Abstracts

FORUM

Trichogramma in Brazil: feasibility of use after twenty years of research

Trichogramma no Brasil: viabilidade de uso após vinte anos de pesquisa

José R.P. Parra; Roberto A. Zucchi

Depto. Entomologia, Fitopatologia e Zoologia Agrícola, ESALQ/USP, C. postal 9, 13418-900, Piracicaba, SP

ABSTRACT

Results of studies with Trichogramma in Brazil are presented, especially those developed at ESALQ/USP in the past two decades (1984-2004). The project involved taxonomy, rearing techniques, biological and behavioral aspects of the pests and parasitoids, pest population dynamics, release techniques, selectivity studies, and efficiency evaluation. It can be considered a model project and has been adopted by other biological control programs in Brazil and Latin America. The program has given rise to a number of publications, allowing the formation of human resources in this area and opening new research areas. The results indicated that the parasitoid can be used to control key pests in cotton, sugarcane, stored grain, vegetables, corn, soybean, and tomato. The perspective of using the parasitoid has stimulated the creation of companies to commercialize it in Brazil, thus more easily transferring this technology to users.

Key words: Trichogrammatidae, egg parasitoid, applied biological control

RESUMO

São apresentados os resultados dos estudos com Trichogramma no Brasil, especialmente aqueles desenvolvidos na ESALQ/USP, nas últimas duas décadas (1984-2004). O projeto, envolvendo desde a taxonomia, técnicas de criação, aspectos biológicos e comportamentais das pragas e dos parasitóides, dinâmica populacional das pragas, técnicas de liberação,estudos de seletividade, avaliação da eficiência, pode ser considerado um modelo e foi seguido por outros programas de controle biológico no Brasil e na América Latina. O programa gerou inúmeras publicações, permitindo a formação de recursos humanos na área, abrindo novas áreas de pesquisa e mostrando que o parasitóide pode ser usado no controle de pragas-chave do algodoeiro, cana-de-açúcar, grãos armazenados, hortaliças, milho, soja e tomateiro. As perspectivas do uso do parasitóides possibilitaram a criação de empresas para comercializá-los no Brasil, transferindo mais facilmente a tecnologia ao usuário.

Palavras-chave: Trichogrammatidae, parasitóide de ovos, controle biológico aplicado

Trichogramma species (Fig. 1) are among the most reared and used natural enemies in the world. Every year, they are released in more than 16 million ha, in annual (for the most part) and perennial crops (Hassan 1997, van Lenteren 2000), even though Smith (1996) had reported their use in 32 million ha. Worldwide, 28 Trichogramma species are released in 28 crops (Hassan 1988). These are among the most frequently studied insects, with several books published about their efficiency in biological control (Wajnberg & Hassan 1994, Parra & Zucchi 1997).

Studies on Trichogramma began in the last century, when Flanders (1927) discovered the possibility of rearing it on a factitious host, Sitotroga cerealella (Oliv.). Since then, there has been great interest in the parasitoid, because of its efficiency and ease of multiplication. In Brazil, studies began in the 1940's to control Neoleucinodes elegantalis (Guenée) in tomato (Gomes 1963). Papers by Moraes et al. (1963) came next, targeted at forest lepidopterans.

In the 20^th century, since the researches by Flanders (1927), several countries, especially the USA, started to use it. However, these initial programs were in general inadequately-planned and isolated projects without inter- and multidisciplinary characteristics. Thus, many errors were committed and the expected efficiency was not achieved because of a lack of knowledge or misevaluation of the following items: (1) egg density of the target pest; (2) Trichogramma species unsuitable for controlling the target pest; (3) quality control of the parasitoid produced; (4) number of parasitoids released, and form of release; (5) pest dynamics and phenology of the plant; (6) competition with other biological control agents; (7) effect of chemical products (selectivity) on Trichogramma in crops where several pests occurred (Parra et al. 2002). These errors led this egg parasitoid to be discredited with regard to its efficiency. However, some important findings in the 1970's, such as the use of male genitalia for species identification (Nagarkati & Nagaraja 1971) and the use of Anagasta kuehniella (Zeller), which was nutritionally more suitable than the moth S. cerealella used until then (Lewis et al. 1976), allowed a reflexion on the use of this parasitoid. With support from IOBC and from Trichogramma work groups, studies were resumed worldwide, under a more modern and adequate view concerning the reality of those times.

Biological control programs for agricultural pests in Brazil began with Jean Voegelé (INRA, Antibes), one of the leaders in studies involving Trichogramma in the 1980's, who taught courses in Brazil, encouraging Brazilian researchers to develop researches with Trichogramma. The interest in developing studies with this egg parasitoid was sparked at ESALQ (Entomology) and, in 1982, the senior author of this work (JRPP) visited the INRA and, in 1984, together with a group of researchers and with the support from a taxonomist (RAZ), the program was started, based on the French model, primarily targeted at the control of Diatraea saccharalis (Fabr.) in sugarcane and Heliothis virescens (Fabr.) and Alabama argillacea (Hueb.) in cotton. This program involved the following stages: collection, identification and maintenance of Trichogramma spp. strains; selection of a factitious host for the mass rearing of the parasitoid; biological and behavioral aspects of Trichogramma spp.; egg dynamics of the target pest; parasitoid release, number of released parasitoids and release points; season and form of release; selectivity of agrochemicals; efficiency evaluation; pest/parasitoid simulation model.

The objective of this paper is to show the advances that occurred 20 years after the program was started, the example followed by other biological control programs, publications produced, human resources formed, new research areas that were initiated in the country, as well as the feasibility to use the parasitoid on different crops in Brazil.

Studies on Trichogramma

Collection, Identification and Maintenance of Trichogramma Strains. Until the 1970's, since Trichogramma was considered nonspecific, individuals of a given species, collected in areas with different climatic characteristics, were used to control pests in geographically distinct regions. Currently, microclimatic specificities are admitted within the same species. For this reason, it is essential to maintain properly labeled, separate strains in the laboratory, in order to ensure the genetic heritage of the initial population.

The presence of a taxonomist is indispensable for this type of program, since an erroneous identification or the lack of it could result in failure of the program. A collection of strains exists at ESALQ, collected from different points of the country. This collection must be supervised by a taxonomist to avoid errors common in the past.

T. minutum Riley was, for a long time, referred as the species parasitizing D. saccharalis in Brazil. Today, it is known that this species does not occur in Brazil (Parra & Zucchi 1986). Due to the minute size of Trichogramma species (0.20 mm), precautions must be taken, such as maintaining the rearing at separate locations; if this measure is neglected, interspecific competition may occur, with predominance of the most aggressive species (Parra et al. 2002).

There are about 200 species of Trichogramma described worldwide (Pinto 1999). In Brazil, 28 species have been recorded, of which one half were described during the studies conducted at ESALQ (Zucchi 1988; Querino & Zucchi 2003 a,b), the characterization of several species based on morphology (Querino & Zucchi 2002 a,b) and on molecular techniques (Ciociola et al. 2001 a,b,c). The most recent list of Trichogramma species in Brazil and their corresponding hosts was prepared by Zucchi & Monteiro (1997). Information about the species contained in the collection at ESALQ is available on the web (Querino & Zucchi 2001).

Selection of Factitious Hosts for Mass Rearing of Parasitoids. Even though several factitious hosts exist for rearing Trichogramma, many researchers still prefer to use S. cerealella, because it is easy to rear, despite being less adequate for the multiplication of the parasitoid in relation to other species, such as A. kuehniella and Corcyra cephalonica (Stainton) (Lewis et al. 1976; Parra et al. 1991, 1997; Gomes 1997; Gomes & Parra 1998; Bernardi et al. 2000). In order to compensate for this lower nutritional quality, T. pretiosum Riley reared on S. cerealella should be released in greater numbers (in relation to those reared on C. cephalonica and A. kuehniella) to control H. virescens in cotton.

Therefore, it is essential to associate the nutritional quality of the host, to produce parasitoids that are competitive with those in nature, with a rearing technique that enables mass production of Trichogramma. Since the Chinese have extensive silkworm rearing at their disposal, they also use Philosamia ricini (Drury) and Antheraea pernyi (Guérin-Méneville) eggs or ovules in Trichogramma rearing.

In general, A. kuehniella has proved to be the most suitable factitious host for the Brazilian species (Parra et al. 1991, Gomes & Parra 1998), although, according to Gomes (1997), C. cephalonica is the best rearing host for T. galloi Zucchi, a predominant parasitoid of D. saccharalis eggs in most of Brazil.

The influence of biotic (mating, oviposition, adult feeding) and abiotic factors (temperature, relative humidity, and photoperiod) was exhaustively studied in a number of papers and book chapters published on the subject (Parra 2002).

Biological and Behavioral Aspects of Trichogramma spp. Basic studies were conducted for the main Trichogramma species collected in Brazil, such as strain selection, temperature requirements for laboratory production and for insect adaptation in the field, humidity requirements, determination of the most suitable parasitoid:host eggs ratio, adaptation to the factitious host (number of generations to achieve it), parasitism capacity, and behavior. In recent years, the findings that Trichogramma has only a single instar (Volkoff et al. 1995, Dahlan & Gordh, 1996) and the influence of symbionts, especially Wolbachia, on the sex ratio of Trichogramma (Stouthamer et. al. 1990), have been taken into account without, however, forsaking the classic papers by Lund (1934), Doutt (1959), Buttler & Lopes (1980), and Calvin et al. (1984), among others.

Quality control must be a constant procedure in laboratory populations, as well as the periodic introduction of wild populations and even rearing the parasitoid on the natural host after a number of generations have been reared on the factitious host. T. pretiosum populations can withstand high inbreeding rates in laboratory rearing, without showing evidence in their biological characteristics of degenerative reflexes that would compromise their quality (Prezotti et al. 2004).

Egg Dynamics of Target Pests. This has been one of the least-studied items in our country, albeit essential for defining the season when the parasitoid should be released, i.e., at the onset of the population of adults and, consequently, of eggs. There have been attempts to use the plant's phenology as groundwork for parasitoid release (Lopes 1988); all other studies are based on the pest's dynamics with the use of pheromone – E. aurantiana (Lima) – or through the on-site evaluation of the plant structure that harbors eggs (most crops).

Parasitoid Release: Numbers, Places, Seasons and Ways. Once the release season has been defined, the parasitoid's dispersal ability must be evaluated (Lopes 1988, Sá et al. 1993, Zachrisson & Parra 1998) to determine the number of release points. The form of release can be very simple, such as releasing the emerged adults from plastic or glass containers by walking through the field, or in a more sophisticated manner, by airplane, using starch capsules (biodegradable) which allow the parasitoids to exit but prevent predation (patented by the INRA, France). In Brazil, "Bug Agentes Biológicos" (an insect-selling company) has developed a similar capsule, which has been largely used. One of the great problems in tropical countries is the predation of pupae after release. Thus, the action of predators must be taken into account when Trichogramma is released in the form of pupae (Pinto 1999). Releases using a center pivot were used in Brazil for T. pretiosum in tomato (Haji et al. 2002).

The number of parasitoids to be released must be defined in laboratory, semi field, and field tests. In sugarcane, Lopes (1988) estimated the proportion of 1.6 parasitoid (Trichogramma spp.) per egg of the pest as ideal. Sá (1991) found a ratio of 10.7 T. pretiosum per egg of Helicoverpa zea (Boddie) in corn, while Zachrisson (1997) found a 5.3 ratio of T. pretiosum per Anticarsia gemmatalis (Hueb.) egg in soybean. In perennial crops, this number can be quite higher (36 parasitoids per egg for the citrus fruit borer, E. aurantiana) (Molina 2003). In general, the literature has recorded, in citrus and other fruit crops, releases varying from 70,000 to 3.8 million parasitoids/ha, or 9,000 to 50,000 parasitoids/plant (Oatman & Platner 1985, Hassan et al. 1988, Newton & Odendaal 1990, Glen & Hoffmann 1997, Mills et al. 2000). In many countries, fixed numbers of parasitoids are released for the sake of ease, regardless of the existing population of the pest. This could be one of the reasons for the lack of success of the parasitoid.

Egg parasitism by Trichogramma on different days in the same release card can extend the period of action of the parasitoid (Pinto 1999).

Selectivity of Agrochemicals. In crops that have a large number of pests and require the application of agrochemicals, selectivity studies must be done to establish an interval between parasitoid release and the application of such products. These selectivity tests must be based on IOBC rules to allow comparisons with other countries (Hassan 1997). Only a few papers on selectivity have been done in Brazil (Foerster 2002, Degrande et al. 2002).

Efficiency Evaluation. Efficiency and cost must be compatible with the crop and comparable to traditional control methods. Thus, a crop such as tomato, with high profitability, allows up to 10 Trichogramma releases per crop cycle. Other crops with a lower profitability, like sugarcane, do not allow more than 3-4 releases in most years.

Pest/Parasitoid Simulation Model. Parasitoid-pest simulation models are developed on specialized software programs, based on biological data about the pest and its natural enemies. These models for Trichogramma can be developed in Brazil, because the number of basic researches involving this parasitoid along these 20 years has been expressive. However, lack of simulation specialists prevents us from having these models at our disposal.

The need of a group of specialists in order to develop an applied biological control program is evident. Thus, a taxonomist, a biologist, an ecologist, a specialist on agrochemicals, an entomologist-economist, and a computer science specialist must work together. Evidently, in this case, the chances of obtaining success are much higher than if the program is developed only by an entomologist with a general formation.

Studies with Trichogramma in Brazil

The research on Trichogramma has spread throughout Brazil, resulting in the appearance of other study groups involved with this subject. However, except by some cases, like the use of T. pretiosum to control Tuta absoluta (Meirick), the releases on cotton by Embrapa Algodão, in Paraíba State, the sporadic use of T. atopovirilia Oatman & Platner and T. pretiosum to control Spodoptera frugiperda (J.E. Smith) in corn and Plutella xylostella L. in cabbage, the project has not reached large areas, due to the difficulty in transferring the technology, and particularly to the lack of good quality insects available for the farmer.

The volume of information and results are very interesting and liable to be used in crops such as cotton, soybean, sugarcane, tomato and other vegetables, corn, stored grain pests, etc. In addition, Garcia (1998) and Molina (2003) demonstrated the potential of use of T. pretiosum in citrus to control E. aurantiana (Lima), the citrus fruit borer; in avocado, the parasitoid is being studied to control Stenoma catenifer Wals. (Hohmann & Meneguim 1993); in agricultural crops, T. pretiosum, T. atopovirilia, and T. galloi have shown the greatest potential of use in our country.

Corn. In both sweet and commercial corn cultivars, results are better on H. zea control, since the egg-laying scales of S. frugiperda, disposed in layers, make parasitism difficult (Sá 1991, Beserra 2000). Trichogramma releases have been recommended in areas attacked by S. frugiperda (Cruz et al. 1999), with good results.

Several basic studies have been performed to control H. zea with T. pretiosum (Sá & Parra 1994a), with natural parasitism by T. pretiosum reaching 95% (Sá & Parra 1994b). When releases are recommended at the rate of 11 parasitoids per H. zea egg, and taking into account the parasitoid's dispersal, such releases should be done at a great number of points per hectare (Sá et al. 1993). Three releases of 100 thousand T. pretiosum adults/ha reduce 26% of H. zea damage on corn (Sá 1991). Rivero (1992) also conducted basic researches with H. zea versus T. pretiosum, indicating that the parasitoid could yield from 4.8 to 8.5 times more generations than the corn earworm. Beserra (2000) observed that egg parasitism in S. frugiperda is low and concentrated at the bottom and medium parts of the plant. Although occurring in smaller numbers in the field, T. atopovirilia shows higher parasitism capacity than T. pretiosum. According to this author, the presence of scales on S. frugiperda egg masses is the main factor that contributes to decrease T. atopovirilia and T. pretiosum potential of parasitism. Also, the distribution of eggs in layers prevents egg parasitism in the lower parts by T. atopovirilia (Table 1).

Cotton. Entomologists at Embrapa Algodão, in Campina Grande, PB, have conducted excellent works with T. pretiosum to control the cotton leafworm, A. argillacea. The potential to use T. pretiosum has also been demonstrated for H. virescens (J.R.P. Parra, unpublished). At Embrapa, a system adapted from the Colombian model for rearing the parasitoid on S. cerealella eggs has been adopted (Almeida 1996).

The pioneering works in Brazil were conducted by Bleicher (1985) involving three T. pretiosum populations aimed at controlling A. argillacea, with parasitoids reared on A. kuehniella.

Life table results confirmed the biological data, and the T. pretiosum population from Iguatu, CE, was the most aggressive (Bleicher & Parra 1990a, 1991). Temperature requirement studies were also carried out, showing that the biology of T. pretiosum is influenced by the collection locality (origin). Between 2.7 and 2.9 Trichogramma generations were observed for each generation of A. argillacea (Bleicher & Parra 1989, 1990b). The parasitoids' longevity was higher when they were given a chance to parasitize and when they were fed pure honey.

Sugarcane. The bulk of the information about Trichogramma is concentrated on this crop. Some of the studies deal with: identification of predominant species in Brazil – especially T. galloi Zucchi and T. distinctum Zucchi – by conventional (Zucchi et al. 1991, Zucchi & Monteiro 1997) and molecular methods (Ciociola Jr. et al. 2001a,b); temperature (Parra et al. 1991) and humidity requirements (Parra & Sales 1994); most suitable factitious host (Gomes & Parra 1998); different developmental stages of the parasitoid's biological cycle (Cônsoli et al. 1999); spatial and temporal distribution of D. saccharalis eggs and their parasitism by T. galloi (Micheletti 1987); effect of natural and fastitious host egg age on the development and parasitism of T. galloi and T. distinctum (Lopes & Parra 1991); effect of constant and fluctuating temperatures on the development and parasitism of T. galloi (Cônsoli & Parra 1995a,b); photoperiod effect on the biology of T. galloi (Cônsoli & Parra 1994); artificial infestation methodology of D. saccharalis eggs for studies with Trichogramma (Lopes et al. 1989); determination of the parasitoid's range of action (10 m) and number of T. galloi to be released per D. saccharalis egg (1.6:1) (Lopes 1988); and efficiency of T. galloi individually or in association with C. flavipes in field evaluations (Botelho et al. 1999). These were followed by parasitism studies of T. galloi in sugarcane varieties, conducted at different row spacing (Botelho et al. 1995a,b). Studies on the selectivity of chemical products to T. galloi were also conducted (Cônsoli et al. 2001), as well as studies on release techniques (Pinto 1999).

Field results were outstanding; the association of one release of Cotesia flavipes (Cameron) with three releases of T. galloi was the most efficient, resulting in a reduction of the infestation intensity of 60.2% relative to the control (Table 2). This treatment (three releases of 200 thousand T. galloi per week and one release of 6 thousand C. flavipes) was much superior to the control, i.e., better than C. flavipes individually, because in this case, the reduction in relation to the area where releases were made was only 16.1%. These results attest the feasibility of using T. galloi, especially in areas where egg predation is low and C. flavipes is not well adapted.

Soybean. In some regions in Brazil, egg parasitism of A. gemmatalis by T. pretiosum is sometimes higher than 90% (Zachrisson 1997). This author showed that the proportion of 5.3 parasitoids per A. gemmatalis egg allows a high parasitism rate, regardless of the phenological stage of soybean.

Aspects related to temperature requirements, plant phenology, parasitism, and strain effect on parasitism, among others, were also studied. The dispersal ability of T. pretiosum to parasitize A. gemmatalis eggs in 24h is 8 m and corresponds to a dispersal area of 77 m² (Fig. 2). Thus, the release of T. pretiosum to control A. gemmatalis should be performed in 130 points per ha, with a control efficiency of 64.8% (Zachrisson & Parra 1998). Its potential to control A. gemmatalis is rather great, since five Trichogramma species have already been collected parasitizing this pest (Foerster & Avanci 1999, Avanci 2004).

Stored Grain. Studies conducted by Inoue (1997) demonstrated the potential of T. pretiosum to control S. cerealella in stored corn, both in bulk and on the ear. The storehouses, being stable ecosystems, experience high temperatures, what favors parasitism, as reported by Inoue & Parra (1998), who found 97.6% of the females parasitizing at 30ºC.

The parasitoids released in the grain mass are capable of parasitizing S. cerealella eggs at depths of up to 40 cm, and the percentage of parasitism decreases, on average, 1.9% for each cm of depth in the corn grain mass. For greater control efficiency, the release should be made in the beginning of the attack by the moth, at a rate of 12 parasitoids per S. cerealella egg. For ear corn, releasing T. pretiosum is efficient to control S. cerealella, reducing the population of adult moths by 60.7% and the damage caused to stored ears in screened warehouses by 63.1%. The association of T. pretiosum with Bracon hebetor Say did not result in advantages in the control of S. cerealella in comparison with the release of T. pretiosum alone; however, numerically there is a tendency for greater efficiency when both species of parasitoids work in association (Inoue 1997) (Table 3).

Tomato. Studies with the moths – T. absoluta and Phthorimaea operculella (Zeller) – and the borer – H. zea and N. elegantalis – in tomato have shown the feasibility to use T. pretiosum in this vegetable crop. Excellent results have been obtained in the greenhouse, in sprawling, and in staked tomatoes.

In the greenhouse, results are very good for T. absoluta, but it is necessary to select strains to control N. elegantalis, the tomato fruit borer; weekly releases are recommended, at the rate of 6.4 parasitoids per T. absoluta egg, based on leaflet samplings. A control of 87% with T. pretiosum releases in relation to the 42% of control was obtained with the growth regulator lufenuron (J.R.P. Parra, unpublished).

In practical terms, the excellent results obtained at Vale do São Francisco, Pernambuco State, in the control of T. absoluta atest the feasibility to use this parasitoid. There are many other basic studies with T. pretiosum for the control of P. operculella and T. absoluta, based on biological studies (Pratissoli & Parra 2000a), fertility life table (Pratissoli & Parra 2000b), and on parasitism capacity (Pratissoli 1995). It has also been demonstrated that cluster analysis is a suitable method for selecting Trichogramma strains (Pratissoli & Parra 2001).

Several selectivity studies on products used in the crop have been conducted (Cônsoli et al. 1998, Carvalho et al. 1999). Papa (1998) recorded excellent results in staked tomato by releasing 800 thousand parasitoids (T. pretiosum) per ha, with a control efficiency similar to the use of insecticides.

For H. zea in processing tomato, Moreira (1999) achieved a control efficiency of 83% in the field, releasing 400 thousand T. pretiosum parasitoids per ha, at weekly intervals, with variable results depending on the strain's origin.

Final Considerations

Twenty years after a steady flow of Trichogramma studies in Brazil began, there are still some problems to be solved. One of them is the availability of insects for the farmer, which is beginning to be resolved presently as insect-selling companies arise, similarly to what is happening in other countries in the world. Even field assays, which for many were considered insufficient, will be improved with such availability of insects. One of the great worldwide hindrances to the development of biological control programs is the technology transfer provided to the user; this worldwide problem could be solved by the above-mentioned companies themselves. It becomes clear that, in these cases, production scale changes would have to be implemented as compared to laboratory research needs. For this reason, it is essential that constant quality control should be maintained by insect suppliers, always under the supervision and follow-up of Research Institutions and Universities.

However, the high volume of information on Trichogramma species, obtained from books, bulletins, congress abstracts, dissertations, theses, and national and international periodicals, has allowed great advances in the area of biological control and, especially, because this study model has served as a basis for researches with other parasitoids and predators in Brazil. Trichogramma study groups were formed in several states, such as Espírito Santo, Rio Grande do Sul, Paraná, Rio de Janeiro, Mato Grosso, Santa Catarina, Paraíba, Pernambuco, Minas Gerais, and groups also found motivation elsewhere in Latin America countries, such as in Chile, Uruguay, Argentina, Paraguay, and Panama.

Concurrently (and in this maybe consists the high proceeds of all basic research that has been developed), cutting-edge studies were conducted in the area of "in vitro" production of Trichogramma (Parra & Cônsoli 1992, Cônsoli & Parra 1999, Cônsoli & Parra 2002) and other parasitoids (Magro & Parra 2004), placing Brazil in the vanguard of this type of study, as a leader in Latin America, and putting the country at a level playing field with developed countries. These studies will enable advances in the area of hosts/natural enemies' relations, allowing greater development in a short period of time, including rearing techniques for parasitoids and predators.

Although still timidly, more intense use of T. pretiosum is being made to control T. absoluta in tomato, both in field areas and under protected cropping, P. xylostella in cabbage, based on results by Barros (1998), and in corn areas with T. atopovirilia and T. pretiosum to control S. frugiperda, though in this case with the previously indicated limitations. In Brazil, between 5 and 10 billion Trichogramma wasps are produced annually, and released in 60,000 ha of corn and in 1,000 to 1,200 ha of tomato and crucifers (cabbage), with a perspective for significant increase in a very near future.

The potential of use in sugarcane, cotton, vegetables, fruit trees, and soybean, among others, as well as the other advantages indicated, give a dimension of the work that has been performed and the progress that has been accomplished along 20 years of studies with Trichogramma spp. in Brazil.

Literature Cited

Publication Dates