Allelopathic research in Brazil

In this article, we review allelopathy studies conducted in Brazil or involving plant species that occur in the country. Conceptions and misconceptions associated with allelopathy, as well as some international criteria to be applied in allelopathic research, are presented and discussed. We observed a sharp increase in the number of papers on this subject conducted in Brazil between 1991 and 2010. However, most studies are conducted under laboratory conditions, lack a clear hypothesis or a solid justification, and typically make use of target species that do not co-exist with the donor species under natural conditions. We also found that most studies do not take the additional steps in order to purify and identify the bioactive molecules. We recommend that further studies be conducted in order to explore the potential of plant biodiversity in Brazil. Such studies could lead to the development of new molecular structures (allelochemicals) that could be used in the control of pests and weeds, thereby reducing the use of the harmful synthetic herbicides that are currently being widely employed.


Introduction
The International Allelopathy Society defines allelopathy as "any process involving secondary metabolites produced by plants, algae, bacteria and fungi that influences the growth and development of agriculture and biological systems" (IAS, 1996).The first recorded use of the word "allelopathy" was by Theophrastus (ca.300 B.C.), a disciple of Aristotle, who observed that, apart from affecting the development of invader plants, chickpea plants (Cicer arietinum L.) did not enrich the soil as other plants did.Since then, numerous studies have described plant interactions characterised as allelopathic (Inderjit & Keating 1999;Fujii & Hiradate 2007;Macías et al. 2004;Reigosa et al. 2006;Blum 2011).Over time, new definitions have been proposed.According to Inderjit & Callaway (2003), allelopathy is the negative effect that one plant can have on another by releasing chemical compounds into the environment.In addition to a certain adaptation of the term for plant-plant interactions, this definition restricts the concept to negative effects of plants on other plants.This is in contrast to the explanation given by Ferreira (2004), who defined allelopathy as the positive or negative influence that secondary metabolites produced by a plant and shed into the environment have on the growth of other plants (Ferreira, 2004).Here, positive effects can also be considered allelopathic.Within this context, the secondary metabolites involved in allelopathic interactions have been designated allelochemicals.
Despite its tarnished history (Reigosa et al. 1999), the scientific study of allelopathy has flourished in recent years.This is due to the fact that, although there has yet to be a true experimental separation of allelopathy from other forms of plant interactions, there is considerable evidence of phenomena that only can be explained in terms of accumulation of allelochemicals in the field.In addition, many authors have found interesting and promising effects of plant residues in the field, as well as of weed-crop interactions that are probably due to intense production and release of such bioactive compounds.
We searched the Scientific Electronic Library Online (SciELO) database (www.scielo.br),using the search terms "allelopathy"' and "allelochemicals", as well as browsing the archives of the Allelopathy Journal, limiting our searches to articles published between 1991 and 2010 and involving plant species found in Brazil.Table 1 shows the characteristics of the studies selected.It is noteworthy that Although they are two different concepts, "allelopathy" can be confused with "competition" (Ferreira, 2004).Competition can be viewed as a type of interaction between organisms which involves the removal of an element from the environment, such as water, light or minerals, by an organism, which in some way affects the growth of neighbours sharing the same habitat.Allelopathy, on the other hand, is a kind of interaction between organisms that involves the production and release of substances into the environment by one organism, which affects the growth of other, nearby organisms (Dakshini et al. 1999).Naturally, in affecting the growth of other organisms, the species that produces and releases such molecules into the environment can benefit in the competition for resources, so that such processes (competition and allelopathy) can occasionally be considered complementary or collaborative (Inderjit & Callaway, 2003).
Allelopathic studies can also be confused with phytotoxicity studies.However, differences in the procedures employed in the extraction of active compounds distinguish one from the other.An allelopathy study is one that is conducted with chemical substances extracted from plant tissue using natural methods such as leaching, exudation and release through the deterioration of plant matter, or even vaporisation; or with allelochemicals extracted under laboratory conditions that replicate a natural process, such as aqueous extraction from plant tissue that has been crushed or decomposed, simulating to some extent the action of rain and dew on parts of the plants in nature.The water to be used for solubilisation of substances can be slightly acidified or alkalinised in order to resemble more closely the pH of the substrate on which the species occur.In contrast, a phytotoxicity study is one that is carried out with substances extracted from plant tissue through any non-natural chemical or physico-chemical procedure, for example, through the use of organic solvents, such as hexane, dichloromethane, and methanol, or by using technologies to extract bioactive compounds from plants, such as ultrasonic, Soxhlet, and high-pressure extraction methods, with the aim of maximising the solubility of chemical substances.Because of these similarities, field and laboratory studies must comply with established criteria and follow basic procedures, as described below, in order to be considered allelopathy studies.

Secondary metabolites and chemical interactions
All plants produce secondary metabolites at various levels of concentration, diversity and composition in a variety of plant tissues (Bonner & Varner 1965;Harbone 1993;Hadacek 2002;Taiz & Zeiger 2010).Although such metabolites can belong to any one of a great number of classes, they generally belong to one of three major categories, specifically terpenes, phenolic compounds and alkaloids (Taiz & Zeiger 2010;Gleason 2012).The production of secondary metabolites by plants is determined by the genetic characteristics of the species producing them and by the environmental conditions in which the plants are found (Hadacek 2002).Variables such as temperature, humidity and light intensity, added to the effects of the biota and the physicochemical structure of the soil, can affect not only the production of metabolites but also the chemical structure and degree of activity of substances released into the environment.
It is of note that the chemical interactions between plants are almost never limited to one compound, but rather to mixtures, often complex ones, of various substances (Gonzalez & Reigosa, 2001).These mixtures can have synergistic or antagonistic effects, which can be modified by the chemical, physical, and biotic properties of the soil (Einhellig 1999).The toxicity of substances and the degree of interaction between organisms depend also on the stage of growth of the donor and recipient species alike (Rice 1984;González & Reigosa 2001).

Allelopathic research in Brazil
Many allelopathy studies are carried out to detect direct effects of allelochemicals produced by one (donor) plant on the growth of a target plant.However, quite often the allelochemical under study has an indirect effect on the organism.For example, Grove et al. (2012) showed that the abundance of ectomycorrhizal fungi is lower on Douglas-fir seedlings grown in forest soils invaded by Cytisus scoparius than on those grown in un-invaded forest soils .Because mycorrhizae help root systems absorb soil nutrients more efficiently, the authors concluded that allelochemicals produced by C. scoparius affect the growth of the conifer by reducing the quantity of ectomycorrhizae associated with its roots.
Allelopathic studies can involve concepts and approaches in various fields such as agronomy, forestry, ecology, physiology, anatomy, plant systematics, cellular biology, molecular biology, and molecular chemistry.Consequently, applying the appropriate criteria in allelopathy studies can require a diversified team of scientists, as well as the equipment and infrastructure required in order to conduct experiments in the laboratory, in the greenhouse, and in the field (Ferreira & Aquila, 2000).

Modes of action of allelochemicals
The extent of allelopathic effects has not been fully demonstrated.Nevertheless, there is evidence that some secondary metabolites, apart from helping the producer plant avoid the effects of insects, fungi, herbivores, etc., can be useful as natural bioherbicides.The increasing need for alternative herbicides is another factor currently driving allelopathy research.Therefore, there are two possible objects of interest: allelochemicals acting as synthetic herbicides do (i.e., with a concrete mode of action); and allelochemicals acting subtly, with multiple modes of action, probably quite dependent on the ecological and physiological stage of the recipient plant.We will examine both, because even these mild, subtle effects can be of agro-ecological interest (Reigosa & Carballeira 1992;Reigosa et al. 1999).
To date, there have been few studies of the ecological role of many secondary metabolites (Field et al. 2006;Duke et al. 2010;Martínez-Peñalver et al. 2012), although many of them can be considered to play pivotal roles in the life of the plant, improving its fitness as defense substances or attracting beneficial organisms.Many secondary metabolites, despite playing a primary role in defending the producer plant against pathogens or herbivores, can be considered to play secondary roles in plant-plant interactions, by which they nevertheless enhance the competitive potential of the producer (Reigosa et al. 1999).The number of studies investigating the modes of action of secondary metabolites is on the rise, which can be mainly attributed to two factors: first, many weeds have evolved resistance to herbicides, and there is some hope that new sites of action, or new modes of action that could alleviate resistance, will be discovered; second, there is hope that, with appropriate management, allelopathic crops will be able to protect themselves from weeds and other pests.This latter approach would be quite good from the ecological and economic point of view, minimising the release of synthetic non-degradable molecules (Dayan et al. 2009).
Under normal circumstances, many different molecules are simultaneously released into the environment.Certain potent molecules have been identified as possibly being responsible for real allelopathic relationships in the field.Good examples are juglone (released by Juglans spp.); sorgoleone (produced by sorghum); avenacin (a triterpene glycoside produced by Avena fatua and potentially responsible for its invasive capacity); momilactone B (released by rice plants), coumarin (released by Arctostaphylos spp.); non-protein amino acids (such as mimosine and tyrosine, produced by Festuca rubra or Leucaena leucocephala); glucosinolates (produced in great quantities by Brassicaceae); and cyanogenic glycosides.Although none of those molecules are potent enough to act individually, they can have significant effects when acting in concert.Most phenolic compounds, for example, are released simultaneously and continuously, perhaps acting synergistically (Reigosa et al., 1999).
To summarize the whole picture of how secondary metabolites can act as allelochemicals, it is worth to mention that these bioactive substances can present different types of effects on plants (Reigosa et al. 1999;Field et al. 2006;Lotina-Hensen et al. 2006).Some allelochemicals have subtle and varied effects on the recipient plant.For example, benzoxazolinone (a quite common hydroxamic acid that is released by several cereals) can simultaneously affect processes such as the cell cycle of the root meristems and the production of reactive oxygen species (and therefore the oxidative stress cycle), as well as membrane permeability, water balance, and osmotic regulation (Baerson et al. 2005;Hussain et al. 2011;Sánchez-Moreiras & Reigosa 2005;Sánchez-Moreiras et al. 2005, 2008, 2009, 2010).This multi--functional activity is quite difficult to investigate, typically requiring the use of numerous high-throughput techniques.These types of allelochemicals are probably best suited for ecologically accepted control of weeds and other pests, only improving the balance between the crop (that produces and releases the allelochemicals) and the target organisms.
There are, of course, some examples of secondary metabolites that have a definite and unique mode of action.In fact, several molecules of natural origin (most from micro-organisms but some also synthesised by higher plants) have been commercialised, either as mixtures of natural compounds or as molecules synthetically modified to enhance their selectivity and capacity of action (Copping & Duke 2007;Duke et al. 2010).

Allelopathy in natural and cultivated systems
Allelopathic research can be divided into two main categories: one following concepts and ecological-based approaches, corresponding to studies on phenomena occurring in natural ecosystems (allelopathy sensu stricto); and the other following commercial and economic criteria and interests, corresponding to studies based on interactions between cultivated species that do not naturally occur in the same habitat (applied allelopathy).In the first category, preliminary or complementary laboratory studies should replicate, under controlled conditions, the expected effects of rainfall or dew on the leaching of substances supposed to occur in the natural environment.Indeed, the species under study should co-exist in the same habitat, and such studies should follow these and other criteria (see below) to be considered "true" (sensu stricto) allelopathic studies.In the second category, the purpose is directed towards understanding how cultivated species interact with each other and to modify such interactions in some way.Those studies may use different kinds of solvents to extract the bioactive compounds involved in such interactions, and the species under study do not naturally co-exist in the same habitat but do co-exist due to commercial interests, irrespective of whether they are native or exotic.From these two categories, a third approach can emerge, which is to look for and identify bioactive molecules that could potentially be used to develop new herbicides and plant growth regulators (Vyvyan, 2002).

Criteria for allelopathy studies
Plants produce and release various bioactive substances into the environment.Such substances have been frequently shown, under laboratory conditions, to stimulate, neutralise or inhibit biological processes in other individuals.However, it can be quite complex and difficult to prove that such released substances are involved in chemical interactions between plants in nature.The same can be said about competition for resources between plants.Plants require resources such as energy, water and nutrients.However, under field conditions, it is not always possible to demonstrate that competition is taking place (Blum 2011).Nevertheless, although competition for resources is a widely accepted concept in the literature, as is that of chemical interactions between plants and insects, between plants and micro--organisms, and between micro-organisms (Harbone 1993), the same cannot be said about of allelopathic interactions (Blum 2011).Perhaps the difficulties in accepting this line of investigation have resulted from the following (Blum 2011): The definition of the term 'allelopathy' has been changed several times.The first allelopathic bioassays have not been clearly stated.
The scepticism of opponents to this kind of research.
The high level of rigour required to demonstrate that such plant interactions are allelopathic.The challenge, therefore, lies perhaps in setting real and applicable criteria that could be used as evidence of allelopathic interaction between plants.In this context, the following criteria have typically been recommended and adopted (Blum 2011): Research studies must show clear patterns of stimulus or inhibition that donor plants exert over the development of recipient plants.Such patterns cannot be explained only by physico--chemical modifications to the environment, by the absorption and use of substances as sources of nutrients or energy, mycorrhizal transfer or grafting, or by biotic factors such as competition for resources, herbivory or acquired disease.Allelopathic plants or their waste products must produce and release organic substances into the environment, which are capable of stimulating or inhibiting the functioning of other, nearby plants.
The recipient plants must be in contact and interact with organic substances produced directly or indirectly by the donor plants (whether modified or not by the environment).Such organic substances must be in appropriate concentrations and the exposure time must be sufficient to modify the functioning of the recipient plant.Although some of these criteria can be met in research conducted under laboratory conditions, or even in the field, observing and demonstrating all of these criteria in nature can be a major challenge for researchers seeking to carry out allelopathy studies.Very few studies have successfully "completed the cycle" or, in other words, have shown the production of a specific metabolite by the allelopathic (donor) plant, its journey through the environment (soil, water, or atmosphere), its arrival at the target, and its influence on the affected (recipient) plant.

Laboratory studies
In studies conducted under controlled conditions, allelopathic activity has frequently been demonstrated.However, as mentioned above, demonstrating that plant tissues from a specific plant produce bioactive substances that affect the functions of the target species in a laboratory experiment cannot be considered proof of allelopathic interaction; for that, it should also be shown that the allelopathic effects are exerted under natural conditions.Such laboratory studies, at most, show only that the donor plant produces bioactive compounds and only suggest that, under natural conditions, those compounds may have some kind of effect on the growth of neighbouring species.
In the literature, there are many experiments purported to be allelopathic that make use of organic solvents, procedures and equipment to optimise the extraction of bioactive substances produced by plant tissues.These procedures maximise the release of substances from plant tissues and can strengthen their biological effects on the development of other plants; consequently, such studies do not in fact represent interaction between neighbouring species in the field and cannot be used as a proof of allelopathic interaction taking place under natural conditions (Inderjit & Weston 2000).
Preliminary studies focused on identifying and elucidating allelopathic interactions should follow a set of criteria for their results to be considered indicative of the existence of allelopathic interaction in the field.Within this context, Inderjit & Weston (2000) established individual components or criteria that should be observed if a study is to be clearly recognized as allelopathic research.According to those authors, bioassays conducted in the laboratory (or even in the field, under controlled conditions) must at least comply with the following criteria (with adaptations): The extracts to be tested must be aqueous, reproducing the effects of rainfall, dew or other forms of aqueous solubilisation occurring naturally in the environment.
Because extraction by organic solvents such as dichloromethane or methanol do not occur in nature, such extraction does not replicate a natural process of compound solubilisation.The substrate used in experiments must be representative of the place where the plants co-exist or at least of the region where they occur.Experiments using filter paper, vermiculite or other types of substrates exclude the biotic and abiotic effects of the substrate on the bioactivity of the substances.Sensitive species must be avoided, because their use can result in an overestimation of allelopathic effects.The sensitivity of exotic or cultivated species to bioactive extract can be completely different from that displayed by native species or by those that occur naturally near the donor plant.Species to be studied must co-exist within their natural ecosystems.Although the use of cultivated or exotic plants as target species can help describe the effects of bioactive substances, or to quantify their activity, they are not representative of what can occur between neighbouring plants in nature.
More than one extract concentration should be tested, and it is recommended that a dose-response test be set up with at least three levels of concentration.Thus, it can be established what levels of allelochemical concentrations are active in the environment.In addition, the concentrations must be compatible with what is expected to occur under natural conditions.
In studies in which the bioactive compounds are purified and identified, their biological activity must be tested individually and as a mixture (crude extract), to determine whether the biological effects are synergistic or individualised.In fact, the effects observed under laboratory conditions should be also evaluated under natural conditions in order to determine the real allelopathic potential of the crude extracts and of the purified compounds.
Each bioassay must be designed to evaluate the allelopathic interactions between species after careful consideration of their growth habits, the biotic and abiotic characteristics of the area, and ecophysiological factors, with reference to key issues, to determine the relevance of a specific laboratory test (Inderjit & Weston, 2000).As with all these criteria, experiments conducted in the laboratory at least serve to predict or postulate allelopathic interactions likely to occur in the field but cannot be used to confirm that the species being studied truly affect the functioning of neighbouring species and, consequently, the dynamics of the local vegetation.

Allelopathy studies carried out in Brazil
There have been several reviews of allelopathy studies conducted in Brazil (Ferreira et al. 1992;Rodrigues et al. 1999;Ferreira & Aquila 2000).More recently, Souza--Filho et al. (2010) carried out a review of experimental procedures used in laboratory-based allelopathy studies.In the last two decades, there has been a proliferation of studies on the allelopathic properties of species that are native to, introduced to or cultivated in Brazil.Our review of the SciELO database showed a clear increase in the number of publications including the terms "allelopathy" or "allelochemicals" (Figure 1).That increase might reflect the growing interest not only in allelopathic interactions in natural ecosystems and agro-ecosystems but also in the products that can be derived from allelochemicals, such as natural herbicides and plant growth regulators.Although the increase likely represents an improvement in this type of research in Brazil, it also generates the need to evaluate the way in which such studies have been conducted.This implies the importance of establishing appropriate criteria and procedures so that data produced are consistent and the study will be recognised for its value as allelopathy research.Hence, it is important to assess whether studies self-designated as allelopathic meet the criteria set forth in the international literature, as discussed above, which allow them to be considered to have made a real contribution to allelopathy.We based our review on the main Brazilian periodicals indexed for the SciELO database and on articles published in the Allelopathy Journal, combining articles in which the terms "allelopathy" or "allelochemicals" were used.In the Allelopathy Journal, only studies by Brazilian authors or conducted in Brazil were considered.We focused on studies involving native Brazilian species, although a few studies involving exotics (occurring in Brazil) were also included.We selected 115 articles, all published between 2000 and 2012, involving a total of 105 species, the majority of which (n = 92) were native to Brazilian biomes.For each article, we evaluated the type of substrate used, the extract concentrations employed, the species tested, as well as other parameters.On the basis of these results, informative and comparative tables were drawn up to analyse the results.The raw data and information emerging from them are fully described in Appendix 1 of this review.
Although we have to consider the possibility that many studies conducted under field conditions were not been identified in this review, the results obtained here allow us to conclude that the majority of allelopathy research has been conducted under laboratory conditions or at least under controlled conditions.Although such studies are certainly important in order to isolate variables and identify the true factors involved in plant-plant interactions, there is a need for complementary studies describing the allelopathic properties of plants under natural conditions.This review also shows that most studies have investigated the allelopathic properties of leaves, followed by those investigating that of other plant parts, such as stems and, to a lesser degree, reproductive structures (Table 2).Most have also made use of two or more plant structures.The preference for leaves might reflect the fact that it is certainly easier to collect leaves than to collect roots, for example, as well as that leaves represent a large part of the litter produced by the vegetation-biomass that directly impacts seedling growth and recruitment in various ways.
Table 2 also shows that more than 70% of these studies have made use of filter paper, and less than 20% have made use of soil as substrate (Table 2).Filter paper has been routinely used in laboratory studies because of its practicality, simplicity and because it is, in theory, an inert substrate.However, for allelopathy studies aimed at recognising and quantifying chemical interactions between plants that co--exist in the field, the substrate to be used is fundamental, as it may change the properties and form of action of the allelochemicals involved (Inderjit 2001).In our review, we noticed that some studies were carried out on filter paper and in soil, thus determining the ways in which the substrate can influence the allelopathic activity.Other authors have performed comparative analyses of sterilised and non--sterilised soil.Those studies are particularly interesting because they permit to separate the effects of the soil per se (physical structure, chemical composition, pH, etc) from the effects of the soil biota on the activity of the allelochemicals (Kaur et al. 2009).It has become clear that a true allelopathic experiment must, at some point, make use of soil, specifically the same soil in which the species involved in the supposed interaction grow, as substrate.
A little more than half of the studies selected made use of water to solubilise bioactive substances, and less than 33% made use of methanol or ethanol as solvent (Table 2).This is a positive feature, since the use of water as solvent is a procedure that approximates what is presumed to take place under natural conditions, such as the leaching of leaves or litter during a rainfall event, or solubilisation of bioactive molecules in wet soil or aquatic environments (Ferreira 2004).Some studies have initially used aqueous extracts for solubilisation of the bioactive substances, then performing procedures involving the use of organic solvents and sophisticated equipment to optimise the extraction and purification of the molecules.The latter procedures cannot be considered strictly allelopathy because such laboratory methods do not reflect processes that occur in nature.In either case, the bioactivity of the extracts should be determined by bioassay.
Among the studies reviewed, the most common measure used to refer the concentrations employed in allelopathic experiments is the percentage (Table 2).In these studies, the weight-volume relationship (the weight or mass of the plant matter per volume of solvent), is the most commonly cited rationale, and the majority of the bioassays are conducted at concentrations of 1-5% (Table 2).As a general rule, the solutions obtained from plant tissues are filtered (typically through filter paper) and diluted to obtain solutions designed to establish dose-response effects.In some cases, when researchers progressed to the purification stage, they used parts per million (ppm) or similar measures (e.g., mg/ml).
Although it may be difficult to determine which concentrations are closest to those occurring naturally in the field, good sense dictates that the use of solutions that are highly concen-  et al. 1992).For example, it has been indicated that seed germination is quite sensitive to solutions over 100 mOsmol (mmol kg −1 ), suggesting that extracts of similar or higher osmolarity can affect seed germination irrespective of any allelopathic property (Leather & Einhellig, 1988).In fact, crude extracts are usually rich in sugars, amino acids and other substances with osmotic potential.Extracts at 3-4% (w/v) would be equivalent to −0.2 MPa of osmotic pressure.Values more negative than that will probably have osmotic effects on the extract solutions (Astarita et al., 1996: Oliveira et al. 2004a).
It should be borne in mind that extremely high concentrations are likely to generate some biological effects by the mere fact that the target plants are subjected to high doses of bioactive substances, which would be phytotoxic rather than allelopathic.Therefore, it is recommended that the extracts be prepared at concentrations that are more comparable to those to which the plants under study are subjected in the natural environment.It would be ideal to determine the concentrations of allelochemicals in the soil, or at least which quantity of plant biomass is produced per unit of soil or covered area, for example.This information would serve as a parameter to establish the concentrations to be used in allelopathy studies under controlled conditions.
The majority of studies reviewed have made use of exotic and or cultivated species as the target species to describe the effects of extracts on plants (Table 3).The most common species used in allelopathy was Lactuca sativa (lettuce), followed by Sesamum indicum (sesame).It appears that these and other species have been chosen because they present fast germination as well as rapid, uniform initial growth, which is certainly desirable when experiments are designed to compare various treatments.In addition, such experiments conducted in Brazil have apparently been modelled on studies carried out in countries where the use of such species is common.However, despite the ease which cultivated species may bring to allelopathic studies, in the majority of cases they represent species which have been introduced and consequently do not present a history of co-existence with the (native) Brazilian species.
The most common physiological parameters used for identifying allelopathic effects are the germination percentage (or rate) and the initial (seedling) growth of the target species (Table 4).Some studies have also described effects of extracts on the morphology of the target species, the effects on the root growth and differentiation being the most frequently cited.Fewer than 15% of the studies evaluated here described some effect on the shoot parts of the target plants; This might be because the allelopathic activity is more pronounced in the roots, or because less attention has been paid to the effects on the shoot parts of target plants.
In most studies, the effects of the plant extracts have been described as inhibitory of the physiological process under investigation (Table 4).This pattern is also observed in the international literature.These results suggest that allelochemicals present in plant tissues are predominantly inhibitory of physiological processes.However, this highlights the importance of screening for substances that promote germination or the initial growth of plant species.In the present review, less than 8% of the studies selected identified plant extracts or substances that stimulate physiological processes.
A little more than 20% of the allelopathy studies progressed toward the identification of the bioactive compounds or at least toward the identification of the main classes of substances (such as phenolic compounds and alkaloids) present in the bioactive fractions of the plant tested (Table 5).
Based on the present review, we can state that none of the allelopathy studies conducted to date in Brazil have adequately identified the allelochemicals involved in allelopathic interactions.In fact, the majority of the studies reviewed did not describe any kind of procedure carried out to identify allelochemicals or even the classes of compounds present in the bioactive plant extracts (Table 5).

Final considerations
There were many purely qualitative studies that were excluded from this review on the basis of the criteria established for a study to be classified as allelopathic research.In most cases, those studies lacked a clear working hypothesis or justification.Another issue was the use of target plants that are cultivated species, which are sometimes useful for comparative analysis but do not naturally co-exist with the donor species.Consequently, studies using such species yield very little information on the allelopathic processes that might occur under natural conditions.There is little or no correspondence between laboratory and field studies in respect to interactions between plant species.In addition, there is a lack of research on the purification and identification of bioactive molecules and their in situ effects.Furthermore, the effects that soil micro-organisms and mycorrhizal fungi have on allelopathic activity constitute an open question in Brazilian research.Additional research is also needed to explore the allelopathic potential of the Brazilian flora in order to develop new molecular structures to be used in the control of pests and invasive weeds, thus reducing the damage caused by the harmful synthetic herbicides currently in use.Moreover, there is a need for genetic and molecular studies of allelopathic plants, in order to increase their protection against competitors, as well to identify allelopathic genes for the transgenic improvement of crop plants.

Figure 1 .
Figure 1.Number of articles, published between 1991 and 2010 and included in the Scientific Electronic Library Online database (www.scielo.br),citing terms associated with allelopathy studies.

Table 2 .
Plant tissues, substrates, solvents and concentrations used in allelopathic studies carried out in Brazil.Data based on articles included in the Scientific Electronic Library Online database and in the Allelopathy Journal.bioassays do not represent what can be observed in nature.Solutions that are highly concentrated can also generate osmotic effects in the bioassay, possibly leading to confusion with the supposed effects of allelochemicals (Wardle

Table 3 .
Target species used with the greatest frequency in allelopathic studies carried out in Brazil.Data based on articles included in the Scientific Electronic Library Online database and in the Allelopathy Journal.

Table 4 .
Physiological effects most commonly reported in allelopathy studies carried out in Brazil.Data based on articles included in the Scientific Electronic Library Online database and in the Allelopathy Journal.

Table 5 .
Studies having identified the bioactive compounds or at least the classes of compounds present in plant extracts that show allelopathic properties.Data based on articles included in the Scientific Electronic Library Online database and in the Allelopathy Journal.Species, plant part, substrate, target species, effects, substance used, cited in literature for Brazilian studies on allelopathy.Between 2000 and 2012.