See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/323979736
Behavioral Repertory of Adult Cylas formicarius (Fabricius)
(Coleoptera: Brentidae)
Article  in  The Coleopterists Bulletin · March 2018
DOI: 10.1649/0010-065X-72.1.85
CITATIONS READS
0 38
3 authors:
Christopher Starr David Wilson
University of the West Indies, St. Augustine University of Ghana
69 PUBLICATIONS   490 CITATIONS    30 PUBLICATIONS   164 CITATIONS   
SEE PROFILE SEE PROFILE
Stanley Kays
University of Georgia
261 PUBLICATIONS   4,885 CITATIONS   
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
none in particular View project
Radiation preservation of foods and agricultural commodities View project
All content following this page was uploaded by Christopher Starr on 07 June 2019.
The user has requested enhancement of the downloaded file.
Behavioral Repertory of Adult Cylas formicarius (Fabricius)
(Coleoptera: Brentidae)
Author(s): Christopher K. Starr, David D. Wilson and Stanley J. Kays
Source: The Coleopterists Bulletin, 72(1):85-92.
Published By: The Coleopterists Society
https://doi.org/10.1649/0010-065X-72.1.85
URL: http://www.bioone.org/doi/full/10.1649/0010-065X-72.1.85
BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the
biological, ecological, and environmental sciences. BioOne provides a sustainable online
platform for over 170 journals and books published by nonprofit societies, associations,
museums, institutions, and presses.
Your use of this PDF, the BioOne Web site, and all posted and associated content
indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/
terms_of_use.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial
use. Commercial inquiries or rights and permissions requests should be directed to the
individual publisher as copyright holder.
BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers,
academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.
The Coleopterists Bulletin, 72(1): 85–92. 2018.
BEHAVIORAL REPERTORY OF ADULT CYLAS FORMICARIUS (FABRICIUS)
(COLEOPTERA: BRENTIDAE)
CHRISTOPHER K. STARR
c/o Department of Life Sciences
St Augustine, TRINIDAD & TOBAGO
ckstarr@gmail.com
DAVID D. WILSON
Department of Zoology, University of Ghana
Legon, GHANA
AND
STANLEY J. KAYS
435 Hampton Court
Athens, GA 30605, USA
ABSTRACT
Adult female andmale sweetpotato weevils,Cylas formicarius (Fabricius), have repertories of at least 19 and 21 behavior
patterns, respectively. These are described and named, including eight self-grooming patterns. Consistent with the species’
slight sexual dimorphism, few behavior patterns are peculiar to one sex. The sequence of self-grooming movements is
analyzed for each sex. Comparative assays of vagility and overall activity do not show any difference between the sexes.
Key Words: Curculionoidea, sweetpotato weevil, sexual behavior, sexual diethism, thanatosis
DOI.org/10.1649/0010-065X-72.1.85
Cylas Latreille is a paleotropical genus of about Valentine (1973) reviewed grooming behavior in
25 known species that form a distinctive group beetles, identifying 17 behavior patterns. Our pur-
within the superfamily of weevils (Coleoptera: pose here is to provide a behavioral catalog of adult
Curculionoidea) and are currently placed in the C. formicarius, with a discussion of sexual diethism.
family Brentidae (Wolfe 1991; Thompson 1992).
The sweetpotato weevil, Cylas formicarius (Fab- MATERIAL AND METHODS
ricius), feeds and breeds on Ipomoea spp. and other
Convolvulaceae (Austin 1991). Worldwide, it is the All weevils were of the “elegantulus” color form,
foremost pest of sweetpotato, Ipomoea batatas (L.). drawn from a laboratory culture at the University of
Because of its economic importance, C. formicarius Georgia, USA. As a safeguard against any genetic
has been the subject of much research, including adaptation to laboratory conditions, wild-caught
several basic bionomic studies (Sutherland 1986 insects from southern Florida were added periodi-
and references therein). Some of these (e.g., cally to the culture. Voucher specimens are de-
Gonzales 1925) include occasional comment on posited in the University of Georgia’s Museum of
behavior. Natural History (UGCA) and the personal collection
Adult C. formicarius are slender, smooth, and of G. William Wolfe.
hard-bodied (Fig. 1). Sexual dimorphism is slight Except where noted, all observations were
and appears limited to the antennae, eyes, and wings made under red light to simulate night. We ob-
(Starr et al. 1997). The antennae are similar in served females and males in a variety of situa-
structure, but much longer in males. Males likewise tions. In each of these, we began with a period of
have substantially larger eyes. Differences in wing qualitative observation in order to form a pre-
size are slight, though statistically significant. Both liminary catalog. This was followed by at least one
sexes fly and appear to be more active at night, with longer period, in which we counted occurrences
males reportedly more vagile than females (Christian (acts) of different behavior patterns and refined
1938; Howard 1982; Proshold 1983). Aside from our descriptions of these. The following are
this, no secondary sexual behavioral differences have the situations employed, in which “segregated”
been noted, as far as we know. means that each chamber contained only same-sex
85
86 THE COLEOPTERISTS BULLETIN 72(1), 2018
served to refine our description of these behavior
patterns and our estimations of their relative frequency.
Because we watched several individuals at once, we
were not always sure that the first grooming motion
notedwas in fact the first of the bout.Where grooming
was apparently disturbed by another individual, the
bout was disregarded in computing bout lengths and
terminal patterns.
To compare the overall activity level between
sexes, we performed two replicates of the following
assay. Five females and five males were isolated in
standard plastic Petri dishes (inner diameter 94 mm
and height 13 mm) and allowed to equilibrate for
30 minutes. At 5-minute intervals in the succeeding
60 minutes, each individual was scored as either
immobile (0) or active (i.e., in motion, 1), resulting
in a cumulative score between zero and 12 for each
of 10 females and 10 males.
To compare vagility between sexes, we per-
formed eight replicates of the following assay. Five
females and five males were segregated in standard
plastic Petri dishes and allowed to equilibrate for
Fig. 1. Male Cylas formicarius in the rearing-up several minutes, after which the lids were removed
posture. Body length 5 6 mm, including proboscis. and the sequence in which individuals first touched
Photograph by Melvyn Yeo. the outside substrate after climbing from the dish
was recorded for five minutes. Effectively, this
amounted to a series of eight races, each with a
individuals, while “integrated” means that each different group of females and males, counting only
chamber contained individuals of both sexes. A “bare” the sequence in which sexed individuals finished
chamber is one with a filter-paper floor and a drop of the race.
water. Numbers in brackets are summed recorded acts Hypothesis testing was performed using Chi-
by females and males, respectively, in the given squared or Mann–Whitney tests where appropri-
treatment, for a total of 1,984 female acts and 2,168 ate. Alpha was set at 0.05.
male acts.
1. Segregated in a bare chamber (510, 519). RESULTS
2. Segregated with sweetpotato after fasting one
day (491, 533). I. Behavioral Catalog. For convenience, be-
3. Integrated with sweetpotato after fasting one havior patterns are grouped according to supposed
day (501, 585). function. Numbers in brackets after each behavior
4. Integrated with potted sweetpotato plant under name indicate the summed number of acts observed
dim white light (245, 274). in females and males, respectively, in observation
5. Segregated in olfactometer chamber with periods that did not focus on a particular class of
sweetpotato leaf odor flow (237, 257). behavior; they do not include data from Table 1. For
example, we recorded 872 acts of walking by fe-
Outside of these five situations, we supplemented males and 845 by males. Where a particular pattern
the qualitative-descriptive treatment of some un- commonly occurs in rhythmic repetition (e.g., steps
common patterns, with emphasis on sexual behavior in walking or the two types of palpation), we treat
and oviposition. an uninterrupted bout as a single act. We charac-
Our observations were in an area in which terize a posture as “sustained” if it typically lasts
C. formicarius is not found in the wild. The necessary more than one second, while a “brief” posture is
quarantine conditions prevented any significant ob- more than momentary but typically no more than a
servations or assays of flight or flight-related behavior. second. Except where indicated, each pattern is
After noting presumed self-grooming patterns in found in apparently identical form in both sexes.
the course of compiling the behavioral catalog, we
observed segregated groups of 2-3-week-old weevils MOBILITY/SENSATION
in bare,white-lit chamberswith attention only to bouts Walk (872, 845): Sweetpotato weevils walk in a
of grooming, i.e., sequences of grooming acts unin- form of the double-tripod gait characteristic of
terrupted by non-grooming behavior. This treatment many insects (Gillott 1980). We observed no other
THE COLEOPTERISTS BULLETIN 72(1), 2018 87
Table 1. Recorded numbers of self-grooming acts/ the mandibles are brought into play. We observed
patterns in female and male Cylas formicarius. Further biting most commonly with respect to sweet-
explanation in text. Rub substrate was observed outside of potato roots, but also stems. In bare-chamber situ-
a recording period. ations, the weevils occasionally bit the underlying
filter paper.
Behavioral pattern Females Males Total Bore (chew) (75, 67): The insect assumes a
Rub antenna 29 20 49 sustained posture similar to biting, with the snout
Rub head 105 33 138 penetrating successively deeper into the substrate.
Rub elytron (leg 2) 50 16 66 In the transition from biting to chewing boring, the
Rub elytron (leg 3) 9 1 10 antennae are lifted off the substrate and gradually
Rub legs 1–2 144 61 205 directed posteriorly, sometimes above the head, not
Rub legs 2–3 58 21 79
Rub legs 3 19 3 22 tucked close below the body as in dropping. The
Rub substrate (snout) 0 0 0 orientation of the snout to the substrate varies from
approximately perpendicular to acutely forward at
Total 414 155 569
about 30°. In our observations, the substrate was
always a sweetpotato root or stem. Boring may
result in a hole little broader than the snout, in which
gait. For purposes of counting acts, we treated an case penetration is at most to the level of the eyes.
individual as walking only while it was not pal- However, where it produces a wider hole, the weevil
pating (see below) for at least one second. may insert its entire head and prothorax.
Drop (0, 1): The weevil falls on its side and re- Bore (push) (15, 5): Theweevil assumes a posture
mains motionless, with the legs moderately flexed similar to chewing boring, with the obvious dif-
and the antennae tucked in close to the body, di- ferences that the substrate is soil and the weevil
rected posteriorly. We noted it only once in our enters completely into it. This pattern has something
observation periods, but it is a common response by more of a pushing-shaking aspect, but our main
both sexes to physical disturbance by the experi- reason for distinguishing it is that we assume there is
menter and thus easy to induce. no use of the mouthparts in boring into soil.
Self-right (0, 1): This is the pattern of rising out of Rest (94, 28): The body lies immobile against the
the dropped posture. The following description is substrate, with the legs moderately drawn in and the
based on responses of 31 females and 31 males that antennae extended posteriorly, although not tucked
were induced to drop. Typically, the weevil starts by in close under the body. This is not simply a ces-
stretching and waving the antennae and then the sation of walking. Individuals sometimes enter the
mid- and hindlegs on the upper side of the body. at-rest posture upon contact with another.
On a flat surface, this is without effect, as neither leg Rear up (59, 110): The insect stands with the body
reaches the substrate, nor does it roll the body into a strongly lifted away from the substrate anteriorly, so
new position. The underlying foreleg is then swept that the head is raised at an angle of roughly 45°,
around past the head and beneath the body to grasp usually with the antennae waving (Fig. 1). All feet
the substrate with the tarsus. That foreleg then remain on the substrate, although the forelegs may
flexes to begin rolling the body up onto its venter, be fully extended. The posture may be held for
as the other legs follow in a similar stretching- several seconds. Males sometimes assume this
grasping-flexing motion. The overlying legs are posture upon being mounted by another male.
then used to grasp the substrate as the weevil rises
to a standing position. From the sweeping of the SELF-GROOMING
underlying foreleg, this is a fluid and apparently Square-bracketed numbers identify Valentine’s
stereotyped set of motions that quickly rights the (1973) description of the behavior pattern.
insect. Rub antenna (56, 67) [6]: The head is lowered to
Palpate (antennae) (396, 379): The weevil, either rest at least the tip of the terminal antennal club
in motion or stationary, taps the substrate or another against the substrate. The corresponding protarsus is
individual with the terminal club of one or (usually) then rubbed distally over the club and sometimes the
both antennae. entire antenna. Males show a greater tendency than
Palpate (snout) (190, 116): The weevil, either in females to lay most or all of the antenna against the
motion or stationary, taps the substrate with the tip substrate.
of the snout. We could not see which mouthparts Rub head (42, 22) [7]: A foreleg is raised and
make contact. rubbed anteriorly over the head or the prothorax and
Bite (31, 8): The tip of the snout is pressed against head, after which the tarsus is commonly drawn
the substrate and held there briefly, with the ter- through the mouthparts in a continuous motion.
minal club of the antennae usually flat against the Depending on whether the leg passes a) above and/
substrate. We assumed, but could not be certain, that or to the side or b) below the head, there appear to be
88 THE COLEOPTERISTS BULLETIN 72(1), 2018
two distinct patterns, which we can characterize as the other. In some cases, the weevil dismounts
rub head above and rub head below. However, these without first probing. In our observations, he rarely
movements are executed quickly and are often climbs down over the other’s head and never backs
difficult to distinguish.We treat them together for all down. Although we observed no behavior that we
quantitive purposes. interpret as an attempt to scrape away a mounting
Rub elytron (leg2) (16, 13) [8]: One or both male, males sometimes are more or less shaken off
midlegs are raised and rubbed along the corre- and fall without an orderly dismount. We do not
sponding elytron posteriorly. count these incidents as dismounts.
Rub elytron (leg3) (3, 5) [9]: With the hind end of
the body tilted upwards, one or both hindlegs are OVIPOSITION
raised and rubbed along the corresponding elytron Probe (abdomen tip) (7, 0): While stationary or
posteriorly. walking slowly, the weevil repeatedly touches the
Rub legs 1–2 (74, 94) [14]:With the body tilted to tip of the abdomen to the substrate. This is pre-
one side, the fore- and midlegs of the other side are sumably exclusively a female pattern.
raised and rubbed across each other distally in Lay egg (10, -): With the tip of her abdomen
alternation. over a bored hole in the substrate, the weevil tilts
Rub legs 2–3 (30, 54) [15]:With the body tilted to posteriorly to apply the tip of her abdomen to the
one side, the mid- and hindlegs of the other side are hole, usually for a sustained period. This is a dis-
raised and rubbed across each other distally in tinct, humped posture, unlike that of resting. She
alternation. may lean away from the hole and back toward it
Rub legs 3 (14, 20) [16]: The weevil stands on its several times, as if adjusting her position. The
fore- and midlegs, with the posterior end of the body position of the antennae is variable. Although the
somewhat elevated, and rubs the hindlegs across actual deposition of an egg is hidden from view, in
each other distally in alternation as they extend all cases where this posture was sustained for many
posteriorly. This sometimes follows fluidly from seconds we subsequently found an egg in the hole.
rubbing elytra with the hindlegs. We also observed apparently incomplete acts of this
Rub substrate (snout) (0, 5): The head is lowered pattern, in which the weevil briefly moved back into
and the snout is rubbed against the substrate, as if in the laying posture and then withdrew and simply
an attempt to scrape away a sticky substance. walked away. This is necessarily exclusively a fe-
male pattern.
SEXUAL Lay-inspect (0, -): The weevil withdraws her
Mount (0, 130): The weevil climbs onto another abdomen from the oviposition hole, turns around,
from the rear and grasps her/him on the sides with and palpates the hole with her antennae for some
his legs. He palpates the other’s head but not her/his seconds, often or always applying her snout to the
antennae with his antennae. Initially, the fore- and/ hole (Starr et al. 2016). This is necessarily
ormidlegs often tap or rub the other’s corresponding exclusively a female pattern.We did not observe this
thoracic terga. A mount together with aedeagal and the next behavior pattern while enumerating
probing (see next) can last several minutes. acts but during other times.
In our interpretation, this is exclusively a male Apply oviposition plug (0, -): With her mouth-
behavior pattern, as are mount-inspecting and dis- parts, the weevil puts macerated plant tissue in the
mounting. On occasion, we observed females hole above the egg, forming a plug (Starr et al. 2016).
climbing onto the backs of other individuals, but with
no distinct front-to-front orientation or palpation of II. Self-grooming Sequences. In only 21 of 167
the head, apparently treating the other simply as part recorded uninterrupted grooming bouts of females
of the substrate. We treat this and other un- were we certain of having observed the first act.
differentiated climbing motions as forms of walking. However, the length-frequency distribution (in
Probe (aedeagus) (-, 104): A mounting weevil bouts) of these 21 is sufficiently similar to that of
slides back to bring the tip of his abdomen close to the remaining 146 to suggest that most observed
the tip of the other’s abdomen and extends his bouts are complete in both females and males.
aedeagus to probe the other’s anal-genital region. Grooming bouts are usually very short (no more
This is necessarily exclusively a male pattern. than a few seconds) (Fig. 2), with no significant
Mount-inspect (0, 2): The mounting weevil ro- difference between the sexes in mean length (Mann-
tates his body to face the other’s posterior end and Whitney Z 5 1.73, P 5 0.08).
palpates the other’s anal-genital region with his The frequencies of patterns observed in periods
antennae. This may be followed by a re-mount and devoted to observations of self-grooming are shown
renewed probing. in Table 1. No significant sexual difference is shown
Dismount (0, 92): The mounting weevil rotates in the relative frequency of the different behavior
his body to one side and walks to the side or rear of patterns (X2 5 9.9, P . 0.10).
THE COLEOPTERISTS BULLETIN 72(1), 2018 89
Fig. 2. Length-frequency distribution of self-grooming bouts in Cylas formicarius. Black bars: 167 bouts in
females. Grey bars: 60 bouts in males. Further explanation in text.
The observed and expected frequencies of within- 1. \\__\_[\\__]
bout transitions between different grooming pat- 2. \__\\\_[\__]
terns are shown in Tables 2–3. There is a highly 3. \__\\_\\[__]
significant departure from randomness in both fe- 4. \____\\\[\_]
males (X2 . 210, P , 0.001) and males (X2 . 79, 5. __\_\\\\[__]
P , 0.001). 6. \____\\[\\_]
Table 4 shows the observed and expected fre- 7. \\\___\\[__]
quencies with which different patterns appear at the 8. \___\\_[\\_]
end of grooming bouts. Here, too, there is a sig-
nificant departure from randomness in both females Treating non-finishers as tied within a trial and
(X2 5 17.6, P , 0.01) and males (X2 5 14.9, summing the ranks of same-sex individuals (e.g.,
P , 0.05). in the first trial females have a sum of 25 and
males a sum of 30), no significant sexual differ-
III. Comparative Overall Activity and Vagility. ence is found between average rank sums (X
2 5
The activity assay produced the following scores in 3.6, 7 df, P . 0.10). Under these conditions, the
ascending order for each sex: minimum and maximum possible sums are 15 and
Females 0, 0, 0, 1, 8, 8, 9, 9, 11, and 12, 40, respectively. The largest difference found (trial
Males 0, 0, 0, 0, 0, 1, 3, 7, 8, and 11. no. 7) was between 21 for females and 34 for
Although females appear on average more active males.
than males, a significant difference is not shown
(Kruskal-Wallis rank sum test, P . 0.10). DISCUSSION
The following is the sequence in which sexed
individuals finished each vagility trial, with those Under the study conditions, the numbers of ob-
that remained in the dish at the end of the 5-minute served acts per behavior pattern do not provide a
test in square brackets: strong index of relative prominence in the life of the
90 THE COLEOPTERISTS BULLETIN 72(1), 2018
Table 2. Observed and expected (from Table 1) transitions between behavior patterns within 181 self-grooming bouts
of femaleCylas formicarius. In each cell, the expected value is below the observed value.A5 rub antenna;H5 rub head;
E(2)5 rub elytron (leg 2); E(3)5 rub elytron (leg 3); L125 rub legs 1–2; L235 rub legs 2–3; L35 rub legs 3. Because
repetitive motions of the same pattern are treated as a single act, no pattern follows itself. Asterisks indicate transitions
whose expected value is much higher than the observed value. Further explanation in text.
Following Preceding A H E(2) E(3) L12 L23 L3
A 12* 1 0 4 0 0
3.2 2.3 0.4 6.5 3.7 0.9
H 2 0 0 61* 1 0
4.6 10.0 1.5 28.1 15.8 3.9
E(2) 0 0 2 1 19* 2
1.6 4.9 0.6 10.0 5.6 1.4
E(3) 1 1 0 0 2 5*
0.6 1.6 1.2 3.3 1.9 0.5
L12 5 22 14 0 17 0
5.4 16.2 11.7 1.8 18.4 0
L23 3 1 10* 2 5 3
1.8 5.4 3.9 0.6 10.9 1.5
L3 1 0 1 0 2 2
0.4 1.1 0.8 0.1 2.2 1.3
sweetpotato weevil. As discussed below, they are frequent under natural conditions. As expected,
more useful as a guide to sexual diethism. None- both sexes walked, palpated features of the sub-
theless, the data provide some fair hypotheses of strate, and self-groomed a great deal in all laboratory
which patterns are most frequent and which are least situations. It is likewise not surprising that sexual
Table 3. Observed and expected (from Table 1) transitions between behavior patterns within 66 self-grooming bouts
of male Cylas formicarius. In each cell, the expected value is below the observed value. A5 rub antenna;H5 rub head;
E(2)5 rub elytron (leg 2); E(3)5 rub elytron (leg 3); L125 rub legs 1–2; L235 rub legs 2–3; L35 rub legs 3. The rare
“rub substrate (snout)” is omitted. Because repetitive motions of the same pattern are treated as a single act, no pattern
follows itself. Asterisks indicate transitions whose expected value ismuch higher than the observed value. Further explanation
in text.
Following Preceding A H E(2) E(3) L12 L23 L3
A 9* 0 0 6 0 0
3.5 2.9 0 4.8 3.3 0.4
H 1 0 0 16* 1 0
1.2 4.3 0 7.0 4.9 0.9
E(2) 0 0 0 0 9* 1
0.7 2.7 0 3.7 2.6 0.3
E(3) 0 0 0 0 0 1
0.1 0.2 0.2 0.3 0.2 0.0
L12 3 6 4 0 4 0
1.3 5.4 4.4 0 5.0 0.6
L23 0 2 9* 0 1 0
0.8 3.4 2.8 0 4.6 0.4
L3 0 0 1 0 0 2
0.2 0.7 0.6 0 0.9 0.7
THE COLEOPTERISTS BULLETIN 72(1), 2018 91
Table4. Observedandexpected (fromTable1) frequencies Although we observed only males rubbing the
of different behavior patterns as the last pattern in self- snout against the substrate, we assume that it is
grooming bouts of adult Cylas formicarius. The rare “rub also part of the female repertory and identical in
substrate (snout)” is omitted. In each cell, the expected form.
value is under the observed value. Further explanation in It is noteworthy that among putative self-
text.
grooming patterns we never observed the rubbing
Behavior pattern Females Males together of the two forelegs or the twomidlegs, each
of which is prominent in wasps of the genus Polistes
Rub antenna 12 3 Latreille (Hymenoptera: Vespidae) (C. Starr, un-
12.2 7.7 published data).
The specifically sexual behavior patterns de-
Rub head 23 7
44.1 12.8 scribed were found with certainty only in males. It is
not surprising, for example, that males were often
Rub elytron (leg 2) 17 5 seen mounting and probing both females and other
21.0 6.2 males, while no female was seen to do either. If
females have patterns specifically associated with
Rub elytron (leg 3) 0 0 courtship or receptivity, these were too uncommon
3.8 0.4 or subtle to claim or attention.
With the possible exception of rearing up, we
Rub legs 1–2 74 37 observed no behavior that seemed specifically di-
60.5 23.6
rected toward dislodging a mounting male. Indeed,
Rub legs 2–3 30 8 even males being mounted appeared to ignore the
24.4 8.1 other male, and we have often seen such pairs of
males moving about for extended periods. There is
Rub legs 3 11 0 thus no evident way to detect a mounted female’s
8.0 1.2 state of receptivity to the male.
Total 167 60 Palpating another individual’s antennae with
one’s own antennae would seem to be an excep-
tionally sure way to sex her/him. It is thus note-
and reproductive behavior patterns were highly worthy that mountingmales were often observed to
context-restricted. palpate the other’s head but not her/his antennae.
Given its context and form, dropping is clearly a We hypothesize that mating attempts most com-
form of thanatosis. The narrow, deep abdomen and monly occur when the other insect is boring-
its smooth, relatively high dorsal curve are evidently chewing. In this situation, she/he will commonly
responsible for the fact that a dropped insect on a be conveniently immobile for an extended time,
level surface lies on its side, rather than on its back. but with the antennae relatively inaccessible for
This in turn evidently accounts for the peculiar form inspection. We presume that a male can sex the
of self-righting, unlike any of those described by other individual by palpating the eyes. Consistent
Chao (1985). with this hypothesis is the observation that males
Bore-pushing seems almost certainly to function sometimes make orderly dismounts without either
in reaching roots or buried stems of food plants, aedeagal probing or antennal palpation of the
presumably in response to odor. However, there is other’s anal-genital area.
no indication that it is a sophisticated or specialized As noted above, no significant difference is
response. Cylas formicarius does not seem espe- shown between the sexes in the frequencies of
cially well suited to burrowing in even loose soil. different grooming patterns. If any such difference
Furthermore, among its known host plants, I. exists, it is likely based on a higher rate of antennal
batatas— with which it has been in contact only in rubbing in males, which accounts for about half of
recent historical times — is unusual in having a the total Chi-squared value for Table 1. This ex-
large, attractive storage root (Austin 1991). pectation is consistent with the greater length of the
We interpret rearing up as primarily a way to antennae in males.
more effectively sample airborne odors, with the We expected to find a consistent pattern in the
possible secondary function of facilitating the re- sequence of groomingmotions, indicative of a very
moval of unwelcome mounting males. Although we restricted set of pathways by which dirt is removed
observed rearing up only by males in this context, from the body. However, we do not detect any such
we hypothesize that unreceptive females behave in clear, unitary pattern in the data on transitions
the same way. within grooming bouts (Tables 2–4). In females,
Rubbing the head presumably functions mainly three consistent partial pathways are apparent
in cleaning the eyes. (Table 2):
92 THE COLEOPTERISTS BULLETIN 72(1), 2018
1. rub antenna → rub head → rub legs 1–2 Chao, J. T. 1985. Comparative righting behavior of in-
2. rub elytron (leg 2) ↔ rub legs 2–3 sects. Bulletin of the Society of Entomology,
3. rub elytron (leg 3) → rub legs 3. National Chung Hsing University (Taichung) 18:
31–36.
The first two are also prominent in males, but the Christian, M. B. 1938. A biological study of the sweet-
data are too few to remark on the third transition potato weevil. MSc thesis, Louisiana State Uni-
(Table 3). Taking these together with data on ter- versity, Baton Rouge, LA.
minal patterns (Table 4), we tentatively suggest that Dawkins, R., and M. Dawkins. 1976. Hierarchical or-
ganization and postural facilitation: rules for
each of these three “partial” pathways is, in fact, grooming in flies. Animal Behaviour 24: 739–755.
complete in itself, so that a given bout serves to Gillott, C. 1980. Entomology. Plenum, New York, NY.
remove dirt either a) from the anterior end of the Gonzales, S. S. 1925. The sweet potato weevil (Cylas
body via the fore- and midlegs, b) from the posterior formicarius, Fabr.). Philippine Agriculturalist 14:
end via the mid- and hindlegs, or c) from the 257–281.
posterior end via the hindlegs. Even if this hy- Howard, F.W. 1982.Diurnal rhythm inCylas formicarius
pothesis is correct, the overall pattern in C. for- elegantulus and some other arthropods in a sweet
micarius is strikingly different from that found potato field. Florida Entomologist 65: 194–195.
in the fly Calliphora vicina Robineau-Desvoidy Proshold, F. I. 1983. Mating activity and movement of
Cylas formicarius elegantulus (Coleoptera: Cur-
(Dawkins and Dawkins 1976) and three species culionidae) on sweet potato. Proceedings of the
of social wasps (C. K. Starr, unpublished data), in American Society of Horticultural Science Trop-
which distinct sets of motions are associated with ical Section 27(B): 81–92.
the anterior and posterior ends of the body, with Starr, C. K., D. D. Wilson, R. F. Severson, and
little within-bout association of the two. S. J. Kays. 1997. Sexual dimorphism in the sweet-
The failure to find sexual differences in overall potato weevil Cylas formicarius (Coleoptera:
activity or vagility must be regarded as a limited Curculionoidea). The Canadian Entomologist 129:
finding, as flight is largely excluded from our as- 61–69.
says. It is, therefore, not in direct contradiction to Starr, C. K., D. D. Wilson, and S. J. Kays. 2016.
Composition of the oviposition plug of Cylas
earlier findings of greater activity/vagility in males. formicarius (Coleoptera: Curculionidae) in host
sweetpotatoes. Journal of Entomological Science
ACKNOWLEDGMENTS 51: 250–251.
Sutherland, J. A. 1986. A review of the biology and
R. K. Jansson provided field-caught specimens control of the sweetpotato weevil Cylas for-
from Florida. Thanks also to G. W. Wolfe for micarius (Fabr.). Tropical Pest Management 32:
confirming the species identification, B. Lauckner 304–315.
for statistical advice, R. Jander, U. Jander, andM. A. Thompson, R. T. 1992. Observations on the morphology
Mullen for criticism of an earlier version of this and classification of weevils (Coleoptera, Curcu-
lionoidea) with a key to major groups. Journal of
paper, and J. Spiers for help with the figures. Fi- Natural History 26: 835–891.
nancial support came from US-AID (Project no. 88- Valentine, B. D. 1973. Grooming behavior in Coleoptera.
CSRS-2-3236, “Facilitating the Development of The Coleopterists Bulletin 27: 63–73.
Resistance to the Sweet Potato Weevil”). Wolfe, G. W. 1991. The origin and dispersal of the pest
species of Cylas, with a key to the pest species
of the world [pp. 13–43]. In: Sweet Potato Pest
REFERENCES CITED Management: A Global Perspective (R. K.
Austin, D. F. 1991.Associations between the plant family Jansson and K. V. Raman, editors). Westview,
Convolvulaceae andCylasweevils [pp. 45–57]. In: Boulder, CO.
Sweet Potato Pest Management: A Global Per-
spective (R. K. Jansson and K. V. Raman, editors). (Received 16 May 2017; accepted 22 November 2017.
Westview, Boulder, CO. Publication date 23 March 2018.)
View publication stats