POTENTIAL ENTOMOPATHOGENIC FUNGI TO CONTROL SCALE INSECT PEST ON CITRUS TANGERINE ( CITRUS SUHUIENSIS TAN . )

Achiving of food self-sufficiency can be done by using of local potential that is by agribusiness in Indonesia. One potential locally owned citrus agribusiness was the use of entomopathogenic fungi to improve the productivity of citrus. Reports showed decrease in productivity due to infestation of scale insect. The experiment was conducted at the Integrated Laboratory of Indonesian Citrus and Subtropical Fruit Research Institute in October 2013 to October 2014. The study began with a survey for scale insect infestation on citrus crops in high land, medium land and low in dry and rainy seasons. Taken from a collection of entomopathogenic fungi associated with scale insect in the field. Collection of fungi isolated from single conidia and its ability to infect selected scale insect. Entomopathogenic fungi were further tested for the viability and pathogenicity against scale insect. The results showed that the sclae insects attacked citrus were types of L.beckii and A.Aurantii. The highest attack occurred at low land during the dry season by L.beckii with population of 4.2 heads increased to 5.5 individuals per 10 cm in the rainy season. Viability test results showed that the isolates had viability above 50% were SKB4K, SKD1K and SBB3K for 73.6, 61.6 and 53% respectively, which were collected during the dry season. While isolates obtained in the rainy season were SBWD2H and SBWD3BH, each with aviability of 77.3 and 78.3% respectively. Pathogenicity test results showed that there were 6 isolates known to have potential as entomopathogenic fungi for controlling scale insect, namely, SBWB2H, SBWD2H, SBWD3BH, SKD1K, SBWD1K and SBB3K which had pathogenicity over 50% up to 14 days.

This study aims to (1) determine the type and population of scale insect that attack citrus plants on three types of elevation during dry season and rainy season. (2) Selection, viability testing, and pathogenicity of fungal isolates against scale insect.

MATERIALS AND METHOD of research
Research conducted at the Integrated Laboratory of Indonesian Citrus and Subtropical FruitsResearch Institute (ICSFRI), Tlekung, Batu. The research was conducted in the dry season and rainy between October 2013 and October 2014. Location forobservation of scale insect population and sampling of fungi associated with scale insectat highland was Bangli Regency, Bali;at medium land was Banyuwangi Regency, East Java; and at lowlandwasSambas Regency, West Kalimantan.
Isolation of fungi associated with scale insect andpropagate single conidia. Technique for isolationof fungiassociated with infestation of scale insect was performed per method of Liu et al. (2005 and2006) and del Prado et al. (2008). Samples of scales section infested with the fungus was sampled using a needle loop and inoculated into petri dishes containing PDA (Potato Dextrose Agar) medium comprising of 50 mg / L teramisin. Fungal cultures incubated at a temperature of 25-30 ° C for approximately 7-14 days until fungal colonies filled the cup. Each colony of fungi that grow was subsequently rejuvenated into a PDA medium and incubated to produce conidia period.
Rearing scale insect. Propagation (rearing) of scale insect used Banjar tangerine as host plant, planted in polybags diameter of 30 cm. 20-30 heads of first instar larvae (crawlers) of brown scale insect still actively moving (aged 1 day) were taken from a citrus plant having severe infestation of brown scale insects and inoculated onto each green leaf. The edges of leaves inoculated with crawler were limited by using wet tissue paper folds to prevent the leaving of the larvae of the leaf. The larvae of brown scale insectwere reared in a screen house at a temperature of about 25 o C until imago reached 30 days and ready for treatment.
Selection of entomopathogenic fungi. Selection of entomopathogenic fungiwas done against isolates that have the phenotype of entomopathogenic and have the ability of Lethal Concentration 50 (LC 50) within 14 days. Pure fungi were isolated on PDA then incubated at 25-30 o C for ± 30 days or until colonies filled the petri dish. The numberof conidia was calculated by Haemocytometer to achieve the density of 107 conidia / ml. Suspension was aseptically put into handsprayer using a micropipette. The percentage of mortality was calculated by formula (Wahyono and Tarin, 2007).
Viability of entomopathogenic fungi conidia. Results obtained from the selection that had the ability above LC 50 at 14 days, meaningthat they were pathogenic against scale insect, continued counting the viability of the fungi. Fungal colonies fromselection results were grown on PDA to fulfill the cup. Conidia were harvested by adding 10 ml of sterile distilled water containing 0.02% Tween 80 into the cup to form a suspension containing cultured conidia masses. The number of conidia on mass suspension was counted with Haemocytometer until reachingdensity of 102 conidia / ml by serial dilution then 0.1 ml was taken using a micropipette and spread onto PDA medium surface in petri dishes by using spread plate method and flattened with dryglassky (Alves et al., 1998 cit. Francisco, 2006). Conidia in PDA medium incubated at room temperature for 24 hours to form conidia germination (Skrobek, 2001).
Fungus pathogenicity testagainst scale insect. Pathogenicity test wasmerely done to isolate of selection results. The experimental design used in the study was Random Block Design factorial with three variables. The first variable was fungal isolates; the second variable was density of application conidia;and the third variable was days of observation. Replications were three times. Total conidia calculated by Haemocytometer. Dilution was to obtain conidia density of 102-107 conidia / ml. Suspension was aseptically put intohandsprayer using a micropipette. The percentage mortality was calculated by formula (Wahyono and Tarin, 2007).

RESULTS AND DISCUSSION
Survey result of scale insect population on tangerine. Scale insect that attacked citrus plants was a type of L. beckii and A. Aurantii. The highest attack occurred on the stems of citrus grown in the lowlands with the scale insect population of L. beckii was 4.2 heads per 10 cm rod (Figure 2). High scale insect populations in the lowlandssupposedly linked to conditions of high humidity on the tangerine plant, where the sample collected in the district of Sambas, West Kalimantan generally known of having rainfall throughout the year for 12 months.
L. beckii liked the dense tree canopy, and severe attacks usually occur in the central part of the tree canopy (Futch et al., 2001;Knapp, 2003;Triwiratno, 2004;Anonymous, 2007). The attacks on the stems, leaves and fruits found on the plant in the field caused typical symptoms of damage on the surface and the appearance of dotted and dull.
Imago L. beckii is dark brown with varied shapes that was long, circular and coma. Scale insect generally has a size of 1.0 to 3.0 mm and has a sort of shield on his back. Scale insect reproduce sexually or parthenogenesis. Most female scale insects can produce 40-80 eggsand placed in groups around the body that will hatch on the eighth day after the egg is produced. In the dry season, the eggs hatch in 15-20 days, while in the rainy season, hatching time is longer (Fasulo and Brooks, 2004). Crawler of scale insectwas white and runs very slowly, usually found on the stems and leaves sidelines. Crawler can survive for three days without food nutrients and can move only a few hours,then settled on a part to develop into adults (Grafton et al., 2000).  Female insect changed skin twice, while the male had four skin changes before they reached the adult stage with wings. In a year, usually there are three or more generations. Scale insect can survive on host plants by sucking fluids from the leaves, fruit, branches, and stems of its host plant, causing chlorosis, leaf drop, incompleted ripening, abscission of fruits, dried branches and plant death (Fasulo & Brooks, 2004).
Observations of the attacks carried out in rainy season was in conjunction with the fruit began to grow. The existence of new growing fruitsstimulated transfer of the crawler to move to get a young plant parts that will be used as a place to live when he became imago. The highest scale insectpopulation was L. beckii at the lowlands and the medium that was between 5 to 5.5 heads per fruit ( Figure 2). While the population on the stems and leaves was lower.
A. Aurantii female mite had a hard body covering, round-shape and was maroon. While the body coveringof the male larvae was oval and smaller. Body shield of the of the female adult had a diameter between 1.5-2 mm. The larvae were brown with a very small body size (Amitaningsih, 2005). According Efendi (2009) Adult female was oval, had diameter of 2 to 2.3 mm, was spherical orange or dark brown, and produce 60-150 crawlers (first instar larvae were active).  Mortality data from the selection trials resulted four species isolates that have the highest pathogenicity, namely, SK D1 K (Sambas), SBW D1 K (Banyuwangi), SB B3 K (Bali) and SK B4 K (Sambas) with a mortality rate of L, beckii was 78.4%, 76.9%, 53.3% and 52.7% respectively in 14 days. Other isolates werevonly able to control L, beckii less than 50% for 14 days (Figure 3). Isolates obtained from the same host insect but of different topographic had different virulence (Fatiha et al, 2007). This was shown from L. beckii mortality data produced by seven different isolates. SK D1 K isolates originating from the lowlands had the highest degree of pathogenicity compared to other isolates. This was because the population of host plant in the lowland was better. In addition, the environmental conditions at the low landcould create the characters of the physiology of fungi that were more virulent than the others originating from the highlands and medium.  (Figure 4). Alavo et al, (2004) asserts that the host range and ecological conditions can influence the genetic diversity that directly affect the virulence of a fungus power. Zhen et al, (2005) states that the virulence of entomopathogenic fungi is influenced by the character of physiology. Meanwhile, physiology character of entomopathogenic fungus closely related to the rate of growth, sporulation, conidia germination and tolerance to temperature differences.
According Prayogo (2006) the concentration of fungi with conidia density of 107 conidia / ml is the standard concentration in testing biological products. Prayogo & Marwoto (2005) also declare that the minimum dose of conidia fungal pathogens that can lead to death of insects is 103 conidia / ml. Previous research has shown that B. bassiana entomopathogenic fungi on the density of 107 conidia / ml could infect termites up to 100% (Desyanti et al., 2005). Viability of selection result of entomopathogenic fungiin dry season and rainy season. A total of seven isolates of the selection result, namely, SB B2 K, SB B3 K, BSD2, SBW D1 K, SBW D2 K, SK B4 K and SK D1 K ( Figure 5) were tested for conidia viability to identify the speed of conidia germinated within 24 hours. The ability of fungi conidia to germinate within a certain time could be seen from the level of its viability. The higher the percentage of viability, the shorter the germination time required. Different isolates types had a significant influence on the value of the viability of conidia of the fungal isolates (P <0.05). Of the seven tested isolates, there were three types of isolates that hadthe highest level of viability of fungal isolates conidia, namely,SK B4 K, SK D1 K and SB B3 K of 73.6%, 61.6% and 53%respectively,while four other isolates had a percentage viability of less than 50% ( Figure 5). Viability of isolates conidia ofSB B3 K was relatively lower than isolates SK D1 K, even though the two isolates were statistically not significantly different (P> 0.05). Based on the results of previous studies, the percentage of fungal conidia viability could reach 90% -100% (Rahayu, 2009). The highest viability was achieved by SBW D2 H and SBW D3B H with viability of 77.3% and 78.3% respectively ( Figure 6). In this viability test, fungal isolates were incubated at room temperature of 25 ° C. The optimum temperature for growth, pathogenicity and survival of entomopathogenic fungus was around 20 -30 o C (Morissey & Osbourn, 1999). Viability of entomopathogenic fungus spore was influenced by temperature, humidity, pH, solar radiation and chemicals, such as nutrients and pesticides (Muller Kongler, 1967 in the Robert & Yendol, 1971;Riyanto, 1993). In this viability test, fungal isolates were incubated at room temperature of 25°C. The optimum temperature for growth, pathogenicity and survival of entomopathogenic fungus was around 20 -30 o C (Morissey & Osbourn, 1999).
Pathogenicity of entomopathogenic fungi from tangerine host plants. A total of six isolates were tested for pathogenicity to determine the level of virulence. All three isolates were selected based on the selection test with mortality, and the highest conidial viability were SBW D1 K, SK D1 K, SB B3 K, SBW D2 H, SBW B2 H, and SBW D3 BH with mortality of L. beckii was 81.1%, 73.6%, 68.8%, 69%, 72%, and 54% respectively, for 14 days at a concentration of 107 conidia / ml. Observation of the first day of treatment showed that all concentration treatment of each type of isolates had not shown L. beckii death.
The increase of L. beckii deaths can be observed on the seventh day and 14th day after the application. Increased mortality rates can be compared with the control treatment. In this case, the control treatment merely contained tween 80. Of all the control treatment, the mortality rate of each isolate was 0%. Differences in density offungus conidia had a significant influence on mortality of L. beckii of each isolatestype (P <0.05).
Isolates of SB B3 K generated from the highlands of Kintamani-Bali had conidial viability rate of 53% on PDA. Although the percentage of the conidia viability was medium, this isolates at selection test could control L. beckii up to 53.3% at a concentration of 107 conidia / ml for 14 days. Based on this result, the SB B3 K isolates was further tested to determine the pathogenicity against L. beckii.
Significant difference between conidia density and observation time indicated an interaction between them (P <0.05). Differences in conidia density gave effect to the increase in the value of the percentage of brown scaleinsect mortality. Isolates of SB B3 K could control L. beckii at the highest concentrations of 10 4 , 10 7 and 10 3 conidia / ml for 70.8%, 68.8% and 59.9% respectively at day 14 th . Isolates of SB B3 K at the concentration of 10 4 conidia / ml could control L. beckii higher than that at the concentration of 10 7 conidia / ml, as well as that at the concentrations of 10 5 and 10 6 conidia / ml could control L. beckii lower than that at the concentrations of 10 4 and 10 3 conidia / ml (Figure 7).
This study indicated that a high conidia density did not always give a high mortality rate. This condition is contrary to previous statement that the higher conidia given, the higher mortality generated. This was caused by the host population density and environmental conditions. LC50 value generated from this isolates SB B3 K was 4,5x10 7 conidia / ml on day 14 th . Isolates of SBW D1 K came from the medium land in Banyuwangi had a viability level of 40% which were grown on PDA. Although the percentage of viability was relatively very low (<50%), this isolates, at the selection phase, could control L. beckii up to 76.9% at a concentration of 10 7 conidia / ml for 14 days. Isolates of SBW D1 K was suspected can germinate well when having directcontact with the host than on PDA medium containing high carbohydrates. According to Nelson et al, (1983), high carbohydrate content causes a loss of viability of entomopathogenic fungi. Based on this, the isolatesof SBW K D1 wasfurther tested to determine the ability of pathogenicity against L. beckii.
Isolatesof SBW D1 K could control L. beckii mostly at the concentrations of 10 7 , 10 6 and 10 5 conidia / ml for 46.2%, 42.4% and 35.4% respectivelyon the seventh day. Mortality valueon the seventh day was different from the fourteenth day. The longer the time the application made, the higher the value of L. beckii mortality resulted from each concentration. On day fourteenth, mortality at concentrations of 10 7 , 10 6 and 10 5 conidia / ml increased by 81.13%, 63.5% and 51.2%respectively,whereas at lower concentrations, that were, 10 4 , 10 3 and 10 2 conidia / ml could control L. beckii less than 50%, i.e. 48.1%, 38.7% and 25% on the fourteenth day. The LC50 value generated from isolates of SBW D1 K was 3,3x106 conidia / ml.
Isolates of SK D1 K came from the lowland in Sambas, West Kalimantan. These isolates had conidial viability value of 61.6% on PDA medium. The resultsof selection test indicated that the mortality rate of L. beckii reached 78.4% at day 14 th . The ability of these isolates to infect L. beckii was high as well as its viability rate, therefore, it was necessary to test the isolate pathogenicity.
On the seventh day, the isolates of SKD1 K could controlL. beckii at concentration of 10 5 , 10 6 and 10 7 conidia / ml for 25.8%, 23.5%, and 20.4%respectively. The value of mortality increasedup to the fourteenth day. On day 14 th , the mortality of L. beckii at concentration of 10 7 and 10 6 conidia / ml increased more rapidly than that at the concentration of 10 5 conidia / ml for 73.6%, 69, 9% and 56% respectively.Whereas the isolate at concentration below 10 4 , 10 3 and 10 2 conidia / ml could control L. beckii less than 50%, i.e. 45.4%, 32.4%, and 28% respectively at day 14 th . The LC50 value generated from isolates of SK D1 K was 3,3x105 conidia / ml. Isolatesof SBW D2 H collected from host plants at the medium land in Banyuwangi, East Java were isolated from leaves during the rainy season. Based on the LC50, isolates of SBW D2 Hwere the most virulent to L. beckii since the conidia of 10 3 conidia / ml on day 14 th could infect 50% of the scale insect. The LC50 value from isolatesof SBW D2 H was 7,2x106 conidia / ml.
Isolatesof SBW B2 H collected from host plants at the medium land in Banyuwangi, East Java were isolated from leaves during the rainy season. Based on the LC50,isolates of SBW D2 H were the most virulent to L. beckii since the conidia of 10 3 conidia / ml on day 14 th could infect 50% of the scale insect. The LC50 value from isolates of SBW B2 H was 1,04x106 conidia / ml.
Isolatesof SBW D3 BH collected from host plants at the medium land in Banyuwangi, East Java were isolated from leaves during the rainy season. Based on LC50, isolates of SBW D3 BH hada lowervirulence than isolates of SBW B2 H, because the conidia of 10 5 conidia / ml on day 7 th couldinfect only 50% of the scale insects, whereas isolates of SBW B2 H could get LC50 at conidiaof 10 4 conidia / ml. The LC50 value from isolatesof SBW D3 BH was 5.3 x104 conidia / ml.

CONCLUSION
Scale insectsthat attacked tangerine (C. Suhuiensis Tan.) at the highlands, medium lands, and low lands during the dry season and the rainy season weretypes of L. beckii and A. Aurantii. The highestpopulation occurred at the low lands during the dry season by L. beckii with a population of 4.2 heads and increased to 5.5 heads per 10 cm in the rainy season. Selection result, viability test and pathogenicity showed that there were six fungal isolates that have potential as entomopathogenic fungi to control scale insects, namely, SBW B2 H, SBW D2 H, SBW D3 BH, SK D1 K, SBW D1 K and SB B3 K.