**6. Conclusions and perspectives**

regarding the decomposers themselves and their consumers, from different trophic levels, associated to the trophic web. The study of these trophic webs necessarily involves two components that may influence the litter decomposition process: (1) the diversity of litter mixture composing the litter; and (2) the diversity and the trophic links among decomposers and detritivores. Once the trophic web related to litter decomposition is, at least partially, donor-controlled – in the case, the leaf litter – it is expected these webs to be controlled predominantly by bottom-up effects, since consumers do not control resource abundance (and,

**Figure 2.** Factors and processes that associate plant diversity in litter mixture to litter decomposition.

In Figure 3 we represent a simplified trophic web of detritivores and decomposers, recognizing three different control pathways of the represented trophic levels. The more direct pathway, symbolized by the dotted line, represents a direct effect of the diversity of litter mixture accounting for higher resource heterogeneity to decomposers (but see Figure 2 and discussion above). This pathway may promote the increase of decomposer diversity and activity, directly augmenting the decomposition process, without the interference of organisms from other trophic levels. However, such effects would only be possible if the effects promoted by the increase of litter diversity on the edaphic biota are positive (see Figure 2 and discussion above).

The activities of the animals that inhabit litter may interfere in the above described patterns in different ways. Following the dashed line in Figure 3, the activity of litter breaking by arthropods may increase the resource availability to the decomposer community, since the

in our example, nor the diversity).

42 Biodiversity - The Dynamic Balance of the Planet

According to the above discussions, it is possible to notice that a single response relating the effects of litter diversity on decomposition does not exist. The interactions among the factors that determine litter decomposition is complex, and several pathways may occur. The effects of litter diversity on decomposition rates are dependent on several features, and the experi‐ mental control of these factors is absolutely necessary to clarify their relative roles and interactions.

Therefore, the studies involving the relationship between the plant diversity composing litter mixtures and its effects on the decomposition process, including the possible effects on the edaphic trophic webs, still need much more study and explanation. The knowledge acquisition in these subjects must necessarily be based on well-designed experiments, to elucidate the relative role of each factor and of the interactions between these factors. We hope to have contributed for future studies by the production of the flowchart of factors and processes,

Biodiversity and Ecosystem Functioning: a Conceptual Model of Leaf Litter Decomposition

http://dx.doi.org/10.5772/57396

45

We thank Sabrina Almeida, Júlio Louzada, Ricardo Ildefonso Campos and Elizabeth Nichols for contributions in manuscript idealization. Financial support of Fundação de Amparo à Pesquisa do Estado de Minas Gerais – FAPEMIG and Conselho Nacional de Desenvolvimento

1 Postgraduate Program in Entomology, Department of Entomology, Federal University of

2 Department of General Biology, Federal University of Viçosa, Viçosa, Minas Gerais, Brazil

[1] Ricklefs R E. A comprehensive framework for global patterns in biodiversity. Ecolo‐

[2] Gaston K J., editor. Biodiversity: a biology of numbers and difference. Oxford: Black‐

[3] Wiegert R G. Holism and reductionism in ecology: hypotheses, scale and systems.

[4] Chapin F S, Sala O E, Burke I C, Grime P J, Hooper D U, Lauenroth W K, Lombard A, Mooney H A, Mosier A R, Naeem S, Pacala S W, Roy J, Steffen W L, Tilman D. Eco‐

system consequences of changing biodiversity. BioScience 1998; 48 45-52.

and Carlos F. Sperber2

facilitating the isolated tests of each of the connections shown in it.

Científico e Tecnológico – CNPq are gratefully acknowledged.

**Acknowledgements**

**Author details**

**References**

Dalana C. Muscardi1\*, José H. Schoereder2

\*Address all correspondence to:

ViçosaViçosa, Minas Gerais, Brazil

gy Letters 2004; 7 1–15.

well Science Ltd; 1996.

Oikos 1988; 53 267-269.

Therefore, studies controlling, simultaneously, litter and edaphic fauna diversities, may give evidence to their joined effects on decomposition rates, allowing a better understanding of the relationship between these two factors. Furthermore, since several elements external to litter diversity (such as soil use, habitat fragmentation and others) may affect the effect of the discussed factors on litter decomposition, the incorporation of such elements would add information to the process. To do this it is necessary to design studies with manipulative control, which would render more informative studies than those that limit to compare decomposition rates in different litter mixtures.

There are some alternative for these studies, and each alternative may achieve an answer that would clarify one or more of the above discussed points. One proposal it is the use of several plant species, among which different number of species are drawn, giving an idea of the relationship between the number of species in litter mixture and decomposition. In the array of litter mixtures other parameters may also be tested, such as arthropod species richness and abundance, as well as microbial biomass and activity. This approach tends to create situations more close to the actual environments studied, since usually the plant species used are the same as those occurring in the study site. On the other hand, this approach may increase very much the variability among repetitions, creating a noise that may impair data interpretation. The alternative approach would be the use of the same plant species in all repetitions, decreasing variability among them and giving more interpretable data, although decreasing their realism towards the studied environments. These studies are particularly common in habitats involving crops, associated or not to other plants, using decomposition litterbags containing leaves of the cultivated plant, alone and together with other plant species.

In order to study the relative role of the trophic web components, manipulative studies are the most usual approach. In these studies, detritivore species are added or removed, and their effect on other species and/or on decomposition rates are observed. This approach suffers from the same problem of the manipulative studies involving litter mixtures: the distancing from biological reality. Furthermore, these studies are virtually impossible to carry out in hyper diversity habitats (as it is usual in most tropical biomes), and when the knowledge regarding taxonomic identity and/or biology of the involved species is weak. In such cases, experiments excluding litter fauna may be carried out through methods as fumigation, or by the use or other biocides, such as naphthalene. These substances, however, may cause impacts in the decomposer community, modifying decomposition process through more than one pathway of the activity of organisms, impairing once more data interpretation. Another option to exclude fauna is decomposition litterbags with different sized mesh, which may exclude selectively the fauna and modify their species composition. Nevertheless, most options to exclude fauna deal with species richness of the edaphic trophic web, and not on its species composition, weakening the interpretation of the effects of the functional and phylogenetic diversity on litter decomposition process.

Therefore, the studies involving the relationship between the plant diversity composing litter mixtures and its effects on the decomposition process, including the possible effects on the edaphic trophic webs, still need much more study and explanation. The knowledge acquisition in these subjects must necessarily be based on well-designed experiments, to elucidate the relative role of each factor and of the interactions between these factors. We hope to have contributed for future studies by the production of the flowchart of factors and processes, facilitating the isolated tests of each of the connections shown in it.
