Journal of Chemical Ecology, Vol. 10, No. 10, 1984
SEARCH
FOR POTENT ATTRACTANTS O N I O N F L I E S 1'2
OF
J. R. MILLER, 3 M. O. HARRIS, 3 and J. A. BREZNAK 4 s Department of Entomology and Pesticide Research Center 4Department of Microbiology and Public Health Michigan State University East Lansing, Michigan 48824 (Received October 26, 1983; revised February 22, 1984) Abstract--Of various chopped vegetables tested, Allium spp. high in propyl-containing alkyl sulfides (e.g., cepa group) caught the most onion flies in trapping tests in the field. Fly catches to chopped onion increased with bait quantity. Attractancy of chopped onion changed dramatically during aging in the field; catch increased over the first few days, peaked at ca. fivefold over fresh material by 3-5 days, and then declined sharply. This age-dependent increase in attraction was not seen for garlic (known to have antimicrobial properties) nor with chopped onion mixed with chopped garlic. These data suggested that attraction of onion flies to onions was strongly influenced by microbial activity associated with decomposing onions. The bacterium Klebsiella pneumoniae was identified as a major colonizer of onions maximally attractive to onion flies. This increased attraction is not due to the previously reported microbially produced volatiles ethyl acetate and tetramethyl pyrazine. Key Words--Delia antiqua, onion fly, Diptera, Anthomyiidae, food attractants, host attractants, microbial attractants, Klebsiella pneumoniae, bacteria, Allium, onion, garlic, chive.
INTRODUCTION
The onion fly, Delia (=Hylemya) antiqua Meigen, is an economically important pest of onions throughout the northern hemisphere (Loosjes, 1976). In North America, control of D. antiqua is achieved mainly by ~Diptera: Anthomyiidae. 2Paper No. 11047 of the Michigan State University Agricultural Experiment Station. 1477 0098-0331/84/1000-1477503.50/0 9 1984 Plenum Publishing Corporation
1478
MILLER ET AL.
organophosphate soil insecticides (e.g., Fonophos TM) applied under onion seed at planting. Although the onion maggot is currently "under control," growers and pest managers are highly concerned that resistance to available insecticides (Harris et al., 1982) will increase in North American populations to the extent that crop failures (Guyer and Wells, 1959) will again occur. We have been examining the onion fly life system from a behavioral and physiological perspective, hoping to identify points of vulnerability to human manipulation. An aspect deserving closer attention is behaviormodifying chemicals. For example, attractants offer the possibility for manipulating D. antiqua spatial distributions to allow population monitoring (via trapping) and increased visitation to insecticidal baits. Since the original finding (Matsumoto, 1970) that the volatile atkyl sulfides from onion elevate trap catches, research on onion fly attractants has been directed toward finding increasingly potent baits. Major onion volatiles like n-dipropyl disulfide (n-Pr2S2) can yield trap catches ca. 10-fold higher than unbaited traps (Dindonis and Miller, 1981a; Vernon et al., 1981); however, these onion synthetics have not proven to be consistently attractive (Miller and Haarer, 1981; Eckenrode et al., 1975). Optimizing release rates of alkyl sulfides (Dindonis and Miller. 1981a) or employing chemical blends (Vernon et al., 1981) led to only modest increases in attraction. Furthermore, baits consisting of intact (Dindonis and Miller, 1980) or freshly cut onion tissues (Dindonis and Miller, 1981b; Vernon et al., 1981; Ishikawa et al., 1981) do not yield catches notably greater than those of synthetics. Combined with direct behavioral observations of onion fly response in the field (Dindonis and Miller, 1980b), these data suggest that, in the setting of commercial fields, onion volatiles alone are not attractive enough to be of much use in managing D. antiqua adults. Attention has been directed to other sources of attractants. Miller and Haarer (1981) found that enzymatic yeast hydrolysate (EYH) is quite attractive to onion and seed flies, yielding catches 17- and 25-fold more, respectively, than unbaited traps. Although maximally attractive to females only during vitellogenesis (Miller, unpublished), EYH has proven to be a more effective and practical bait for D. antiqua than onion synthetics. Another potential source of onion fly attractants is microorganisms. It is not unusual for muscid flies to be attracted to the volatiles emanating from microbially infested organic matter (Hwang et al., 1978). In addition to possible nutritional benefit to the adult onion fly, bacteria are known to speed larval development (Schneider et al., 1983, and references therein). Hence, there is ample reason to search for microbially generated attractants of D. antiqua. Several groups, working independently, have documented that traps baited with decomposing onions catch more flies than do those baited with
| 479
ATTRACTANTS OF ONION FLIES
freshly damaged onion tissues. Dindonis and Miller (1981 b) found that after 4 days, bacteria-inoculated onion halves became significantly more attractive than freshly cut onion halves. Furthermore, they hypothesized that the optimal D. antiqua attractant would be a mixture of onion volatiles combined with volatiles from particular microorganisms. Vernon et al. (1981) also reported that onion attractancy to D. antiqua increased with time and showed this change was correlated with qualitative and quantative changes in the profile of volatiles as determined by GLC. Following the little-known lead of Tomioka (1977), Ishikawa et al. (I 981) provided the best documentation, to date, for the increased attractancy of decomposing onion and the time course for its development. In this paper we: (1) compare the attractiveness of fresh onion tissue with that of other alliums, (2) present further data on the time course of changes in onion tissue attractiveness as it decomposes, (3) identify a major microorganism from optimally decomposing onion, and (4) determine whether the increased attractiveness of decomposing onions is due to the microbially produced volatiles ethyl acetate and tetramethyl pyrazine as reported by Ikeshoji et al. (1980). METHODS AND MATERIALS
Experiment 1--Relative Attractiveness of Alliums. The attractiveness of onion (Allium cepa) was compared to that of other alliums to determine whether A. cepa is uniquely attractive to D. antiqua and whether A. cepa tissue is, in fact, the best starting material upon which to try to improve attraction. The vegetables selected for this test were chopped (mean particle weight ca. I00 mg), dispensed in 100- to 150-g portions in 9-cm petri dishes covered with 9 mesh/cm cheesecloth, and deployed under cone traps (Dindonis and Miller, 1980) in post-harvest commercial onion field in Stockbridge, Michigan. The precise cultivar of these market-purchased vegetables was unknown. Onion (A. cepa), shallot (A. ascalonicum), and garlic (A. sativum) were purchased as bulbs. Leek (A. porrum) and chive (A. schoenoprasum) came as intact maturing plants, but only the nonchlorophyllose portions were chopped. Radish (Raphanus sativus) bulbs a n d potato (Solanum tuberosum) tubers were included in the test as non-Allium controls. The experimental design was randomized complete block, and trap spacing within the five linear blocks was c a 5 m. Baits were replaced with freshly chopped materials rerandomized every 24 hr during the first half of the test (total duration September 21, to October 3, 1979) and then every 48 hr. Experiment 2--Fly Catch as Affected by Bait Quantity. Cone traps baited with precise quantities (0, 15, 25, 100, and 300 g) of chopped onion
1480
MILLER ET AL.
(Michigan-grown Yellow Globe cultivar) and garlic (the most and least attractive alliums, respectively, from Experiment 1 were deployed to test the hypothesis that catch varies with bait quantity. Chopped materials were dispensed in 9-cm-diam x 10-cm-high aluminum-screen (7 mesh/cm) cylinders capped on the bottom and top with a 9-cm Petri dish. The top cap was painted brown to reduce solar incidence. This test was conducted from May 28, to June 1, 1980, on the edges of a commercial onion field in Stockbridge, Michigan. The design was that of experiment 1, but with four blocks. Every 24 hr, baits were replaced with fresh material and traps within blocks were rerandomized. Experiment 3--Fly Catch as Affected by Bait Aging. The Allium materials generated during Experiment 2 were saved and set apart from experiment 1 to age. Thus, sets of chopped onion and garlic, incrementally aged from 0-1 days to 4-5 days, were readily obtained for concurrent testing for attractiveness to D. antiqua. Only the 300-g portions oi" chopped material were used in experiment 3, since this quantity of freshly chopped onion was maximally attractive in experiment 2. This test was conducted as per experiment 2 and ran from June 3 to 5, 1980. During this time, freshly chopped material was rotated into the experiment while material older than 4-5 days was rotated out. Experiment 4--Inhibitory Effect of Garlic on Onion Attraction. In Experiment 3, the attraction of chopped onion increased dramatically with time while that of chopped garlic did not. As judged by human sight and smell, the aged onion material decomposed while garlic did not. Since garlic has strong antimicrobial properties (A1-Delaimy and All, 1970), we hypothesized that, over time, a chopped onion-garlic mixture would not become as attractive as chopped onion alone. This idea was tested using the procedures of experiment 3. The treatments were 150 g chopped onion, 150 g chopped garlic, 150 g of each thoroughly mixed, and 150 g of chopped onion placed directly over but not in physical contact with 150 g of chopped garlic. The five age classes of each treatment were tested concurrently on August 22-28, 1980, in Grant, Michigan.
Experiment 5--Identification of Predominant Microorganism from Optimally Decomposing Onion. Chopped onion tissue 3-5 days old and ca. five fold more attractive in the field than fresh material was immediately transported to the laboratory for microbial analysis. Fifty grams fresh weight of onion tissue were placed in 200 ml of sterile buffer (0.05 M potassium phosphate buffer, pH 7.0, containing 0.15 M NaC1) and homogenized for 5 min by using a sterile Waring blender. Serial decimal dilutions of the homogenate were made in buffer, and 0.1-ml portions of each dilution were spread on the surface of solid culture medium containing (%): nutrient agar (Difco), 2.3; glucose, 0.5; and Difco yeast ex-
ATI'RACTANTS OF ONION FLIES
1481
tract, 0.5. Plates were then incubated aerobically at 25 __+ I~ A bacterial isolate, derived from the dominant colony type appearing on isolation plates, was then examined by morphological, physiological, and biochemical tests (Holdeman and Moore, 1972; Lennette et al 1974). Assignment of the isolate to a taxon was made by consulting contemporary classification keys (Lennette et al., 1974; Buchanan and Gibbons, 1974).
Experiment 6--Attractiveness of Ethyl Acetate and Tetramethyl Pyrazine. Ikeshoji et al. (1980) identified ethyl acetate and tetramethyl pyrazine among the volatiles over Klebsiella sp. inoculated onion slices. These chemicals elevated oviposition when added to certain levels of nPhS2; however, it was not clear whether these "synergists" primarily attracted more flies to decomposing onion or stimulated increased oviposition once flies had found the resource. Here, we tested whether these chemicals elevated trap catches in the field, either when dispensed alone or in combination with n-PhS2, ethanol, and acetic acid. The 20 treatments are given in Table 2. Each of the n-Pr2S2, ethanol, and ethyl acetate treatments was comprised of two No. 130 Beem Capsules, size 00 (Ted Pella Co., P.O. Box 510, Tustin, California 92680), half filled with the respective liquid. Acetic acid was dispensed as a 33% aqueous solution in a 7-ml (1.6-cm-diam) glass miniscintillation vial with a 6-ram hole punched through the polyethylene snap-on cap. Tetramethyl pyrazine was dispensed as 750 mg of dry crystals in each vial with the 6-mm hole in the cap. Odors from each treatment were readily detectable by human olfaction. Treatments were placed in 10-cm-diam clay pots containing enough washed silica sand to hold all capsules and vials cap-upright. Onion was dispensed as per experiment 2. This test was conducted July 17-25, 1981, at Eaton Rapids, Michigan, using four linear blocks along the edges of a commercial onion field in .which no pesticides were used. Trap spacing was 3 m, and treatments within blocks were rerandomized three times during the test. Onion materials and acetic acid treatments were replaced every 2 days. Capsules with little liquid in reservoir were replaced as appropriate. RESULTS AND DISCUSSION
Experiment 1--Relative Attractiveness of Alliums. This test demonstrated that onion flies are differentially responsive to plant volatiles generated by the selected vegetables (Figure 1). Onion, shallot, and leek fell into a grouping of alliums most attractive to D. antiqua adults. In this test, chive was less attractive than onion, and garlic was the least attractive Allium. Chopped potato and radish, as rich as the alliums in water and other nutrients, yielded catches statistically indistinguishable from the very low
1482
MILLER ET AL.
40
o W w 9.J
o 30 < o #.. t10 #eq/ml) of acetate, but only minor amounts of lactate, succinate, and formate. These data suggested that JM 1 was a member of the Enterobacteriaceae. Results of further tests (Table 1) indicated that JM1 was a strain of Klebsiellapneumoniae.
Experiment 6--Attractiveness of Ethyl Acetate and Tetramethyl Pyrazine. In this test, none of the synthetics tested singly or in any combination
1486
MILLER ET AL.
TABLE 1. BIOCHEMICALCHARACTERISTICSOF Klebsiellapneumoniae STRAIN J M 1
Test or substrate TSIb Slant Base Gas H2S Indole Methyl red Voges-Proskauer Citrate Urease Arginine dihydrolase Lysine decarboxylase Ornithine decarboxylase Fermentation of Glucose Glycerol Adonitol Dulcitol Inositol Sorbitol Polypectate liquefaction Pigmentation
Reaction~ A A + + + + + + + + + + +
aA, acid production; +, positive reaction; +, weak positive reaction;-, negative reaction. Triple sugar iron agar (Difco). caught significantly more onion flies than unbaited traps (Table 2), while aged onion again caught more than 50 times more flies than control, In similar tests, where ethyl acetate and tetramethyl pyrazine were added to freshly chopped onion or EYH, no significant increase in trap catch was realized (Harris and Miller, unpublished). These experiments do not support the hypothesis that ethyl acetate and tetramethyl pryrazine are onion fly attractants. However, Harris (1982) has substantiated that ethyl acetate does elevate D. antiqua oviposition in response to both propyl and allyl disulfides. In summary, chopped onions which are exposed to the range of naturally occurring microorganisms in an onion field become highly attractive to onion flies of both sexes, but only at a particular stage of microbial succession. The observation that both male and female onion flies and other insects feed heavily upon this conditioned plant material suggests that it is evoking strong food-finding responses in addition to increased oviposition responses by D. antiqua females (Ikeshoji et al., 1980). The
ATTRACTANTSOF ONION FLIES
1487
TABLE 2. ONION FLY CATCH IN TRAPS BAITED WITH AGED ONION AND VARIOUS SYNTHETICS, INCLUDING ETHYL ACETATE AND TETRAMETHYL PYRAZ1NE
Mean u trap catch (5:SD)
Treatment I. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
n-Dipropyl disulfide Ethanol Ethyl acetate Tetramethyl pyrazine acetic acid 1+ 2 1+ 3 1 +4 1+5 1 +2+3 2+3 2+3+4+5 1 +3+4+5 1+2+4+5 1+2+3+5 1+2+3+4 1+2+3+4+5 Chopped onion (0-2 days) Chopped onion (2-4 days) Unbaited
6.3 _+ 1.5 cde 5.0 :h 2.9 cde 3.8 J: 3.0 de 7.5 + 4.5 cde 4.3 + 2.2 cde 8.8 + 6.2 cde 6.5 J: 3.3 cde 7 . 3 + 1.0cde 5.5 + 1.3 cde 7.8 5 4 . 2 c d e 2.5_+ 1.7e 5 . 3 5 : 1 . 5 cde 7.3 5: 4.6cde I1.3 5 6.1 cde 12.0 5:4.4 cd 8.8 5:3.8 cde 15.3 + 11.1 c 71.0 5:21.7 b b 184.0 5:55.7 a b 3.3 5:1.3 cde
"Means followed by a c o m m o n letter are not statistically different at P_
SEARCH
FOR POTENT ATTRACTANTS O N I O N F L I E S 1'2
OF
J. R. MILLER, 3 M. O. HARRIS, 3 and J. A. BREZNAK 4 s Department of Entomology and Pesticide Research Center 4Department of Microbiology and Public Health Michigan State University East Lansing, Michigan 48824 (Received October 26, 1983; revised February 22, 1984) Abstract--Of various chopped vegetables tested, Allium spp. high in propyl-containing alkyl sulfides (e.g., cepa group) caught the most onion flies in trapping tests in the field. Fly catches to chopped onion increased with bait quantity. Attractancy of chopped onion changed dramatically during aging in the field; catch increased over the first few days, peaked at ca. fivefold over fresh material by 3-5 days, and then declined sharply. This age-dependent increase in attraction was not seen for garlic (known to have antimicrobial properties) nor with chopped onion mixed with chopped garlic. These data suggested that attraction of onion flies to onions was strongly influenced by microbial activity associated with decomposing onions. The bacterium Klebsiella pneumoniae was identified as a major colonizer of onions maximally attractive to onion flies. This increased attraction is not due to the previously reported microbially produced volatiles ethyl acetate and tetramethyl pyrazine. Key Words--Delia antiqua, onion fly, Diptera, Anthomyiidae, food attractants, host attractants, microbial attractants, Klebsiella pneumoniae, bacteria, Allium, onion, garlic, chive.
INTRODUCTION
The onion fly, Delia (=Hylemya) antiqua Meigen, is an economically important pest of onions throughout the northern hemisphere (Loosjes, 1976). In North America, control of D. antiqua is achieved mainly by ~Diptera: Anthomyiidae. 2Paper No. 11047 of the Michigan State University Agricultural Experiment Station. 1477 0098-0331/84/1000-1477503.50/0 9 1984 Plenum Publishing Corporation
1478
MILLER ET AL.
organophosphate soil insecticides (e.g., Fonophos TM) applied under onion seed at planting. Although the onion maggot is currently "under control," growers and pest managers are highly concerned that resistance to available insecticides (Harris et al., 1982) will increase in North American populations to the extent that crop failures (Guyer and Wells, 1959) will again occur. We have been examining the onion fly life system from a behavioral and physiological perspective, hoping to identify points of vulnerability to human manipulation. An aspect deserving closer attention is behaviormodifying chemicals. For example, attractants offer the possibility for manipulating D. antiqua spatial distributions to allow population monitoring (via trapping) and increased visitation to insecticidal baits. Since the original finding (Matsumoto, 1970) that the volatile atkyl sulfides from onion elevate trap catches, research on onion fly attractants has been directed toward finding increasingly potent baits. Major onion volatiles like n-dipropyl disulfide (n-Pr2S2) can yield trap catches ca. 10-fold higher than unbaited traps (Dindonis and Miller, 1981a; Vernon et al., 1981); however, these onion synthetics have not proven to be consistently attractive (Miller and Haarer, 1981; Eckenrode et al., 1975). Optimizing release rates of alkyl sulfides (Dindonis and Miller. 1981a) or employing chemical blends (Vernon et al., 1981) led to only modest increases in attraction. Furthermore, baits consisting of intact (Dindonis and Miller, 1980) or freshly cut onion tissues (Dindonis and Miller, 1981b; Vernon et al., 1981; Ishikawa et al., 1981) do not yield catches notably greater than those of synthetics. Combined with direct behavioral observations of onion fly response in the field (Dindonis and Miller, 1980b), these data suggest that, in the setting of commercial fields, onion volatiles alone are not attractive enough to be of much use in managing D. antiqua adults. Attention has been directed to other sources of attractants. Miller and Haarer (1981) found that enzymatic yeast hydrolysate (EYH) is quite attractive to onion and seed flies, yielding catches 17- and 25-fold more, respectively, than unbaited traps. Although maximally attractive to females only during vitellogenesis (Miller, unpublished), EYH has proven to be a more effective and practical bait for D. antiqua than onion synthetics. Another potential source of onion fly attractants is microorganisms. It is not unusual for muscid flies to be attracted to the volatiles emanating from microbially infested organic matter (Hwang et al., 1978). In addition to possible nutritional benefit to the adult onion fly, bacteria are known to speed larval development (Schneider et al., 1983, and references therein). Hence, there is ample reason to search for microbially generated attractants of D. antiqua. Several groups, working independently, have documented that traps baited with decomposing onions catch more flies than do those baited with
| 479
ATTRACTANTS OF ONION FLIES
freshly damaged onion tissues. Dindonis and Miller (1981 b) found that after 4 days, bacteria-inoculated onion halves became significantly more attractive than freshly cut onion halves. Furthermore, they hypothesized that the optimal D. antiqua attractant would be a mixture of onion volatiles combined with volatiles from particular microorganisms. Vernon et al. (1981) also reported that onion attractancy to D. antiqua increased with time and showed this change was correlated with qualitative and quantative changes in the profile of volatiles as determined by GLC. Following the little-known lead of Tomioka (1977), Ishikawa et al. (I 981) provided the best documentation, to date, for the increased attractancy of decomposing onion and the time course for its development. In this paper we: (1) compare the attractiveness of fresh onion tissue with that of other alliums, (2) present further data on the time course of changes in onion tissue attractiveness as it decomposes, (3) identify a major microorganism from optimally decomposing onion, and (4) determine whether the increased attractiveness of decomposing onions is due to the microbially produced volatiles ethyl acetate and tetramethyl pyrazine as reported by Ikeshoji et al. (1980). METHODS AND MATERIALS
Experiment 1--Relative Attractiveness of Alliums. The attractiveness of onion (Allium cepa) was compared to that of other alliums to determine whether A. cepa is uniquely attractive to D. antiqua and whether A. cepa tissue is, in fact, the best starting material upon which to try to improve attraction. The vegetables selected for this test were chopped (mean particle weight ca. I00 mg), dispensed in 100- to 150-g portions in 9-cm petri dishes covered with 9 mesh/cm cheesecloth, and deployed under cone traps (Dindonis and Miller, 1980) in post-harvest commercial onion field in Stockbridge, Michigan. The precise cultivar of these market-purchased vegetables was unknown. Onion (A. cepa), shallot (A. ascalonicum), and garlic (A. sativum) were purchased as bulbs. Leek (A. porrum) and chive (A. schoenoprasum) came as intact maturing plants, but only the nonchlorophyllose portions were chopped. Radish (Raphanus sativus) bulbs a n d potato (Solanum tuberosum) tubers were included in the test as non-Allium controls. The experimental design was randomized complete block, and trap spacing within the five linear blocks was c a 5 m. Baits were replaced with freshly chopped materials rerandomized every 24 hr during the first half of the test (total duration September 21, to October 3, 1979) and then every 48 hr. Experiment 2--Fly Catch as Affected by Bait Quantity. Cone traps baited with precise quantities (0, 15, 25, 100, and 300 g) of chopped onion
1480
MILLER ET AL.
(Michigan-grown Yellow Globe cultivar) and garlic (the most and least attractive alliums, respectively, from Experiment 1 were deployed to test the hypothesis that catch varies with bait quantity. Chopped materials were dispensed in 9-cm-diam x 10-cm-high aluminum-screen (7 mesh/cm) cylinders capped on the bottom and top with a 9-cm Petri dish. The top cap was painted brown to reduce solar incidence. This test was conducted from May 28, to June 1, 1980, on the edges of a commercial onion field in Stockbridge, Michigan. The design was that of experiment 1, but with four blocks. Every 24 hr, baits were replaced with fresh material and traps within blocks were rerandomized. Experiment 3--Fly Catch as Affected by Bait Aging. The Allium materials generated during Experiment 2 were saved and set apart from experiment 1 to age. Thus, sets of chopped onion and garlic, incrementally aged from 0-1 days to 4-5 days, were readily obtained for concurrent testing for attractiveness to D. antiqua. Only the 300-g portions oi" chopped material were used in experiment 3, since this quantity of freshly chopped onion was maximally attractive in experiment 2. This test was conducted as per experiment 2 and ran from June 3 to 5, 1980. During this time, freshly chopped material was rotated into the experiment while material older than 4-5 days was rotated out. Experiment 4--Inhibitory Effect of Garlic on Onion Attraction. In Experiment 3, the attraction of chopped onion increased dramatically with time while that of chopped garlic did not. As judged by human sight and smell, the aged onion material decomposed while garlic did not. Since garlic has strong antimicrobial properties (A1-Delaimy and All, 1970), we hypothesized that, over time, a chopped onion-garlic mixture would not become as attractive as chopped onion alone. This idea was tested using the procedures of experiment 3. The treatments were 150 g chopped onion, 150 g chopped garlic, 150 g of each thoroughly mixed, and 150 g of chopped onion placed directly over but not in physical contact with 150 g of chopped garlic. The five age classes of each treatment were tested concurrently on August 22-28, 1980, in Grant, Michigan.
Experiment 5--Identification of Predominant Microorganism from Optimally Decomposing Onion. Chopped onion tissue 3-5 days old and ca. five fold more attractive in the field than fresh material was immediately transported to the laboratory for microbial analysis. Fifty grams fresh weight of onion tissue were placed in 200 ml of sterile buffer (0.05 M potassium phosphate buffer, pH 7.0, containing 0.15 M NaC1) and homogenized for 5 min by using a sterile Waring blender. Serial decimal dilutions of the homogenate were made in buffer, and 0.1-ml portions of each dilution were spread on the surface of solid culture medium containing (%): nutrient agar (Difco), 2.3; glucose, 0.5; and Difco yeast ex-
ATI'RACTANTS OF ONION FLIES
1481
tract, 0.5. Plates were then incubated aerobically at 25 __+ I~ A bacterial isolate, derived from the dominant colony type appearing on isolation plates, was then examined by morphological, physiological, and biochemical tests (Holdeman and Moore, 1972; Lennette et al 1974). Assignment of the isolate to a taxon was made by consulting contemporary classification keys (Lennette et al., 1974; Buchanan and Gibbons, 1974).
Experiment 6--Attractiveness of Ethyl Acetate and Tetramethyl Pyrazine. Ikeshoji et al. (1980) identified ethyl acetate and tetramethyl pyrazine among the volatiles over Klebsiella sp. inoculated onion slices. These chemicals elevated oviposition when added to certain levels of nPhS2; however, it was not clear whether these "synergists" primarily attracted more flies to decomposing onion or stimulated increased oviposition once flies had found the resource. Here, we tested whether these chemicals elevated trap catches in the field, either when dispensed alone or in combination with n-PhS2, ethanol, and acetic acid. The 20 treatments are given in Table 2. Each of the n-Pr2S2, ethanol, and ethyl acetate treatments was comprised of two No. 130 Beem Capsules, size 00 (Ted Pella Co., P.O. Box 510, Tustin, California 92680), half filled with the respective liquid. Acetic acid was dispensed as a 33% aqueous solution in a 7-ml (1.6-cm-diam) glass miniscintillation vial with a 6-ram hole punched through the polyethylene snap-on cap. Tetramethyl pyrazine was dispensed as 750 mg of dry crystals in each vial with the 6-mm hole in the cap. Odors from each treatment were readily detectable by human olfaction. Treatments were placed in 10-cm-diam clay pots containing enough washed silica sand to hold all capsules and vials cap-upright. Onion was dispensed as per experiment 2. This test was conducted July 17-25, 1981, at Eaton Rapids, Michigan, using four linear blocks along the edges of a commercial onion field in .which no pesticides were used. Trap spacing was 3 m, and treatments within blocks were rerandomized three times during the test. Onion materials and acetic acid treatments were replaced every 2 days. Capsules with little liquid in reservoir were replaced as appropriate. RESULTS AND DISCUSSION
Experiment 1--Relative Attractiveness of Alliums. This test demonstrated that onion flies are differentially responsive to plant volatiles generated by the selected vegetables (Figure 1). Onion, shallot, and leek fell into a grouping of alliums most attractive to D. antiqua adults. In this test, chive was less attractive than onion, and garlic was the least attractive Allium. Chopped potato and radish, as rich as the alliums in water and other nutrients, yielded catches statistically indistinguishable from the very low
1482
MILLER ET AL.
40
o W w 9.J
o 30 < o #.. t10 #eq/ml) of acetate, but only minor amounts of lactate, succinate, and formate. These data suggested that JM 1 was a member of the Enterobacteriaceae. Results of further tests (Table 1) indicated that JM1 was a strain of Klebsiellapneumoniae.
Experiment 6--Attractiveness of Ethyl Acetate and Tetramethyl Pyrazine. In this test, none of the synthetics tested singly or in any combination
1486
MILLER ET AL.
TABLE 1. BIOCHEMICALCHARACTERISTICSOF Klebsiellapneumoniae STRAIN J M 1
Test or substrate TSIb Slant Base Gas H2S Indole Methyl red Voges-Proskauer Citrate Urease Arginine dihydrolase Lysine decarboxylase Ornithine decarboxylase Fermentation of Glucose Glycerol Adonitol Dulcitol Inositol Sorbitol Polypectate liquefaction Pigmentation
Reaction~ A A + + + + + + + + + + +
aA, acid production; +, positive reaction; +, weak positive reaction;-, negative reaction. Triple sugar iron agar (Difco). caught significantly more onion flies than unbaited traps (Table 2), while aged onion again caught more than 50 times more flies than control, In similar tests, where ethyl acetate and tetramethyl pyrazine were added to freshly chopped onion or EYH, no significant increase in trap catch was realized (Harris and Miller, unpublished). These experiments do not support the hypothesis that ethyl acetate and tetramethyl pryrazine are onion fly attractants. However, Harris (1982) has substantiated that ethyl acetate does elevate D. antiqua oviposition in response to both propyl and allyl disulfides. In summary, chopped onions which are exposed to the range of naturally occurring microorganisms in an onion field become highly attractive to onion flies of both sexes, but only at a particular stage of microbial succession. The observation that both male and female onion flies and other insects feed heavily upon this conditioned plant material suggests that it is evoking strong food-finding responses in addition to increased oviposition responses by D. antiqua females (Ikeshoji et al., 1980). The
ATTRACTANTSOF ONION FLIES
1487
TABLE 2. ONION FLY CATCH IN TRAPS BAITED WITH AGED ONION AND VARIOUS SYNTHETICS, INCLUDING ETHYL ACETATE AND TETRAMETHYL PYRAZ1NE
Mean u trap catch (5:SD)
Treatment I. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
n-Dipropyl disulfide Ethanol Ethyl acetate Tetramethyl pyrazine acetic acid 1+ 2 1+ 3 1 +4 1+5 1 +2+3 2+3 2+3+4+5 1 +3+4+5 1+2+4+5 1+2+3+5 1+2+3+4 1+2+3+4+5 Chopped onion (0-2 days) Chopped onion (2-4 days) Unbaited
6.3 _+ 1.5 cde 5.0 :h 2.9 cde 3.8 J: 3.0 de 7.5 + 4.5 cde 4.3 + 2.2 cde 8.8 + 6.2 cde 6.5 J: 3.3 cde 7 . 3 + 1.0cde 5.5 + 1.3 cde 7.8 5 4 . 2 c d e 2.5_+ 1.7e 5 . 3 5 : 1 . 5 cde 7.3 5: 4.6cde I1.3 5 6.1 cde 12.0 5:4.4 cd 8.8 5:3.8 cde 15.3 + 11.1 c 71.0 5:21.7 b b 184.0 5:55.7 a b 3.3 5:1.3 cde
"Means followed by a c o m m o n letter are not statistically different at P_
