Abstract
To evaluate the influence of Beauveria bassiana (Balsamo) Vuill and Metarhizium anisopliae (Metsch) Sorok on the survival of the predator Euborellia annulipes Lucas (Dermaptera: Anisolabididae) were used product concentrations 5.00x109; 7.50x109; 10.00x109; 12.50x109 conidia/L from isolates of B. bassiana and M. anisopliae which were obtained, respectively, through products Boveril® and Metarril® and control (0.0 conidia/L). High rates of hatching nymphs of E. annulipes have been recorded in clutches that received applications of M. anisopliae. Moreover, B. bassiana has affected negatively the hatching rate of nymphs of E. annulipes. A slightly harmful effect has been observed in all concentrations of conidia B. bassiana in clutches of E. annulipes. The survival was 100% in all tested concentrations of M. anisopliae in females of E. annulipes. The females which received topical application of B. bassiana have shown a survival rate ranging from 80.30 (12.50x109 conidia) to 100% (5.00x109 conidia). In males of E. annulipes, the survival rate in insects treated with B. bassiana ranged from 95.00 (12.50x109 conidia) to 100% (5.00x109 conidia), while with M. anisopliae applications this rate was 96.02 (12.50x109 conidia) to 100% (5.00x109 conidia). The entomopathogenic fungi B. bassiana and M. anisopliae did not affect the mortality of nymphs and adults of E. annulipes. However, careful measures should be adopted in applications of B. bassiana directly into clutches of E. annulipes.
Introduction
Earwigs or Dermaptera constitute several species with high predation capacity due to their aggresive behavior. Among these species, Euborellia annulipes (Lucas, 1847) (Dermaptera: Anisolabididae) is an important biological control agent of pests in agriculture. This insect species occurs throughout the neotropical zoogeographical region and has been an potential predator of several insect pests, particularly of eggs and immature stages of insects of the Lepidoptera, Hemiptera, Coleoptera and Diptera (Lemos et al. 1998, 1999). In northeast Brazil it is common occurrence of this insect preying on immature stages of pests that occur in soil or in plant structures lying on the ground as case of the larvae and pupae of boll weevil at field of cotton.
Entomopathogens are prevalent in natural systems and should receive greater attention in life-history studies (Hesketh et al. 2010; Roy et al. 2009). Mycoinsecticides compose only a small component of the biopesticide market, but their use is gaining acceptance as an alternative to the use of traditional contact insecticides for pest control (Ludwig and Oetting 2001). Beauveria bassiana (Balsamo) Vuill and Metarhizium anisopliae (Metsch) Sorok are among the most studied species of entomopathogenic and are used as biocontrol agents of agricultural pests (Alves 1992; Alves et al. 1998). In biological control of pests an important pathogen is that one who affects only the target insect and not their natural enemies (Flexner et al. 1986; Generoso 2002).
The severity of the effects of microbial insecticides depends on the concentration of spores applied and the stage of the insect (Ginsberg et al. 2002). The use of low concentrations of alternative insecticides may increase the selectivity and help reduce costs of production (Oliveira 2008). In recent studies conducted by Oliveira et al. (2010), it was observed that low concentrations of mycoinsecticides isolates of B. bassiana and M. anisopliae promoted high efficiency control of larvae and pupae of fruit flies. However the direct effects of entomopathogenic fungi in natural enemies have not been well studied and probably have been overestimated. It is also necessary to develop techniques to express accuracy in doses of microbial insecticides to achieve maximum physiological compatibility. Due to the increased use of biological control in integrated pest management and considering that information regarding the selectivity of entomopathogenic fungi in biological control agents is incipient, the aim of this research is to evaluate the influence of M. anisopliae and B. bassiana in the survival of the predator E. annulipes.
Material and Methods
Results and Discussion
In our test with C. capitata, it has been recorded a mortality rate approximately 98% and 97% caused by fungi M. anisopliae and B. bassiana, respectively. Therefore, the survival of E. annulipes was not because the lack of fungi virulence.
High rates of hatching nymphs of E. annulipes have been recorded in clutches that received applications of M. anisopliae. The hatching percentage nymphs confirmed was 96.23, 92.47, 91.58 and 91.50% in the concentrations of 5.00x109, 7.50x109, 10.00x109 and 12.50x109 conidia/L of M. anisopliae, respectively. Moreover, B. bassiana affected the hatching rate of nymphs of E. annulipes, the survival ranged from 31.13 (12.50x109) to 60.06% (5.00x109) (Fig. 1). Results obtained by Duso et al. (2008) with the predatory mite Phytoseiulus persimilis (Athias-Henriot, 1957) (Acari: Phytoseiidae) have shown high hatching rate of this predator species after application of B. bassiana, although this researchers have reported mortality of up to 43% in females.

Hatching percentage Euborellia annulipes nymphs with conidia concentrations/L of Metarhizium anisopliae (
Ma) (y=0.92x2-6.108x+101.32. R2=0.98) and Beauveria bassiana (
Bb) (y=0.4408x2-11.188x+103.55. R2=0.91). Observed ± S.E. (dots) and predict results (line models).
In the other hand, M. anisopliae and B. bassiana did not affect the survival of nymphs. There was overall survival (100%) of nymphs of 1st and 2nd instars in all tested concentrations of both the fungi B. bassiana and M. anisopliae (data not shown in figures). Results obtained by Kohno et al. (2007) are in agreement with those obtained in this study. These authors verified no effect of B. bassiana to nymphs of 2nd instar Labidura riparia (Pallas, 1773) (Dermaptera, Labiduridae).
The fungi M. anisopliae and B. bassiana did not affect the survival of adults of E. annulipes. In all tested concentrations of M. anisopliae in female E. annulipes, survival was 100%. The females which received topical application of B. bassiana have been shown a survival rate ranging from 80.3% (12.10x109 conidia) up to 100% (5.00x109 conidia); whereas, for males of E. annulipes, the survival rate in insects treated with B. bassiana ranged from 95% (12.10x109 conidia) to 100% (5.00x109 conidia) and from 96.02% (12.10x109 conidia) to 100% (5.00x109 conidia) after receiving applications of M. anisopliae (Figs. 2a and 2b).

Relationship between the application of different concentrations of conidia of fungi Metarhizium anisopliae (
Ma) (y=0.025x2-0.045x+96.375. R2=0.96) and Beauveria bassiana (
Bb) (y=0.004x2-0.002x+96.395. R2=0.976) on the survival of males of E. annulipes A. and the different concentrations of conidia of fungi M. anisopliae (
Ma) (y=100) and B. bassiana (
Bb) (y=0.0016x2-0.0744x+80.916. R2=0.958) on the survival of females of E. annulipes B. Observed ± S.E. (dots) and predict results (line models).
Efficient selectivity is based on physiological differences between pests and their natural enemies, being that pest killed at a concentration of product that does not affect individual benefits. A slightly harmful effect (survival> 30% and <70%) has been observed in all the concentrations of conidia B. bassiana in clutches of E. annulipes (Fig. 1). The microbial insecticides tested were not harmful (survival> 70%) for adults (Figs. 2A and 2B) and nymphs of first and second instar. Therefore, the two mycoinsecticides can be used to control pests without significantly affecting nymphs or adults of E. annulipes. The greater susceptibility of eggs of E. annulipes in relation to nymphs or adults should be the interaction that occurs due to pathophysiologic effects on the cuticle of nymphs or adults of the predator, which hindered the process of infection and lessens the potential for invasion of microorganisms for these stages of E. annulipes. Moreover, the ability of this insect to survive in areas of land, soil inhabitant galleries can to be an important factor to provide a greater hardening of their cuticle. Other factors such as intestinal pH, biochemical or immunological defense-physiological may be related to the tolerance of nymphs or adults of E. annulipes concentrations of mycoinseticidas used in this study.
According Kiselek (1975), B. bassiana was harmless to adults Cryptolaemus montrouzieri Mulsant, 1853 (Coccinellidae: Coleoptera) while the effect was lethal to larvae (50%) that consumed insects infected by this fungus. In this study, nymphs and adults of E. annulipes were not affected by concentrations tested in this study adopted of B. bassiana and M. anisopliae. In other words, the results of this study have revealed that B. bassiana caused a reduction in the rates of hatched nymphs, unlike M. anisopliae. Moreover, insecticide applications that are not directly to the soil may enable the adoption of that microorganism in programs of integrated pest management in agroecosystems in which E. annulipes is present, because this kind of predator tends to deposit their eggs in soil galleries built by adults (Bharadwaj 1966).
Other studies have also demonstrated the selectivity of entomopathogenic organisms to natural enemies, like B. bassiana with Orius insidiosus (Say, 1832) (Hemiptera: Anthocoridae) (Cavalcanti 2006); M. anisopliae and B. bassiana with parasitoid Oomyzus sokolowskii (Kurdjumov, 1912) (Hymenoptera: Eulophidae) (Santos Jr. et al. 2006). The concentrations adopted in this study of entomopathogenic fungi B. bassiana and M. anisopliae did not affect the mortality of nymphs and adults of E. annulipes. Neither we found ovicidal activity of M. anisopliae to E. annulipes. However, careful measures should be adopted in applications from 5.0x109 conidia/L of B. bassiana directly into clutches of E. annulipes, because from this concentration we found a significant reduction in hatching rate this predator. We also suggest the implementation of other studies, as the effects these bio-insecticides on the fitness of E. annulipes.
Footnotes
Acknowledgements
The authors thank the Brazilian agency Coordenadoria de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) to concession scholarship to first author. The authors would like to thank Robério de Oliveira for his help in the laboratory. We especially thank Tricia Hornsby (Ohio State University) for her valuable help with an important review of English in this manuscript.
