Journal of Chemical Ecology, Vol. 22, No. 1, 1996
IDENTIFICATION OF SEX PHEROMONE OF Tetanolita mynesalis (LEPIDOPTERA: NOCTUIDAE), A PREY SPECIES OF BOLAS SPIDER, Mastophora hutchinsoni
KENNETH
F.
HAYNES,
j'* KENNETH
V.
JOCELYN
G.
MILLAR,
z and
B. C H A S T A I N
BONNIE
YEARGAN,
I I
IDepartment of Entomology University of Kentucky Lexington, Kentucky. 40546 2Department of Entomology Universit3, of California Riverside, California 92521 (Received May 1, 1995; accepted August 28, 1995) Abstract--The bolas spider, Mastophora hutchinsoni, attracts adult males of four species of nocturnally active Lepidoptera through aggressive chemical mimicry of those species' sex pheromones. Here we report the identification of the sex pheromone of one prey species, the smoky tetanolita (Tetanolita mynesalis). In sex pheromone gland extracts, only two peaks stimulated an electrophysiological response as measured by a coupled gas chromatographyelectroantennographic detection analysis. These two peaks had retention times identical to (3Z,6Z,9Z)-heneicosatriene (3Z,6Z,9Z-21:H ) and (3Z,9Z)-cis-6,7epoxy-heneicosadiene (3Z,9Z-cis-6,7-epoxy-21:H), respectively, and mass spectra identical to these two compounds. It was determined that 0.23 +__0.16 and 0.56 5 : 0 . 2 6 ng of 3Z,6Z,9Z-21 :H and 3Z,9Z-cis-6,7-epoxy-21 :H, respectively, were present in pheromone gland extracts from individual females. A 1 : 1 blend of 3Z,6Z,9Z-21 :H and 3Z,9Z-6S,7R-epoxy-21 : H was an effective attractant for adult males from feral populations. Blend ratios of these two components from 2 : 1 to 1 : 2 were equally effective as attractants. Greater deviation from the optimal blends resulted in diminished trap catches. The enantiomer 3Z,9Z-6R,7S-epoxy-21:H not only was not effective in attracting males, its presence in the effective blend shut down trap catches. These results indicate that the pheromone blend consists of 3Z,6Z,9Z-21 :H and 3Z,9Z-6S,7R-epoxy-21 : H. This is the first report of a hydrocarbon/epoxide pheromone for a prey species of this bolas spider. Sex attractants or pheromones for the other three prey species are composed of aldehydes or acetates. *To whom correspondence should be addressed. 75 00984~331/96/01I]0-0075509.50•0 © 1996 PlenumPublishingCorporation
76
HAYNES, YEARGAN, M|LLAR, AND CHASTAIN Key Words--Chemical mimicry, pheromone. Mastophora hutchinsoni, bolas spider. Tetanolita mynesalis, 3Z,6Z,9Z-heneicosatriene, 3Z.OZ-6S,7R-epoxyheneicosadiene,
INTRODUCTION
Late-stadia and adult females of bolas spiders in the genus Mastophora hunt by aggressive chemical mimicry of the sex pheromones of their lepidopterous prey (Eberhard, 1977; Stowe et al., 1987; Yeargan, 1988, 1994). Only adult males of a limited subset of available prey species are attracted to volatile chemicals emitted by these spiders (Stowe, 1988; Yeargan, 1994). The bolas spider, Mastophora hutchinsoni Gertsch, has been found to capture only four species of nocturnally active Lepidoptera, including three noctuid moths--the smoky tetanolita, Tetanolita mynesalis (Walker), the bristly cutwoma, Lacinipotia renigera (Stephens), and the bronzed cutworm, Nephelodes minians Guenre--and one pyralid moth, the bluegrass webworm, Parapediasia teterrella (Zincken) (Yeargan, 1988). Sex pheromones or attractants for the latter three species have been shown to be unsaturated long-chain aliphatic acetates or aldehydes (Clark and Haynes, 1990; Haynes, 1990; Underhill et al., 1977). There are no known attractants or pheromone components for T. mynesalis. However, this species belongs to the Herminiinae, a subfamily of Noctuidae within which monoepoxide derivatives of triene or diene hydrocarbons have been shown to be attractants for several species (Wong et al., 1985; Millar et al.. 1991). This information concerning the known sex pheromones and attractants for three of the prey species, and phylogenetic inferences regarding potential attractants for one of the most frequently captured prey species, T. mynesalis, suggested that this bolas spider may be capable of producing a biosynthetically diverse set of allomones. The objective of the research described herein was to identify the femaleproduced sex attractant pheromone of 7". mynesalis, thus providing additional indirect evidence that addresses this hypothesis.
METHODS AND MATERIALS
Insects. Adult T. mynesalis were collected at blacklight traps in suburban backyards in Lexington and Richmond, Kentucky, during June-August 1994. Captured adults were transferred to 473-ml paper cartons with nylon screen lids. Eggs were removed from the inner surface of the carton and transferred to a 10-cm-diam. × 1.5-cm Petri dish containing a damp 7.5-cm piece of filter paper. Hatched larvae were transferred to plastic containers (15 cm diam. x 7.5 cm) lined with moist paper towels containing a mulch consisting primarily
T. mynesalis PHEROMONE
77
of white oak, Quercus alba L., leaves that had fallen to the ground the previous autumn. Covell (1984) indicated the larval food for T, mynesalis was probably dead leaves. With this suggestion as our starting point, we selected white oak leaves as the larval food when these leaves proved to be adequate to rear T. mynesalis to the adult stage. Larvae were reared under laboratory conditions, without any effort to control light or temperature cycles. After pupation, male and female pupae were separated by sex based on external genital characteristics (Mosher, 1969). Pupae were kept in Petri dishes with moist filter paper. On emergence, adults were transferred to 473-ml paper cartons. Pupae and adults were held in an environmental chamber with a t 6 L : 8 D light cycle and temperature cycle of 25°C: 18°C. Chemicals. The n-pentadecane and n-tricosane used as internal standards were obtained from Alltech (Deerfield, Illinois). Z9-14: Ac and ZI 1-16: A1, which were used in standard solutions for GC-EAD, were obtained from the Institute of Pesticide Research (Wageningen, The Netherlands) and Scentry (Buckeye, Arizona), respectively. 3Z,9Z-6S,7R-Epoxy-21 :H (94.15% enantiomeric purity, or 88.3 % enantiomeric excess) and 3Z,9Z-6R,7S-epoxy-21: H (94.10% enantiomeric purity, or 88.2% enantiomeric excess) were synthesized previously (Millar and Underhill, 1986). 3Z,6Z-9Z-21:H was 96.4% pure and was synthesized as described by Underhill et at. (1983). Extraction of Pheromone Glands. When females were 2-5 days old (primarily 3-4 days old), they were removed from the environmental chamber at midscotophase. Gentle pressure was applied to the abdomen of each female until the ovipositor and the associated pheromone gland emerged. The ovipositor and gland were removed with a pair of fine dissecting scissors, and placed into a 100-/zl conical-bottom vial containing 20 t~l of methylene chloride. For pooled extracts, several glands were extracted per vial. For analyses intended to document variation in pheromone content among individuals, only one gland was extracted per vial. Each vial was either analyzed immediately or sealed and stored at - 8 0 ° C until needed. Before analysis, a vial was removed from the freezer and allowed to warm to room temperature. For the study of individual variation, 10/~1 of an internal standard solution containing 1 ng/tzl of n-pentadecane (15 : H ) and n-tricosane (23 : H ) were added to each vial. Previously it was determined that these internal standards would not coelute with components of the gland extracts. The volume of the solution was reduced under Nz to 1 to 2/zl before injection. Chemical and Electrophysiological Analysis. The GC-etectroantennographic detection system was similar to that described by Struble and Am (1984) and Leal et al. (1992). A DB-5 column (30 m × 0.25 mm ID, J&W Scientific Folsom, California) was installed in a Hewlett-Packard 5890 Series II GC. Injections were splitless. The effluent from the column was split at a 1 : 1 ratio between a flame ionization detector ( F I D ) and a electroantennographic detector
78
HAYNES, YEARGAN, MILLAR, AND CHASTAIN
(EAD). The oven temperature was held at 80°C for 2 min and then increased to 260°C at a rote of 20°C/min. Just prior to the split, a makeup gas (nitrogen) flow rate of 120 ml/min was introduced via a VSIS-5 T connector (Scientific Glass Engineering, Austin, Texas). The effluent flowed out of this T connector through a 5-cm section of deactivated column (0.53 mm), which was inserted into a deactivated Universal Y Press-Tight connector (Resteck, Bellefonte, Pennsylvania). A 10-cm section of deactivated column was used to make a connection of one branch of the Y to the FID. An identical section of column connected the other branch of the Y through a separate heated zone into the side of a modified glass condenser. Air flowed through the system at 1 liter/rain and was humidified by bubbling through water before it entered the condensor. Chilled water circulated through the outer jacket of the condenser, cooling the humidified air to 22°C. This air flowed over the end of the GC column and delivered the GC effluent to the antennal preparation, which was 5 cm downstream from the end of the GC column, The antennal preparation was similar to that of Leal et al. (1992) with the following exceptions. The base of the moth's excised head was placed into a drop of Pringle's saline (Pringle, 1938). The terminal segment of one antenna was removed, and the cut tip placed into a separate drop of saline. Silver-silver chloride electrodes were placed into each drop of saline and connected through a Grass P16 high-impedance probe to a Grass P16 amplifier (Quincy, Massachusetts). Baseline drift was controlled by the passive high-pass filter (capacitor = 10~000 fzF, resistor = 130 f~) described by Struble and Am (1984). The amplified EAD and FID signals were sent to a Dataq model DI-420 Signal Conditioning Module (Akron, Ohio). Both signals were recorded and analyzed using Dataq MCA-CODAS software (Akron, Ohio). A Hewlett-Packard 5890 Series II gas chromatograph equipped with either a DB-5 or a DB-WAX column (both columns 30 m × 0.25 mm ID, J&W Scientific, Folsom, California), linked with a Hewlett-Packard 5972 mass selective detector (MSD), was used for analysis of gland extracts. GC operating conditions for both columns were the same with an initial oven temperature of 50°C for 2 min, increasing to 240°C at a rate of 10°C/min. The flow rate of helium through the column was maintained at 1 ml/min. The MSD was set to scan from 40 to 400 amu, to obtain electron impact (70 eV) spectra. Field Tests. For the first field exF riment, the quantity of 3Z,9Z-6S,7Repoxy-21 :H applied to a red rubber septum (A.H. Thomas, Swedesboro, New Jersey) that had previously been extracted two times in hexane was kept constant at 100 ~g and the quantity of 3Z,6Z,9Z-21 :H was varied to yield 1:1, 4: 1, and 10 : 1 blend ratios of this epoxide relative to the hydrocarbon. Fifty microliters of one of these three blends in hexane or hexane only was loaded onto a septum. Each rubber septum was hung by a paper clip from inside the top of a Pherocon IC trap (Trrcr, Salinas, California). These traps were suspended from
T. mynesalis PHEROMONE
79
limbs of trees so that the trap opening was approximately 1.8 m above the ground, which corresponds to a normal height at which the bolas spider M. hutchinsoni has been observed to catch male T. mynesalis. Traps were separated by at least 5 m. The assignment o f a particular treatment to one of these trap positions was determined at random. The positions of treatments were rerandomized after each trapping period (one or two nights). A priori we had decided not to include any blocks of the experiment in which no males were caught in any of the four treatments (unsuccessful replications). A total of 13 successful replications were conducted between 23 and 28 September 1994 at the University of Kentucky's South Farm (N = 2), Coldstream Farm (N = 2), Spindletop Farm (N = 2), a suburban backyard in Lexington, Kentucky (backyard site 1) (N = 2), and suburban backyard in Richmond, Kentucky (backyard site 2) (N = 5). After each trapping period, sticky bottoms of the traps were returned to the laboratory and the numbers of male T. mynesalis were counted. A KruskalWallis nonparametric anlaysis of variance was performed for this and subsequent experiments (Statistix 4.0, St. Paul Minnesota). This statistical approach was necessary because often control traps (and for some experiments other treatments) caught no moths; thus the homogeneity of variance assumption of the more powerful standard parametric analysis of variance was violated (with or without transformation). Means were separated with an experimentwise error rate set at ct = 0.05. In a second field experiment, the following blend ratios of 3Z,9Z-6S,7Repoxy-21 :H, 3Z,6Z,9Z-21 :H, and 3Z,9Z-6R,7S-epoxy-21 :H were tested: (a) 0 : 0 : 1, (b) 0 : 1 : I, (c) 1:1: 1, (d) 1 : 1 : 0 , and (e) 1 : 0 : 0 . The quantity of each compound when it was part of the blend to be applied to a rubber septum was kept at 100/zg. Control septa were loaded with hexane only. Because a methodological experiment conducted after our first experiment indicated that traps set at heights of about 30 cm were more effective in capturing male T. mynesalis than those set at 1.8 m (Haynes and Yeargan, unpublished data), we conducted this second and all subsequent tests at the lower height, Other details of this experiment were similar to our first experiment except that a trapping period was two or three nights. A total of eight successful replications were conducted between 3 and 12 October 1994 at the University of Kentucky's South Farm (N = 2), Spindletop Farm (N = 2), backyard site 1 (N = 2), and backyard site 2 (N = 2). In the third field experiment, the following blend ratios of 3Z,9Z,-6S,7Repoxy-21 :H and 3Z,6Z,9-Z-21 :H were evaluated: (a) 1 : 1, (b) 1:4, (c) 1: 10, and (d) 0 : 1 . The quantity of 3Z,6Z,9Z-21:H applied to a rubber septum was kept at 100/~g, and the quantity of 3Z,9Z-6S,7R-epoxy-21 :H was varied (100, 25, 10, and 0/~g). Control septa were loaded with hexane only. Other details of this experiment were similar to our second experiment, except that the Spindletop site was dropped because of the apparent low population density of
80
HAYNES, YEARGAN, MILLAR, AND CHASTAIN
7". mynesalis found there, and one additional location was added at both the South Farm and Richmond (backyard site 2) sites. A total of 14 successful replications were conducted between 12 and 19 October 1994 at the University of Kentucky's South Farm ( N = 6), backyard site 1 (N = 2), and backyard site 2 (N = 6). In the last field experiment, the following blend ratios of 3Z,9Z-6S,7Repoxy-21 :H and 3Z,6Z,9Z-21 : H were evaluated: (a) 2: 1, (b) 1 : 1, and (c) 1:2. The total quantity of 3Z,9Z-6S,7R-epoxy-21 :H plus 3Z,6Z,9Z-21 :H applied to a rubber septum was kept constant at 200 #g. Thus, for example, the 2 : 1 blend ratio consisted of 133 tzg 3Z,9Z-6S,7R-epoxy-21 : H plus 67/zg 3Z,6Z,9Z-21 : H. An additional septum was loaded with hexane only. Other details of this experiment were similar to our third experiment. A total of 14 successful replications were conducted between 19 and 26 October 1994 at the University of Kentucky's South Farm (N = 6), backyard site 1 (N = 2), and backyard site 2 (N = 6). Voucher specimens of male T. mynesalis trapped with a 1 : 1 blend of 3Z,9Z-6S,7R-epoxy-21 :H and 3Z,6Z,9Z-21 :H, and males reared from eggs laid by females caught in a black light trap have been deposited in the University of Kentucky, Department of Entomology, Collection. RESULTS AND DISCUSSION
Chemical and Electrophysiological Analysis. The GC-EAD analysis indicated there were two compounds present in gland extracts of individual female 7". mynesalis that stimulated mate 7". mynesalis antennae (Figure 1A). The EAD activity was observed at 12.00 and 13.20 min (DB-5), but there were no prominent FID peaks at these retention times. Other peaks were present in the gland extracts, but these did not stimulate an EAD response. The retention times of the straight-chain hydrocarbon standards 2 0 : H , 21: H, 2 2 : H , and 2 3 : H were 11.52, 12.11, 12.77 and 13.52 min, respectively. Because monoepoxide derivatives of triene hydrocarbons have been shown to be attractants for several species in the same subfamily (Wong et al., 1985), we evaluated standard solutions of 17- to 21-carbon triene (3Z,6Z,9Z-17-21 : H ) and 17- to 22-carbon monoepoxide derivatives (6Z,9Z-cis-3,4-epoxy-17 to 2 2 : H ; 3Z,9Z-cis-6,7epoxy- 17 to 22 : H; and 3Z,6Z-cis-9,10-epoxy- 17 to 22 : H). The 21-carbon hydrocarbon and epoxides gave the closest retention time matches to the EAD active compounds. Synthetic 3Z,6Z,9Z-21: H and 3Z,9Z-6S,7R-epoxy-21: H had retention times identical to the EAD active compounds (12.00 and 13.20 min, respectively) and stimulated strong EAD responses (Figure 1B). Other compounds (e.g., aldehydes and acetates) that have been determined to be pheromone components or sex attractants for other prey species of M. hutchinsoni did not stimulate EAD activity in male T. mynesalis antennae (Figure 1B).
T. mynesatis PHEROMONE
81
A. Gland extract
J 9
10
11
12
13
14
15
Time (rain)
B. Synthetic blend
d
b
I
I
9
10
e
I
I
I
11 12 13 Time ('rain)
I
I
14
15
FIG. 1. (A) GC-EAD analysis of the pheromone gland extract of one female T. mynesalis. The top trace shows the response o f the flame ionization detector (FID). The
bottom trace shows the electroantennogram (EAD) response. Peaks o f EAD activity were seen at retention times (R,) of 12.00 and 13.20 min (B). A synthetic blend consisting of 0.5 ng (a) 15:H (R, = 8.45), (b) Z 9 - 1 4 : A c (R, = 10.37), (c) Z l l - 1 6 : A I (R, = 10.45), (d) 3Z,6Z,9Z-21 :H (R t = 12.00), (e) 3Z,9Z-6S,7R-epoxy-21 :H (Rt = 13.20), and (f) 23 : H (R~ = 13.52) was analyzed by GC-EAD. The top trace shows the response of the FID. EAD responses (bottom trace) were seen at R, of 12.00 and 13.20 min.
82
HAYNES, YEARGAN, MILLAR, AND CHASTAIN
GC-MSD analysis of the pheromone extract using the DB-5 column indicated that two minor compounds present in individual gland extracts corresponded exactly in retention times and mass spectra with those of synthetic 3Z,6Z,9Z-21:H (20.09 min) and 3Z,9Z-6S,7R-epoxy-21:H (21.66 min). From pooled extracts (5 or 10 female equivalents), the first peak and 3Z,6Z,9Z-21 :H were characterized by prominent ions at m/z 55, 67, 79, and 108, as well as less abundant ions at 234 and 290 (the molecular ion for 3Z,6Z,9Z-21 : H). The less abundant ions were reliably detected only in pooled extracts, while the prominent ions were consistently found in individual and pooled extracts. The diagnostic ions (Underhill et al., 1983) at m/z 55 (CH3CH2CH=CH ÷) and t08 [CH3CH2(CH=CH)3H+], located the first of the three double bonds at C-3, and indicated that all three double bonds are clustered at one end of the structure. A further diagnostic ion at m/z 234 [CH3(CH2)10(CH=CH)3H +] located a second double bond at C-9 (Underhill et al., 1983). On the basis of exact matches in retention times and spectra, and on the reasonable assumption that conjugation of the third double bond with either one of the other two would significantly change the retention time, the triene was tentatively assigned the 3Z,6Z,9Z21 :H structure. Further evidence for this structure was provided by exact retention time matches of the insect-produced compound with a synthetic standard on a different GC column (DB-WAX). The second EAD-active peak had a retention time identical to synthetic 3Z,9Z-cis-6,7-epoxy-2t :H on two columns (DB-5 and DB-WAX). Its mass spectrum exactly matched that of a synthetic standard of 3Z,9Z-6S,7R-epoxy21 :H and data reported in the literature (Millar and Underhill, 1986). In particular, the mass spectrum was characterized by prominent ions at m/z 55, 67, 83, 95, and 111, as well as less abundant ions at 223, 277, and 306 (the molecular ion for 3Z,9Z-6S,7R-epoxy-21:H). The prominent ion at m/z 111 is diagnostic for a structure of this type with the epoxide in the 6,7 position (cleavage of C - C bond adjacent to the epoxide; T6th et al., 1992). However, the enantiomer 3Z,9Z-6R,7S-epoxy-21 : H cannot be distinguished from 3Z,9Z6S,7R-epoxy-21 :H by retention time or mass spectra. In individual gland extracts of five females analyzed by GC-MS on the DB-5 column, the quantity of 3Z,6Z,9Z-21 :H ranged from 0.12 to 0.51 ng with a mean of 0.23 + 0.16 ng/gland. The quantity of 3Z,9Z-cis-6,7-epoxy-21:H ranged from 0.30 to 0.86 ng with a mean of 0.56 + 0.26 ng. The range in blend ratios of 3Z,9Z-cis-6,7-epoxy-21 :H to 3Z,6Z,9Z-21 :H was 1.59:1 to 3.28:1 (blend ratio defined as the quantity of the epoxide divided by the quantity of the triene hydrocarbon) with a mean of 2.66 + 0.68 : 1. Results were similar for independent analyses of five glands conducted on a DB-WAX column. The quantity of 3Z,6Z,9Z-21 :H ranged from 0.19 to 0.65 ng with a mean of 0.36 + 0.17 ng/gland. The quantity of 3Z,9Z-cis-6,7-epoxy-21 :H ranged from 0.39 to 1.11 ng, with a mean of 0.71 + 0.32 ng. The range in blend ratios of 3Z,9Z-
T. mynesalis
PHEROMONE
83
cis-6,7-epoxy-21 : H to 3Z,6Z,9Z-21 : H was 1.31 : 1 to 3.01 : 1, with a mean o f 2.12 + 0.77: 1.
FieM Tests. Following a preliminary field test in which traps baited with 100 ~tg of 3Z,9Z-6S,7R-epoxy-21 : H caught four male T. mynesalis, and the enantiomer 3Z,9Z-6R,7S-epoxy-21 : H caught no males, we opted to focus the first full field experiment on varying blend ratios of 3Z,9Z-6S,7R-epoxy-21:H and 3Z,6Z,9Z-21 :H. In this experiment, the 1 : 1 blend of 3Z,9Z-6S,7R-epoxy2 1 : H and 3Z,6Z,9Z-21 : H was the most attractive blend for trapping male T. mynesalis (Figure 2). Traps baited with this blend caught significantly more mate T. mynesalis than control traps [Kruskal-Wallis statistic ( H ) = 17.73, P < 0.001, experimentwise error rate set at a = 0.05 for mean comparisons]. The mean numbers of males caught in traps baited with the 4 : 1 and 10 : 1 blends of 3Z,9Z-6S,7R-epoxy-21 : H and 3Z,6Z,9Z-21 : H were intermediate between and not significantly different from the numbers caught in control traps (zero) and the 1 : 1 blend, using this conservative statistical test. In the second field experiment, traps baited with a 1:1 blend of 3Z,9Z,6S,7R-epoxy-21 :H and 3Z,6Z,9Z-21 : H caught the largest number of
~4 e~
ab
1
/ Ratio 3Z,9Z-6S,7R-epoxy-21:H to 3Z,6Z,9Z-21:H
FIG. 2. Captures of male T. mynesalis moths in traps baited with blend ratios of 3Z,9Z6S,7R-epoxy-21 :H and 3Z,6Z,9Z-21 :H, or a hexane-treated control. The quantity of 3Z,gZ-6S,7R-epoxy-21:H was kept constant at 100 p.g. Bars represent standard errors. Numbers are the mean number of males caught per trap per trapping interval (see Methods and Materials). Means that do not share a letter in common are significantly different (experimentwise error rate c¢ = 0.05, Kruskal-Wallis nonparametric analysis of variance).
84
H A Y N E S , Y E A R G A N , M I L L A R , AND CHASTAIN
males and significantly more than all other treatments except for traps baited with 3Z,9Z-6S,7R-epoxy-21 : H alone (Figure 3) ( H = 32.42, P < 0.0001). The enantiomer, 3Z,9Z-6R,7S-epoxy-21 : H was ineffective in attracting males to traps either by itself or as part of a 1 : 1 blend with 3Z,6Z,9Z-21 : H. Addition of this enantiomer to the otherwise effective attractant o f a 1 : I blend of 3Z,9Z6S,7R-epoxy-21 : H and 3Z,6Z,9Z-21:H shut down the response to the lure. In the third field experiment, traps baited with 3Z,6Z,9Z-21 :H alone or a 1 : 10 blend of 3Z,9Z-6S,7R-epoxy-21 : H and 3Z,6Z,9Z-21 :H were significantly less effective in catching male T. mynesalis than traps baited with the 1 : 1 blend ratio (Figure 4) ( H = 45.39, P < 0.0001). The hydrocarbon by itself and the 1 : 1 0 blend of 3Z,9Z-6S,7R-epoxy-21 :H and 3Z,6Z,9Z-21 : H were no more effective than control traps in capturing males in traps. In the fourth trapping experiment, we focused on a narrower range of blend ratios of 3Z,9Z-6S,7R-epoxy-21: H and 3Z,6Z,9Z-21:H. The number of males caught in traps baited with blend ratios of 2 : 1, 1 : 1, and 1 : 2 of 3Z,9Z-6S,7R-
e~
~6 e~ eq
E4
Z2 b T
b
b
b
3Z,9Z-6S,7R-epoxy-21 :H to 3Z,6Z,9Z-21 :H to 3Z,9Z-6R,7S-epoxy-21 :H FIG. 3. Captures of male T~ mynesalis moths in traps baited with blend ratios of 3Z,9Z6S.7R-epoxy-21 : H, 3Z,6Z,9Z-21 : H, and 3Z,9Z-6R,7S-epoxy-21 : H, or a hexane-treated control. The quantity of each compound was kept at 100 p.g. Bars represent standard errors. Numbers are the mean number of males caught per trap per trapping interval (see Methods and Materials). Means that do not share a letter in common are significantly different (experimentwise error rate m = 0.05, Kruskal-Wallis nonparametric analysis of variance).
T. mynesalis PHEROMONE
85
20 m.
L. 15
ab E ~ 10 o
E
bc
0
3 Z , 9 Z - 6 S , 7 R - e p o x y - 2 1 :H to 3Z,6Z,9Z-21 :H
FIG. 4. Captures of male T. mynesalis moths in traps baited with blend ratios of 3Z,9Z6S,7R-epoxy-21 : H and 3Z,6Z,9Z-21:H, or a hexane-treated control. The quantity of 3Z,6Z,9Z-21:H was kept constant at 100 p~g. Bars represent standard errors. Numbers are the mean number of males caught per trap per trapping interval (see Methods and Materials). Means that do not share a letter in common are significantly different (experimentwise error rate c~ = 0.05, KruskaI-Wallis nonparametric analysis of variance).
epoxy-21 : H and 3Z,6Z,9Z-21 : H were not significantly different from each other, but each of these blend ratios was significantly more attractive than hexane alone (Figure 5) ( H = 29.59, P < 0.0001). Blend ratios from 2 : t to 1:2 o f 3Z,9Z-6S,7R-epoxy-21:H and 3Z,6Z,9Z2 1 : H were effective in capturing male T. mynesalis in sticky traps. The range of blend ratios of 3Z,9Z-cis-6,7-epoxy-21:H to 3Z,6Z,9Z-21: H found in pheromone gland extracts (!.31 : 1 to 3.28: 1) overlapped this range. The chromatographic techniques we employed did not allow us to distinguish between the enantiomers 3Z,9Z-6S,7R-epoxy-21:H and 3Z.9Z-6R,7S-epoxy-21:H, and to date, there is no reported analytical chemistry method for distinguishing these two enantiomers at nanogram levels. However, the results from the second field test indicate that 100/zg of 3Z,9Z-6R,7S-epoxy-21:H is not effective in attracting males when presented with 100 /~g of 3Z,6Z,9Z-21 :H. In addition, the presence of 100/~g of 3Z,6Z,9Z-6R,7S-epoxy-21: H within an otherwise effective 1:1 blend of 3Z,9Z-6S,7R-epoxy-21:H and 3Z,6Z,9Z-21:H (100/zg each) prevented captures of any males, and thus at this level 3Z,9Z-6R,7S-epoxy21 : H is a behavioral antagonist. The enantiomeric purity of the synthetic 3Z,9Z-
86
HAYNES, YEARGAN, MILLAR, AND CHASTAIN
12 a
a
e~
9 gh
6 A3
S 3
0 3Z,9Z-6S,7R-epoxy-21 :H to 3Z,6Z,9Z-21 :H
FIG. 5. Captures of male T. mynesalis moths in traps baited with blend ratios of 3Z,9Z6S,7R-epoxy-2t : H and 3Z,6Z,9Z-21 :H, or a hexane-treated control. The total quantity of 3Z,9Z-6S,7R-epoxy-21:H plus 3Z,6Z,9Z-21 :H applied to a rubber septum was kept constant at 200 ~g. Bars represent standard errors. Numbers are the mean number of males caught per trap per trapping interval (see Methods and Materials). Means that do not share a letter in common are significantly different (experimentwise error rate c¢ = 0.05, Kruskal-Wallis nonparametric analysis of variance). 6S,7R-epoxy-21 :H was approximately 94% (Millar et al., 1986). This leaves open the possibility that even this 6% level of 3Z,9Z-6R,7S-epoxy-21 :H within 3Z,9Z-6S,7R-epoxy-21 :H had a negative (or possibly a positive) impact on captures of males. This cannot be tested until material of higher enantiomeric purity is available. The identification of a blend of 3Z,9Z-6S,7R-epoxy-21 : H and 3Z,6Z,9Z21 : H as a sex pheromone for T. mynesalis is the first such identification for a species in the Herminiinae subfamily of the Noctuidae (Am et al., 1992). However, in field trials, monoepoxide dienes or trienes (19-21 carbons) have been reported to be sex attractants for other Herminiinae, including Bleptina caradrinalis Guenre, Idia aemula (Hfibner), 1. americalis (GuenEe), and Palthis angulalis (Hiibner) ( W o n g et al., 1985; Millar et al., 1991). Every pheromone blend that has been identified to date for prey species of bolas spiders is composed of unsaturated straight-chain aliphatic aldehydes, alcohols, or acetates (see Yeargan, 1994; A m et al., 1992). No diene or triene hydrocarbons or their monoepoxide derivatives have been discovered to be sex attractants for any of these prey species. Four noctuid prey species (three species
T. mynesalis PHEROMONE
87
belonging to Catocalinae and one to Herminiinae) of M. bisaccata and two prey species (Geometridae: Larentiinae) of M. ph~.nosoma, however, are closely related to species that have been attracted to specific diene or triene hydrocarbons or monoepoxide derivatives in field screening tests of these compounds. This taxonomic relationship suggests it is very likely that the pheromone blends of a subset of the prey species of these two bolas spiders will be hydrocarbons or epoxides, as is the case for M. hutchinsoni. Prior to the present study, sex attractants or pheromones were known for three of the four species of moths that are captured by M. hutchinsoni. Previously, Haynes (1990) determined that female bristly cutworm moths, Lacinipolia renigera (Noctuidae: Hadeninae), emit Z 9 - 1 4 : A c and Z , E - 9 , 1 2 - 1 4 : A c (3.8% of Z 9 - 1 4 : A c ) and that synthetic blends of these two components are attractive to males. Males of the bronzed cutworm, Nephelodes minians (Noctuidae: Hadeninae), are attracted to Z11-16 : AI (Underhill et al., 1977). An optimized attractant for male bluegrass webworms, Parapediasia teterrella (Pyralidae: Crambinae), consists of a 20 : 1 blend of Z11-16 : AI and Z9-16 : AI (Clark and Haynes, 1990). Yeargan (1988) found that T. mynesalis (52.6%) and L. renigera (40.0%) accounted for the vast majority of the 460 moths caught by M. hutchinsoni in a two-year study. Thus, the two most frequently captured prey use pheromones that are composed of different classes of chemicals. The identification of a blend of 3Z,9Z-6S,7R-epoxy-21 : H and 3Z,6Z,9Z21 : H as a sex pheromone for T. mynesalis suggests that female M. hutchinsoni may be capable of synthesizing a diverse set of allomones, including both straight-chain aliphatic acetates and aldehydes and triene hydrocarbons and monoepoxide derivatives. Stowe (1988) discussed the evolution of chemical mimicry in plants and animals, including the case of bolas spiders and moths. At the time of his review, no prey species of bolas spiders had been shown to use diene or triene hydrocarbons or their monoepoxide derivatives in their sex pheromones, but the phylogenetic relationships of some prey to species for which these pheromones or attractants had been found suggested that these compounds were likely candidates to be pheromone components of certain prey. Stowe (1988) offered several possible explanations of how bolas spiders might capture moth species that use chemically diverse pheromone components; given the results of our current study, three alternative hypotheses seem most plausible to explain this diversity. First, there may be variation in the attractants produced among individual bolas spiders, resulting in specialization by individual spiders but yielding a more diverse assemblage of prey at the population level. Second, the blend produced by an individual spider might change over time. Last, each spider might produce a consistent single blend of compounds, if individual components do not interfere with attraction of the other species of prey. Yeargan (1988) found definitive evidence that precludes the possibility of interindividual variation in M. hutchinsoni leading to extreme specialization of
88
HAYNES, YEARGAN, MILLAR, AND CHASTA1N
individual spiders on a single prey species. He found that the majority of spiders (56.3%) that caught two or more prey items on a single night also caught two or more species of prey. In fact, some individual spiders caught all four of the known prey species over the course of his study, It is possible that a spider changes the attractant blend it produces over the course of the scotophase, because the two principal prey of M. hutchinsoni are captured at different times of night. L. renigera is captured before 10:30 PM, while T. mynesalis is captured after 11:00 PM, except on very cool nights (Yeargan, 1988). Preliminary evidence suggests that temporal variation may occur in the composition of the blend emitted by M. cornigera, the only species of bolas spider for which allomonal components have been identified from the spider itself (Stowe et al., 1987). An alternative to temporal variation in the allomonal blend produced by M. hutchinsoni is that the spider, while hunting, may produce a relatively constant blend of the attractants for the two principal prey species. This would be possible if the chemical constituents do not interfere with the effectiveness of either attractant. The G C - E A D results (Figure 1) indicate that the male T. mynesalis's antenna does not show a measurable EAD response to Z 9 - 1 4 : A c (the major pheromone component of L. renigera) or Z 1 1 - 1 6 : AI (the major sex attractant component for P. teterrella). Thus, it is possible that these heterospecific pheromone blends are not even detected by T. mynesalis. Either the second hypothesis (temporal variation) or the third hypothesis (consistent, compatible blend) could account for the ability of M. hutchinsoni to capture moths that use chemically diverse pheromones. We plan to investigate this chemical mimicry system to ascertain which hypothesis is correct.
Acknowledgments--WalterLeal provided helpful advice concerning the setup of the GC-EAD. F.W. Knapp and P.A. Weston kindly reviewed an earlier draft of the manuscript. This work was supported in part by State and Federal Hatch funds to the Kentucky Agricultural ExperimentStation (paper no. 95-08-062). REFERENCES ARN. H., TOTH,M., and PRIESNER,E. 1992. List of Sex Pheromones of Lepidoptera and Related Attractants, 2nd ed., International Organization for Biological Control, Montfavet, France. CLARK, J.D., and HAYNES,K.F. 1990. Sex attractant for the bluegrass webworm (Lepidoptera: Pyralidae). J. Econ. Entomol. 83:856-859. COVEt.L,C.V., JR. 1984. A Field Guide to the Moths of Eastern North America. Houghton Mifflin, Boston. EBERHARD,W.G. 1977. Aggressive chemical mimicry by a bolas spider. Science 198:1173-1175. HAYNES, K.F. 1990. Identification of sex pheromone of bristly cutworm, Lacinipolia renigera (Stephens). J. Chem. Ecol. 16:2615-2621, LEAL, W.S., MOCHIZUKI,F., WAKAMURA,S., and YASUDA,T. 1992. Electroantennographicdetec-
T. mynesalis PHEROMONE
89
tion ofAnomala cuprea Hope (Coleoptera: Scarabaeidae) sex pheromone. Appl. Entomol. Zool. 27:289-291. MILLAR, J.G., and UNDERHmL, E.W, 1986. Synthesis of chiral bis-homoallylic epoxides. A new class of lepidopteran sex attractants, J, Org. Chem. 51:4726-4728. MILLAR, J.G., GIBLIN, M., BARTON, D,, and UNDERHILL,E.W. 1991. Synthesis and field screening of chiral monounsaturated epoxides as lepidopteran sex attractants and sex pheromone components. J. Chem. Ecol. 17:911-929. MOSHER, E. 1969. Lepidoptera Pupae. Entomological Reprint Specialists, East Lansing, Michigan. PRINGLE, J.W.S. 1938, Proprioception in insects I. A new type of mechanical receptor from the palps of the cockroach. J. Exp. Biol, 15:101-113. STOWE, M.K. 1988. Chemical mimicry, pp. 513-580. in K.C. Spencer (ed.). Chemical Mediation of Coevolution. Academic Press, San Diego, STOWE, M.K., TUMLINSON,J.H., and HEATH, RR. 1987. Chemical mimicry: Bolas spiders emit components of moth prey species sex pheromones. Science 236:964-967. STRUBLE, D.L., and ARN, H. 1984, Combined gas chromotography and electroantennogram recording of insect olfactory responses, pp. 161-178, #7 H.E, Hummel and T.A. Miller (eds_). Techniques in Pheromone Research. Springer-Vedag, New York. TOTH, M,, BUSER, H.R., GUERIN, P.M., ARN, H., SCHM|DT, F., FRANCKE, W,, and Szocs, G, 1992. Abraxas grossulariata L. (Lepidoptera: Geometridae): Identification of (3Z,6Z,9Z)3,6,9-heptadecatriene and (6Z,9Z)-6,9-cis-3,4-epoxyheptadecadiene in the female sex pheromone. J. Chem Ecol. 18:13-25, UNDERHILL, E.W., CmSHOLM, M.D., and STECI'L W. 1977. Olefinic aldehydes as constituents of sex attractants for noctuid moths, Envircm. EntomoL 6:333-337, UNDERHILL, E.W., PALANISWAMY,P,, ABRAMS,S.R,, BAILEY,B.K., STECK, W.F., and CHISHOLM, M.D. 1983. Triunsaturated hydrocarbons, sex pheromone components of Caenurgina erechtea, J. Chem. Ecol. 9:1413-1423. WONG, J.W., UNOERHJLL, E.W., MACKE:.NZm,S.L., and CmSHOLM, M.D. 1985. Sex attractants for geometrid and noctuid moths; field trapping and electroantennographic responses to triene hydrocarbons and monoepoxydiene derivatives. J. Chem. Ecol. 11:727-756. YEARGAN, K.V. 1988. Ecology of a bolas spider. Mastophora hutchinsoni: Phenology, hunting tactics, and evidence for aggressive chemical mimicry. Oecologia (Berlin) 74:524-530. YEARGAN, K.V. 1994. Biology of bolas spiders, Annu. Rev. Entomol. 39:81-99.
IDENTIFICATION OF SEX PHEROMONE OF Tetanolita mynesalis (LEPIDOPTERA: NOCTUIDAE), A PREY SPECIES OF BOLAS SPIDER, Mastophora hutchinsoni
KENNETH
F.
HAYNES,
j'* KENNETH
V.
JOCELYN
G.
MILLAR,
z and
B. C H A S T A I N
BONNIE
YEARGAN,
I I
IDepartment of Entomology University of Kentucky Lexington, Kentucky. 40546 2Department of Entomology Universit3, of California Riverside, California 92521 (Received May 1, 1995; accepted August 28, 1995) Abstract--The bolas spider, Mastophora hutchinsoni, attracts adult males of four species of nocturnally active Lepidoptera through aggressive chemical mimicry of those species' sex pheromones. Here we report the identification of the sex pheromone of one prey species, the smoky tetanolita (Tetanolita mynesalis). In sex pheromone gland extracts, only two peaks stimulated an electrophysiological response as measured by a coupled gas chromatographyelectroantennographic detection analysis. These two peaks had retention times identical to (3Z,6Z,9Z)-heneicosatriene (3Z,6Z,9Z-21:H ) and (3Z,9Z)-cis-6,7epoxy-heneicosadiene (3Z,9Z-cis-6,7-epoxy-21:H), respectively, and mass spectra identical to these two compounds. It was determined that 0.23 +__0.16 and 0.56 5 : 0 . 2 6 ng of 3Z,6Z,9Z-21 :H and 3Z,9Z-cis-6,7-epoxy-21 :H, respectively, were present in pheromone gland extracts from individual females. A 1 : 1 blend of 3Z,6Z,9Z-21 :H and 3Z,9Z-6S,7R-epoxy-21 : H was an effective attractant for adult males from feral populations. Blend ratios of these two components from 2 : 1 to 1 : 2 were equally effective as attractants. Greater deviation from the optimal blends resulted in diminished trap catches. The enantiomer 3Z,9Z-6R,7S-epoxy-21:H not only was not effective in attracting males, its presence in the effective blend shut down trap catches. These results indicate that the pheromone blend consists of 3Z,6Z,9Z-21 :H and 3Z,9Z-6S,7R-epoxy-21 : H. This is the first report of a hydrocarbon/epoxide pheromone for a prey species of this bolas spider. Sex attractants or pheromones for the other three prey species are composed of aldehydes or acetates. *To whom correspondence should be addressed. 75 00984~331/96/01I]0-0075509.50•0 © 1996 PlenumPublishingCorporation
76
HAYNES, YEARGAN, M|LLAR, AND CHASTAIN Key Words--Chemical mimicry, pheromone. Mastophora hutchinsoni, bolas spider. Tetanolita mynesalis, 3Z,6Z,9Z-heneicosatriene, 3Z.OZ-6S,7R-epoxyheneicosadiene,
INTRODUCTION
Late-stadia and adult females of bolas spiders in the genus Mastophora hunt by aggressive chemical mimicry of the sex pheromones of their lepidopterous prey (Eberhard, 1977; Stowe et al., 1987; Yeargan, 1988, 1994). Only adult males of a limited subset of available prey species are attracted to volatile chemicals emitted by these spiders (Stowe, 1988; Yeargan, 1994). The bolas spider, Mastophora hutchinsoni Gertsch, has been found to capture only four species of nocturnally active Lepidoptera, including three noctuid moths--the smoky tetanolita, Tetanolita mynesalis (Walker), the bristly cutwoma, Lacinipotia renigera (Stephens), and the bronzed cutworm, Nephelodes minians Guenre--and one pyralid moth, the bluegrass webworm, Parapediasia teterrella (Zincken) (Yeargan, 1988). Sex pheromones or attractants for the latter three species have been shown to be unsaturated long-chain aliphatic acetates or aldehydes (Clark and Haynes, 1990; Haynes, 1990; Underhill et al., 1977). There are no known attractants or pheromone components for T. mynesalis. However, this species belongs to the Herminiinae, a subfamily of Noctuidae within which monoepoxide derivatives of triene or diene hydrocarbons have been shown to be attractants for several species (Wong et al., 1985; Millar et al.. 1991). This information concerning the known sex pheromones and attractants for three of the prey species, and phylogenetic inferences regarding potential attractants for one of the most frequently captured prey species, T. mynesalis, suggested that this bolas spider may be capable of producing a biosynthetically diverse set of allomones. The objective of the research described herein was to identify the femaleproduced sex attractant pheromone of 7". mynesalis, thus providing additional indirect evidence that addresses this hypothesis.
METHODS AND MATERIALS
Insects. Adult T. mynesalis were collected at blacklight traps in suburban backyards in Lexington and Richmond, Kentucky, during June-August 1994. Captured adults were transferred to 473-ml paper cartons with nylon screen lids. Eggs were removed from the inner surface of the carton and transferred to a 10-cm-diam. × 1.5-cm Petri dish containing a damp 7.5-cm piece of filter paper. Hatched larvae were transferred to plastic containers (15 cm diam. x 7.5 cm) lined with moist paper towels containing a mulch consisting primarily
T. mynesalis PHEROMONE
77
of white oak, Quercus alba L., leaves that had fallen to the ground the previous autumn. Covell (1984) indicated the larval food for T, mynesalis was probably dead leaves. With this suggestion as our starting point, we selected white oak leaves as the larval food when these leaves proved to be adequate to rear T. mynesalis to the adult stage. Larvae were reared under laboratory conditions, without any effort to control light or temperature cycles. After pupation, male and female pupae were separated by sex based on external genital characteristics (Mosher, 1969). Pupae were kept in Petri dishes with moist filter paper. On emergence, adults were transferred to 473-ml paper cartons. Pupae and adults were held in an environmental chamber with a t 6 L : 8 D light cycle and temperature cycle of 25°C: 18°C. Chemicals. The n-pentadecane and n-tricosane used as internal standards were obtained from Alltech (Deerfield, Illinois). Z9-14: Ac and ZI 1-16: A1, which were used in standard solutions for GC-EAD, were obtained from the Institute of Pesticide Research (Wageningen, The Netherlands) and Scentry (Buckeye, Arizona), respectively. 3Z,9Z-6S,7R-Epoxy-21 :H (94.15% enantiomeric purity, or 88.3 % enantiomeric excess) and 3Z,9Z-6R,7S-epoxy-21: H (94.10% enantiomeric purity, or 88.2% enantiomeric excess) were synthesized previously (Millar and Underhill, 1986). 3Z,6Z-9Z-21:H was 96.4% pure and was synthesized as described by Underhill et at. (1983). Extraction of Pheromone Glands. When females were 2-5 days old (primarily 3-4 days old), they were removed from the environmental chamber at midscotophase. Gentle pressure was applied to the abdomen of each female until the ovipositor and the associated pheromone gland emerged. The ovipositor and gland were removed with a pair of fine dissecting scissors, and placed into a 100-/zl conical-bottom vial containing 20 t~l of methylene chloride. For pooled extracts, several glands were extracted per vial. For analyses intended to document variation in pheromone content among individuals, only one gland was extracted per vial. Each vial was either analyzed immediately or sealed and stored at - 8 0 ° C until needed. Before analysis, a vial was removed from the freezer and allowed to warm to room temperature. For the study of individual variation, 10/~1 of an internal standard solution containing 1 ng/tzl of n-pentadecane (15 : H ) and n-tricosane (23 : H ) were added to each vial. Previously it was determined that these internal standards would not coelute with components of the gland extracts. The volume of the solution was reduced under Nz to 1 to 2/zl before injection. Chemical and Electrophysiological Analysis. The GC-etectroantennographic detection system was similar to that described by Struble and Am (1984) and Leal et al. (1992). A DB-5 column (30 m × 0.25 mm ID, J&W Scientific Folsom, California) was installed in a Hewlett-Packard 5890 Series II GC. Injections were splitless. The effluent from the column was split at a 1 : 1 ratio between a flame ionization detector ( F I D ) and a electroantennographic detector
78
HAYNES, YEARGAN, MILLAR, AND CHASTAIN
(EAD). The oven temperature was held at 80°C for 2 min and then increased to 260°C at a rote of 20°C/min. Just prior to the split, a makeup gas (nitrogen) flow rate of 120 ml/min was introduced via a VSIS-5 T connector (Scientific Glass Engineering, Austin, Texas). The effluent flowed out of this T connector through a 5-cm section of deactivated column (0.53 mm), which was inserted into a deactivated Universal Y Press-Tight connector (Resteck, Bellefonte, Pennsylvania). A 10-cm section of deactivated column was used to make a connection of one branch of the Y to the FID. An identical section of column connected the other branch of the Y through a separate heated zone into the side of a modified glass condenser. Air flowed through the system at 1 liter/rain and was humidified by bubbling through water before it entered the condensor. Chilled water circulated through the outer jacket of the condenser, cooling the humidified air to 22°C. This air flowed over the end of the GC column and delivered the GC effluent to the antennal preparation, which was 5 cm downstream from the end of the GC column, The antennal preparation was similar to that of Leal et al. (1992) with the following exceptions. The base of the moth's excised head was placed into a drop of Pringle's saline (Pringle, 1938). The terminal segment of one antenna was removed, and the cut tip placed into a separate drop of saline. Silver-silver chloride electrodes were placed into each drop of saline and connected through a Grass P16 high-impedance probe to a Grass P16 amplifier (Quincy, Massachusetts). Baseline drift was controlled by the passive high-pass filter (capacitor = 10~000 fzF, resistor = 130 f~) described by Struble and Am (1984). The amplified EAD and FID signals were sent to a Dataq model DI-420 Signal Conditioning Module (Akron, Ohio). Both signals were recorded and analyzed using Dataq MCA-CODAS software (Akron, Ohio). A Hewlett-Packard 5890 Series II gas chromatograph equipped with either a DB-5 or a DB-WAX column (both columns 30 m × 0.25 mm ID, J&W Scientific, Folsom, California), linked with a Hewlett-Packard 5972 mass selective detector (MSD), was used for analysis of gland extracts. GC operating conditions for both columns were the same with an initial oven temperature of 50°C for 2 min, increasing to 240°C at a rate of 10°C/min. The flow rate of helium through the column was maintained at 1 ml/min. The MSD was set to scan from 40 to 400 amu, to obtain electron impact (70 eV) spectra. Field Tests. For the first field exF riment, the quantity of 3Z,9Z-6S,7Repoxy-21 :H applied to a red rubber septum (A.H. Thomas, Swedesboro, New Jersey) that had previously been extracted two times in hexane was kept constant at 100 ~g and the quantity of 3Z,6Z,9Z-21 :H was varied to yield 1:1, 4: 1, and 10 : 1 blend ratios of this epoxide relative to the hydrocarbon. Fifty microliters of one of these three blends in hexane or hexane only was loaded onto a septum. Each rubber septum was hung by a paper clip from inside the top of a Pherocon IC trap (Trrcr, Salinas, California). These traps were suspended from
T. mynesalis PHEROMONE
79
limbs of trees so that the trap opening was approximately 1.8 m above the ground, which corresponds to a normal height at which the bolas spider M. hutchinsoni has been observed to catch male T. mynesalis. Traps were separated by at least 5 m. The assignment o f a particular treatment to one of these trap positions was determined at random. The positions of treatments were rerandomized after each trapping period (one or two nights). A priori we had decided not to include any blocks of the experiment in which no males were caught in any of the four treatments (unsuccessful replications). A total of 13 successful replications were conducted between 23 and 28 September 1994 at the University of Kentucky's South Farm (N = 2), Coldstream Farm (N = 2), Spindletop Farm (N = 2), a suburban backyard in Lexington, Kentucky (backyard site 1) (N = 2), and suburban backyard in Richmond, Kentucky (backyard site 2) (N = 5). After each trapping period, sticky bottoms of the traps were returned to the laboratory and the numbers of male T. mynesalis were counted. A KruskalWallis nonparametric anlaysis of variance was performed for this and subsequent experiments (Statistix 4.0, St. Paul Minnesota). This statistical approach was necessary because often control traps (and for some experiments other treatments) caught no moths; thus the homogeneity of variance assumption of the more powerful standard parametric analysis of variance was violated (with or without transformation). Means were separated with an experimentwise error rate set at ct = 0.05. In a second field experiment, the following blend ratios of 3Z,9Z-6S,7Repoxy-21 :H, 3Z,6Z,9Z-21 :H, and 3Z,9Z-6R,7S-epoxy-21 :H were tested: (a) 0 : 0 : 1, (b) 0 : 1 : I, (c) 1:1: 1, (d) 1 : 1 : 0 , and (e) 1 : 0 : 0 . The quantity of each compound when it was part of the blend to be applied to a rubber septum was kept at 100/zg. Control septa were loaded with hexane only. Because a methodological experiment conducted after our first experiment indicated that traps set at heights of about 30 cm were more effective in capturing male T. mynesalis than those set at 1.8 m (Haynes and Yeargan, unpublished data), we conducted this second and all subsequent tests at the lower height, Other details of this experiment were similar to our first experiment except that a trapping period was two or three nights. A total of eight successful replications were conducted between 3 and 12 October 1994 at the University of Kentucky's South Farm (N = 2), Spindletop Farm (N = 2), backyard site 1 (N = 2), and backyard site 2 (N = 2). In the third field experiment, the following blend ratios of 3Z,9Z,-6S,7Repoxy-21 :H and 3Z,6Z,9-Z-21 :H were evaluated: (a) 1 : 1, (b) 1:4, (c) 1: 10, and (d) 0 : 1 . The quantity of 3Z,6Z,9Z-21:H applied to a rubber septum was kept at 100/~g, and the quantity of 3Z,9Z-6S,7R-epoxy-21 :H was varied (100, 25, 10, and 0/~g). Control septa were loaded with hexane only. Other details of this experiment were similar to our second experiment, except that the Spindletop site was dropped because of the apparent low population density of
80
HAYNES, YEARGAN, MILLAR, AND CHASTAIN
7". mynesalis found there, and one additional location was added at both the South Farm and Richmond (backyard site 2) sites. A total of 14 successful replications were conducted between 12 and 19 October 1994 at the University of Kentucky's South Farm ( N = 6), backyard site 1 (N = 2), and backyard site 2 (N = 6). In the last field experiment, the following blend ratios of 3Z,9Z-6S,7Repoxy-21 :H and 3Z,6Z,9Z-21 : H were evaluated: (a) 2: 1, (b) 1 : 1, and (c) 1:2. The total quantity of 3Z,9Z-6S,7R-epoxy-21 :H plus 3Z,6Z,9Z-21 :H applied to a rubber septum was kept constant at 200 #g. Thus, for example, the 2 : 1 blend ratio consisted of 133 tzg 3Z,9Z-6S,7R-epoxy-21 : H plus 67/zg 3Z,6Z,9Z-21 : H. An additional septum was loaded with hexane only. Other details of this experiment were similar to our third experiment. A total of 14 successful replications were conducted between 19 and 26 October 1994 at the University of Kentucky's South Farm (N = 6), backyard site 1 (N = 2), and backyard site 2 (N = 6). Voucher specimens of male T. mynesalis trapped with a 1 : 1 blend of 3Z,9Z-6S,7R-epoxy-21 :H and 3Z,6Z,9Z-21 :H, and males reared from eggs laid by females caught in a black light trap have been deposited in the University of Kentucky, Department of Entomology, Collection. RESULTS AND DISCUSSION
Chemical and Electrophysiological Analysis. The GC-EAD analysis indicated there were two compounds present in gland extracts of individual female 7". mynesalis that stimulated mate 7". mynesalis antennae (Figure 1A). The EAD activity was observed at 12.00 and 13.20 min (DB-5), but there were no prominent FID peaks at these retention times. Other peaks were present in the gland extracts, but these did not stimulate an EAD response. The retention times of the straight-chain hydrocarbon standards 2 0 : H , 21: H, 2 2 : H , and 2 3 : H were 11.52, 12.11, 12.77 and 13.52 min, respectively. Because monoepoxide derivatives of triene hydrocarbons have been shown to be attractants for several species in the same subfamily (Wong et al., 1985), we evaluated standard solutions of 17- to 21-carbon triene (3Z,6Z,9Z-17-21 : H ) and 17- to 22-carbon monoepoxide derivatives (6Z,9Z-cis-3,4-epoxy-17 to 2 2 : H ; 3Z,9Z-cis-6,7epoxy- 17 to 22 : H; and 3Z,6Z-cis-9,10-epoxy- 17 to 22 : H). The 21-carbon hydrocarbon and epoxides gave the closest retention time matches to the EAD active compounds. Synthetic 3Z,6Z,9Z-21: H and 3Z,9Z-6S,7R-epoxy-21: H had retention times identical to the EAD active compounds (12.00 and 13.20 min, respectively) and stimulated strong EAD responses (Figure 1B). Other compounds (e.g., aldehydes and acetates) that have been determined to be pheromone components or sex attractants for other prey species of M. hutchinsoni did not stimulate EAD activity in male T. mynesalis antennae (Figure 1B).
T. mynesatis PHEROMONE
81
A. Gland extract
J 9
10
11
12
13
14
15
Time (rain)
B. Synthetic blend
d
b
I
I
9
10
e
I
I
I
11 12 13 Time ('rain)
I
I
14
15
FIG. 1. (A) GC-EAD analysis of the pheromone gland extract of one female T. mynesalis. The top trace shows the response o f the flame ionization detector (FID). The
bottom trace shows the electroantennogram (EAD) response. Peaks o f EAD activity were seen at retention times (R,) of 12.00 and 13.20 min (B). A synthetic blend consisting of 0.5 ng (a) 15:H (R, = 8.45), (b) Z 9 - 1 4 : A c (R, = 10.37), (c) Z l l - 1 6 : A I (R, = 10.45), (d) 3Z,6Z,9Z-21 :H (R t = 12.00), (e) 3Z,9Z-6S,7R-epoxy-21 :H (Rt = 13.20), and (f) 23 : H (R~ = 13.52) was analyzed by GC-EAD. The top trace shows the response of the FID. EAD responses (bottom trace) were seen at R, of 12.00 and 13.20 min.
82
HAYNES, YEARGAN, MILLAR, AND CHASTAIN
GC-MSD analysis of the pheromone extract using the DB-5 column indicated that two minor compounds present in individual gland extracts corresponded exactly in retention times and mass spectra with those of synthetic 3Z,6Z,9Z-21:H (20.09 min) and 3Z,9Z-6S,7R-epoxy-21:H (21.66 min). From pooled extracts (5 or 10 female equivalents), the first peak and 3Z,6Z,9Z-21 :H were characterized by prominent ions at m/z 55, 67, 79, and 108, as well as less abundant ions at 234 and 290 (the molecular ion for 3Z,6Z,9Z-21 : H). The less abundant ions were reliably detected only in pooled extracts, while the prominent ions were consistently found in individual and pooled extracts. The diagnostic ions (Underhill et al., 1983) at m/z 55 (CH3CH2CH=CH ÷) and t08 [CH3CH2(CH=CH)3H+], located the first of the three double bonds at C-3, and indicated that all three double bonds are clustered at one end of the structure. A further diagnostic ion at m/z 234 [CH3(CH2)10(CH=CH)3H +] located a second double bond at C-9 (Underhill et al., 1983). On the basis of exact matches in retention times and spectra, and on the reasonable assumption that conjugation of the third double bond with either one of the other two would significantly change the retention time, the triene was tentatively assigned the 3Z,6Z,9Z21 :H structure. Further evidence for this structure was provided by exact retention time matches of the insect-produced compound with a synthetic standard on a different GC column (DB-WAX). The second EAD-active peak had a retention time identical to synthetic 3Z,9Z-cis-6,7-epoxy-2t :H on two columns (DB-5 and DB-WAX). Its mass spectrum exactly matched that of a synthetic standard of 3Z,9Z-6S,7R-epoxy21 :H and data reported in the literature (Millar and Underhill, 1986). In particular, the mass spectrum was characterized by prominent ions at m/z 55, 67, 83, 95, and 111, as well as less abundant ions at 223, 277, and 306 (the molecular ion for 3Z,9Z-6S,7R-epoxy-21:H). The prominent ion at m/z 111 is diagnostic for a structure of this type with the epoxide in the 6,7 position (cleavage of C - C bond adjacent to the epoxide; T6th et al., 1992). However, the enantiomer 3Z,9Z-6R,7S-epoxy-21 : H cannot be distinguished from 3Z,9Z6S,7R-epoxy-21 :H by retention time or mass spectra. In individual gland extracts of five females analyzed by GC-MS on the DB-5 column, the quantity of 3Z,6Z,9Z-21 :H ranged from 0.12 to 0.51 ng with a mean of 0.23 + 0.16 ng/gland. The quantity of 3Z,9Z-cis-6,7-epoxy-21:H ranged from 0.30 to 0.86 ng with a mean of 0.56 + 0.26 ng. The range in blend ratios of 3Z,9Z-cis-6,7-epoxy-21 :H to 3Z,6Z,9Z-21 :H was 1.59:1 to 3.28:1 (blend ratio defined as the quantity of the epoxide divided by the quantity of the triene hydrocarbon) with a mean of 2.66 + 0.68 : 1. Results were similar for independent analyses of five glands conducted on a DB-WAX column. The quantity of 3Z,6Z,9Z-21 :H ranged from 0.19 to 0.65 ng with a mean of 0.36 + 0.17 ng/gland. The quantity of 3Z,9Z-cis-6,7-epoxy-21 :H ranged from 0.39 to 1.11 ng, with a mean of 0.71 + 0.32 ng. The range in blend ratios of 3Z,9Z-
T. mynesalis
PHEROMONE
83
cis-6,7-epoxy-21 : H to 3Z,6Z,9Z-21 : H was 1.31 : 1 to 3.01 : 1, with a mean o f 2.12 + 0.77: 1.
FieM Tests. Following a preliminary field test in which traps baited with 100 ~tg of 3Z,9Z-6S,7R-epoxy-21 : H caught four male T. mynesalis, and the enantiomer 3Z,9Z-6R,7S-epoxy-21 : H caught no males, we opted to focus the first full field experiment on varying blend ratios of 3Z,9Z-6S,7R-epoxy-21:H and 3Z,6Z,9Z-21 :H. In this experiment, the 1 : 1 blend of 3Z,9Z-6S,7R-epoxy2 1 : H and 3Z,6Z,9Z-21 : H was the most attractive blend for trapping male T. mynesalis (Figure 2). Traps baited with this blend caught significantly more mate T. mynesalis than control traps [Kruskal-Wallis statistic ( H ) = 17.73, P < 0.001, experimentwise error rate set at a = 0.05 for mean comparisons]. The mean numbers of males caught in traps baited with the 4 : 1 and 10 : 1 blends of 3Z,9Z-6S,7R-epoxy-21 : H and 3Z,6Z,9Z-21 : H were intermediate between and not significantly different from the numbers caught in control traps (zero) and the 1 : 1 blend, using this conservative statistical test. In the second field experiment, traps baited with a 1:1 blend of 3Z,9Z,6S,7R-epoxy-21 :H and 3Z,6Z,9Z-21 : H caught the largest number of
~4 e~
ab
1
/ Ratio 3Z,9Z-6S,7R-epoxy-21:H to 3Z,6Z,9Z-21:H
FIG. 2. Captures of male T. mynesalis moths in traps baited with blend ratios of 3Z,9Z6S,7R-epoxy-21 :H and 3Z,6Z,9Z-21 :H, or a hexane-treated control. The quantity of 3Z,gZ-6S,7R-epoxy-21:H was kept constant at 100 p.g. Bars represent standard errors. Numbers are the mean number of males caught per trap per trapping interval (see Methods and Materials). Means that do not share a letter in common are significantly different (experimentwise error rate c¢ = 0.05, Kruskal-Wallis nonparametric analysis of variance).
84
H A Y N E S , Y E A R G A N , M I L L A R , AND CHASTAIN
males and significantly more than all other treatments except for traps baited with 3Z,9Z-6S,7R-epoxy-21 : H alone (Figure 3) ( H = 32.42, P < 0.0001). The enantiomer, 3Z,9Z-6R,7S-epoxy-21 : H was ineffective in attracting males to traps either by itself or as part of a 1 : 1 blend with 3Z,6Z,9Z-21 : H. Addition of this enantiomer to the otherwise effective attractant o f a 1 : I blend of 3Z,9Z6S,7R-epoxy-21 : H and 3Z,6Z,9Z-21:H shut down the response to the lure. In the third field experiment, traps baited with 3Z,6Z,9Z-21 :H alone or a 1 : 10 blend of 3Z,9Z-6S,7R-epoxy-21 : H and 3Z,6Z,9Z-21 :H were significantly less effective in catching male T. mynesalis than traps baited with the 1 : 1 blend ratio (Figure 4) ( H = 45.39, P < 0.0001). The hydrocarbon by itself and the 1 : 1 0 blend of 3Z,9Z-6S,7R-epoxy-21 :H and 3Z,6Z,9Z-21 : H were no more effective than control traps in capturing males in traps. In the fourth trapping experiment, we focused on a narrower range of blend ratios of 3Z,9Z-6S,7R-epoxy-21: H and 3Z,6Z,9Z-21:H. The number of males caught in traps baited with blend ratios of 2 : 1, 1 : 1, and 1 : 2 of 3Z,9Z-6S,7R-
e~
~6 e~ eq
E4
Z2 b T
b
b
b
3Z,9Z-6S,7R-epoxy-21 :H to 3Z,6Z,9Z-21 :H to 3Z,9Z-6R,7S-epoxy-21 :H FIG. 3. Captures of male T~ mynesalis moths in traps baited with blend ratios of 3Z,9Z6S.7R-epoxy-21 : H, 3Z,6Z,9Z-21 : H, and 3Z,9Z-6R,7S-epoxy-21 : H, or a hexane-treated control. The quantity of each compound was kept at 100 p.g. Bars represent standard errors. Numbers are the mean number of males caught per trap per trapping interval (see Methods and Materials). Means that do not share a letter in common are significantly different (experimentwise error rate m = 0.05, Kruskal-Wallis nonparametric analysis of variance).
T. mynesalis PHEROMONE
85
20 m.
L. 15
ab E ~ 10 o
E
bc
0
3 Z , 9 Z - 6 S , 7 R - e p o x y - 2 1 :H to 3Z,6Z,9Z-21 :H
FIG. 4. Captures of male T. mynesalis moths in traps baited with blend ratios of 3Z,9Z6S,7R-epoxy-21 : H and 3Z,6Z,9Z-21:H, or a hexane-treated control. The quantity of 3Z,6Z,9Z-21:H was kept constant at 100 p~g. Bars represent standard errors. Numbers are the mean number of males caught per trap per trapping interval (see Methods and Materials). Means that do not share a letter in common are significantly different (experimentwise error rate c~ = 0.05, KruskaI-Wallis nonparametric analysis of variance).
epoxy-21 : H and 3Z,6Z,9Z-21 : H were not significantly different from each other, but each of these blend ratios was significantly more attractive than hexane alone (Figure 5) ( H = 29.59, P < 0.0001). Blend ratios from 2 : t to 1:2 o f 3Z,9Z-6S,7R-epoxy-21:H and 3Z,6Z,9Z2 1 : H were effective in capturing male T. mynesalis in sticky traps. The range of blend ratios of 3Z,9Z-cis-6,7-epoxy-21:H to 3Z,6Z,9Z-21: H found in pheromone gland extracts (!.31 : 1 to 3.28: 1) overlapped this range. The chromatographic techniques we employed did not allow us to distinguish between the enantiomers 3Z,9Z-6S,7R-epoxy-21:H and 3Z.9Z-6R,7S-epoxy-21:H, and to date, there is no reported analytical chemistry method for distinguishing these two enantiomers at nanogram levels. However, the results from the second field test indicate that 100/zg of 3Z,9Z-6R,7S-epoxy-21:H is not effective in attracting males when presented with 100 /~g of 3Z,6Z,9Z-21 :H. In addition, the presence of 100/~g of 3Z,6Z,9Z-6R,7S-epoxy-21: H within an otherwise effective 1:1 blend of 3Z,9Z-6S,7R-epoxy-21:H and 3Z,6Z,9Z-21:H (100/zg each) prevented captures of any males, and thus at this level 3Z,9Z-6R,7S-epoxy21 : H is a behavioral antagonist. The enantiomeric purity of the synthetic 3Z,9Z-
86
HAYNES, YEARGAN, MILLAR, AND CHASTAIN
12 a
a
e~
9 gh
6 A3
S 3
0 3Z,9Z-6S,7R-epoxy-21 :H to 3Z,6Z,9Z-21 :H
FIG. 5. Captures of male T. mynesalis moths in traps baited with blend ratios of 3Z,9Z6S,7R-epoxy-2t : H and 3Z,6Z,9Z-21 :H, or a hexane-treated control. The total quantity of 3Z,9Z-6S,7R-epoxy-21:H plus 3Z,6Z,9Z-21 :H applied to a rubber septum was kept constant at 200 ~g. Bars represent standard errors. Numbers are the mean number of males caught per trap per trapping interval (see Methods and Materials). Means that do not share a letter in common are significantly different (experimentwise error rate c¢ = 0.05, Kruskal-Wallis nonparametric analysis of variance). 6S,7R-epoxy-21 :H was approximately 94% (Millar et al., 1986). This leaves open the possibility that even this 6% level of 3Z,9Z-6R,7S-epoxy-21 :H within 3Z,9Z-6S,7R-epoxy-21 :H had a negative (or possibly a positive) impact on captures of males. This cannot be tested until material of higher enantiomeric purity is available. The identification of a blend of 3Z,9Z-6S,7R-epoxy-21 : H and 3Z,6Z,9Z21 : H as a sex pheromone for T. mynesalis is the first such identification for a species in the Herminiinae subfamily of the Noctuidae (Am et al., 1992). However, in field trials, monoepoxide dienes or trienes (19-21 carbons) have been reported to be sex attractants for other Herminiinae, including Bleptina caradrinalis Guenre, Idia aemula (Hfibner), 1. americalis (GuenEe), and Palthis angulalis (Hiibner) ( W o n g et al., 1985; Millar et al., 1991). Every pheromone blend that has been identified to date for prey species of bolas spiders is composed of unsaturated straight-chain aliphatic aldehydes, alcohols, or acetates (see Yeargan, 1994; A m et al., 1992). No diene or triene hydrocarbons or their monoepoxide derivatives have been discovered to be sex attractants for any of these prey species. Four noctuid prey species (three species
T. mynesalis PHEROMONE
87
belonging to Catocalinae and one to Herminiinae) of M. bisaccata and two prey species (Geometridae: Larentiinae) of M. ph~.nosoma, however, are closely related to species that have been attracted to specific diene or triene hydrocarbons or monoepoxide derivatives in field screening tests of these compounds. This taxonomic relationship suggests it is very likely that the pheromone blends of a subset of the prey species of these two bolas spiders will be hydrocarbons or epoxides, as is the case for M. hutchinsoni. Prior to the present study, sex attractants or pheromones were known for three of the four species of moths that are captured by M. hutchinsoni. Previously, Haynes (1990) determined that female bristly cutworm moths, Lacinipolia renigera (Noctuidae: Hadeninae), emit Z 9 - 1 4 : A c and Z , E - 9 , 1 2 - 1 4 : A c (3.8% of Z 9 - 1 4 : A c ) and that synthetic blends of these two components are attractive to males. Males of the bronzed cutworm, Nephelodes minians (Noctuidae: Hadeninae), are attracted to Z11-16 : AI (Underhill et al., 1977). An optimized attractant for male bluegrass webworms, Parapediasia teterrella (Pyralidae: Crambinae), consists of a 20 : 1 blend of Z11-16 : AI and Z9-16 : AI (Clark and Haynes, 1990). Yeargan (1988) found that T. mynesalis (52.6%) and L. renigera (40.0%) accounted for the vast majority of the 460 moths caught by M. hutchinsoni in a two-year study. Thus, the two most frequently captured prey use pheromones that are composed of different classes of chemicals. The identification of a blend of 3Z,9Z-6S,7R-epoxy-21 : H and 3Z,6Z,9Z21 : H as a sex pheromone for T. mynesalis suggests that female M. hutchinsoni may be capable of synthesizing a diverse set of allomones, including both straight-chain aliphatic acetates and aldehydes and triene hydrocarbons and monoepoxide derivatives. Stowe (1988) discussed the evolution of chemical mimicry in plants and animals, including the case of bolas spiders and moths. At the time of his review, no prey species of bolas spiders had been shown to use diene or triene hydrocarbons or their monoepoxide derivatives in their sex pheromones, but the phylogenetic relationships of some prey to species for which these pheromones or attractants had been found suggested that these compounds were likely candidates to be pheromone components of certain prey. Stowe (1988) offered several possible explanations of how bolas spiders might capture moth species that use chemically diverse pheromone components; given the results of our current study, three alternative hypotheses seem most plausible to explain this diversity. First, there may be variation in the attractants produced among individual bolas spiders, resulting in specialization by individual spiders but yielding a more diverse assemblage of prey at the population level. Second, the blend produced by an individual spider might change over time. Last, each spider might produce a consistent single blend of compounds, if individual components do not interfere with attraction of the other species of prey. Yeargan (1988) found definitive evidence that precludes the possibility of interindividual variation in M. hutchinsoni leading to extreme specialization of
88
HAYNES, YEARGAN, MILLAR, AND CHASTA1N
individual spiders on a single prey species. He found that the majority of spiders (56.3%) that caught two or more prey items on a single night also caught two or more species of prey. In fact, some individual spiders caught all four of the known prey species over the course of his study, It is possible that a spider changes the attractant blend it produces over the course of the scotophase, because the two principal prey of M. hutchinsoni are captured at different times of night. L. renigera is captured before 10:30 PM, while T. mynesalis is captured after 11:00 PM, except on very cool nights (Yeargan, 1988). Preliminary evidence suggests that temporal variation may occur in the composition of the blend emitted by M. cornigera, the only species of bolas spider for which allomonal components have been identified from the spider itself (Stowe et al., 1987). An alternative to temporal variation in the allomonal blend produced by M. hutchinsoni is that the spider, while hunting, may produce a relatively constant blend of the attractants for the two principal prey species. This would be possible if the chemical constituents do not interfere with the effectiveness of either attractant. The G C - E A D results (Figure 1) indicate that the male T. mynesalis's antenna does not show a measurable EAD response to Z 9 - 1 4 : A c (the major pheromone component of L. renigera) or Z 1 1 - 1 6 : AI (the major sex attractant component for P. teterrella). Thus, it is possible that these heterospecific pheromone blends are not even detected by T. mynesalis. Either the second hypothesis (temporal variation) or the third hypothesis (consistent, compatible blend) could account for the ability of M. hutchinsoni to capture moths that use chemically diverse pheromones. We plan to investigate this chemical mimicry system to ascertain which hypothesis is correct.
Acknowledgments--WalterLeal provided helpful advice concerning the setup of the GC-EAD. F.W. Knapp and P.A. Weston kindly reviewed an earlier draft of the manuscript. This work was supported in part by State and Federal Hatch funds to the Kentucky Agricultural ExperimentStation (paper no. 95-08-062). REFERENCES ARN. H., TOTH,M., and PRIESNER,E. 1992. List of Sex Pheromones of Lepidoptera and Related Attractants, 2nd ed., International Organization for Biological Control, Montfavet, France. CLARK, J.D., and HAYNES,K.F. 1990. Sex attractant for the bluegrass webworm (Lepidoptera: Pyralidae). J. Econ. Entomol. 83:856-859. COVEt.L,C.V., JR. 1984. A Field Guide to the Moths of Eastern North America. Houghton Mifflin, Boston. EBERHARD,W.G. 1977. Aggressive chemical mimicry by a bolas spider. Science 198:1173-1175. HAYNES, K.F. 1990. Identification of sex pheromone of bristly cutworm, Lacinipolia renigera (Stephens). J. Chem. Ecol. 16:2615-2621, LEAL, W.S., MOCHIZUKI,F., WAKAMURA,S., and YASUDA,T. 1992. Electroantennographicdetec-
T. mynesalis PHEROMONE
89
tion ofAnomala cuprea Hope (Coleoptera: Scarabaeidae) sex pheromone. Appl. Entomol. Zool. 27:289-291. MILLAR, J.G., and UNDERHmL, E.W, 1986. Synthesis of chiral bis-homoallylic epoxides. A new class of lepidopteran sex attractants, J, Org. Chem. 51:4726-4728. MILLAR, J.G., GIBLIN, M., BARTON, D,, and UNDERHILL,E.W. 1991. Synthesis and field screening of chiral monounsaturated epoxides as lepidopteran sex attractants and sex pheromone components. J. Chem. Ecol. 17:911-929. MOSHER, E. 1969. Lepidoptera Pupae. Entomological Reprint Specialists, East Lansing, Michigan. PRINGLE, J.W.S. 1938, Proprioception in insects I. A new type of mechanical receptor from the palps of the cockroach. J. Exp. Biol, 15:101-113. STOWE, M.K. 1988. Chemical mimicry, pp. 513-580. in K.C. Spencer (ed.). Chemical Mediation of Coevolution. Academic Press, San Diego, STOWE, M.K., TUMLINSON,J.H., and HEATH, RR. 1987. Chemical mimicry: Bolas spiders emit components of moth prey species sex pheromones. Science 236:964-967. STRUBLE, D.L., and ARN, H. 1984, Combined gas chromotography and electroantennogram recording of insect olfactory responses, pp. 161-178, #7 H.E, Hummel and T.A. Miller (eds_). Techniques in Pheromone Research. Springer-Vedag, New York. TOTH, M,, BUSER, H.R., GUERIN, P.M., ARN, H., SCHM|DT, F., FRANCKE, W,, and Szocs, G, 1992. Abraxas grossulariata L. (Lepidoptera: Geometridae): Identification of (3Z,6Z,9Z)3,6,9-heptadecatriene and (6Z,9Z)-6,9-cis-3,4-epoxyheptadecadiene in the female sex pheromone. J. Chem Ecol. 18:13-25, UNDERHILL, E.W., CmSHOLM, M.D., and STECI'L W. 1977. Olefinic aldehydes as constituents of sex attractants for noctuid moths, Envircm. EntomoL 6:333-337, UNDERHILL, E.W., PALANISWAMY,P,, ABRAMS,S.R,, BAILEY,B.K., STECK, W.F., and CHISHOLM, M.D. 1983. Triunsaturated hydrocarbons, sex pheromone components of Caenurgina erechtea, J. Chem. Ecol. 9:1413-1423. WONG, J.W., UNOERHJLL, E.W., MACKE:.NZm,S.L., and CmSHOLM, M.D. 1985. Sex attractants for geometrid and noctuid moths; field trapping and electroantennographic responses to triene hydrocarbons and monoepoxydiene derivatives. J. Chem. Ecol. 11:727-756. YEARGAN, K.V. 1988. Ecology of a bolas spider. Mastophora hutchinsoni: Phenology, hunting tactics, and evidence for aggressive chemical mimicry. Oecologia (Berlin) 74:524-530. YEARGAN, K.V. 1994. Biology of bolas spiders, Annu. Rev. Entomol. 39:81-99.
