Journal of Chemical Ecology, Vol. 15, No. 1, 1989
PHEROMONE EMISSION A N D BLEND PERCENTAGES IN E o r e u m a loftini I DETERMINED BY TWO METHODS
D.C.
ROBACKER 2 and K.J.R.
J O H N S O N 3'4
2Subtropical Fruit Insects Research USDA-ARS, Weslaco, Texas 78596 3Texas A & M University Agricultural Research and Extension Center Weslaco, Texas 78596 (Received July 23, 1987; accepted October 29, 1987) Abstract--Calling behavior and pheromone emission by virgin female E. lofiini moths were investigated in laboratory experiments. Calling peaked during the last three hours of the night. Three-day-old moths called more than older or younger moths and began calling earlier in the night than younger moths. Female emissions were collected in flasks without airflow and cylinders with airflow. Control tests indicated that the pheromone components (Z)-ll-hexadecenyl acetate (HDA) and (Z)-13-octadecenylacetate (ODA) were 69 and 54% adsorbed on moths, respectively, and the component (Z)13-octadecenal (ODL) was 92-99 % adsorbed depending on its concentration, when put into flasks with noncalling females for 4 hr. Pheromone exposed to moths for less than 4 hr was adsorbed less. After correction for adsorption, the pheromone blend from females calling in flasks was 9 : 42 : 49 % of HDA/ ODL/ODA with an overall emission rate of 58 ng/female/4 hr. Three-dayold females emitted mo~'e pheromone than 0- to 2- or 4- to 5-day-old moths, in flasks. Little or no pheromone put into cylinders either downwind or upwind from a male moth was adsorbed by the moth. The pheromone blends from females calling in cylinders, corrected using downwind and upwind control test results, respectively, were 15 : 35 : 50 and 13 : 40 : 48% of HDA/ODL/ ODA with overall emission rates of 32 and 35 ng/female/night. Key Words--Sex pheromones, pheromone adsorption, age, pheromone emission, Lepidoptera, Pyralidae, Crambinae, Eoreuma loftini, calling behavior, (Z)-ll-hexadecenyl acetate, (Z)-13-octadecenyl acetate, (Z)-13octadecenal.
t Lepidoptera: Pyralidae. 4 Present address: Oregon Department of Agriculture, 635 Capitol Street NE, Salem, Oregon 97310.
105
106
ROBACKER AND JOHNSON INTRODUCTION
Three components of the sex pheromone of Eoreuma loftini (Dyar) have been identified by Shaver et al. (1988). The three chemicals, (Z)-I 1-hexadecenyl acetate (HDA), (Z)-13-octadecenal (ODL), and (Z)-13-octadecenyl acetate (ODA), proved active as sexual stimulants and attractants in laboratory and field experiments when tested in ratios found in abdomens of calling females. However, amounts in abdomens do not necessarily reflect blends emitted by the moths (Tumlinson et al., 1982). In particular, aldehydes are frequently emitted in greater blend percentages than are present in glands (Baker et al., 1980; Ramaswamy and Cardr, 1984). Our work reports when female E. loftini of various ages call and how much of each component they emit, as determined by two methods.
METHODS AND MATERIALS
General. Pupae were from cultures that originated in the Lower Rio Grande Valley (LRGV) of Texas. Insects were maintained on a 14:10 light-dark cycle similar to natural summer conditions in the LRGV. Laboratory temperatures ranged between 20 and 27°C and relative humidity between 40 and 75 %. Eclosion began after three to five days under these conditions. Female moths of each age were held in separate wood-frame, screened cages (20 cm/side). Cages were sprayed with water each day during the light period. Calling Behavior Observations. The numbers of "calling" virgin females of various ages were recorded every half hour during the last 7 hr of the night using night-vision goggles. Calling behavior was described by Brown et al. (1988). Ages were 0, I, 2, 3, and 4-5 days posteclosion. Since most moths eclosed at night, 0 days posteclosion indicates moths that emerged earlier on the test night. Between 62 and 122 moths of each age were observed over three test nights. Pheromone Emission in Closed Flasks without Airflow. We collected pheromone emitted by females of various ages in flasks with no airflow using a method similar to that of Baker et al. (1980). Ages were 0, 1, 2, 3, 4, and 5 days posteclosion. For each replication, 25 virgin females of one age were put into a 500-ml round-bottom flask with a glass stopper during the last 4 hr of the night (N = 10 replications/age). At the onset of the light period, the flask was chilled at 0°C for 1 h to freeze the pheromone onto the sides of the flask. The dead moths were then discarded from the cold flask. The flask was extracted by shaking for 5 min with each of two 20-ml portions of pentane. The portions were combined, filtered, and concentrated to 100/xl under a stream of nitrogen with low heat.
PHEROMONE EMISSION
107
To control for sample-handling losses, 100 ng of each component were added to empty 500-ml flasks 4 hr before extraction (N = 40). To control for pheromone losses due to adsorption on moths, three additional experiments were conducted. In the first experiment, 0.4, 0.6, and 0.6/~g of HDA, ODL, and ODA, respectively, and 25 males were put into 500-ml flasks 4 hr before extraction (N = 6). In the second experiment, various amounts of the three pheromone components and 25 noncalling females were put into flasks 4 hr before extraction. Noncalling females were virgin females used during the photophase. Since these moths moved little compared to calling females, flasks were shaken slightly each hour to stimulate activity. Amounts of HDA, ODL, and ODA, respectively, added to flasks were: 0, 0, and 0/zg (N = 2); 0.1, 0.15, and 0.15 /zg (N = 3); 0.4, 0.6, and 0.6/zg (N = 8); 0.7, 1.0, and 1.0 ~g (N = 5); 1.0, 1.5, and 1.5/~g (N = 3); 1.3, 2.0, and 2.0/zg (N = 6); and 2.0, 3.0, and 3.0 /zg (N = 3). In the third experiment, 25 noncalling females were put into flasks and 0.4, 0.6, and 0.6 tzg of the three components were introduced in four equal increments at 4, 3, 2, and 1 hr before extraction (N = 8). This experiment was to measure adsorption when pheromone was introduced in a way more like calling females would emit it. Extractions of flasks in control experiments were performed exactly as in tests with calling females. Pheromone Emission in Cylinders with Airflow. This method was used to minimize adsorption of pheromone on calling females. Females of ages 0, 1, 2, 3, 4, and 5 days posteclosion were tested (N = 2/age). For each replication, one virgin female was put into a vertical cylinder (20 cm long x 0.85 cm ID) with its abdomen oriented downward. The tapered bottom of the cylinder contained glass wool covered with 20 mg of Porapak Q (Supelco, Inc., Bellefonte, Pennsylvania) (50/80 mesh). A 5-cm section containing the moth was located 10 cm above the Porapak Q and was delineated above and below by plugs of glass wool. Another 50 mg of Porapak Q covered the glass wool over the moth section. A humidified airflow of ca. 2-5 cm/sec was directed downward through the cylinder. Moths remained in the cylinders for the entire 10-hr scotophase, except 0-day-old moths, which were put into the cylinders for the last 4 hr of the night. Cylinders containing the bottom 20 mg of Porapak Q were cleaned with successive rinses of 100 ml of acetone, I00 ml of 10% acetone in pentane, and 50 ml of pentane before testing. To extract pheromone, the cylinder, the bottom 20 mg of Porapak Q, and all glass wool downwind from the moth were rinsed with 20 ml of 5% acetone in pentane. The extract was concentrated to ca. 200 p.l under nitrogen at low heat. Several control experiments were conducted. First, 40, 60, and 60 ng of HDA, ODL, and ODA, respectively, were introduced onto the walls of a cylinder 1 cm downwind of a male moth 4 hr before extraction (N = 2). Second, 40, 60, and 60 ng were introduced on the glass wool upwind from a male moth
108
ROBACKER AND JOHNSON
4 hr before extraction (N = 9). The second experiment was also conducted with no moth in the cylinder to control for adsorption onto the upwind glass wool (N = 2). Finally, the experiment was conducted with a male moth present but no pheromone introduced to the cylinder (N = 4). Extractions were performed exactly as with cylinders containing calling females. Chemical Analyses. Extracts were prepared for gas chromatography (GLC) using a Waters liquid chromatograph (HPLC) (Waters Associates, Milford, Massachusetts) to remove hydrocarbons that interfere with detection of the pheromones. Two methods were used, as dictated by availability of materials. For flasks with calling females and flask controls without moths, the extract was injected onto a NOVA-PAK C18 Radial-PAK HPLC cartridge (4/~m particle size) in an RCM-100 module using a mobile phase of acetonitrile at 2 ml/ min. Detection was at 205 nm (model 490 Programmable Multiwavelength Detector). Eluent fractions were collected from 2 ml of elution volume before the retention volumes of the three pheromones until 2 ml of elution volume after the pheromone retention volumes, as determined by standards. The k' values were ca. 4, 4, and 7 for ODL, HDA, and ODA, respectively. The fractions were combined and reduced to 5-10 ~tl under a stream of nitrogen at low heat. Gas chromatography was conducted using a Shimadzu GC-9A instrument (Shimadzu Scientific Instruments Inc., Columbia, Maryland) with a flame ionization detector. Analyses were on an SPB-5 capillary column (Supelco) (60 m, fused silica, 0.32 mm ID, 1 tzm film). Helium carrier gas linear velocity was 25 cm/sec. A splitless injection (0.5 min split valve closed) of 1-2/zl was used. The column oven conditions were: 50°C for 5 min; 10°C/min increase until 250°C; 250°C isothermal. For all other pheromone emission and control experiments, extracts were injected onto a Resolve silica Radial-PAK HPLC cartridge (5/~m particle size) in an RCM-100 module. The mobile phase was 2 ml/min of the following: hexane for 2 min; 0.5% isopropyl alcohol in hexane f o r 4 min; and 1% isopropyl alcohol in hexane for 14 min. Eluent was collected from 18-30 mt of elution volume since ODL eluted between 20-24 ml and HDA and ODA eluted between 26-28 ml. The fraction was concentrated to 5-7 /zl, all of which was injected onto a DB-225 Megabore column (J & W Scientific, Rancho Cordova, Califomia) (30 m, fused silica, 0.53 mm ID, 1 tzm film). Helium carrier gas linear velocity was 60 cm/sec. A splitless injection was used as before. The column oven conditions were: 50°C for 5 min; 10°C/min increase until 170°C; 2°C/ min increase until 190°C; 190°C isothermal. The two chemistry methods proved identical by tests with standards. Pheromone amounts were quantitated by comparing peak heights to those of standards. Standard HDA and ODL (Sigma Chemical Company, St. Louis Missouri) were greater than 95 % pure. Standard ODA (Bedoukian Research, Danbury, Connecticut) was 99% pure. Known amounts of (Z)-I 1-tetradecenyl
109
PHEROMONE EMISSION
acetate and methyl arachidonate were added to the syringe and coinjected with pheromone standards and extracts. Elution of (Z)-I 1-tetradecenyl acetate preceded the earliest-eluting pheromone component, and elution of methyl arachidonate followed the latest-eluting pheromone. The two purposes for these chemicals were to correct for retention-time shifts and response-factor changes over the course of numerous analyses with varying injection volumes. Statistical Analyses. The calling experiment was a randomized complete block design with three replications of a two-way classification of five age and 15 time treatments. Proportions of calling moths were transformed to arcsins of the square roots for analysis of variance. Age, time, and age/time interactions were included in the model. Means and groups of means were compared by t tests and single-degree-of-freedom F contrasts, respectively, using the error term from the analysis of variance. The flask pheromone-collection experiment was completely randomized. Analyses of variance tested effects of age on total emission (HDA + ODL + ODA) and on blend percentages, both uncorrected for adsorption on moths. Flask control experiments were completely randomized. For tests with noncalling females in which pheromone was introduced 4 hr before extraction, analyses of variance tested effects of amounts added to flasks on the percent recoveries of the three components. The treatment effect was partitioned into linear regression and deviations. Effects were tested using the experimental error term. Estimations of variances of derived results calculated by multiplication and/ or division of two or more means were conducted by addition of relative variances (Skoog and West, 1969).
RESULTS AND DISCUSSION
Calling Behavior. Few females called during the first half of the 10-hr night according to preliminary observations. Most calling occurred during the last 3 hr of the night (Figure l). In general, older moths began calling earlier. This was demonstrated by a large age/time interaction (P < 0.001) resulting from earlier rises in the curves of the older moths. Kanno (1979) and Das and Islam (1982) also found that older Pyralidae females called earlier than younger moths. More older moths called at most times during the peak calling period. More 1-day-old than 0-day-old moths, more 2-day-old than 1-day-old moths, and more 3- and 4- to 5-day-old than 2-day-old moths called (P < 0.01). An exception was that fewer 4- to 5-day-old moths than 3-day-old moths called during the peak hours (P < 0.01), possibly due to senescence (Tamaki, 1985). Similar effects of age on maximum calling rate have been observed in other
110
ROBACKER AND JOHNSON
t~ ta/a
70
ii i
(.9
_z 60 J J
4-5
•
.J .--\\
•
//".~
/I
/
\
Z'-'-+ -~'~ //'~,
//,,'-[\
(..3
17
cn 50 LJ
_..I
m k.L 40 0
"f ~' 30 pZ
o 20 ~A 13._
10
_.t 7
/// 6
5
4
3
2
1
0
HOURS BEFORE LIGHT PERIOD
FIG. 1. Calling behavior of virgin female Eoreuma Io35ini moths of different ages.
Pyralidae (Kanno and Sato, 1978; Kanno, 1979; Das and Islam, 1982; Hirai, 1982). Calling behavior declined just before the beginning of the light period (0 hr = 2-3 min before the lights came on) (Figure 1). Significant decreases occurred during the last hour for 0-, 1-, 3-, and 4- to 5-day-old moths (P < 0.05). This suggests that cessation, as well as onset, of calling behavior is a circadian rhythm rather than simply a direct response to light. Pheromone Recover 3, in Flask Control Experiments. Percent recoveries of pheromone components from flasks without moths were: HDA, 49.9%; ODL, 42.1%; and ODA, 49.5 % (Table 1). Nearly all losses occurred during the final concentrating procedure and syringe loading before GLC analysis. Pheromone amounts extracted from flasks with noncalling females to which no pheromone was added were: HDA, 0.4 ng/25 females; ODL, 0.3 ng; and ODA, 5.0 ng. It is important that the amount of ODL extracted from these flasks was low since experiments discussed below showed that ODL recovery
PHEROMONEEMISSION
111
TABLE 1. PERCENTRECOVERIESOF PHEROMONEFROMFLASKCONTROLEXPERIMENTS COMPARINGPRESENCEOF MOTHS, SEX, AND TIMINGOF PHEROMONEINTRODUCTION Percent recoveries (ME)~
Without mothsh With males~ With noncalling females' Noncalling females, incrementsa
HDA
ODL
ODA
49.9 (3.2) d 9.6 (1.5) a
42.1 (3.0) c 0.33 (0.07) a
49.5 (3.3) c 15.8 (2.2) a
14.9 (1.2) b 20.0 (2.0) c
0.56 (0.11) ab 1.56 (0.51) b
24.4 (2.0) b 29.2 (3.0) b
"HDA = (Z)-I l-hexadecenyl acetate; ODL = (Z)-13-octadecenal;ODA = (Z)-13-octadecenyl acetate. Means in the same vertical column followed by the same letter are not significantlydifferent at the 5% level by t tests. ~'0.1 /~g each of HDA, ODL, ODA introduced4 hr before extraction. "0.4 ttg HDA, 0.6 #g ODL, 0.6 ttg ODA, introduced4 hr before extraction. d0.4 #g HDA, 0.6 t~g ODL, 0.6 p,g ODA, introducedin four incrementsat 4, 3, 2, and 1 hr before extraction.
rates depend on amounts in flasks. The extracted 0.3 ng ODL translates into 0.016 ttg in the flask before losses, according to methods discussed in a later section. This amount is small relative to amounts of ODL introduced to flasks ( 0 . 1 5 - 3 . 0 / z g ) and probably little affected recovery rates. U n k n o w n compounds with retention times within 0 . 1 - 0 . 3 min of the pheromone peaks were also found in flasks with noncalling females to which no pheromone was added. These unknowns interfered with pheromone quantitation. Therefore, to correct data from tests in which pheromone was added to flasks containing noncalling females, we subtracted the following amounts from those recovered: HDA, 1.4 ng/25 females; ODL, 1.8 ng; and ODA, 8.1 ng. These amounts include pheromone from noncalling females and the unknown peaks. Recoveries from flasks containing 0.4 #g HDA, 0.6 tzg ODL, 0.6 ~zg ODA and moths were lower than recoveries from flasks without moths (P < 0.05) (Table 1), indicating adsorption onto moths that were discarded before flask extraction. Pheromone recoveries from flasks containing males, to which pheromone was introduced 4 hr before extraction, were: HDA, 9.6%; ODL, 0.33 %; and ODA, 15.8%. Recoveries from flasks containing noncalling females, to which pheromone was also introduced 4 hr before extraction, were: HDA, 14.9%; ODL, 0.56%; and ODA, 24.4%. Male/female differences in recoveries (HDA, ODA: P < 0.05; ODL: P = 0.10) may be related to behavior as males were more active during tests. Recoveries from flasks containing noncalling
112
ROBACKER AND JOHNSON
females to which pheromone was introduced in four increments were: HDA, 20.0%; ODL, 1.56%; and ODA, 29.2%. These recoveries were higher than from flasks with noncalling females to which all pheromone was added 4 hr before extraction (HDA, P < 0.05; ODL, P = 0.08; ODA, P = 0.21). Pheromone introduced at 1, 2, and 3 hr before extraction apparently did not have as much time to adsorb on moths as that added 4 hr before extraction. The greater the adsorption affinity, the more the incremental technique increased recovery rate. Thus, recovery of ODL increased more (178%) than recoveries of HDA (34%) and ODA (20%). Since the incremental technique is similar to pheromone emission by females, these results suggest recoveries from flasks with calling females should also be 20.0%, 1.56%, and 29.2% for HDA, ODL, and ODA, respectively. Note, however, that these values are accurate only for HDA and ODA. Recovery of ODL depends on the amount of ODL in flasks (0.6 ~g in this test) as reported below. Flask experiments with noncalling females and various amounts of pheromone added 4 hr before extraction indicated that percent recovery was independent of amount added for HDA and ODA (regression coefficients were not significant) but not for ODL. Mean recoveries of HDA and ODA, over all amounts added to flasks, were ca. 16 and 23%, respectively. Regression of percent recovery of ODL on amounts added to flasks gave the equation: Recovery (%) = 1.38 (amount added to flask) - 0.17 The coefficient, 1.38, was significant (P < 0.05). The correlation coefficient, r, was 0.46. Linear regression (1 df) accounted for 84% of the treatment effect (5 df), indicating a good fit to linearity. The low r was due to variability associated with the greater amounts of ODL introduced to flasks (1.5, 2.0, 3.0/zg). Recovery rates ranged from 0.2 to 10.1% for these treatments compared to 0.11.8% for the three lower treatments. Estimation of Percent Adsorption on Moths. Percent adsorptions were estimated for flask control experiments with noncalling females in which various amounts of pheromone were added 4 hr before extraction (Table 2). Percent adsorption was calculated by dividing the percent recovery from flasks with moths by the percent recovery from flasks without moths, subtracting the quotient from 1, and multiplying by 100. This calculation assumed that handling losses from flasks without moths does not depend on amounts added to flasks. Table 2 shows that HDA and ODA adsorption rates on moths did not vary with amounts of HDA and ODA added, but ODL adsorption increased with decreasing amounts of ODL added to flasks (P < 0.05 by regression). Mean adsorptions were: HDA, 69.0%; ODL, 96.2%; and ODA, 54.1%. Similar adsorption trends were reported for Grapholitha molesta (Lepidoptera: Tortricidae) with two components of its pheromone (Baker et al., 1980). Effects of Age on Pheromone Emission by Females in Closed Flasks. Blend
PHEROMONE EMISSION
1 13
TABLE 2. PERCENT ADSORPTION OF PHEROMONE BY NONCALLING FEMALE MOTHS IN FLASK CONTROL EXPERIMENTS WITH VARIOUS AMOUNTS OF INTRODUCED PHEROMONE
Percent adsorbed (ME)" Amount introduced to flask (/.tg)
HDA
0.10 0.15 0.40 0.60 0.70 1.0 1.3 1.5 2.0 3.0 Overall Mean
71.5 (4.4) -70.2 (2.4) -77.6 (3.2) 72.3 (4.4) 58.8 (6.4) -66.4 (7.2) -69.0 (2.1)
ODL ab
-99.4 (0.2) -98.7 (0.3) -98.6 (0.7) -93.6 (5.0) 92.6 (3.7) 92.2 (5.9) 96.2 (1.1)
ab b ab a ab
ODA -56.0 (6.0) -50.7 (4.0) -59.7 (5.5) -60.0 (4.6) 48.8 (5.6) 52.7 (7.7) 54.1 (2.2)
a a a a a a
a a a a a a
"HDA = (Z)-I l-hexadecenyl acetate; ODL = (Z)-13-octadecenal; ODA = (Z)-13-octadecenyl acetate. Means in the same vertical column followed by the same letter are not significantly different at the 5 % level by LSD.
percentages
of the three components,
corrected for handling losses but not for
a d s o r p t i o n o n m o t h s , c h a n g e d w i t h m o t h a g e ( T a b l e 3). A p p a r e n t b l e n d s b e c a m e r i c h e r in O D A
a n d p o o r e r in H D A
to 3 d a y s ( P
PHEROMONE EMISSION A N D BLEND PERCENTAGES IN E o r e u m a loftini I DETERMINED BY TWO METHODS
D.C.
ROBACKER 2 and K.J.R.
J O H N S O N 3'4
2Subtropical Fruit Insects Research USDA-ARS, Weslaco, Texas 78596 3Texas A & M University Agricultural Research and Extension Center Weslaco, Texas 78596 (Received July 23, 1987; accepted October 29, 1987) Abstract--Calling behavior and pheromone emission by virgin female E. lofiini moths were investigated in laboratory experiments. Calling peaked during the last three hours of the night. Three-day-old moths called more than older or younger moths and began calling earlier in the night than younger moths. Female emissions were collected in flasks without airflow and cylinders with airflow. Control tests indicated that the pheromone components (Z)-ll-hexadecenyl acetate (HDA) and (Z)-13-octadecenylacetate (ODA) were 69 and 54% adsorbed on moths, respectively, and the component (Z)13-octadecenal (ODL) was 92-99 % adsorbed depending on its concentration, when put into flasks with noncalling females for 4 hr. Pheromone exposed to moths for less than 4 hr was adsorbed less. After correction for adsorption, the pheromone blend from females calling in flasks was 9 : 42 : 49 % of HDA/ ODL/ODA with an overall emission rate of 58 ng/female/4 hr. Three-dayold females emitted mo~'e pheromone than 0- to 2- or 4- to 5-day-old moths, in flasks. Little or no pheromone put into cylinders either downwind or upwind from a male moth was adsorbed by the moth. The pheromone blends from females calling in cylinders, corrected using downwind and upwind control test results, respectively, were 15 : 35 : 50 and 13 : 40 : 48% of HDA/ODL/ ODA with overall emission rates of 32 and 35 ng/female/night. Key Words--Sex pheromones, pheromone adsorption, age, pheromone emission, Lepidoptera, Pyralidae, Crambinae, Eoreuma loftini, calling behavior, (Z)-ll-hexadecenyl acetate, (Z)-13-octadecenyl acetate, (Z)-13octadecenal.
t Lepidoptera: Pyralidae. 4 Present address: Oregon Department of Agriculture, 635 Capitol Street NE, Salem, Oregon 97310.
105
106
ROBACKER AND JOHNSON INTRODUCTION
Three components of the sex pheromone of Eoreuma loftini (Dyar) have been identified by Shaver et al. (1988). The three chemicals, (Z)-I 1-hexadecenyl acetate (HDA), (Z)-13-octadecenal (ODL), and (Z)-13-octadecenyl acetate (ODA), proved active as sexual stimulants and attractants in laboratory and field experiments when tested in ratios found in abdomens of calling females. However, amounts in abdomens do not necessarily reflect blends emitted by the moths (Tumlinson et al., 1982). In particular, aldehydes are frequently emitted in greater blend percentages than are present in glands (Baker et al., 1980; Ramaswamy and Cardr, 1984). Our work reports when female E. loftini of various ages call and how much of each component they emit, as determined by two methods.
METHODS AND MATERIALS
General. Pupae were from cultures that originated in the Lower Rio Grande Valley (LRGV) of Texas. Insects were maintained on a 14:10 light-dark cycle similar to natural summer conditions in the LRGV. Laboratory temperatures ranged between 20 and 27°C and relative humidity between 40 and 75 %. Eclosion began after three to five days under these conditions. Female moths of each age were held in separate wood-frame, screened cages (20 cm/side). Cages were sprayed with water each day during the light period. Calling Behavior Observations. The numbers of "calling" virgin females of various ages were recorded every half hour during the last 7 hr of the night using night-vision goggles. Calling behavior was described by Brown et al. (1988). Ages were 0, I, 2, 3, and 4-5 days posteclosion. Since most moths eclosed at night, 0 days posteclosion indicates moths that emerged earlier on the test night. Between 62 and 122 moths of each age were observed over three test nights. Pheromone Emission in Closed Flasks without Airflow. We collected pheromone emitted by females of various ages in flasks with no airflow using a method similar to that of Baker et al. (1980). Ages were 0, 1, 2, 3, 4, and 5 days posteclosion. For each replication, 25 virgin females of one age were put into a 500-ml round-bottom flask with a glass stopper during the last 4 hr of the night (N = 10 replications/age). At the onset of the light period, the flask was chilled at 0°C for 1 h to freeze the pheromone onto the sides of the flask. The dead moths were then discarded from the cold flask. The flask was extracted by shaking for 5 min with each of two 20-ml portions of pentane. The portions were combined, filtered, and concentrated to 100/xl under a stream of nitrogen with low heat.
PHEROMONE EMISSION
107
To control for sample-handling losses, 100 ng of each component were added to empty 500-ml flasks 4 hr before extraction (N = 40). To control for pheromone losses due to adsorption on moths, three additional experiments were conducted. In the first experiment, 0.4, 0.6, and 0.6/~g of HDA, ODL, and ODA, respectively, and 25 males were put into 500-ml flasks 4 hr before extraction (N = 6). In the second experiment, various amounts of the three pheromone components and 25 noncalling females were put into flasks 4 hr before extraction. Noncalling females were virgin females used during the photophase. Since these moths moved little compared to calling females, flasks were shaken slightly each hour to stimulate activity. Amounts of HDA, ODL, and ODA, respectively, added to flasks were: 0, 0, and 0/zg (N = 2); 0.1, 0.15, and 0.15 /zg (N = 3); 0.4, 0.6, and 0.6/zg (N = 8); 0.7, 1.0, and 1.0 ~g (N = 5); 1.0, 1.5, and 1.5/~g (N = 3); 1.3, 2.0, and 2.0/zg (N = 6); and 2.0, 3.0, and 3.0 /zg (N = 3). In the third experiment, 25 noncalling females were put into flasks and 0.4, 0.6, and 0.6 tzg of the three components were introduced in four equal increments at 4, 3, 2, and 1 hr before extraction (N = 8). This experiment was to measure adsorption when pheromone was introduced in a way more like calling females would emit it. Extractions of flasks in control experiments were performed exactly as in tests with calling females. Pheromone Emission in Cylinders with Airflow. This method was used to minimize adsorption of pheromone on calling females. Females of ages 0, 1, 2, 3, 4, and 5 days posteclosion were tested (N = 2/age). For each replication, one virgin female was put into a vertical cylinder (20 cm long x 0.85 cm ID) with its abdomen oriented downward. The tapered bottom of the cylinder contained glass wool covered with 20 mg of Porapak Q (Supelco, Inc., Bellefonte, Pennsylvania) (50/80 mesh). A 5-cm section containing the moth was located 10 cm above the Porapak Q and was delineated above and below by plugs of glass wool. Another 50 mg of Porapak Q covered the glass wool over the moth section. A humidified airflow of ca. 2-5 cm/sec was directed downward through the cylinder. Moths remained in the cylinders for the entire 10-hr scotophase, except 0-day-old moths, which were put into the cylinders for the last 4 hr of the night. Cylinders containing the bottom 20 mg of Porapak Q were cleaned with successive rinses of 100 ml of acetone, I00 ml of 10% acetone in pentane, and 50 ml of pentane before testing. To extract pheromone, the cylinder, the bottom 20 mg of Porapak Q, and all glass wool downwind from the moth were rinsed with 20 ml of 5% acetone in pentane. The extract was concentrated to ca. 200 p.l under nitrogen at low heat. Several control experiments were conducted. First, 40, 60, and 60 ng of HDA, ODL, and ODA, respectively, were introduced onto the walls of a cylinder 1 cm downwind of a male moth 4 hr before extraction (N = 2). Second, 40, 60, and 60 ng were introduced on the glass wool upwind from a male moth
108
ROBACKER AND JOHNSON
4 hr before extraction (N = 9). The second experiment was also conducted with no moth in the cylinder to control for adsorption onto the upwind glass wool (N = 2). Finally, the experiment was conducted with a male moth present but no pheromone introduced to the cylinder (N = 4). Extractions were performed exactly as with cylinders containing calling females. Chemical Analyses. Extracts were prepared for gas chromatography (GLC) using a Waters liquid chromatograph (HPLC) (Waters Associates, Milford, Massachusetts) to remove hydrocarbons that interfere with detection of the pheromones. Two methods were used, as dictated by availability of materials. For flasks with calling females and flask controls without moths, the extract was injected onto a NOVA-PAK C18 Radial-PAK HPLC cartridge (4/~m particle size) in an RCM-100 module using a mobile phase of acetonitrile at 2 ml/ min. Detection was at 205 nm (model 490 Programmable Multiwavelength Detector). Eluent fractions were collected from 2 ml of elution volume before the retention volumes of the three pheromones until 2 ml of elution volume after the pheromone retention volumes, as determined by standards. The k' values were ca. 4, 4, and 7 for ODL, HDA, and ODA, respectively. The fractions were combined and reduced to 5-10 ~tl under a stream of nitrogen at low heat. Gas chromatography was conducted using a Shimadzu GC-9A instrument (Shimadzu Scientific Instruments Inc., Columbia, Maryland) with a flame ionization detector. Analyses were on an SPB-5 capillary column (Supelco) (60 m, fused silica, 0.32 mm ID, 1 tzm film). Helium carrier gas linear velocity was 25 cm/sec. A splitless injection (0.5 min split valve closed) of 1-2/zl was used. The column oven conditions were: 50°C for 5 min; 10°C/min increase until 250°C; 250°C isothermal. For all other pheromone emission and control experiments, extracts were injected onto a Resolve silica Radial-PAK HPLC cartridge (5/~m particle size) in an RCM-100 module. The mobile phase was 2 ml/min of the following: hexane for 2 min; 0.5% isopropyl alcohol in hexane f o r 4 min; and 1% isopropyl alcohol in hexane for 14 min. Eluent was collected from 18-30 mt of elution volume since ODL eluted between 20-24 ml and HDA and ODA eluted between 26-28 ml. The fraction was concentrated to 5-7 /zl, all of which was injected onto a DB-225 Megabore column (J & W Scientific, Rancho Cordova, Califomia) (30 m, fused silica, 0.53 mm ID, 1 tzm film). Helium carrier gas linear velocity was 60 cm/sec. A splitless injection was used as before. The column oven conditions were: 50°C for 5 min; 10°C/min increase until 170°C; 2°C/ min increase until 190°C; 190°C isothermal. The two chemistry methods proved identical by tests with standards. Pheromone amounts were quantitated by comparing peak heights to those of standards. Standard HDA and ODL (Sigma Chemical Company, St. Louis Missouri) were greater than 95 % pure. Standard ODA (Bedoukian Research, Danbury, Connecticut) was 99% pure. Known amounts of (Z)-I 1-tetradecenyl
109
PHEROMONE EMISSION
acetate and methyl arachidonate were added to the syringe and coinjected with pheromone standards and extracts. Elution of (Z)-I 1-tetradecenyl acetate preceded the earliest-eluting pheromone component, and elution of methyl arachidonate followed the latest-eluting pheromone. The two purposes for these chemicals were to correct for retention-time shifts and response-factor changes over the course of numerous analyses with varying injection volumes. Statistical Analyses. The calling experiment was a randomized complete block design with three replications of a two-way classification of five age and 15 time treatments. Proportions of calling moths were transformed to arcsins of the square roots for analysis of variance. Age, time, and age/time interactions were included in the model. Means and groups of means were compared by t tests and single-degree-of-freedom F contrasts, respectively, using the error term from the analysis of variance. The flask pheromone-collection experiment was completely randomized. Analyses of variance tested effects of age on total emission (HDA + ODL + ODA) and on blend percentages, both uncorrected for adsorption on moths. Flask control experiments were completely randomized. For tests with noncalling females in which pheromone was introduced 4 hr before extraction, analyses of variance tested effects of amounts added to flasks on the percent recoveries of the three components. The treatment effect was partitioned into linear regression and deviations. Effects were tested using the experimental error term. Estimations of variances of derived results calculated by multiplication and/ or division of two or more means were conducted by addition of relative variances (Skoog and West, 1969).
RESULTS AND DISCUSSION
Calling Behavior. Few females called during the first half of the 10-hr night according to preliminary observations. Most calling occurred during the last 3 hr of the night (Figure l). In general, older moths began calling earlier. This was demonstrated by a large age/time interaction (P < 0.001) resulting from earlier rises in the curves of the older moths. Kanno (1979) and Das and Islam (1982) also found that older Pyralidae females called earlier than younger moths. More older moths called at most times during the peak calling period. More 1-day-old than 0-day-old moths, more 2-day-old than 1-day-old moths, and more 3- and 4- to 5-day-old than 2-day-old moths called (P < 0.01). An exception was that fewer 4- to 5-day-old moths than 3-day-old moths called during the peak hours (P < 0.01), possibly due to senescence (Tamaki, 1985). Similar effects of age on maximum calling rate have been observed in other
110
ROBACKER AND JOHNSON
t~ ta/a
70
ii i
(.9
_z 60 J J
4-5
•
.J .--\\
•
//".~
/I
/
\
Z'-'-+ -~'~ //'~,
//,,'-[\
(..3
17
cn 50 LJ
_..I
m k.L 40 0
"f ~' 30 pZ
o 20 ~A 13._
10
_.t 7
/// 6
5
4
3
2
1
0
HOURS BEFORE LIGHT PERIOD
FIG. 1. Calling behavior of virgin female Eoreuma Io35ini moths of different ages.
Pyralidae (Kanno and Sato, 1978; Kanno, 1979; Das and Islam, 1982; Hirai, 1982). Calling behavior declined just before the beginning of the light period (0 hr = 2-3 min before the lights came on) (Figure 1). Significant decreases occurred during the last hour for 0-, 1-, 3-, and 4- to 5-day-old moths (P < 0.05). This suggests that cessation, as well as onset, of calling behavior is a circadian rhythm rather than simply a direct response to light. Pheromone Recover 3, in Flask Control Experiments. Percent recoveries of pheromone components from flasks without moths were: HDA, 49.9%; ODL, 42.1%; and ODA, 49.5 % (Table 1). Nearly all losses occurred during the final concentrating procedure and syringe loading before GLC analysis. Pheromone amounts extracted from flasks with noncalling females to which no pheromone was added were: HDA, 0.4 ng/25 females; ODL, 0.3 ng; and ODA, 5.0 ng. It is important that the amount of ODL extracted from these flasks was low since experiments discussed below showed that ODL recovery
PHEROMONEEMISSION
111
TABLE 1. PERCENTRECOVERIESOF PHEROMONEFROMFLASKCONTROLEXPERIMENTS COMPARINGPRESENCEOF MOTHS, SEX, AND TIMINGOF PHEROMONEINTRODUCTION Percent recoveries (ME)~
Without mothsh With males~ With noncalling females' Noncalling females, incrementsa
HDA
ODL
ODA
49.9 (3.2) d 9.6 (1.5) a
42.1 (3.0) c 0.33 (0.07) a
49.5 (3.3) c 15.8 (2.2) a
14.9 (1.2) b 20.0 (2.0) c
0.56 (0.11) ab 1.56 (0.51) b
24.4 (2.0) b 29.2 (3.0) b
"HDA = (Z)-I l-hexadecenyl acetate; ODL = (Z)-13-octadecenal;ODA = (Z)-13-octadecenyl acetate. Means in the same vertical column followed by the same letter are not significantlydifferent at the 5% level by t tests. ~'0.1 /~g each of HDA, ODL, ODA introduced4 hr before extraction. "0.4 ttg HDA, 0.6 #g ODL, 0.6 ttg ODA, introduced4 hr before extraction. d0.4 #g HDA, 0.6 t~g ODL, 0.6 p,g ODA, introducedin four incrementsat 4, 3, 2, and 1 hr before extraction.
rates depend on amounts in flasks. The extracted 0.3 ng ODL translates into 0.016 ttg in the flask before losses, according to methods discussed in a later section. This amount is small relative to amounts of ODL introduced to flasks ( 0 . 1 5 - 3 . 0 / z g ) and probably little affected recovery rates. U n k n o w n compounds with retention times within 0 . 1 - 0 . 3 min of the pheromone peaks were also found in flasks with noncalling females to which no pheromone was added. These unknowns interfered with pheromone quantitation. Therefore, to correct data from tests in which pheromone was added to flasks containing noncalling females, we subtracted the following amounts from those recovered: HDA, 1.4 ng/25 females; ODL, 1.8 ng; and ODA, 8.1 ng. These amounts include pheromone from noncalling females and the unknown peaks. Recoveries from flasks containing 0.4 #g HDA, 0.6 tzg ODL, 0.6 ~zg ODA and moths were lower than recoveries from flasks without moths (P < 0.05) (Table 1), indicating adsorption onto moths that were discarded before flask extraction. Pheromone recoveries from flasks containing males, to which pheromone was introduced 4 hr before extraction, were: HDA, 9.6%; ODL, 0.33 %; and ODA, 15.8%. Recoveries from flasks containing noncalling females, to which pheromone was also introduced 4 hr before extraction, were: HDA, 14.9%; ODL, 0.56%; and ODA, 24.4%. Male/female differences in recoveries (HDA, ODA: P < 0.05; ODL: P = 0.10) may be related to behavior as males were more active during tests. Recoveries from flasks containing noncalling
112
ROBACKER AND JOHNSON
females to which pheromone was introduced in four increments were: HDA, 20.0%; ODL, 1.56%; and ODA, 29.2%. These recoveries were higher than from flasks with noncalling females to which all pheromone was added 4 hr before extraction (HDA, P < 0.05; ODL, P = 0.08; ODA, P = 0.21). Pheromone introduced at 1, 2, and 3 hr before extraction apparently did not have as much time to adsorb on moths as that added 4 hr before extraction. The greater the adsorption affinity, the more the incremental technique increased recovery rate. Thus, recovery of ODL increased more (178%) than recoveries of HDA (34%) and ODA (20%). Since the incremental technique is similar to pheromone emission by females, these results suggest recoveries from flasks with calling females should also be 20.0%, 1.56%, and 29.2% for HDA, ODL, and ODA, respectively. Note, however, that these values are accurate only for HDA and ODA. Recovery of ODL depends on the amount of ODL in flasks (0.6 ~g in this test) as reported below. Flask experiments with noncalling females and various amounts of pheromone added 4 hr before extraction indicated that percent recovery was independent of amount added for HDA and ODA (regression coefficients were not significant) but not for ODL. Mean recoveries of HDA and ODA, over all amounts added to flasks, were ca. 16 and 23%, respectively. Regression of percent recovery of ODL on amounts added to flasks gave the equation: Recovery (%) = 1.38 (amount added to flask) - 0.17 The coefficient, 1.38, was significant (P < 0.05). The correlation coefficient, r, was 0.46. Linear regression (1 df) accounted for 84% of the treatment effect (5 df), indicating a good fit to linearity. The low r was due to variability associated with the greater amounts of ODL introduced to flasks (1.5, 2.0, 3.0/zg). Recovery rates ranged from 0.2 to 10.1% for these treatments compared to 0.11.8% for the three lower treatments. Estimation of Percent Adsorption on Moths. Percent adsorptions were estimated for flask control experiments with noncalling females in which various amounts of pheromone were added 4 hr before extraction (Table 2). Percent adsorption was calculated by dividing the percent recovery from flasks with moths by the percent recovery from flasks without moths, subtracting the quotient from 1, and multiplying by 100. This calculation assumed that handling losses from flasks without moths does not depend on amounts added to flasks. Table 2 shows that HDA and ODA adsorption rates on moths did not vary with amounts of HDA and ODA added, but ODL adsorption increased with decreasing amounts of ODL added to flasks (P < 0.05 by regression). Mean adsorptions were: HDA, 69.0%; ODL, 96.2%; and ODA, 54.1%. Similar adsorption trends were reported for Grapholitha molesta (Lepidoptera: Tortricidae) with two components of its pheromone (Baker et al., 1980). Effects of Age on Pheromone Emission by Females in Closed Flasks. Blend
PHEROMONE EMISSION
1 13
TABLE 2. PERCENT ADSORPTION OF PHEROMONE BY NONCALLING FEMALE MOTHS IN FLASK CONTROL EXPERIMENTS WITH VARIOUS AMOUNTS OF INTRODUCED PHEROMONE
Percent adsorbed (ME)" Amount introduced to flask (/.tg)
HDA
0.10 0.15 0.40 0.60 0.70 1.0 1.3 1.5 2.0 3.0 Overall Mean
71.5 (4.4) -70.2 (2.4) -77.6 (3.2) 72.3 (4.4) 58.8 (6.4) -66.4 (7.2) -69.0 (2.1)
ODL ab
-99.4 (0.2) -98.7 (0.3) -98.6 (0.7) -93.6 (5.0) 92.6 (3.7) 92.2 (5.9) 96.2 (1.1)
ab b ab a ab
ODA -56.0 (6.0) -50.7 (4.0) -59.7 (5.5) -60.0 (4.6) 48.8 (5.6) 52.7 (7.7) 54.1 (2.2)
a a a a a a
a a a a a a
"HDA = (Z)-I l-hexadecenyl acetate; ODL = (Z)-13-octadecenal; ODA = (Z)-13-octadecenyl acetate. Means in the same vertical column followed by the same letter are not significantly different at the 5 % level by LSD.
percentages
of the three components,
corrected for handling losses but not for
a d s o r p t i o n o n m o t h s , c h a n g e d w i t h m o t h a g e ( T a b l e 3). A p p a r e n t b l e n d s b e c a m e r i c h e r in O D A
a n d p o o r e r in H D A
to 3 d a y s ( P
