Journal of Chemical Ecology, Vol. 12, No. 9, 1986

SEX PHEROMONE OF FALL ARMYWORM, Spodoptera frugiperda (J.E. SMITH) 1 Identification of Components Critical to Attraction in the Field

J.H.

TUMLINSON,

E.R.

MITCHELL,

and L.J.

P.E.A.

TEAL,

R.R.

HEATH,

MENGELKOCH

Insect Attractants, Behavior, and Basic Biology Research Laboratory Agricultural Research Service, U.S. Department of Agriculture Gainesville, Florida 32604 (Received July 30, 1985; accepted November 19, 1985)

Abstract--Analyses of extracts of pheromone glands and of volatiles from calling female fall armyworm moths, Spodoptera frugiperda (J.E. Smith), revealed the presence of the following compounds: dodecan-l-ol acetate, (Z)7-dodecen-l-ol acetate, ll-dodecen-l-ol acetate, (Z)-9-tetradecenal, (Z)-9tetradecen- 1-ol acetate, (Z)- 1 l-hexadecenal, and (Z)- 11-hexadecen- 1-ol acetate. The volatiles emitted by calling females differed from the gland extract in that the two aldehydes were absent. Field tests were conducted with sticky traps baited with rubber septa formulated to release blends with the same component ratios as those emitted by calling females. These tests demonstrated that both (Z)-7-dodecen-1-ol acetate and (Z)-9-tetradecen-1-ol acetate are required for optimum activity and that this blend is a significantly better lure than either virgin females or 25 mg of (Z)-9-dodecen-l-ol acetate in a polyethylene vial, the previously used standard. Addition of the other three acetates found in the volatiles did not significantly increase the effectiveness of the two-component blend as a bait for Pherocon 1C or International Pheromones moth traps.

Key Words--Spodoptera frugiperda, fall armyworm, Lepidoptera, Noctuidae, (Z)-7-dodecen-l-ol acetate, (Z)-9-tetradecen-l-ol acetate, pheromone, attractant, sex pheromone.

Mention of a commercial or proprietary product does not constitute an endorsement by the USDA.

19O9 0098-0331/86/0900-1909505.009 1986PlenumPublishingCorporation

19t0

TUMLINSON ET AL. INTRODUCTION

The genus Spodoptera includes several insect species that cause substantial damage to agricultural crops throughout the world. Three species, S. exigua (Hfibner), S. fi'ugiperda (J.E. Smith), and S. eridania (Cramer), are serious pests of many crops in the United States. The fall armyworm (FAW), S. frugiperda, is an important pest of corn and other grass crops in the United States and occurs throughout the Caribbean basin, where it is known as the whorlworm, and attacks a variety of crops (Andrews, 1980). Components of the FAW pheromone have been identified in previous investigations. Sekul and Sparks (1967) used a laboratory bioassay in which a mating response was evoked in FAW males to monitor the isolation and identification of a sex pheromone, (Z)-9-tetradecen-l-ol acetate (Z9-14: Ac), from FAW females. Subsequent tests showed this compound to be an ineffective lure for FAW males in the field (Mitchell and Doolittle, 1976; Sparks, 1980). However, this compound did reduce mating by FAWs when it was evaporated into the atmosphere in disruption tests (Mitchell and McLaughlin, 1982). A second compound, (Z)-9-dodecen- 1-ol acetate (Z9-12 : Ac), was isolated and identified from FAW females (Sekul and Sparks, 1976), and this compound either alone or with small quantities of Z 9 - 1 4 : A c added is a good practical lure for fall armyworm males when used in sticky traps (Mitchell, 1978; Jones and Sparks, 1979). However, fairly large quantities of Z9-12 :Ac are required for effectiveness (5-10 mg on a rubber septum), and the baits are effective in the field for only one to two weeks (Mitchell et al., 1983). Sparks (1980) reported that two additional compounds had been identified from washes of FAW female ovipositors, but he did not give their identity and he stated that they did not improve the effectiveness of Z 9 - 1 2 : A c as a lure. Descoins and coworkers (personal communication) analyzed the FAW female-produced pheromone and found (Z)-ll-hexadecen-l-ol acetate ( Z l l 16 : Ac), in addition to the two compounds already reported. However, in field tests conducted in Florida, we could not find a blend of these three compounds that was significantly better than Z9-12 : Ac alone in luring FAW males to sticky traps (Mitchell et al., 1983). Although Z9-12 : Ac can be used as a lure for monitoring FAW males, it is obvious that our knowledge of the pheromone system of this insect is incomplete and that the complete pheromone is needed to achieve maximum effectiveness in pheromone monitoring and control programs. Therefore, we decided to reinvestigate the pheromone system of this species to determine the precise blend emitted by calling FAW females and the optimum blend for trapping FAW males in the field.

1911

PHEROMONE OF FALL ARMYWORM

METHODS

AND MATERIALS

Rearing. The majority of the fall armyworm females from which pheromone was obtained were reared in the laboratory on an artificial diet (Burton, 1969). Insects were sexed in the pupal stage and placed in 0.5-liter paper cartons that were then placed in 23 x 23-cm Plexiglas cages for eclosion. Male and female pupae were maintained in separate rooms under a reversed light cycle of 14 hr light and 10 hr dark (65% relative humidity, 26~ Daily sequential transfers of uneclosed pupae were made to uninhabited cages so each holding cage contained adult insects of a single age and sex. Additionally, fourth- to sixth-instar fall armyworm larvae were collected in the field from sorghum plants and reared to pupation in the laboratory on artificial diet. Subsequent sexing and handling of these insects were conducted the same as for the laboratory-reared insects. Pheromone Extraction and Collection. Actively calling 2- to 3-day-old FAW females were used for all pheromone extractions and volatile collections. In the laboratory the calling activity of the female FAW peaked about 4 hr after the beginning of the dark period. Sex pheromone glands from both laboratoryreared and wild females were excised and extracted. Excision of the sex pheromone glands was performed during the dark period by gently squeezing the lateral-posterior abdominal section of the female, causing protrusion of the terminal abdominal segments, which were then clipped as a unit. The excised glands were then placed in 50 #1 of isooctane (Fisher Scientific Co., 99 mol % pure) in a microvial. Typically, when 25 glands were accumulated in the solvent (about 15 min), the walls of the microvial were rinsed with an additional 20 t~l of isooctane and the solvent was removed from the glands and transferred to a clean vial. The extract was then concentrated by careful evaporation with gentle warming to a final concentration of about five female equivalents (FE) per microliter. The number of glands excised during any collection period varied with availability of calling females within a particular batch. Gland extract was stored at - 6 0 ~ in microvials with Teflon-lined screwcaps until analysis. An apparatus similar to the ones described by Cross (1980) and Tumlinson et al. (1982) was used for all volatile collections. Zero-grade nitrogen (99.99 % minimum purity, 0.5 ppm maximum hydrocarbons) or, when live females were aerated, "hydrocarbon-free" compressed air, was delivered through a stainlesssteel regulator, a gas purifier (Alltech Associates), a methylene chloride-extracted charcoal (6-14 mesh, Fisher Scientific Co.) prefilter (9 cm x 2 cm OD), and then through a glass Y-tube to two aeration chambers simultaneously. Each aeration chamber was constructed from two glass tubes (7 cm • 2.8 cm OD) connected by a ball joint (35/25). A coarse-glass frit was sealed into the upwind

1912

TUMLINSONET AL.

tube to provide a laminar flow of air (or N2) over the aeration subject. The downwind tube was the aeration chamber, and its exit narrowed to a tube 0.6 cm in outside diameter to which was connected a small charcoal filter similar to the one described by Grob and Ztircher (1976). With this apparatus, volatiles can be collected from any desired subject in one chamber and simultaneously a ~ blank" can be collected in the other chamber under identical conditions. Four or five FAW females were selected during their peak calling activity and placed in an aeration chamber. An airflow of about 1.0 liter/min was maintained over the females for 1.5 hr. When the aeration was complete, the collected volatiles were eluted from the charcoal collection filters with three aliquots (20, 15, and 15 /A) of redistilled dichloromethane (HPLC grade, J.T. Baker), and the eluates from two aerations, representing 9-10 females, were combined. The initial collections were concentrated by gentle wanning to a volume of 2-5 ~1. Then 5 ~1 of isooctane was added to rinse the walls of the microvial and the solution was concentrated to a final volume of 1-2 tA for analysis by capillary gas chromatography. After the initial analyses to establish the location of the peaks of interest on a particular stationary phase, 5 ng of an internal standard in 1/xl of isooctane was added to the CH2C12 filter eluate before concentration. Pentadecan-l-ol acetate (S-15:Ac) and hexadecan-l-ol acetate (S16 :Ac) were used as internal standards for analyses on OV-1 and cyanosilicone columns, respectively. The system blank filters were extracted and the extract concentrated and analyzed in an identical manner on the same column and on the same day as that of the companion filters containing female volatiles. Chemical Analyses. Gas chromatographic (GC) analyses were conducted on a Varian model 3700 GC and a Hewlett-Packard model 5710A GC, both equipped with splitless capillary injector systems and flame ionization detectors. A Perkin Elmer chromatographic data system was used for data collection, storage, and subsequent analysis. Either nitrogen (linear flow velocity 9.8 cm/sec) or helium (linear flow velocity 19 cm/sec) was used as a carrier gas. Columns used for initial analyses of gland extracts and volatiles and conditions for each column were: 42-m x 0.25-mm ID glass capillary column coated with SP2340, operated at 80~ for 2 min, then temperature programmed (TP) at 30~ min to 170~ 60-m x 0.25-mm ID fused silica column coated with 0.20 ~m film of SP-2330, initial temp 80~ for 1 min, TP at 30~ to 170~ 29m x 0.25-ram ID glass column coated with 0.4% OV-1 over 1% Superox, initial temp 80 ~ for 2 rain, TP at 32~ to 180~ Although the three columns initially used for these analyses were adequate for separation of most lepidopteran pheromones, they did not provide the separation and resolution required to unequivocally assign the double bond posi-

P H E R O M O N E OF F A L L A R M Y W O R M

1913

tions in the 12-carbon acetates present in volatiles collected from FAW females. Therefore, three columns with superior resolution and with three different, but complementary, separation mechanisms were selected for analysis of the volatiles. They were: 50-m • 0.25-ram ID fused silica CPS-1 (cyanopropyl methyl silicone, Quadrex Corp., New Haven, Connecticut), initial temperature 80~ for 1 rain, TP at 10~ to 165~ 50-m x 0.25-mm ID fused silica OV101, initial temperature 80~ for 1 rain, TP at 10~ to 180~ 13.6-m x 0.25-ram ID glass cholesteryl p-chlorocinnamate (liquid crystal) (Heath et al., 1979; Heath and Doolittle, 1983) column coupled to 0.46-m • 0.33-m ID fused silica BP-I (SGE) precolumn, initial temperature 60~ for 1 min, TP at 30~ min to 150~ All injections were made in the splitless mode. Samples also were analyzed by GC-mass spectrometry (MS) with either a Finnigan model 3200 chemical ionization mass spectrometer or a Nermag model R1010 mass spectrometer in the chemical ionization mode. The SP-2340 and OV-101 capillary columns used in previous analyses were used in the GC-MS analyses with He carrier gas. Methane and isobutane were used as the reagent gases in the mass spectrometers. Spectra of the natural products were compared with those of candidate synthetic compounds. All synthetic standards used in this study were obtained from commercial sources and were purified by high-performance liquid chromatography on a 25 • 2.5-cm (OD) AgNO3-coated silica column eluted with toluene (Heath et al., 1977). These compounds were analyzed on both polar and nonpolar capillary GC columns described previously and determined to be greater than 99 % pure. Formulation. Synthetic blends and individual synthetic compounds were formulated on 5 x 9-ram rubber septa (A.H. Thomas Co.) for all biological tests. Septa were Soxhlet extracted with CH2C12 for 24 hr and air dried prior to loading. Desired release ratios of the components of blends were obtained by loading the septa with mixtures containing calculated percentages by weight of each component of the blend. The percentage of a component in the loading mixture was calculated on the basis of its relative volatility determined from retention indices on liquid crystal capillary GC columns (Heath and Tumlinson, 1986) and a method developed to predict release ratios of components of a blend from rubber septa (Heath et al., 1986). Each septum was loaded with 200/~1 of a hexane solution of the blend pipetted into the well on the large end of the septum. Septa were aired for two days at room temperature before use and were then used for a maximum of 14 days in the field. In field tests conducted in 1982, the quantity of Z 9 - 1 4 : A c , the major component, in all blends was 2 mg/septum. In these initial tests, which compared the effectiveness of four-, five-, and six-component blends with live females and Z9-12 : Ac in a polyethylene vial, the relative amount of each component was adjusted to keep the release ratio of the blend components as constant

19 ~4

TUMLINSONET AL.

as possible for all blends. In the 1983 and 1984 field tests, each septum was loaded with 2 mg of the total blend dissolved in 200/xl of hexane, and after airing for two days at room temperature, was used for 4-11 days. To verify that ratios released from the septa were the same as those calculated for a blend, septa loaded with the various blends and aired for two days were aerated in the same apparatus used to collect volatiles from glands and live females, and the volatiles collected on charcoal were eluted and analyzed by capillary GC on SP-2330. Two septa were placed in the aeration chamber of the apparatus, and zero-grade nitrogen was blown over them at flow rates of 100, 200, 400, 600, and 800 cc/min for 1.5 hr. An internal standard, S-16: Ac, was added to the eluate from the charcoal filter for quantitative analysis. Field Tests. Trapping experiments were conducted in Alachua County, Florida, in late summer 1982. The rubber septa containing the test blends were placed in Pherocon 1C sticky traps supported ca. 1 m above the ground on metal poles. Treatments were arranged in three randomized complete blocks in and around corn fields. The traps were set ca. 30 m apart in lines perpendicular to the prevailing wind and were checked every one to two days for captured FAW moths. Trap liners were replaced when more than five moths were captured or on every second visit. Whenever a trap had five or fewer moths, the insects were removed carefully to leave the sticky trapping surface intact; traps were rerandomized after every collection; thus, each collection was considered a te replicate. For statistical analysis, the data were transformed to 0.5 and subjected to analysis of variance (Steel and Torrie, 1960). Mean separations were achieved by Duncan's (1955) multiple-range test. In 1983, field-trapping experiments were conducted in late September with blends and ratios found in volatiles collected from calling females. The experimental design and data analysis were as described for the 1982 experiment. In 1984, the field trapping studies were conducted in August and September using International Pheromones moth traps (IPM traps) positioned ca. 1 m above the ground in and around sorghum fields. The IPM traps capture moths using a funnel-bucket system. Moths entering the bucket receptacle through the funnel were killed with a volatile insecticide, Vapona. The switch to the IPM trap was based upon preliminary experiments which indicated that the IPM trap was more effective in capturing large numbers of FAW moths than the Pherocon 1C sticky traps used in 1982 and 1983. The experimental design was as described for the 1982 and 1983 field tests. Due to the large numbers and extreme variations in total numbers of moths captured in different blocks on the same or different dates in the 1984 field tests, the data were converted to percentages using the formula:

~

Number of moths caught in treatment Total number of moths caught in block

x 100

PHEROMONE OF FALL ARMYWORM

1915

The data then were converted to arcsin x/X and subjected to analysis of variance (Steel and Torrie, 1960). Means were separated using Duncan's (1955) multiple-range test. RESULTS AND DISCUSSION

Gland Extracts. Initially the pheromone glands from calling FAW females were excised and extracted to determine what compounds were present, although we realized from previous experience and from literature reports (Cross et al., 1976; Hill et al., 1975) that the components and ratios present in the gland may differ from those released by calling females. Analysis of extracts of pheromone glands excised from laboratory-reared and wild, calling female FAW, by capillary GC on SP-2330 and SP-2340, indicated that there were five peaks consistently present above the background peaks of impurities in the solvent. Four of these peaks coincided in retention times with the following authentic standards: Z 7 - 1 2 : A c , Z9-14:A1, Z 9 - 1 4 : A c , and Z l l - 1 6 : A c . The other peak was very close in retention time to that of Z9-12 :Ac. Furthermore, a peak coincidental in retention time with Z l l - 1 6 : A 1 on SP-2330 also appeared, but it was not present consistently in quantities large enough to distinguish it from background impurities. Analysis on OV-1 confirmed the presence of Z9-14 :Ac, the major peak, and Z l 1 - 1 6 : Ac. Additionally, peaks coincidental in retention time with Z7-12 : Ac and Z9-14:A1 were present, but since Z 9 - 1 2 : A c coeluted with Z9-14:A1 on this column, confirmation of the presence of the latter two was not possible. Methane ionization mass spectral analysis of gland extracts from laboratory-reared FAW females confirmed that the major peak, coincidental in retention time with Z9-14: Ac on SP-2340, SP-2330, and OV-1 capillary columns, was a monounsaturated 14-carbon acetate with diagnostic ions at m/e 255 (M + 1), 195 (M + 1 - 60), 89, and 61. Also, the presence of peaks consistent in retention time and mass spectral diagnostic ions with Z7-12 :Ac, Z9-14 : A1, and Z 11-16 : Ac was confirmed. Furthermore, small peaks, eluting just prior to and just after Z7-12 : Ac on the SP-2340 column, had mass spectra consistent with dodecan-l-ol acetate (S-12:Ac) and a monounsaturated 12:Ac, respectively. Thus the presence of Z7-12 : Ac, Z9-14 :A1, Z9-14 : Ac, and Z 11-16 : Ac in pheromone glands of calling FAW females was firmly established by GCMS and by retention times on OV-1, SP-2340, and SP-2330 capillary GC columns; the cyanosilicone GC phases have previously been demonstrated to separate both geometrical and positional isomers of most olefinic aliphatic primary acetates (Heath et al., 1980; Tumlinson et al., 1982). Additionally, S-12:Ac, Z9-12 : Ac, and Z11-16:A1 appeared to be present in small quantities, but this could not be confirmed.

1916

TUMLINSONET AL.

Gland extracts were prepared from five batches of laboratory-reared FAW females and from two batches of wild FAW females from each of two locations in Alachua County, Florida. There was considerable variability in gland contents, in both ratios and total quantities between laboratory-reared and wild females and also among batches of wild females collected from different locations. The relative amounts of Z9-14 : A1, Z11-16 :A1, and Z11-16 : Ac in particular varied more than those of the other components. Thus, "the most representative ratio" of components found in the glands was 4 : 2 : 13 : 69 : 3 : 9 for Z7-12 :Ac, Z9-12 :Ac (subsequently identified as 11-12 : Ac), Z9-14 :A1, Z 9 14 : Ac, Z I 1-16: A1, and Z11-16 :Ac, respectively, but the relative amounts of the two aldehydes and Z l l - 1 6 : A c varied from these values by as much as 100 %. However, as noted earlier, the actual pheromone blend emitted by the female often differs significantly, both in compounds present and in ratio, from the contents of the gland. Therefore, we proceeded to collect volatiles to more accurately assess the composition of the FAW pheromone. Volatile Collections. Analysis on the SP-2330 capillary column of the volatiles collected from laboratory-reared, calling FAW females indicated the presence of five peaks not present in the system blank analyzed under identical conditions (Figure 1). Four of these peaks coincided in retention times on this column with peaks representing S-12 :Ac, Z7-12 : Ac, Z9-14 :Ac, and Z11t 6 : A c . Again, the third peak in the chromotogram (Figure 1) had a retention 4

J

i

i

I.S. S-16:Ac

:!

1

!i!

Sys,o=Fema'Jiev~li, ,,

!i/

l,

,. . . . . . . .

............

Blank 1

13

i

i

15

TIME

i

i

17

i

i

T---

19

(min)

FIG. 1. Analysis of volatiles collected from calling, laboratory-reared fall armyworm female moths on a 60-m fused silica SP-2330 capillary gas chromatographic column. Peaks not present in the blank correspond in retention time to: (1) S-12 :Ac, (2) Z712:Ac, (3) ll-12:Ac, (4) Z9-14:Ac, and (5) Zll-16:Ac.

P H E R O M O N E OF F A L L A R M Y W O R M

1917

time very close to, but not identical with that of Z 9 - 1 2 : A c . The ratio (mean of three replicates, ca. 10 females per replicate) of these compounds present in the volatiles was 4.9 : 3.1 : 1.7 : 86.9 : 3.5, respectively. Although Z9-12 :Ac was previously reported to be a component of the FAW pheromone, its presence in the volatiles released by calling females could not be confirmed. Additionally, the quantity of S-12 : Ac in the volatiles could not be accurately determined because of interference of an impurity in the system. Therefore, an analytical procedure was devised, employing three highresolution capillary GC columns and mass spectrometry, that was capable of resolving all possible 12-carbon acetate candidate compounds and confirming their identities. The CPS-1 (cyanosilicone) column separated the S-12 : Ac from the interfering impurity and resolved the other isomers, although 11-12 :Ac was not completely resolved from Z 9 - 1 2 : A c (Figure 2A). However, the OV-101 column completely resolved 11-12 : Ac from Z9-12 : Ac, although S-12 : Ac was now incompletely resolved from Z9-12 :Ac (Figure 2B). The liquid crystal column resolved all candidate compounds except 11-12 :Ac and S-12 :Ac (Figure 2C). Volatiles were collected from 44 batches of laboratory-reared, calling FAW females (five females per batch) and pooled for analysis. Aliquots of this sample were analyzed on all three high-resolution capillary columns and by GC-MS. As indicated in Figure 2, the volatile components not present in the system blank were coincidental on all three columns with S-12:Ac, Z 7 - 1 2 : A c , and 11-12:Ac. Z 9 - 1 2 : A c did not coincide in retention time with any of the candidate pheromone peaks on any of the these columns. Methane/isobutane ionization mass spectra of the volatile peaks were consistent with these identities. Z 9 - 1 4 : A c and Z 11-16:Ac also were present in these volatiles and confirmed by all the data. The ratio of the components in the volatiles was 1 . 9 : 3 . 2 : 2 . 2 : 9 0 . 1 : 2 . 6 for S-12:Ac, Z 7 - 1 2 : A c , l l - 1 2 : A c , Z 9 - 1 4 : A c , and Z l l - 1 6 : A c , respectively. The quantity of Z 9 - 1 4 : A c collected per hour per female was about 2 ng. Formulation. Volatiles released from rubber septa formulated for field tests were collected and analyzed by capillary GC to verify that actual release ratios were approximately the same as the calculated release ratios. Examples of release ratios measured in this way for the five-component blend and a four-component blend are given in Table 1. Release rates, but not blend ratios, from the septa varied with the flow rate through the aeration chamber. For example, the mean release rates of Z9-14: Ac from a septum loaded with 200/zg of the fivecomponent blend at flows of 100,200, and 400 cc/min were 1.7, 2.5, and 4.3 ng/hr/septum, respectively. Because release rates from septa vary with the velocity of airflow over the septa, as well as temperature, it is very difficult to control release rates in the field. Therefore, it is more practical to compare captures of septa loaded with different doses of pheromone (see later) whose relative release rates can be measured accurately.

A ~

~

SystemBlank

.. ,: ' ::

' .: :.. :

9

;S~

FemaleVolatiles

~Z8

t

q

T

14,6

~

14.8

I

r

I

15.0

I

r

15.2

15.4

11zg~Standard 12:Ac I

,

i

:

15.6

I

15.8

E

16.0

B

FemaleVolatiles .... .................................... "

Standard 12:A0 i

.........

/ zTV 11~//z9 s~ F

16.6

9

............... .....

i

16.8

i

17.0

17.2

17.4 I

C Female Volatiles

;

:. ....: ! :: I

-

I

\

"~.,...

Standard ~

1Q

j

i

8

~

v i

8.5

9

T

9.5

TIME (min) FIG. 2. Sections of chromatograms showing separation of 12-carbon acetates and analysis of volatiles collected from calling, laboratory-reared fall armyworm female moths on three-capillary GC columns (A) CPS-1; (B) OV-101; (C) cholesteryl-p-chlorocinnamate (liquid crystal). In (C), the symbol I indicates impurities found in the system blank.

1919

PHEROMONE OF FALL ARMYWORM

TABLE 1. CHROMATOGRAPHICALLY MEASURED RELEASE RATIOS OF COMPONENTS IN TWO PHEROMONE BLENDS FORMULATED ON RUBBER SEPTA

Blend 1 Theoretical Load release (%) (%) S-12:Ac Z7-12:Ac Z9-12:Ac zg-14:Ac Zll-16:Ac

1.0 0.5 0.3 79.5 18.8

4.9 3.1 1.7 86.9 3.5

release (%, X + SD)" 6.2 _+ 0.4 3.0_+0.2 1.9 ___0.1 86.9 _+ 0.6 2.0 _. 0.2

Blend 2 Theoretical Load release (%) (%) -0.5 0.3 81.6 17.7

release (% X + SD)~

-3.0 1.7 91.9 3.4

-4.0_+ 1.3 1.1 _+ 0.1 90.8 +_. 3.4 4.1 _+ 2.2

~Mean of three replications. Each septum loaded with 2 mg of the total blend. Field Tests. The field tests conducted in 1982 were designed to evaluate blends of the compounds identified in pheromone gland extracts. All experiments included 25 mg of Z 9 - 1 2 : A c dispensed from a polyethylene vial, the previously used standard (Mitchell, 1978), and three virgin females for comparison. In the first experiment in which four different ratios of components in the six-component blend (Z7-12 : Ac, Z9-12 : Ac, Z9-14 :A1, Z9-14 :Ac, Z1116:A1, Z l l - 1 6 : A c ) and two different ratios in a four-component blend (Z914 :A1, Z9-14 :Ac, Z11-16 :A1, Z 1 1 - 1 6 : A c ) were tested (40 replicates/treatment), the ratios of components in these blends were selected to represent ratios found in gland extracts from laboratory-reared and different batches of wild F A W females. All the six-component blends tested were equivalent in luring F A W males to traps (means of 21.5-23.2 males/trap per night), and they were not significantly different from the standard 25 mg of Z9-12 : Ac in a polyethylene vial (mean of 26.5). However, it is very unlikely that the attractancy of these blends was due solely to the presence of Z9-12 : A c because the qauntity of Z9-12 : A c loaded on a septum was only 68/~g at the most. Traps baited with both four-component blends tested captured significantly fewer males (mean of 7.4) than either the standard Z 9 - 1 2 : A c or females (mean of 13.8) (5% level, Duncan's multiple range test). Thus either Z 7 - 1 2 : A c or Z 9 - 1 2 : A c or both appeared to be necessary for optimum activity. In a subsequent experiment, we compared a six-component blend with fivecomponent blends, each of which was missing a different component. The results (Table 2) clearly establish that Z 7 - 1 2 : A c is necessary for activity and suggested that none of the other components except Z9-14 : Ac are required. A separate experiment established that deletion of Z9-14 : Ac from the blends resulted in trap captures not significantly different from a hexane blank. It also was interesting to note that deletion of Z 9 - 1 4 : A 1 increased the trap capture above that of all other blends and of the standard 2,9-12 : Ac. This suggests that

0

2 (0.3)

2 (0.3)

2 (0.4)

4 (0.5)

5 (0.5)

4 (0.5)

4 (0.9)

14 (4.2)

13 (2.6)

0

13 (2.5)

14 (2.5)

13 (2.5)

Z9-14:A1

76 (88.6)

71 (55.0)

79 (54.0)

70 (53.0)

72 (53.0)

69 (52.8)

Z9-14 : Ac

3 (5.9)

0

3 (3.6)

3 (3.5)

3 (3.5)

3 (3.5)

Zll-16:A1

0

9 (41.9)

10 (41.5)

9 (40.5)

9 (40.6)

9 (40.4)

Z11-16:Ac

c

14.5 • 1.7 b 1.2 • 0.3 d 13.0 • 1.6 b 20.7 • 2.2 a 14.9 _+ 1.8 b 16.2 • 2.0 b 14.4 • 1.8 b 7.0 • 1.3

"Numbers under each compound in each row indicate the approximate percentage of that component released in the blend. Numbers in parentheses indicate the percent of each compound loaded to obtain the desired release ratio. The quantity of Z 9 - 1 4 : A c loaded onto the septum was 2 mg for all blends. ~'Forty replicates/treatment; means not followed by the same letters differ significantly at the 1% level, Duncan's multiple-range test. r a polyethylene vial.

3 Females

2 (0.3)

0

25 mg"

2 (0.3)

Z9-12 :Ac

4 (0.5)

Z7-12:Ac

Mean/trap per night (• SE) j'

T a b l e 2. COMPARISON OF STICKY TRAP CATCHES OF Spodopterafrugiperda MALES WITH FIVEAND SIx-CoMPONENT BLENDS OF SYNTHETIC COMPOUNDS FORMULATED ON RUBBER SEPTA, GAINESVILLE, FLORIDA, SEPTEMBER 1 5 - 2 9 , 1982 a

bo Q

PHEROMONE OF FALL ARMYWORM

192 1

Z9-14 : A1 may be an inhibitor and that, although it is present in the pheromone gland, it may not be a pheromone component. Since the volatiles collected from calling females contained only five acetates and did not contain either of the aldehydes, we designed our 1983 field experiments to evaluate specific blends of the acetates (Table 3). Our primary objective was to determine whether all five compounds were required for optimum trap captures and whether any of the components were critical for biological activity. At this point Z9-12 : Ac was included as a blend component, although we had not yet confirmed whether or not it was released by females. Blends of two, three, four, and five components were all equally attractive in our 1983 field trapping experiments (Table 3). As demonstrated in our 1982 field tests, both Z 7 - 1 2 : Ac and Z9-14 : Ac were required for maximum activity. However, in the 1983 tests with blends formulated to release the same ratio emitted by calling females, all the blends were significantly more active than 25 mg of Z 9 - 1 2 : A c dispensed from a polyethylene vial. In fact Z 9 - 1 2 : A c , formulated in the same manner and quantity as the blends, was not different significantly from the blank.

Table 3. COMPARISON OF STICKY TRAP CAPTURES OF S. frugiperda MALES WITH BLENDS OF SYNTHETIC COMPOUNDS FORMULATED ON RUBBER SEPTA TO RELEASE RATIOS CORRESPONDING TO THOSE FOUND IN VOLATILES COLLECTED FROM CALLING FEMALES, GAINESVILLE, FLORIDA, SEPTEMBER 19-27, 1983"

SI2:Ac 0 0 0 5.0 (1.0) 4.9 (1.0)

Z7-12:Ac 3.4 3.3 3.0 3.2 3.1

(0.6) (0.5) (0.4) (0.5) (0.5)

Z9-12:Ac 0 0 1.7 (0.3) 0 1.7 (0.3) 25 rag"

Z9-14:Ac 96.6 (99.4) 92.9 (80.3) 91.9 (81.6) 88.3 (79.5) 86.9 (79.5) 100.0

100.0 Hexane blank

Zll-16:Ac 0 3.8 3.4 3.6 3.5

(19.2) (17.7) (19.0) (18.7)

Mean/trap per night ( • SE) ~' 14.5 + 1.5 a 15.3 _+ 1.6 a 14.9 + 1.6 a 14.4 + 1.8 a 12.3 + 2.0 a 3.9 _+ 0.8 b 1.8 +_ 0.7 c 0.9 _+ 0.6 cd 0.1 + 0.1 d

~Numbers under each compound in each row indicate the approximate percentage of that component released in the blend. Numbers in parentheses indicate the percent of each compound loaded to obtain the desired release ratio. Each septum was loaded with 2 mg of the total blend dissolved in 200 #1 of hexane. bEighteen replicates per treatment; treatments in blocks (N = 3) were randomized after each collection (N = 6); means not followed by the same letters differ significantly at the 5% level, Duncan's (1955) multiple-range test. ' Twenty-five mg of Z9-12 : Ac in a polyethylene vial.

3.3 3.2 3.2 3.4 3.0 3.3 3.3 3.2

(0.5) (0.5) (0.5) (0.6) (0.4) (0.6) (0.6) (0.5)

Z7-12 :Ac

2.3 (0.4) 2.2 (0.4)

1.7 (0.3)

Z9-12: Ac

2.3 (0.5)

2.2 (0.5) 2.2 (0.5)

All-12:Ac 92.9 90.1 91.9 96.6 91.9 94.4 94.4 90.1

(80.3) (84.4) (84.5) (99.4) (81.6) (98.9) (99.0) (84.5)

Z9-14:Ac

2.6 (14.2)

3.4 (17.7)

3.8 (19.2) 2.6 (14.2) 2.7 (14.5)

Zll-16:Ac + + + 444-

0.0 b

19.7 18.2 15.5 15.3 15.0 13.4

1.8 2.3 1.7 1.8 1.8 1.8

a a a a a a

Test 1 (August, 9-20)

b

16.9 + 16.9 + 12.6 417.2 413.5 + 8.6 412.8 40.0 b

2.2 1.0 1.8 2.9 1.3 1.2 1.9

a a a a a a a

Test 2 (August, 21-29)

Total catch (% •

aNumbers under each compound in each row indicate the approximate percentage of the component released in the blend. Numbers in parentheses indicate the percent of each compound loaded to obtain the desired release ratio. Each septum was loaded with 2 mg of the total blend dissolved in 200/zl of hexane. bFor analysis of variance, percentages were transformed to arcsin x/X and mean separations were achieved using Duncan's (1955) multiple-range test. Treatments were replicated 21 (7 randomizations) and 15 (5 randomizations) times in tests 1 and 2, respectively. Means (reconverted to percentages) followed by different letters differ significantly at the 1% level. Totals of 10,077 and 28,114 moths were captured in tests 1 and 2, respectively.

1.9 (0.4) Hexane control

1.9 (0.4)

S-12:Ac

Release ratios (%) of pheromone components

COMPOUNDS FORMULATED ON RUBBER SEPTA TO RELEASE RATIOS CORRESPONDING TO THOSE FOUNDS IN VOLATILES COLLECTED FROM CALLING FEMALES~ GAINESVILLE, FLORIDA, 1984 a

Table 4. CAPTURE OF MALE FALL ARMYWORM MOTHS IN TRAPS BAITED WITH DIFFERENT BLENDS OF SYNTHETIC

z

t-

,-d

bo t..9

PHEROMONE OF FALL ARMYWORM

1923

50

c 40

2 COMPONENT

I

I

-r o 30

5 COMPONENT

J l-C] .~

20

b b I0 o

o

O. 20

O. 08

OOSE

2. O0

(MG)

FIG. 3. Capture of male FAW moths in IPM traps baited with different dosages of two blends of synthetic sex pheromone on rubber septa, September 10-17, 1984 (14 replicates per treatment; treatments randomized four times per test period). Two-component blend and approximate release ratios: Z7-12 : Ac, 3.4% and Z9-14 : Ac, 96.6%. Fivecomponent blend and approximate release ratios: S-12 : Ac, 1.9%, Z7-12 :Ac, 3.2%, l l - 1 2 : A c , 2.2%, Z9-14:Ac, 90.1%, and Z l l - 1 6 : A c , 2.6%. Percentages were converted to arcsin ~ for statistical analysis. A total of 4668 FAW males were captured. Bars (reconverted to percentages) with different letters differ significantly at the 1% level, Duncan's (1955) multiple-range test. Narrow lines above bars represent SE of the mean. Our 1984 field trapping tests were designed to evaluate blends containing the compounds and ratios found in our latter analyses and to compare these blends with some of the previously tested blends containing Z9-12 : A c rather than 1 1 - 1 2 : A c . The blends tested and the results are given in Table 4. The numbers of moths captured per trap per night were much greater than in 1983. However, as the data in Table 4 illustrate, it is still difficult to determine which blend, if any, is optimum as a trap bait. There are indications that some blends are better, although they cannot be validated statistically. For example, the fivecomponent blend containing 11-12 : Ac, which is identical to the volatile blend collected from calling females, is consistently near the top in percentage of males captured. Also, the two-component blend is always in the middle to low range in percent captured although, statistically, it is always equal to the best blend.

! 924

T U M L I N S O N ET AL.

Thus a field trapping test was conducted in 1984 to compare the effectiveness of the two-component and five-component blends. IPM traps, each baited with a rubber septum containing 0.06, 0.2, or 2 mg of either the two-component or the five-component blend, were deployed in randomized complete blocks as previously described (14 replicates). Additionally, volatiles were collected from septa loaded with each of these blends at the doses indicated and analyzed by capillary GC to determine the release rates of the blends. The septa were aerated at flow rates of 1 and 2 liters/m. At any given dose and flow rate, the release rate of the two-component blend from a septum was the same as that of the five-component blend within experimental error. Additionally, the results of the field trapping experiment (Figure 3) indicate that both blends are equivalent as trap baits at each of the three doses tested. CONCLUSIONS These data support the conclusion that the sex pheromone of the FAW consists of at least two components, Z 7 - 1 2 : A c and Z 9 - 1 4 : A c . Three other components also may have a role in the pheromonal communication system of this species, but behavioral studies will be required to define this pheromone system more accurately and precisely. The reason for the capture of male F A W in traps baited with Z9-12 : Ac is not known. Although we could not detect this compound in gland extracts or volatiles from F A W females, Jones and Sparks (1979) found that traps baited with 500 tzg of Z9-12 : Ac on a dental wick captured twice as many FAW males as did F A W females. However, we found that traps baited with 2 mg of Z91 2 : A c on a rubber septum captured no more F A W males than traps baited with a hexane blank and that much higher doses of Z 9 - 1 2 : A c are required for attraction equal to 2 mg of the blends containing both Z 7 - 1 2 : A c and Z9-14 :Ac. It is likely that cotton wicks release these compounds at a much higher rate than rubber septa (although we have no data to substantiate this), and that Z 9 - 1 2 : A c is active only at concentrations much higher than that of the pheromone produced by a calling female. Thus, we now have identified a pheromone blend that is a more powerful attractant in the field than anything available previously. It is already proving useful in monitoring programs and for studying migration by the fall armyworm (Mitchell et al., 1985). Acknowledgments--We thank R. Hines and W. Copeland for their assistance in conducting field tests, and V. Bauder for assistance in formulating synthetic pheromone blends for field tests.

REFERENCES AYDREWS,K.L. 1980. The whorlwonn, Spodopterafrugiperda, in Central America and neighboring areas. Fla. Entomol. 63:456-467.

PHEROMONE OF FALL ARMYWORM

] 925

BURTON, R.L. 1969. Mass rearing the corn earworm in the laboratory. U.S. Dep. Agric. Tech. Bull. ARS-33-134, 8 pp. CROSS, J.H. 1980. A vapor collection and thermal desorption method to measure semiochemical release rates from controlled release formulations. J. Chem. Ecol. 6:781-787. CROSS, J.H., BYLER, R.C., CASS1DY,R,F., SILVERSTEIN,R.M., GREENBLATT,R.E., BURKHOLDER, W.E., LEVlNSON, A.R., and LEVINSON, H.Z. 1976. Porapak-Q collection of pheromone components and isolation of (Z)- and (E)-14-methyl-8-hexadecenal, sex pheromone components from the females of four species of Trogoderma (Coleoptera: Dermestidae). J. Chem. Ecol. 2:457-468. DUNCAN, J.H. 1955. Multiple-range and multiple-F tests. Biometrics 11:1-42. GROt~, K., and Z~RCHER, F. 1976. Stripping of trace organic substances from water. Equipment and procedures. J. Chromatogr. 117:285-294. HEATH, R.R., and DOOLITTLE, R.E. 1983. Derivatives of cholesteryl cinnamate: A comparison of the separations of geometrical isomers when used as gas chromatographic stationary phases. J. HRC CC 6:16-19. HEATH, R.R., and TUMLINSON, J.H. 1986. Correlation of retention times on a liquid crystal capillary column with reported vapor pressures and half-lives of compounds used in pheromone formulations. J. Chem. Ecol. 12:2081-2088. HEATH, R.R., TUMLINSON,J.H., DOOLITTLE, R.E., and DUNCAN, J.H. 1977. Analytical and preparative separation of geometrical isomers by high efficiency silver nitrate liquid chromatography. J. Chromatogr. Sci. 15:10-13. HEATH, R.R., JORDAN, J.R., SONNET, P.E., and TUMLINSON, J.H. 1979. Potential for the separation of insect pheromones by gas chromatography on columns coated with cholesteryl cinnamate, a liquid-crystal phase. J. HRC CC 2:712-714. HEATH, R.R., BURNSED, G.E., TUMLINSON, J.H., and DOOLITTLE, R.E. 1980. Separation of a series of positional and geometrical isomers of olefinic aliphatic primary alcohols and acetates by capillary gas chromatography. J. Chromatogr., 189:199-208. HEATH, R.R., TEAL, P.E.A., TUMLINSON, J.H., and MENGELKOCH, L.J. 1986. Prediction of release of multicomponent pheromone blends from rubber septa: A first approximation using relative vapor pressure calculated from retention indices on a liquid crystal column. J. Chem. Ecol. 12:2133-2143. HILL, A.S., CARDt~,R., KIDO, A., and ROELOFS, W.L. 1975. Sex pheromone of the orange tortrix moth Argyrotaemia citrana (Lepidoptera: Tortricidae). J. Chem. Ecol. 1:215-224. JONES, R.L., and SPARKS, A.N. 1979. (Z)-9-Tetradecen-l-ol acetate: A secondary sex pheromone of the fall armyworm, Spodopterafrugiperda (J.E. Smith). J. Chem. Ecol. 5:721-725. MITCHELL, E.R. 1978. Monitoring adult populations of the fall armyworm. Fla. Entomol. 62:9198. MITCHELL,E.R., and DOOLITTLE, R.E. 1976. Sex pheromones of Spodoptera exigua, S. eridania, and S. frugiperda: Bioassay for field activity. J. Econ. EntomoL 69:324-326. MITCHELL, E.R., and MCLAUGHL1N, J.R. 1982. Suppression of mating and oviposition by fall armyworm and mating by corn earworm in corn, using the air permeation technique. J. Econ. Entomol. 75:270-273. MrrCHELL, E.R., SUGIE, H., and TUMLtNSON, J.H. 1983. Rubber septa as a dispenser for the fall armyworm sex attractant. J. Environ. Sci. Health A18:463-470. MITCHELL,E.R., TUMLINSON,J.H., and McNEIL, J.N. 1985. Field evaluation of commercial pheromone formulations and traps using a more effective sex pheromone blend for the fall armyworm (Lepidoptera: Noctuidae). J. Econ. EntomoL 78:1364-1369. SEKUL, A.A., and SPARKS, A.N. 1967. Sex pheromone of the fall armyworm moth: Isolation, identification, and synthesis. J. Econ. Entomol. 60:1270-1272. SEKUL, A.A., and SPARKS,A.N. 1976. Sex attractant of the fall armyworm moth. USDA Tech. Bull. 1542, 6 pp.

1926

TUMLINSON ET AL.

SPARKS,A.N. 1980. Pheromones: Potential for use in monitoring and managing populations of the fall armyworm. Fla. Entomol. 63:406-410. STEEL, R.G.D., and TORRIE, J.H. t960. Principles and Procedures of Statistics. McGraw-Hill, New York, 481 pp. TUMLtNSON, J.H., HEATH, R.R., and TEAL, P.E.A. 1982. Analysis of chemical communications systems of Lepidoptera, pp. 1-25, in B.A. Leonhardt and M. Beroza (eds.). Insect Pheromone Technology: Chemistry and Applications. American Chemical Society Symposium Series, No. 190.