964. land use systems of Colombia. Eur. J. Soil Biol. 38:267–271. Byzov B. A., N. V.Khomyakov, S. A. Kharin, and A. V.Kurakov. 2007. Fate of Decaëns T., L. Mariani, N. Betancourt, and J. J. Jimenez. 2003. Seed dispersion soil bacteria and fungi in the gut of earthworms. Eur. J. Soil Biol. 43: by surface casting activities of earthworms in Colombian grasslands. Acta S149–S156. Oecol. Int. J. Ecol. 24:175–185. Cai H. J., B. G. Zarda, R. Mattison, F. Schonholzer, and D. Hahn. 2002. Fate of Decaëns T., J. J. Jimenez, and J. P. Rossi. 2009. A null-model analysis of the Protozoa transiting the digestive tract of the earthworm Lumbricus terrestris spatio-temporal distribution of earthworm species assemblages in Colom- L. Pedobiologia. 46:161–175. bian grasslands. J. Trop. Ecol. 25:415–427.

Degens B. P., G. P. Sparling, and L. K. Abbott. 1996. Increasing the length of Hastings A., J. E. Byers, J. A. Crooks, K. Cuddington, C. G. Jones, J. G. hyphae in a sandy soil increases the amount of water-stable aggregates. Lambrinos, T. S. Talley, and W. G. Wilson. 2007. Ecosystem engineering App. Soil Ecol. 3:149–159. in space and time. Ecol. Lett. 10:153–164. Dijkstra F. A., Y. Carrillo, E. Pendall, and J. A. Morgan. 2013. Rhizosphere Hedde M., J. Joffre, P.Lavelle, J. J. Jiménez, and T. Decaëns. 2005. Specific functional priming: A nutrient perspective. Front. Microbiol. 4:216. signature in soil macro-invertebrates bio-structures. Funct. Ecol. 19:785–793. Dokuchaev V.V.1889. The zones of Russia (in Russian). Akad, Nauk Moscow. Hiltner R. 1904. Uber neuere Erfahrungen und Probleme auf dem Gebiet DrakeH.L.,andM.A.Horn.2007.Asthewormturns:Theearthwormgut der Bodenbakteriologie und unter besonderer Berücksichtigung der – as a transient habitat for soil microbial biomes. Annu. Rev. Microbiol. Gründüngung und Brache. Arb. Deut. Landw. Ges. 98:59 78. 61:169–189. Hinsinger P.,A. G. Bengough, D. Vetterlein, and I. M. Young. 2009. Rhizosphere: Biophysics, biogeochemistry and ecological relevance. PLSO. 321:117–152. Duchaufour P. 1977. Pedology: Pedogenesis and Classification. George Allen, Unwin, London. Hoelzer G. A., E. Smith, and J. W.Pepper. 2006. On the logical relationship between natural selection and self-organization. J. Evol. Biol. 19:1785–1794. Dunham A. E. 2011. Soil disturbance by vertebrates alters seed predation, movement and germination in an African rain forest. J. Trop. Ecol. Holdsworth A. R., L. E. Frelich, and P. B. Reich. 2007. Effects of earth- 27:581–589. worm invasion on plant species richness in northern hardwood forests. Conserv. Biol. 21:997–1008. Edwards J., C. Johnson, C. Santos-Medellin, E. Lurie, N. K. Podishetty, S. Bhatnagar, J. A. Eisen, and V. Sundaresan. 2015. Structure, variation, and Holling C. S. 2001. Understanding the complexity of economic, ecological and assembly of the root-associated microbiomes of rice. Proc. Natl. Acad. social systems. Ecosystems. 4:390–405. Sci. U. S. A. 112:E911–E920. Hölldobler B., and E. O. Wilson. 1990. The Ants. Harvard University Press, ErensH.,M.Boudin,F.Mees,B.B.Mujinya,G.Baert,M.vanStrydonck, Boston, MA. P. Boeckx, and E. van Ranst. 2015. The age of large termite mounds- Holt J. A. 1987. Carbon Mineralization in semi arid North Eastern Australia. radiocarbon dating of Macrotermes falciger mounds of the Miombo The role of termites. J. Trop. Ecol. 3:255–263. woodland of Katanga, DR Congo. Palaeog. Palaeoclim. Palaeoeco. Huerta E., D. Brunet, E. Velazquez, and P. Lavelle. 2013. Identifying earth- 435:265–271. worm’s organic matter signatures by near infrared spectroscopy in different Ettema C. H., and D. A. Wardle. 2002. Spatial soil ecology. Tr. Ecol. Evol. land-use systems in Tabasco, Mexico. App. Soil Ecol. 69:49–55. – 17:177 183. James A. I., D. J. Eldridge, and B. M. Hill. 2009. Foraging animals create fertile Feeney D. S., J. W. Crawford, T. Daniell, P. D. Hallett, N. Nunan, K. Ritz, M. patches in an Australian desert shrubland. Ecography. 32:723–732. Rivers, and I. M. Young. 2006. Three-dimensional microorganization of Jana U., S. Barot, M. Blouin, P. Lavelle, D. Laffray, and A. Repellin. 2010. the soil-root-microbe system. Microb. Ecol. 52:151–158. Earthworms influence the production of above- and belowground biomass Folgarait P.J. 1998. Ant biodiversity and its relationship to ecosystem function- and the expression of genes involved in cell proliferation and stress re- ing: A review. Biodiv. Cons. 7:1221–1244. sponses in Arabidopsis thaliana. Soil Biol. Biochem. 42:244–252. Folgarait P. J., S. Perelman, N. Gorosito, R. Pizzio, and J. Fernandez. 2002. Jenkinson D. S. 1966. The priming action. J. Appl. Radiat. Isotopes Suppl, Effects of Camponotus punctulatus ants on plant community composi- 199–208. tion and soil properties across land-use histories. Pl. Ecol. 163:1–13. Jenny H. 1941. Factors of Soil Formation—A System of Quantitative Pedology. Fontaine S., A. Mariotti, and L. Abbadie. 2003. The priming effect of or- McGraw-Hill, New York. ganic matter: A question of microbial competition? Soil Biol. Biochem. Jimenez J. J., J. P. Rossi, and P. Lavelle. 2001. Spatial distribution of earthworms in 35:837–843. acid-soil savannas of the eastern plains of Colombia. App. Soil Ecol. 17:267–278. Forey E., S. Barot, T. Decaëns, E. Langlois, K. R. Laossi, P. Margerie, S. Jimenez J. J., T. Decaëns, and J. P. Rossi. 2006. Stability of the spatio-temporal Scheud, and N. Eisenhauer. 2011. Importance of earthworm-seed interactions for the composition and structure of plant communities: A review. distribution and niche overlap in neotropical earthworm assemblages. Acta – Acta Oecol. Int. J. Ecol. 37:594–603. Oecol. Int. J. Ecol. 30:299 311. Furlong M. A., D. R. Singleton, D. C. Coleman, and W.B. Whitman. 2002. Mo- Jimenez J. J., T. Decaëns, and J. P.Rossi. 2012. Soil environmental heterogene- lecular and culture-based analyses of prokaryotic communities from an ag- ity allows spatial co-occurrence of competitor earthworm species in a gal- ‘ ’ – ricultural soil and the burrows and casts of the earthworm Lumbricus lery forest of the Colombian Llanos . Oikos. 121:915 926. rubellus. Appl. Environ. Microbiol. 68:1265–1279. Jimenez J.J., T. Decaens, P.Lavelle, and J.-P.Rossi. 2014. Dissecting the multi- Gisiger T. 2001. Scale invariance in biology: Coincidence or footprint of a uni- scale spatial relationship of earthworm assemblages with soil environmental versal mechanism? Biol. Rev. 76:161–209. variability. BMC. Ecol. 14:26. Gorosito N. B., P. Curmi, V. Hallaire, P. J. Folgarait, and P. M. Lavelle. 2006. Johnson D. L., and F. D. Hole. 1994. Soil formation theory: A summary of its Morphological changes in Camponotus punctulatus (Mayr) anthills of dif- principal impacts on geography, geomorphology, soil-geomorphology, qua- ferent ages. Geoderma. 132:249–260. ternary geology and paleopedology. In: Factors of Soil Formation: A Fiftieth – Grimaldi M., M. Sarrazin, A. Chauvel, F. Luizao, N. Nunes, M. R. Lobato Anniversary Retrospective. SSSA, Madison, WI, pp. 111 126. Rodriguez, P. Amblard, and D. Tessier. 1993. Effets de la déforestation Jones C. G., J. H. Lawton, and M. Shachak. 1994. Organisms as ecosystem et des cultures sur la structure des sols argileux d’Amazonie brésilienne. engineers. Oikos. 69:373–386. Cah Agri. 2:36–47. Jouquet P., L. Ranjard, M. Lepage, and J. C. Lata. 2005. Incidence of fungus- Grohmann C., J. Oldeland, D. Stoyan, and K. E. Linsenmair. 2010. Multi- growing termites (Isoptera, Macrotermitinae) on the structure of soil scale pattern analysis of a mound-building termite species. Ins. Soc. microbial communities. Soil Biol. Biochem. 37:1852–1859. 57:477–486. Jouquet P.,J. Dauber, J. Lagerlof, P.Lavelle, and M. Lepage. 2006. Soil inverte- Haichar F. Z., C. Santaella, T. Heulin, and W. Achouak. 2014. Root exudates brates as ecosystem engineers: Intended and accidental effects on soil and mediated interactions belowground. Soil. Biol. Biochem. 77:69–80. feedback loops. App. Soil Ecol. 32:153–164. Haken H. 2008. Self organization. Scholarpedia. 3:1401. Jouquet P., N. Bottinelli, J. C. Lata, P. Mora, and S. Caquineau. 2007. Role Hallaire V.,P. Curmi, A. Duboisset, P. Lavelle, and B. Pashanasi. 2000. Soil struc- of the fungus-growing termite Pseudacanthotermes spiniger (Isoptera, ture changes induced by the tropical earthworm Pontoscolex corethrurus and Macrotermitinae) in the dynamic of clay and soil organic matter con- organic inputs in a Peruvian ultisol. Eur. J. Soil Biol. 36:35–44. tent. An experimental analysis. Geoderma. 139:127–133.

Jouquet P.,S. Traore, C. Choosai, C. Hartmann, and D. Bignell. 2011. Influence Liebeke M., N. Strittmatter, S. Fearn, A. J. Morgan, P.Kille, J. Fuchser, D. Wal- of termites on ecosystem functioning. Ecosystem services provided by ter- lis, V. Palchykov, J. Robertson, E. Lahive, D. J. Spurgeon, D. McPhail, Z. mites. Eur. J. Soil Biol. 47:215–222. Takats, and J. G. Bundy. 2015. Unique metabolites protect earthworms – Jouquet P., N. Bottinelli, G. Kernels, T. Henry-des-Tureaux, T. D. Thuy, O. against plant polyphenols. Nat. Com. 6:7869 7869. Planchon, and D. T. Toan. 2013. Surface casting of the tropical Metaphire Lobry de Bruyn L. A., and A. J. Conacher. 1990. The role of termites and ants in posthuma increases soil erosion and nitrate leaching in a laboratory experi- soil modification: A review. Aust J Soil Res. 28:55–93. – ment. Geoderma. 204:10 14. Loranger G., J. F. Ponge, E. Blanchart, and P. Lavelle. 1998. Impact of earth- Korb J., and K. E. Linsenmair. 2001. The causes of spatial patterning of mounds worms on the diversity of microarthropods in a vertisol (Martinique). Biol. of a fungus-cultivating termite: Results from nearest-neighbour analy- Fertil. Soils. 27:21–26. – sis and ecological studies. Oecologia. 127:324 333. Lovelock J. 1988. The Ages of Gaia A Biography of Our Living Earth. Oxford Korb J. 2011. Termite mound architecture, from function to construction. In: University Press, Oxford. Bignell D. E., Y. Roisin, and N. Lo (ed.). Biology of Termites, a Modern Lu M., M. Davidescu, R. S. Sukri, and J. H. Daskin. 2013. Termites facilitate root – Synthesis. Springer, Dordrecht. 349 373. foraging by trees in a Bornean tropical forest. J. Trop. Ecol. 29:563–566. Kuzyakov Y., and G. Domanski. 2000. Carbon input by plants into the soil. Re- Lu S.-G., Z. Malik, D.-P.Chen, and C.-F. Wu. 2014. Porosity and pore size dis- – view. J. Plant. Nutr. Soil Sci. 163:421 431. tribution of Ultisols and correlations to soil iron oxides. Catena. 123:79–87. Kuzyakov Y., and E. Blagodatskaya. 2015. Microbial hotspots and hot moments Margulis L. 1981. Symbiosis in Cell Evolution: Life and Its Environment on the – in soil: Concept & review. Soil Biol. Biochem. 83:184 199. Early Earth. W. H. Freeman, San Francisco, CA. Lange M., N. Eisenhauer, C. A. Sierra, H. Bessler, C. Engels, R. I. Griffiths, P. Marichal R., M. Grimaldi, A. M. Feijoo, J. Oszwald, C. Praxedes, D. H. R. G. Mellado-Vazquez, A. A. Malik, J. Roy, S. Scheu, S. Steinbeiss, B. C. Cobo, M. P. Hurtado, T. Desjardins, M. L. da Silva, L. G. D. Costag, I. S. Thomson, S. E. Trumbore, and G. Gleixner. 2015. Plant diversity increases Miranda, M. N. D. Oliveira, G. G. Brown, S. Tselouiko, M. B. Martins, soil microbial activity and soil carbon storage. Nat. Com. 6. T. Decaëns, E. Velasquez, and P. Lavelle. 2014. Soil macroinvertebrate Lavelle P. 1975. Consommation annuelle de terre par une population naturelle de communities and ecosystem services in deforested landscapes of Amazo- vers de terre (Millsonia anomala Omodeo, Acanthodrilidae, Oligochaeta) nia. App. Soil Ecol. 83:177–185. dans la savane de Lamto (Côte d’Ivoire). Rev. Ecol. Biol. Sol. 12(1):11–24. Marinissen J. C. Y., and J. Bok. 1988. Earthworm amended soil structure. Lavelle P. 1978. Les Vers de Terre de la savane de Lamto (Côte d’Ivoire): It’s influence on Collembola populations in grassland. Pedobiologia. Peuplements, populations et fonctions dans l’écosystème. Thèse d’Etat. 32:243–252. Paris VI. Pub. Lab. Zool, ENS. Marschner P., D. Crowley, and C. H. Yang. 2004. Development of specific rhi- Lavelle P. 1984. The soil system in the humid tropics. Biol. Int. 9:2–15. zosphere bacterial communities in relation to plant species, nutrition and soil type. PLSO. 261:199–208. Lavelle P. 1986. Associations mutualistes avec la microflore du sol et richesse spécifique sous les tropiques : L’hypothèse du premier maillon. C. R. Acad. Martin A., J. Cortez, I. Barois, and P. Lavelle . 1987. Les mucus intestinaux de Sci. Paris T. 302, Sér. III, 11–14. Ver de Terre, moteur de leurs interactions avec la microflore. Rev. Ecol. Biol. Sol. 24: 549–558. Lavelle P.,C. Lattaud, D. Trigo, and I. Barois. 1995. Mutualism and biodiversity in soils. PLSO. 170:23–33. Martin A. 1991. Short- and long-term effects of the endogeic earthworm Millsonia anomala (Omodeo) (Megascolecidae, Oligochaeta) of tropical Lavelle P. 1997. Faunal activities and soil processes: Adaptive strategies savannas, on soil organic matter. Biol. Fertil. Soils. 11:234–238. that determine ecosystem function. In: Begon M., and A. H. Fitter (eds.). Adv. Ecol. Res. 27, pp. 93–132. Martin A., and J. C. Y. Marinissen. 1993. Biological and physico-chemical processes in excrements of soil animals. Geoderma. 56:331–347. Lavelle P. 2000. Ecological challenges for soil science. Soil. Sci. 165:73–86. McKey D., S. Rostain, J. Iriarte, B. Glaser, J. J. Birk, I. Holst, and D. Renard. Lavelle P., and A. V.Spain. 2001. Soil Ecology. Kluwer Sci. Pub, Amsterdam. 2010. Pre-Columbian agricultural landscapes, ecosystem engineers, – Lavelle P. 2002. Functional domains in soils. Ecol Res. 17:441 450. and self-organized patchiness in Amazonia. Proc. Natl. Acad. Sci. U. Lavelle P., C. Rouland, F. Binet, M. Diouf, and A. Kersanté. 2005. Regu- S. A. 107:7823–7828. lation of microbial activities by roots and soil invertebrates. In:Mi- Menendez I., J. Caniego, J. F. Gallardo, and K. Olechko. 2005. Use of fractal croorganisms in Soils: Roles in Genesis and Functions. Soil Biology scaling to discriminate between and macro- and meso-pore sizes in forest Series,vol.3.BuscotF.,andA.Varma(eds.).SpringerVerlag,Berlin, soils. Ecol. Model. 182:323–335. pp. 291–305. Milleret R., R. C. Le Bayon, F. Lamy, J. M. Gobat, and P.Boivin. 2009. Impact Lavelle P.,T. Decaëns, M. Aubert, S. Barot, M. Blouin, F. Bureau, F. Margerie, of roots, mycorrhizas and earthworms on soil physical properties as P. Mora, and J. P. Rossi. 2006. Soil invertebrates and ecosystem services. assessed by shrinkage analysis. J. Hydrol. 373:499–507. Eur. J. Soil Biol. 42:3–15. Mitchell J., W. Dorney, R. Mayer, and J. McIlroy. 2007. Ecological impacts of Le Bayon R. C., and F. Binet. 2001. Earthworm surface casts affect soil ero- feral pig diggings in north Queensland rainforests. Wild Res. 34:603–608. sion by runoff water and phosphorus transfer in a temperate maize crop. Mora P., E. Miambi, J. J. Jimenez, T. Decaëns, and C. Rouland. 2005. Functional Pedobiologia. 45:430–442. complement of biogenic structures produced by earthworms, termites and Lee K. E., and T. G. Wood. 1971. Termites and Soils. Academic Press, London. ants in the neotropical savannas. Soil Biol. Biochem. 37:1043–1048. Lee K. E. 1985. Eartwhworms. Their Ecology and Relationships with Soils and Mujinya B. B., M. Adam, F. Mees, J. Bogaert, I. Vranken, H. Erens, G. Baert, Land Use. Academic Press, Orlando. M. Ngongo, and E. van Ranst. 2014. Spatial patterns and morphology Lepage M. G., and J. Darlington. 1984. Observations on the ant Carebara viuda of termite (Macrotermes falciger) mounds in the Upper Katanga, DR Smith preying on termites in Kenya. J. Nat. Hist. 18:293–302. Congo. Catena. 114:97–106. Levick S. R., G. P. Asner, T. Kennedy-Bowdoin, and D. E. Knapp. 2010. The Mulder C., H. den Hollander, T. Schouten, and M. Rutgers. 2006. Allornetry, spatial extent of termite influences on herbivore browsing in an African sa- biocomplexity, and web topology of hundred agro-environments in the vanna. Biol. Cons. 143:2462–2467. Netherlands. Ecol. Compl. 3:219–230. Lewis D. H. 1985. Symbiosis and mutualism: Crisp concepts and soggy se- Mulder C., A. Boit, M. Bonkowski, P.C. de Ruiter, G. Mancinelli, M. G. A. van mantics. In: Biology of Mutualism. Boucher D. H. (ed.). Croom Helm, der Heijden, H. J. van Wijnen, J. A. Vonk, and M. Rutgers. 2011. A below- Beckenham, pp. 29–39. ground perspective on dutch agroecosystems: How soil organisms interact

to support ecosystem services. In: Woodward, G. (ed.). Adv. Ecol. Res. Rouelle J. 1983. Introduction of amoebæ and Rhizobium japonicum into the gut 44:277–357. of Eisenia foetida (Sav.) and Lumbricus terrestris L. In: Earthworm Ecol- – Nooren C. A. M., N. Breemen, J. J. van Stoorvogel, and A. G. Jongmans. 1995. ogy. Satchell J. E. (ed.). Chapman and Hall, London, pp 375 381. The role of earthworms in the formation of sandy surface soils in a tropical RyanP.R.,Y.Dessaux,L.S.Thomashow,andD.M.Weller.2009.Rhizo- forest in Ivory Coast. Geoderma. 65:135–148. sphere engineering and management for sustainable agriculture. PLSO. – Ohashi M., J. Kilpelainen, L. Finer, A. C. Risch, T. Domisch, S. Neuvonen, and 321:363 383. P.Niemela. 2007. The effect of red wood ant (Formica rufa group) mounds Salmon S., J. J. Geoffroy, and J. F. Ponge. 2005. Earthworms and collembola re- on root biomass, density, and nutrient concentrations in boreal managed for- lationships: Effects of predatory centipedes and humus forms. Soil Biol. ests. J For Res. 12:113–119. Biochem. 37:487–495. Orgiazzi A., R. Bardgett, E. Barrios, V. Behan-Pelletier, M. J. I. Briones, J.-L. Salmon S., and J. F. Ponge. 2001. Earthworm excreta attract soil springtails: Lab- Chotte, G. B. de Deyn, P. Eggleton, N. Fierer, T. Fraser, K. Hedlund, S. oratory experiments on Heteromurus nitidus (Collembola: Entomobryidae). Jeffery, N. C. Johnson, A. Jones, E. Kandeler, N. Kaneko, P. Lavelle, P. Soil Biol. Biochem. 33:1959–1969. Lemanceau, L. Miko, L. Montanarella, F. M. S. Moreira, K. S. Ramirez, Sanderson M. G. 1996. Biomass of termites and their emissions of methane and S. Scheu, B. K. Singh, J. Six, W. H. van der Putten, and D. H. Wall (eds). carbon dioxide: A global database. Glob. Biogeochem. Cyc. 10:543–557. 2016. Global Soil Biodiversity Atlas. In: European Commission. Publica- tions Office of the, European Union, Luxembourg, pp. 176. Scheu S. 1991. Mucus excretion and carbon turnover of endogeic reathworms. Biol. Fertil. Soils. 12:217–220. Otieno D. O., G. O. K’Otuto, B. Jakli, P.Schroettle, J. N. Maina, E. Jung, and J. C. Onyango. 2011. Spatial heterogeneity in ecosystem structure and pro- Schmidt M. W.I., M. S. Torn, S. Abiven, T. Dittmar, G. Guggenberger, I. A. ductivity in a moist Kenyan savanna. Pl. Ecol. 212:769–783. Janssens,M.Kleber,I.Koegel-Knabner,J.Lehmann,D.A.C.Man- ning, P. Nannipieri, D. P. Rasse, S. Weiner, and S. E. Trumbore. 2011. Ott D., C. Digel, B. Klarner, M. Maraun, M. Pollierer, B. C. Rall, S. Scheu, G. Persistence of soil organic matter as an ecosystem property. Nature. Seelig, and U. Brose. 2014. Litter elemental stoichiometry and biomass 478:49–56. densities of forest soil invertebrates. Oikos. 123:1212–1223. Schutz M., C. Kretz, L. Dekoninck, M. Iravani, and A. C. Risch. 2008. Impact Paine R. T. 1969. A note on trophic complexity and community stability. Am. of Formica exsecta Nyl. on seed bank and vegetation patterns in a subalpine – Nat. 103:91 93. grassland ecosystem. J. Appl. Entom. 132:295–305. – Perry D. A. 1995. Self-organizing systems across scales. TREE. 10:241 245. ShahzadT.,C.Chenu,P.Genet,S.Barot,N.Perveen,C.Mougin,andS. Postma J., M. T. Schilder, J. Bloem, and W. K. van Leeumen-Haagsma. 2008. Fontaine. 2015. Contribution of exudates, arbuscular mycorrhizal fungi Soil suppressiveness and functional diversity of the soil microflora in or- and litter depositions to the rhizosphere priming effect induced by ganic farming systems. Soil Biol. Biochem. 40:2394–2406. grassland species. Soil Biol. Biochem. 80:146–155. Pozo M. J., and C. Azcon-Aguilar. 2007. Unraveling mycorrhiza-induced resis- ShusterW.D.,M.J.Shipitalo,S.Subler,S.Aref,E.L.McCoy,andC. tance. Curr. Op. Pl. Biol. 10:393–398. McKeegan. 2004. Earthworm additions affect leachate production and Price P. W. 1984. Alternative paradigms in Community Ecology. In:A nitrogen losses in typical midwestern agroecosystems. J. Env. Qual. – New Ecology: Novel Approaches to Interactive Systems. Price P. 33:1946 1946. W., C. N. Slobodchikoff, and W. S. Gaud (eds.). John Wiley, New Spain A. V., D. F. Sinclair, and P.J. Diggle. 1986. Spatial distributions of the York, pp. 1–14. mounds of harvester and forager termites (Isoptera: Termitidae) at four Puga-Freitas R., S. Barot, L. Taconnat, J.-P.Renou, and M. Blouin. 2012. Signal locations in tropical north-eastern Australia. Act. Oecol./Oecol. Gener. – Molecules Mediate the Impact of the Earthworm Aporrectodea caliginosa 7:335 352. on growth, development and defense of the plant Arabidopsis thaliana. Spain A. V.,and J. G. McIvor. 1988. The nature of herbaceous vegetation asso- PLoS One. 7. ciated with termitaria in north-eastern Australia. J. Ecol. 76:181–191. Puga-Freitas R., and M. Blouin. 2015. A review of the effects of soil or- Targulian V. O., and P. V. Krasilnikov. 2007. Soil system and pedogenetic ganisms on plant hormone signalling pathways. Env. Exp. Bot. processes: Self-organization, time scales and environmental signifi- 114:104–116. cance. Catena. 71:373–381. Pulleman M. M., and J. C. Y. Marinissen. 2004. Physical protection of mineral- Tisdall J. M., and J. M. Oades. 1982. Organic matter and water-stable aggregates izable C in aggregates from long-term pasture and arable soil. Geoderma. in soils. J. Soil. Sci. 3:141–161. – 120:273 282. Tisdall J. M. 1994. Possible role of soil microorganisms in aggregation in soils. Reddell P. R., and A. V. Spain. 1991. Earthworms as vectors of viable propa- Plant Soil. 159:115–121. – gules of mycorrhizal fungi. Soil Biol. Fert. 27:767 774. Topoliantz S., J.-F. Ponge, and P. Viaux. 2000. Earthworm and enchytraeid ac- Renard D., J. Iriarte, J. J. Birk, S. Rostain, B. Glaser, and D. McKey. 2012. Eco- tivity under different arable farming systems, as exemplified by biogenic logical engineers ahead of their time: The functioning of pre-Columbian structures. Plant Soil. 225:39–51. raised-field agriculture and its potential contributions to sustainability to- Traore S., M. Tigabu, P. Jouquet, S. J. Ouedraogo, S. Guinko, and M. Lepage. – day. Ecol. Eng. 45:30 44. 2015. Long-term effects of Macroternies termites, herbivores and annual Resner K., K. Yoo, S. D. Sebestyen, A. Aufdenkampe, C. Hale, A. Lyttle, and A. early fire on woody undergrowth community in Sudanian woodland, Blum. 2015. Invasive earthworms deplete key soil inorganic nutrients (Ca, Burkina Faso. Flora. 211:40–50. Mg, K, and P) in a Northern Hardwood forest. Ecosystems. 18:89–102. TrigoD.,I.Barois,M.H.Garvin,E.Huerta,S.Irisson,andP.Lavelle. Rice E. L. 1979. Allelopathy—update. Bot. Rev. 45:15–109. 1999. Mutualism between earthworms and soil microflora. Pedobiologia. – Rillig M. C., and D. L. Mummey. 2006. Mycorrhizas and soil structure. New 43:866 873. Phytol. 171:41–53. Uren N. C. 1993. Mucilage secretion and its interaction with soil, and contact – Rossi J. P.,A. Albrecht, and P.Lavelle. 1996. Relationships between spatial pat- reduction. PLSO. 156:79 82. tern of the endogeic earthworm Polypheretima elongata and soil heteroge- van Breemen N. 1993. Soils as biotic constructs favouring net primary produc- neity in a tropical pasture of Martinique (French West Indies). Soil Biol. tion. Geoderma. 57:183–211. – Biochem. 29:485 488. van der Putten W. H., P.C. de Ruiter, T. M. Bezemer, J. A. Harvey, M. Wassen, Rossi J. P.2003. The spatiotemporal pattern of a tropical earthworm species as- and V.Wolters. 2004. Trophic interactions in a changing world. Bas. App. semblage and its relationship with soil structure. Pedobiologia. 47:497–503. Ecol. 5:487–494.

van der Putten W.H., R.D. Bardgett, P.C. de Ruiter, W. H. G. Hol, K. M. WellerD.M.,J.M.Raaijmakers,B.B.M.Gardener,andL.S.Thomashow. Meyer, T. M. Bezemer, M. A. Bradford, S. Christensen, M. B. 2002. Microbial populations responsible for specific soil suppressiveness Eppinga, T. Fukami, L. Hemerik, J. Molofsky, M. Schaedler, C. to plant pathogens. Ann. Rev. Phyto. 40:309. Scherber,S.Y.Strauss,M.Vos,andD.A.Wardle2009.Empirical Whipps J. M. 2001. Microbial interactions and biocontrol in the rhizo- and theoretical challenges in aboveground-belowground ecology. sphere.J.Exp.Bot.52:487–511. Oecologia 161: 1–14. White G. 1789. The Natural History of Selborne. Benjamin White, London. van Loon L., P. Bakker, W. van der Heijdt, D. Wendehenne, and A. Wilkinson M. T., P.J. Richards, and G. S. Humphreys. 2009. Breaking ground: Pugin. 2008. Early responses of tobacco suspension cells to Pedological, geological, and ecological implications of soil bioturbation. rhizobacterial elicitors of induced systemic resistance. Mol. Pl. Micr. Earth Sci. Rev. 97:257–272. Int. 21:1609–1621. Wood T. G., and W. A. Sands. 1978. The Role of Termites in Ecosystems. Velasquez E., C. Pelosi, D. Brunet, M. Grimaldi, M. Martins, A. C. Rendeiro, E. Cambridge University Press, Cambridge, UK, pp., Production Ecology of Barrios, and P. Lavelle. 2007a. This ped is my ped: Visual separation and Ants and Termites., pp. 245–292. NIRS spectra allow determination of the origins of soil macro-aggregates. Pedobiologia. 51:75–87. Yair A., and J. Rutin. 1981. Some aspects of the regional variation in the amount of available sediment produced by Isoipods and porcupines, Northern Velasquez E., S. J. Fonte, S. Barot, M. Grimaldi, T. Desjardins, and P. Negev, Israel. Earth Surf. Proc. Landf. 6:221–234. Lavelle. 2012. Soil macrofauna-mediated impacts of plant species composition on soil functioning in Amazonian pastures. App. Soil Young I. M., and J. W.Crawford. 2004. Interactions and self-organisation in the – Ecol. 56:43–50. soil-microbe complex. Science. 304:1634 1637. Velasquez E., P. Lavelle, and M. Andrade. 2007b. GIQS: A multifunctional in- Zangerle A., A. Pando, and P.Lavelle. 2011. Do earthworms and roots cooper- dicator of soil quality. Soil Biol. Biochem. 39:3066–3080. ate to build soil macroaggregates? A microcosm experiment. Geoderma. 167–68:303–309. Villarreal D., K. L. Clark, L. C. Branch, J. L. Hierro, and M. Machicote. 2008. Alteration of ecosystem structure by a burrowing herbivore, the plains viz- Zangerle A., C. Hissler, M. Blouin, and P. Lavelle. 2014. Near infrared cacha (Lagostomus maximus). J. Mam. 89:700–711. spectroscopy (NIRS) to estimate earthworm cast age. Soil Biol. Biochem. 70:47–53. de Vries F. T., E. Thebault, M. Liiri, K. Birkhofer, M. A. Tsiafouli, L. Bjornlund, H. B. Jorgensen, M. V.Brady, S. Christensen, P.C. de Ruiter, Zangerlé A., C. Hissler, D. McKey, and P.Lavelle. 2016. Using near infrared re- T. d’Hertefeldt,J.Frouz,K.Hedlund,L.Hemerik,W.H.G.Hol,S. flectance spectroscopy (NIRS) to identify the contribution of earthworms to Hotes, S. R. Mortimer, H. Setala, S. P. Sgardelis, K. Uteseny, W. H. soil macroaggregation in field conditions». In: App. Soil Ecol. in press. van der Putten, V. Wolters, and R. D. Bardgett. 2013. Soil food web Zhuang X., J. Gao, A. Ma, S. Fu, and G. Zhuang. 2013. Bioactive molecules properties explain ecosystem services across European land use sys- in soil ecosystems: Masters of the Underground. Int. J. Mol. Sci. tems. Proc. Natl. Acad. Sci. U. S. A. 110:14296–14301. 14:8841–8868.