Pheromone components of the wheat stem sawfly: Identification, electrophysiology, and field bioassay

Article abstract of DOI:10.1023/A:1017946527376

Volatiles collections and cuticular extracts of the wheat stem sawfly, Cephus cinctus Norton (Hymenoptera: Cephidae), were analyzed by coupled gas chromatographic-electroantennographic detection (GC-EAD) in order to detect possible pheromone components of

Full text of DOI:10.1023/A:1017946527376

Journal of Chemical Ecology, Vol. 28, No. 2, February 2002 (�C 2002)
PHEROMONE COMPONENTS OF THE WHEAT STEM
SAWFLY: IDENTIFICATION, ELECTROPHYSIOLOGY, AND
1
FIELD BIOASSAY
´
2,
2
3
ALLARD A. COSSE, * ROBERT J. BARTELT, DAVID K. WEAVER,
and BRUCE W. ZILKOWSKI2
2USDA Agricultural Research Service
National Center for Agricultural Utilization Research
Crop Bioprotection Research Unit
1
815 N. University Street
Peoria, Illinois 61604
3Montana State University
Department of Entomology
3
33 Leon Johnson Hall, P.O. Box 173020
Bozeman, Montana 59717-3020
(Received August 31, 2000; accepted October 5, 2001)
Abstract—Volatiles collections and cuticular extracts of the wheat stem sawfly,
Cephus cinctus Norton (Hymenoptera: Cephidae), were analyzed by coupled
gas chromatographic–electroantennographic detection (GC-EAD) in order to
detectpossiblepheromonecomponentsofthisspecies. Volatilescollectionsfrom
maleandfemalesawfliescontainedthesame13GC-EAD-activecompoundsthat
stimulated both male and female antennae. GC-EAD-active compounds were
identified by mass spectrometry, and the identities were verified by acquiring
or synthesizing authentic standards with identical GC retention times and mass
spectra. 9-Acetyloxynonanal gave the strongest EAD response. Additional GC-
EAD-active compounds included 13-acetyloxytridecanal, aldehydes with 9–16
carbon chain lengths, acids with 8–10 carbon chain lengths, and phenylacetic
acid. The last was instantly detectable by human nose when several males were
allowed to interact, but not from isolated insects. On a per-insect basis, grouped
males overall released higher amounts of the GC-EAD-active compounds than
grouped females. Notable quantitative differences between the sexes were rel-
atively higher amounts of 9-acetyloxynonanal, phenylacetic acid, and tetrade-
canal from males and hexadecanal from females. In the absence of an absolute,
qualitative difference between sexes, these relative differences might be im-
*
To whom correspondence should be addressed. E-mail: cosseaa@ncaur.usda.gov
1
Mention of trade names or commercial products in this article is solely for the purpose of providing
specific information and does not imply recommendation or endorsement by the US Department of
Agriculture.
4
07
0
098-0331/02/0200-0407/0 �C 2002 Plenum Publishing Corporation

4
08
COSS E´ , BARTELT, WEAVER, AND ZILKOWSKI
portant for sex recognition. Field bioassay demonstrated the attractiveness of
9
-acetyloxynonanal for male and female C. cinctus. The major compounds of
male and female cuticular extracts showed no GC-EAD activity, but ozonolysis
of extracts gave dramatically increased amounts of GC-EAD-active material.
Hence, double-bond-containing cuticular compounds are suggested as precur-
sors of the aldehydic pheromone components.
Key Words—Sawfly, aggregation pheromone, 9-acetyloxynonanal, phenyl-
acetic acid, gas chromatographic–electroantennographic detection, cuticular ex-
tracts, Cephus cinctus, Cephidae, Hymenoptera.
INTRODUCTION
Thewheatstemsawfly, CephuscinctusNorton(Hymenoptera:Cephidae), hasbeen
a major pest of wheat, Triticum aestivum L, in the northern Great Plains of the
United States and the Canadian prairies since cultivation began (Weiss and Morrill,
1992). In Montana, the insect has been estimated to cause annual losses in excess
of $25 million in recent years (Blodgett et al., 1996, 1997), and losses in the entire
area impacted by the wheat stem sawfly are estimated to exceed $100 million (US
Department of Agriculture, 2000). The biology of the insect is described in Ainslie
(1920), and population dynamics are given in Holmes (1982). Briefly, short-lived
adults emerge from wheat stubble in the spring and fly to cereals or other grasses,
where they oviposit into the interior of growing stems. The larvae feed and develop
cryptically within the stem during the summer. When crop maturity is reached and
the host plant desiccates, the larvae descend to the base of the plant and girdle the
stem near the soil interface, resulting in lodged plants. The mature larva seals the
tip of the cut stem with a uniform plug and then descends to near the root where
it overwinters as a mature larva within a translucent hibernaculum. Dispause is
broken in the spring, and metamorphosis is completed. Adults chew a hole in the
plug at the tip of the cut stem and move to the surrounding green vegetation.
The female-produced sex pheromones of sawflies belonging to the family
Diprionidae and one species belonging to the family Tenthredinidae are well doc-
umented (Anderbrant 1999 and references therein). In addition, studies on several
species belonging to the family Pamphiliidae have shown that female-produced sex
pheromones are used for mate location (Borden et al., 1978). However, no studies
have been published on pheromones in the remaining 11 families of hymenopteran
sawflies, including Cephidae. Understanding the chemical ecology of C. cinctus
could lead to new and environmentally friendly approaches for reducing agricul-
tural damage, especially since insecticides are ineffective against larvae inside the
wheat stem and adult emergence is difficult to predict (Morrill et al., 1992).
This paper and the companion paper (Bartelt et al., 2002) describe the iden-
tification, electrophysiological, chemical, and behavioral aspects of pheromonal
communication in the wheat stem sawfly.

WHEAT STEM SAWFLY PHEROMONE
409
METHODS AND MATERIALS
Insects. Overwintering stage prepupae of C. cinctus were collected at several
field sites in Montana. After the 1998 harvest, collections were made in September
and October in Broadwater and Cascade counties, and again in March and April,
1999, from the same sites. Stubble was removed from these heavily infested fields
with a shovel and taken to the laboratory. The uninfested residue was separated
from the sawfly-cut stems, and stems containing sawfly larvae were rinsed to
remove the soil, spotted dry with a paper towel, and stored in lots of 100 in
plastic Ziploc bags. To allow the larvae to complete diapause, bags containing fall-
�
collected larvae were stored at 4 C for at least 100 days, while material collected
in late winter and early spring was refrigerated until used. Upon arrival in Peoria,
Illinois, sawfly-containing wheat stems were placed individually into glass vials
�
capped with a moist piece of cloth-covered cotton. Adults emerged at 25 C under
a 16L:8D photoperiod regime using a fluorescent light source.
Initial Cuticular Extracts. Groups of 25 male and 25 female sawflies were
killed by freezing when 1–3 days old and cuticular extracts were obtained by
�
soaking insects for 20 min in 1 ml hexane. Hexane extracts were stored at � 20 C.
Volatiles Collections. Volatiles collections were made from individuals or
groups of two to five male or female insects. The 1- to 3-day-old adults were
placed in 75-ml silanized (dichlorodimethylsilane) (Bartelt and Zilkowski, 1998)
Pyrex glass tubes fitted with corks at both ends, into which Super Q porous polymer
(80–100 mesh, Alltech, Deerfield, Illinois) traps were inserted. The Super Q traps
were prepared from 10-cm � 0.5-cm-ID glass tubing. A piece of stainless steel
screen (325 mesh) was sealed into one end by heating. Traps were filled to a depth
of 0.5 cm with Super Q and topped with a plug of silanized glass wool. Newly
prepared and used traps were conditioned by several rinses of CH2Cl2 followed by
hexane and dried under a gentle stream of nitrogen. Traps used to collect volatiles
from male and female insects were kept separated. Insects had access to water from
2
ml cotton plugged vials, which were placed inside the collection tubes. Volatiles
collections from tubes with water vials, but without insects, served as controls. A
vacuum was applied to one of the Super Q traps so that volatiles from the tube were
drawn into the trap. The second Super Q trap then served as a filter for incoming
�
air. The tubes were kept under fluorescent light in an incubator (16L:8D) at 27 C
and 40% relative humidity with an airflow through the tubes adjusted to 30 ml/min.
Two- to three-day volatiles collections were started 2–3 hr into the photophase. To
extract volatiles from the Super Q traps, each trap was rinsed with 250 �l hexane.
Samples were concentrated to approx. 10 �l under a gentle stream of nitrogen and
1
�l was analyzed. Approximately 160 volatiles collections were made.
Solid-phase microextraction (SPME) was used as a second volatiles col-
lection method. Individuals and groups of up to five adults were placed in the
above-mentioned silanized Pyrex glass tubes. The upwind end of a glass tube was

4
10
COSS E´ , BARTELT, WEAVER, AND ZILKOWSKI
covered with nylon mesh and the downwind end was fitted with a cork, into which
a glass tube (0.4 cm ID) with a side arm was inserted. A SPME fiber (100 �m poly-
dimethylsiloxane. Supelco) was inserted though a septum placed into the other end
of the tube, extending past the side arm junction. Vacuum was applied to the side
arm so that air from the outside passed though the tube with the sawflies and then
over the SPME fiber. Volatiles collections were made 2–3 hr into the photophase
�
at 25 C for 30 min with airflow of 10 ml/min. Insects could not contact the SPME
fiber.
Electrophysiology. Coupled gas chromatographic–electroantennographic
(GC-EAG) analyses were made by methods and equipment generally described
by Coss e´ and Bartelt (2000).
Pheromone Identification. Volatiles collections were analyzed by gas chro-
matography with flame ionization detection (GC-FID), coupled GC-mass spec-
trometry (GC-MS), and GC-EAD. The GC was a Hewlett Packard 6890 instrument
fitted with a 30-m EC-1 capillary column (0.25 mm ID, 0.25 �m film thickness,
�
�
�
Alltech). Temperature programs were from 50 C to 275 C at 10 C/min. (GC-
�
�
�
FID) or from 50 C to 275 C at 15 C/min. (GC-EAD). Injector temperatures were
maintained at 250 C, and the GC-EAD effluent interface from the post-colum split-
ter was kept at 275 C. Injectors were fitted with SPME injector liners (Supelco,
�
�
Bellefonte, Pennsylvania) for SPME analysis. Mass spectrometry was performed
by using a Hewlett Packard 5973 instrument (electron impact, 70 eV). Chem-
ical ionization spectra (methane) were obtained on a Finnigan 4535 quadrupole
mass spectrometer with GC inlet. Most EAD-active compounds were identified by
searches of the mass spectral library (Wiley, 1995). Identifications were confirmed
with authentic standards that had the same GC retention times and mass spectra.
Ozonolysis of cuticular extracts was done as described by Beroza and Bierl
(
1967), andtheproductswerecomparedwiththecompoundsinthevolatilescollec-
�
tions. Samples were ozonized in 0.5 ml dichloromethane for about 2 min at � 70 C.
Ozone was formed by electrical discharge through a stream of oxygen. The gas
was bubbled gently through the solution until a faint bluish color developed, in-
dicative of excess ozone. Ozonides were reduced with excess triphenylphosphine.
Samples were analyzed by GC-MS and GC-EAD.
Chemicals. Nonanal and (Z)-9-dodecen-1-ol acetate were purchased from
Bedoukian (Danbury, Connecticut), and the remaining commercially available
chemicalsusedinthisstudywereobtainedfromAldrichChemicalCo. (Milwaukee,
Wisconsin).
Synthesis of 9-Acetyloxynonanal. A solution of (Z)-9-dodecen-1-ol acetate
(100 mg, 0.44 mmol) in 5 ml hexane was cooled in a dry ice–ethanol bath. Ozone
was bubbled through the solution until a faint bluish color developed, indicative of
excess ozone. The solution was purged with N2 until the blue color disappeared.
The reaction mixture was reduced by adding dimethylsulfide (0.66 mmol) and
allowed to warm to room temperature. After addition of 0.5 ml H2O and separation

WHEAT STEM SAWFLY PHEROMONE
411
of layers, the organic phase was dried (Na2SO4) and evaporated (N2) to remove
propanal and excess dimethylsulfide, followed by dilution by hexane. Part of this
sample was used for subsequent field studies.
Field Trial with 9-Acetyloxynonanal. Synthetic 9-acetyloxynonanal was ap-
plied in six selected doses (0.3, 1, 3, 10, 100, or 1000 �g) to rubber septa in Peoria,
�
Illinois, and sent to Montana, where they were stored at � 80 C until use. Septa
without added compound were the seventh (control) treatment. Beginning on June
23, 1999, a 15-day field test was conducted by one of us (D.K.W.) at a site 17 km
northwest of Three Forks, Montana, near the county line between Jefferson and
�
0
�
0
Broadwater counties (46 2.19 N, 111 38.39 W). The test was conducted as the
sawfly population was beginning to decline. The field site consisted of a fence
line that separated a standing crop of wheat (to the west) from a fallow field (to
the east) that had been moderately infested with wheat stem sawflies during the
1998 growing season. This field was the source of the sawflies that would infest
the growing crop. Prevailing winds were westerly; thus, movement into the new
crop would generally be upwind. Twenty-one traps were spaced at 30-m intervals
along this fence line. Each trap consisted of a 10-cm � 10-cm square of #20 mesh
metal hardware cloth, coated with Tangle-Trap on both sides. The trap was stapled
to a 60-cm wooden stake driven into the ground so that the bottom of the sticky
surface was approximately 5 cm above the soil substrate. A rubber septum trap
bait was penetrated with a paper clip and attached to the center of the upwind
side of the sticky surface. Daily insect counts were taken and the trapped insects
were removed. Lures were replaced and the traps were re-randomized every three
days. Each of the seven dose treatments was replicated three times in the random
trap array. The trap catch data were analyzed by Poisson regression, a method that
is appropriate for small counts, using the Statistix analysis package (Analytical
Software, 1998). Details of the models are given in the Results section.
RESULTS
Cuticular Extracts. GC-EAD analyses of male and female cuticular extracts
were obtained with male and female antennae. As shown in Figure 1, GC-EAD
analyses of female cuticular extracts and the simultaneously recorded EAD of male
antennae showed no activity with any of the higher molecular weight compounds,
such as the major cuticular hydrocarbons (Bartelt et al., 2002). However, com-
pounds of lower molecular weight did stimulate male antennae. Similar results
were obtained with female antennae and with male cuticular extracts and male
and female antennae. Further research was focused on the analysis of volatiles
collections due to the EAD activity of compounds of lower molecular weight.
Volatiles Collections from Individual Insects. Initial GC-EAD profiles of
Super Q-collected volatiles from individual males and females revealed variable

4
12
COSS E´ , BARTELT, WEAVER, AND ZILKOWSKI
FIG. 1. Simultaneously recorded gas chromatogram (GC) and electroantennogram re-
sponses (EAD) of a typical male antenna to a cuticular hexane extract of female C. cinctus.
and relatively small amounts of insect-specific material during two- to three-day
collection periods. In male-collected volatiles, the major compounds were nonanal
and (Z)-9-tricosene (Bartelt et al., 2002). The same compounds, at lower amounts,
were present in female-collected volatiles. With males, these compounds were
present at 43 ng (� 14 SD, N = 8) and 96 ng (� 76 SD, N = 8) per sawfly per day,
respectively. From females, collected amounts of nonanal were 26 ng (� 13 SD,
N = 5) per sawfly per day. Amounts of (Z)-9-tricosene from females were smaller
(e.g., Figure 2), because females have less of this hydrocarbon than males (Bartelt
et al., 2002). These initial male and female collections consistently elicited a no-
table depolarization on both male and female antennae (Figure 2). In addition to
the depolarization elicited by the compound eluting at 10.42 min, several smaller
depolarizations were noted. However, these depolarizations were less consistent,
which was probably due to the variable and relatively small amounts of insect-
specific material.
Identification of 9-Acetyloxynonanal. The mass spectrum of the compound
giving the relatively strong depolarization (10.42 min) is shown in Figure 3. No

WHEAT STEM SAWFLY PHEROMONE
413
FIG. 2. Simultaneously recorded gas chromatogram (GC) and electroantennogram re-
sponses (EAD) of a typical male antenna to a volatiles collection from an individual female
C. cinctus. Arrow in cutout denotes the location of 9-acetyloxynonanal.
molecular ion was visible, and the mass spectral library did not provide a probable
identification. However, the fragment at m/z 61 was informative and suggested
that the compound was a primary acetate. The largest significant ion in the chem-
ical ionization (CI) mass spectrum had m/z 201, suggesting it was the protonated
molecular ion; thus, the molecular weight was 200. The CI spectrum also had two
major fragments. The first (m/z 141) represented a loss of 60 from the molecular
ion and corresponded to a neutral loss of acetic acid, as would be anticipated for
an acetate. The second fragment (m/z 123) represented an additional loss of 18,
corresponding to a loss of water. Together, these features suggested the presence
of three (or more) oxygen atoms for a possible molecular formula of C11H20O3,
with two sites of unsaturation. If two oxygens and one site of unsaturation were
accounted for by an acetate function, then one additional oxygen and one double-
bond equivalent would remain to be explained. Example spectra from the mass
spectral library gave some guidance: in primary acetates with another functional
group, the m/z 61 fragment peak is relatively large, as in the sawfly compound, only
when the two functional groups are distant from each other in the structure. Thus,
a nine-carbon chain with an acetate group on one end and an epoxy, aldehyde, or
ketone group near the other end was suggested.

4
14
COSS E´ , BARTELT, WEAVER, AND ZILKOWSKI
FIG. 3. Electron impact and chemical ionization mass spectra for 9-acetyloxynonanal.
Amounts of this compound collected from males and females were variable
and were in the range of 1–5 ng per sawfly per day, with lower amounts produced by
females overall. The amounts of material on hand were so small that other spectra
or microchemical tests were not deemed feasible. The approach for continuing
was to synthesize model compounds for comparison of mass spectral details and
GC retention times. 9-Acetyloxynonanal was the first compound prepared because
it was accessible in a single step (ozonolysis) from (Z)-9-dodecen-1-ol acetate,
which was on hand as an example moth pheromone. Fortuitously, the synthetic and
natural compounds matched perfectly, by GC retention time and mass spectrum.
A larger amount of this compound was synthesized for subsequent laboratory and
field use.
Volatiles Collections from Groups of Insects. Single males appeared odorless
to the human nose and stayed relatively inactive when placed in collection tubes.
However, placing several males together in a collection tube almost instantly trig-
gered wing fanning of all the males, which was accompanied by the release of a
relatively strong, distinctive odor. This odor was absent with single females and
barely detectable with grouped females. Due to the detectable odor of grouped
males, additional GC-EAD analyses were performed on volatiles from groups of
insects, pooling collections to increase amounts of compound to be analyzed.

WHEAT STEM SAWFLY PHEROMONE
TABLE 1. WHEAT STEM SAWFLY (Cephus cinctus) VOLATILES
415
IDENTIFIED FROM GROUPED SAWFLIES WITH ELECTROPHYSIOLOGICAL
ACTIVITY ON MALE AND FEMALE ANTENNAE AND THEIR RELATIVE
ABUNDANCE
b
Relative abundance (mean � SD)
Peak
Compounda
Nonanalc
Octanoic acid
Decanal
Phenylacetic acid
Nonanoic acid
Decanoic acid
Dodecanal
9-Acetyloxynonanal
Tridecanal
Per male/day
Per female/day
1
2
3
4
5
6
7
8
9
100 � 0
29 � 10
66 � 9
53 � 43
53 � 18
35 � 12
12 � 1
10 � 6
9 � 2
100 � 0
27 � 6
63 � 26
3 � 3
47 � 12
40 � 31
10 � 2
3 � 2
7 � 3
10
11
12
13
Tetradecanal
Pentadecanal
Hexadecanal
35 � 20
5 � 1
14 � 8
9 � 4
19 � 13
trace
32 � 8
trace
13-Acetyloxytridecanal
aIdentification is based on comparison of mass spectra and retention times of the
natural materials with those of authentic synthetic standards.
b
Two- to three-day volatiles collections of groups of 5 males or females (N = 5).
c
Male: 74 ng (� 3 SD), Female: 37 ng (� 7 SD).
GC analyses of volatiles from groups of females and the simultaneously
recorded EAD activity of male and female antennae revealed 12 compounds
(Table 1), that consistently stimulated the antennae (Figure 4). The same 12
compounds were present in volatiles collected from groups of males, and sub-
sequent GC-EAD analyses confirmed their activity on male and female antennae
(Figure 5).
The GC profiles differed notably between the sexes. Groups of males pro-
duced relatively more of the EAD-active compounds than groups of females on
a per sawfly basis. The distinct odor released by grouped males was identified as
phenylacetic acid, which was clearly more abundant with volatiles from males than
from females. Ratios of aldehydes also differed. Male volatiles collections showed
a relatively higher abundance of tetradecanal, whereas female collections showed
higher amounts of hexadecanal. In addition. 9-acetyloxynonal was approximately
three times more abundant in male collections than in female collections.
SPME-Collected Volatiles. The profile of SPME-collected volatiles from the
first 30 min after placing five males together is shown in Figure 6. In males, the most
abundant compounds were nonanal and tetradecanal. In females (not shown), hex-
adecanal replaced tetradecanal as the abundant larger aldehyde. GC-EAD profiles

4
16
COSS E´ , BARTELT, WEAVER, AND ZILKOWSKI
FIG. 4. Simultaneously recorded gas chromatogram (GC) and electroantennogram re-
sponses (EAD) of typical male and female antenna to volatiles collection on Super Q
from a group of five female C. cinctus.
of SPME-collected volatiles confirmed the EAD-activity of the 12 compounds
collected with Super Q.
Ozonolysis of Cuticular Extracts. The extract shown in Figure 1 showed rela-
tively small amounts of EAD-active materials. However, this same extract showed
a dramatic increase in EAD-active compounds after ozonolysis. The presence of
these EAD-active compounds was confirmed by GC-MS and GC-EAD analysis
(Figure 7). The most abundant compound after ozonolysis was nonanoic acid. 13-
Acetyloxytridecanal, which was present in trace amounts in some of the volatiles
collections and was without detectable antennal activity, consistently stimulated
male and female antennae in ozonized extracts (peak 13, Figure 7).
Field Trial with 9-Acetyloxynonanal. The trap catch totals and means for the
various doses of 9-acetyloxynonanal are shown in Table 2. The sex ratio was nearly
1:1, with 179 females and 164 males being captured. No pattern could be discerned
between sex ratio and trap treatment; thus, the data were pooled over sex. The

WHEAT STEM SAWFLY PHEROMONE
417
FIG. 5. Simultaneously recorded gas chromatogram (GC) and electroantennogram re-
sponses (EAD) of typical male and female antenna to volatiles collection on Super Q
from a group of five male C. cinctus.
initial Poisson regression model applied to the trap catch data (315 observations)
included terms for differences among days of the experiment (14 df ), differences
among treatments (6 df), and the constant term (1 df ). The resulting likelihood
2
ratio statistic (G ) for the model was 409.6 (294 df, P < 0.0001, compared to
2
the � distribution). Thus the overall residual variance was larger than expected
for a Poisson process. Examination of residuals revealed that two of the catches
for the 1000-�g treatment were inconsistently large: 59 sawflies in one trap on
day 2 and 13 sawflies in one trap on day 5. When these two data points were
2
removed as outliers, the model fitted adequately (G = 303.9, 292 df, P = 0.30),
and conditional tests (Fienberg, 1977) for differences among treatments or days
were then justified. The overall test for differences among treatments was highly
2
significant (G = 147.2, 6 df, P < 0.00001). Differences among days were also
2
highly significant (G = 146.4, 14 df, P < 0.00001). Averaged over all treatments
and replications, mean daily trap catches for the 15 days of the study ranged from

4
18
COSS E´ , BARTELT, WEAVER, AND ZILKOWSKI
FIG. 6. Simultaneously recorded gas chromatogram (GC) and electroantennogram re-
sponses (EAD) of a typical male antenna to volatiles collected with solid phase microex-
traction (SPME) of the first 30 min when five male C. cinctus were grouped together.
0.11 to 1.23; it was necessary to allow for differences among days in the model.
Daily sawfly activity is greatly influenced by weather conditions (temperature,
wind, and cloud cover; D. K. Weaver, personal observation).
Differences among treatments were further explored. A linear relationship
was fitted for log catch versus log dose; a separate term (1 df ) was simultaneously
fitted for the control because the logarithm for dose = 0 would be meaning-
2
less. The linear trend was highly significant (G = 107.4, 1 df, P � 0.00001),
and the value of the slope corresponded to a 1.70-fold increase in trap catch for
each 10-fold increase in septum dose. Deviations from the linear trend were also
2
significant (G = 11.45, 4 df, P = 0.022), but these were relatively minor and
did not suggest that any simple curve would describe the data better than the
straight line. A series of 1-df comparisons were made between individual doses
and the control to define the response threshold. As shown in Table 2, all doses

WHEAT STEM SAWFLY PHEROMONE
419
FIG. 7. Simultaneously recorded gas chromatogram (GC) and electroantennogram re-
sponses (EAD) of a typical male antenna to a cuticular hexane extract of female C. cinctus
after ozonolysis.
of 9-acetyloxynonanal � 3 �g/septum were significantly more attractive than
the control. This data set demonstrated a consistent, although fairly subtle, field
activity of 9-acetyloxynonanal. Although the highest dose, 1000 �g, was clearly
the most active, activity became detectable at 3 �g.
TABLE 2. SUMMARY OF FIELD TRAPPING EXPERIMENT WITH
9
-ACETYLOXYNONANAL
Septum dose
�g)
Total catch
Mean
catch
G2 for difference from
control (P value, 1 df )
(
(N)
0
0
1
3
1
1
1
(control)
.3
12 (45)
20 (45)
15 (45)
33 (45)
39 (45)
43 (45)
106 (43)a
0.27
0.44
0.33
0.73
0.87
0.95
2.47
2.02 (0.16)
0.33 (0.57)
10.19 (0.0014)
15.05 (0.0001)
18.54 (0.00002)
94.18 (� 0.00001)
0
00
000
aTwo outlier data points were removed, which had values of 59 and
3 (see text).
1

4
20
COSS E´ , BARTELT, WEAVER, AND ZILKOWSKI
The conclusion from the analysis was that the effect of 9-acetyloxynonanal
was real, even after two of the highest trap catches had been removed as outliers.
An alternative analysis that would legitimately make use of all trap data (but that
would probably be more complex than Poisson regression) would be expected to
demonstrate the effect of the compound even more strongly.
DISCUSSION
Both males and females release the same EAD-active compounds. Although
there was no evidence for any sex-specific compounds, overall release of EAD-
active compounds was higher with grouped males than with grouped females. This
was especially noted for phenylacetic acid, which was instantly detectable by the
human nose when male sawflies were put together. In addition, male and female
volatiles collections differed in the ratios of 9-acetyloxynonanal, tetradecanal, and
hexadecanal. In the absence of qualitative differences, these different male and
female ratios might be important clues for sex recognition.
The EAD-active 9-acetyloxynonanal is part of the pheromone of C. cintus.
This compound alone attracted both male and female sawflies in the field bioas-
say. However, as is evident from our results, the complete pheromonal system of
C. cintus appears to be more complicated. Our volatiles collections showed that
males and females release 13 EAD-active compounds and, with the exception of
phenylacetic acid, all could be derived through oxidation of cuticular compounds
(Bartelt et al., 2002). Higher molecular weight compounds in male and female
cuticular extracts showed no GC-EAD activity and the relatively small amounts
of compounds of lower molecular weight showed relatively weak EAD responses.
However, extracts after ozonolysis showed dramatically increased amounts of
aldehydic GC-EAD-active compounds and absence of the unsaturated, higher-
molecular-weight materials, giving credibility to the notion of an oxidative origin
of the GC-EAD active compounds. In particular, unsaturated cuticular compounds
could function as the precursors for the EAD-active compounds by oxidative cleav-
age of their double bonds (Bartelt et al., 2002). One might expect the oxidation of
cuticular monoolefins to be a relatively slow process. However, EAD-active mate-
rials could be detected within minutes rather than hours using SPME, suggesting
that an oxidative system might be fast enough to be significant to the sawflies’
sexual communications under field conditions.
A multitude of factors can trigger and influence the abiotic oxidation of
unsaturated cuticular compounds, making this a complicated process to study
(Bartelt et al., 2002). For example, the action of ultraviolet light and the effect
of temperature, as well as premating behaviors such as wing fanning and rubbing
of the abdomen (D. K. Weaver, personal observations) could all influence this
oxidation and will need further attention. Preliminary experiments showed that

WHEAT STEM SAWFLY PHEROMONE
421
abioticoxidationofalkenedioldiacetates, presentonthecuticleoftheinsects, could
yield the biologically active acetyloxyaldehydes (Bartelt et al., 2002). However,
enzymatic involvement in C. cinctus pheromone system has not been ruled out,
especially with regard to the occurrence of phenylacetic acid.
All the pheromones of sawfly species have thus far been reported as being
female-produced. Forexample, sexattractioninthegeneraDiprionandNeodiprion
depends to a large extent on esters derived from various optical isomers of female-
produced 3,7-dimethyl-2-pentadecanol and from homologous alcohols with car-
bon chains of different length (Anderbant, 1999, and references therein). A differ-
ent pheromone system has been reported for the tenthredinid, Pikonema alaskensis
(Rohwer), still female-produced, but derived from the oxidation of sex-specific
straight-chain hydrocarbon dienes found on the cuticle of the insects (Bartelt and
Jones, 1983; Bartelt et. al., 1983). Although restricted to one sex, the oxidative
component in the pheromone system of P. alaskensis resembles that of C. cinctus.
Field observations have shown that males form in-flight aggregations and
that female sawflies near these active aggregations are rapidly recognized by male
C. cinctus, and synthetic 9-acetyloxynonanal released from rubber septa can induce
this aggregation (D. K. Weaver, unpublished). Furthermore, from the electrophys-
iological data, it is likely that more dramatic responses would result if additional
antennally active compounds were included in the trap baits. Finally, trap design
was a critical issue in this study. A number of trap types were tried (D. K. Weaver,
unpublished) but were unsuccessful either because the insects would not enter a
trap with interior shadows or because the trapping surface was so large and accessi-
ble that frequent accidental captures overwhelmed the treatment effects. While the
trap design employed here was suitable for the present study, future improvements
would be desirable.
9-Acetyloxynonanal has not been reported as an insect attractant before, but
this and other acetyloxyaldehydes were analytically useful as cleavage products
of epoxides in a method to locate the double-bond position in long-chain acetates
(Bierl-Leonhardt et al., 1980). The mass spectrum of 9-acetyloxynonanal reported
by Bierl-Leonhardt et al. (1980) did not agree with ours. However, this discrepancy
might be due to a typographical error in the above-mentioned paper, as the pub-
lished spectrum was also inconsistent with the spectra of homologous aldehyde
acetates in that report.
The behavioral role of phenylacetic acid has yet to be determined. Its presence
was noted in volatiles collections from both sexes, and its odor was instantly
detected by the human nose when males were grouped together. Phenylacetic acid
has not been reported as an insect attractant. Laboratory bioassays have shown
that female C. cinctus were attracted to volatiles emitted from groups of males,
but not to volatiles from single males (D. K. Weaver, unpublished). Whether this
attraction to groups of males is due to the release of phenylacetic acid or due to a
threshold effect is not clear and will need further study.

4
22
COSS E´ , BARTELT, WEAVER, AND ZILKOWSKI
Individually, none of the identified EAD-active compounds would account
for recognition of females by males. Therefore, future research will focus on more
behaviorally driven assays of collections in order to collect a possible distinct
female-specific chemical signal. In addition, emphasis will be given to elucidate the
behavioral significance of additional EAD-active compounds and the importance
of different ratios of EAD-active volatiles between the sexes.
Acknowledgments—We thank Ronald D. Plattner of NCAUR, Peoria, Illinois, for assistance in
obtaining CI mass spectra.
REFERENCES
AINSLIE, C. N. 1920. The western grass stem sawfly. United States Department of Agriculture Bulletin
8
41. US Government Printing Office, Washington, D.C.
ANALYTICAL SOFTWARE. 1998. Statistix for Windows User’s Manual. Analytical Software,
Tallahassee, Florida.
ANDERBRANT, O. 1999. Sawflies and seed wasps. In J. Hardie and A. K. Minks (eds.). Pheromones of
Non-Lepidopteran Insects Associated with Agricultural Plants. CABI Publishing, Wallingford,
United Kingdom.
BARTELT, R. J. and JONES, R. L. 1983. (Z)-10-Nonadecenal: A pheromonally active air oxidation
product of (Z,Z)-9-19 dienes in yellowheaded spruce sawfly. J. Chem. Ecol. 9:1333–1341.
BARTELT, R. J. and ZILKOWSKI, B. W. 1998. A versatile and quantitative volatile-delivery system for
laboratory bioassays. J Chem. Ecol. 24:535–558.
BARTELT, R. J., JONES, R. L., and KRICK, T. P. 1983. (Z)-5-Tetradecen-1-ol: A secondary pheromone
of the yellowheaded spruce sawfly, and its relationship to (Z)-10-nonadecenal. J. Chem. Ecol.
9
:1343–1352.
BARTELT, R. J, COSSE´, A. A., PETROSKI, R. J., and WEAVER, D. K. 2002. Cuticular hydrocarbons
and novel alkenediol diacetates from wheat stem sawfly (Cephus cinctus): Natural oxidation to
pheromone components. J. Chem. Ecol. 28:385–405.
BEROZA, M. and BIERL, B. A. 1967. Rapid determination of olefin positions in organic compounds in
microgram range by ozonolysis and gas chromatography. Anal. Chem. 39:1131–1135.
BIERL-LEONHARDT, B. A., DEVILBISS. E. D., and PLIMMER, J. R. 1980. Location of double-bond
position in long-chain aldehydes and acetates by mass spectral analysis of epoxide derivatives.
J. Chromatogr. Sci. 18:364–367.
BLODGETT, S. L., JOHNSON, G. D., LANIER, W. T., GOOSEY, H. B., LENSSEN, A. W., WATERS, D. K.,
TALBERT, L. E., and BRUCKNER, P. L. 1996. Special issue: Wheat stem sawfly. Mont. Crop Health
Rep. 9(11):1–7.
BLODGETT, S. L., JOHNSON, G. D., LANIER, W. T., MORRILL, W. L., WEAVER, D. K., GOOSEY, H. B.,
LENSSEN, A. W., SING, S. E., WATERS, D. K., BRUCKNER, P. L., TALBERT, L. E., LANNING,
S., BERG, J., BRASTRUP, R., CARLSON, G., CROUCH, W., ECKHOFF, J., GRAHAM, P., KIRBY, R.,
KUSHNAK, G. D., MATTA, J., RANNEY, J., ROOS, R., WARGO, J., and WICHMAN, D. 1997. Special
edition: Wheat stem sawfly. Mont. Crop Health Rep. 9(11):1–8.
BORDEN, J. H., BILLANY, D. J., BRADSHAW, J. W. S., EDWARDS, M., BAKER, R., and EVANS, D. A.
1
978. Pheromone response and sexual behavior of Cephalcia lariciphila Wachtl. (Hymenoptera:
Pamphiliidae). Ecol. Entomol. 3:13–24.
COSSE´, A. A. andBARTELT, R. J. 2000. Male-producedaggregationpheromoneofColopterustruncatus:
Structure, electrophysiological and behavioral activity. J. Chem. Ecol. 26:1735–1748.

WHEAT STEM SAWFLY PHEROMONE
423
FIENBERG, S. E. 1977. The Analysis of Cross-Classified Categorical Data. MIT Press, Cambridge,
Massachusetts.
HOLMES, N. D. 1982. Population dynamics of the wheat stem sawfly, Cephus cinctus (Hymenoptera:
Cephidae) in wheat. Can. Entomol. 114:775–788.
MORRILL, W. L., GABOR, J. W., and KUSHNAK, G. D. 1992. Wheat stem sawfly (Hymenoptera:
Cephidae) damage and detection. J. Econ. Entomol. 85:2413–2417.
US DEPARTMENT OF AGRICULTURE. 2000. World-Wide-Web page at http://www.sidney.ars.usda.gov/
scientists/tshanower/wss/sawfly.html., composed by T. G. Shanower, USDA-ARS, Northern
Plains Agricultural Research Laboratory, Sidney, Montana.
WEISS, M. J. and MORRILL, W. L. 1992. Wheat stem sawfly (Hymenoptera: Cephidae) revisited. Am.
Entomol. 38:241–245.
WILEY. 1995. Spectral library, 6th ed. (CD-ROM). John Wiley & Sons, New York.