2010 SOUTHEASTERN NATURALIST 9(3):427–434
Sex-specific Attraction of Moth Species to Ultraviolet Light Traps
Heath W. Garris1,* and John A. Snyder2
Abstract - Phototactic behavior toward ultraviolet light varies among nocturnal flying
insects. By recording sex ratios of 28 southeastern US moth species, we tested
the commonly held belief that UV light-trap collections of moths are significantly
skewed toward males. Twelve species demonstrated a statistically significant male
preponderance, but a wide range of sex ratios was found. Two of the 28 species
demonstrated both significant male and female bias during different observation
periods, illustrating the need to collect over the entire flight period. Since the sex
ratio of collected organisms varies by species and by time, this must be taken into
consideration when using light-trap collection to make population estimates and to
gather information for conservation or control of any particular species.
Adults of many moth species demonstrate phototaxis toward ultraviolet
(UV) light in a range of wavelengths. Attraction to artificial light sources
has become progressively more important for insect populations as urban
lighting has increased (Frank 1988). This taxis has long been exploited by
entomologists through the use of traps that include emitters of UV light.
Light traps are used for a variety of purposes, such as gaining information
on diversity (Thomas and Thomas 1994), geographic ranges, and migration
patterns (Gregg et al. 1994), estimating the density of populations (Thomas
1989), and controlling populations of agricultural pests in fields (with variable
success; Cantelo et al. 1972, Frank 1988).
It is important to know whether the male-female ratio of moths captured
in light traps is representative of the actual ratio in the population. Limited
observations in other studies (e.g., Levine 1989, Steinbauer 2003, Worth
and Muller 1979, Yathom 1981) and anecdotal records have indicated that
males dominate light-trap catches. A number of explanations are possible.
For instance, the sexes might perceive and respond to UV light in different
ways, one sex might have a more limited flight range, the sex ratio might
be skewed from 1:1 at eclosion of adults, or some combination of these and
other factors might be occurring. In this study, we document the adult malefemale
ratio for 28 nocturnal moth species attracted to UV light in Greenville
County, SC and changes in that ratio during the observation period.
Over 90% of specimens were collected from a single location
(34°55'20"N, 82°22'38"W, elevation = 342 m), 300 m south of the boundary
1Department of Integrated Bioscience, University of Akron, Akron, OH 44325.
2Department of Biology, Furman University, Greenville, SC 29613. *Corresponding
author - email@example.com.
428 Southeastern Naturalist Vol. 9, No. 3
of Paris Mountain State Park, Greenville County, SC. Situated in eastern
temperate forest, the habitat is comprised of anthropogenically maintained
open grass immediate to the trap (within 20 m, all of which was shorter
than the trap collection bucket), surrounded by mixed forest typical for this
region—hardwoods, e.g., Quercus spp. (oaks), Liriodendron tulipifera L.
(American Tulip Tree), and Liquidambar styraciflua L. (Sweetgum), and
softwoods, e.g., Pinus strobus L. (Eastern White Pine) and Pinus virginiana
L. (Virginia Pine). This area is progressively suburban-urban to the south
and west toward the city of Greenville and is abruptly less so to the immediate
north, constituting the State Park. An area encompassing 0.40 km2
with the primary collection site at the center yielded 74% forest cover, 24%
cover devoted to anthropogenic activities (right of ways, roads, dwellings,
and disturbed or early successional scrub-brush habitats), and 2% visible
lake surface determined using ArcGIS® (ESRI of Redlands, CA) from Landsat
7 ortho-imagery, May 2005 (USGS 2005). The second sample location
(34°54'35.74"N, 82°24'41.77"W, elevation = 311m; 3.3 km southwest of
the primary sampling site) was situated on open grass overlooking a small
pond, with broad-scale forest cover, anthropogenic cover, and lake surface
components similar in composition to the primary site.
Moths were collected with ultraviolet lights at two locations within Greenville
County, SC, between 13 June and 19 August 2005, and between 21 May
and 20 July 2006. The light trap consisted of a UV light source (PestWest Quantum
BL UV bulb with output peak at 365 nm) vertically attached at the center of
4 metal vanes positioned over a collecting bucket charged with ethyl acetate as
a killing agent. In each case, the bulb was illuminated beginning at dusk for approximately
12 h on most dates throughout the observation period.
Each specimen was identified, and its sex was determined either by gross
exterior anatomy or by dissection. A representative of each species was
prepared as a voucher specimen and deposited in the Furman University
Sex differences were examined for species with aggregates of 10 or more
individuals over the period of observation. The sex ratio of each species,
summed across sites and its observed flight period, was evaluated using a
Yates-corrected chi-squared test (Hassard 1991). A sequential Bonferroni
correction for multiple comparisons (Rice 1989) was applied. We also tested
the hypothesis that sex ratio changes progressively throughout the majority
of the flight period, by correlating percent males with the date of capture
(Spearman rank correlation). Any correlation from these evaluations might
describe, to some extent, patterns that were hidden when data were summed
from the entire observation period for the chi-squared analysis. The species
chosen were the subset of those used for analyzing gross sex ratios which
had sample sizes that were sufficiently large to provide some confidence in
correlative statistical analysis and for which the sample sizes of individual
sexes were termed sufficiently balanced to impart analytical relevance.
2010 H.W. Garris and J.A. Snyder 429
A total of 1101 individuals representing 56 species were examined. An
aggregate of 843 males and 258 females was found. Of the 28 species for
which at least 10 individuals had been collected, 12 demonstrated a signifi-
cant sex bias toward males, 1 species demonstrated a significant sex-bias
toward females, and the sex ratio of 15 species did not differ significantly
from 1:1 following a sequential Bonferroni correction (Table 1). Figure 1
shows the distribution of male percentages among the 28 species.
When we analyzed possible sex differences in observed flight periods
by comparing the percentage of males captured over the observation period,
there were no significant changes in male bias for Atteva punctella, Desmia
funeralis, Halysidota sp., Polygrammate hebraeicum, and Spodoptera ornithogalli.
One species, Thioptera nigrofimbria (Fig. 2) showed a significant
trend toward females caught at light traps as flight days progressed (P =
0.0005, rs = -0.7049), although the total catch showed no significant bias
toward either sex. Despite exhibiting a significant overall male bias in the
chi-squared analysis (P = 1.31 x 10-4), Tetanolita mynesalis (Fig. 3) also
Table 1. Yates corrected chi-squared analyses performed for 28 species collected at UV light.
Numbers represented in the ID column reflect species number designations in Figure 1. * denotes
statistical significance retained at the 2-tailed P = 0.05 level after sequential Bonferroni
correction was applied (Rice 1989).
ID Family Species ♂ ♀ χ2 value P-value
21 Acrolophidae Acrolophus sp. 18 1 13.47 2.42 x 10-4 *
8 Yponomeutidae Atteva punctella (Cramer) 30 21 1.26 0.2620
22 Tortricidae Pandemis limitata (Robinson) 42 2 34.57 4.11 x 10-9*
7 Limacodidae Prolimacodes badia (Hübner) 7 6 0.00 1.0000
12 Crambidae Desmia funeralis (Hübner) 12 5 2.12 0.1450
15 Pyralidae Dolichomia olinalis (Guenée) 10 3 2.77 0.0960
14 Geometridae Epimecis hortaria (Fabricius) 13 4 3.76 0.0520
1 Hypagyrtis unipunctata (Haworth) 0 18 18.00 2.21 x 10-5*
26 Euchlaena amoenaria (Guenée) 10 0 8.10 4.43 x 10-3
19 Lasiocampidae Malacosoma americanum (Fabricius) 26 4 14.70 1.26 x 10-4*
27 Saturniidae Dryocampa rubicunda (Fabricius) 13 0 13.00 3.11 x 10-4*
17 Anisota stigma (Fabricius) 23 5 13.32 2.63 x 10-4*
23 Notodontidae Datana perspicua Grote & Robinson 42 2 34.57 4.11 x 10-9*
20 Nadata gibbosa (J.E. Smith) 103 7 83.78 5.53 x 10-20*
16 Arctiidae Hypoprepia fucosa Hübner 10 3 2.77 0.0960
10 Pyrrharctia isabella (J.E. Smith) 7 4 0.35 0.5540
25 Spilosoma latipennis Stretch 65 2 57.37 3.61 x 10-14*
24 Apantesis vittata (Fabricius) 150 6 131.08 2.38 x 10-30*
28 Grammia parthenice (W. Kirby) 23 0 21.04 4.50 x 10-6*
5 Halysidota sp. 17 20 0.11 0.7400
11 Noctuidae Tetanolita mynesalis (Walker) 72 32 14.63 1.31 x 10-4*
18 Hypena scabra (Fabricius) 24 5 11.17 8.31 x 10-4*
9 Thioptera nigrofimbria (Guenée) 28 19 1.36 0.2440
6 Acronicta haesitata (Grote) 5 5 0.10 0.7520
3 Polygrammate hebraeicum Hübner 7 14 1.71 0.1910
4 Spodoptera ornithogalli (Guenée) 10 18 1.75 0.1860
2 Agrotis ipsilon (Hufnagel) 4 15 5.26 0.0218
13 Xestia dolosa Franclemont 14 5 3.37 0.0664
430 Southeastern Naturalist Vol. 9, No. 3
demonstrated a significant increase (P = 0.0174, rs = -0.5680) in females
captured as the study period progressed. Spodoptera ornithogalli showed
Figure 1. Distribution of evaluated species (labeled 1–28) in ascending order from
0% males to 100% males captured at a UV light source. Species 1, Hypagyrtis unipunctata,
reflects 0% males as all of the individuals captured were female.
Figure 2. The percentage of captured males over time in days beginning 13 June
2005 and ending 19 August 2005 for Thioptera nigrofimbria (Noctuidae). Analysis of
Spearman’s rank correlation coefficient revealed a significant trend toward females
caught at light traps as flight days progressed (P = 0.0005, rs = -0.7049).
2010 H.W. Garris and J.A. Snyder 431
Figure 3. The percentage of captured males over time in days beginning 13 June 2005
and ending 19 August 2005 for Tetanolita mynesalis (Noctuidae). Analysis of Spearman’s
rank correlation coefficient revealed a significant trend toward females caught
at light traps as flight days progressed (P = 0.0174, rs = -0.5680).
a similar yet not statistically significant change (P = 0.0580, rs = -0.51763)
toward females. None of the remaining species analyzed yielded significant
correlations signifying a change in sample sex-bias over time.
Half of the evaluated species exhibited more males than females at a UV
light trap. This result matches the limited published data for other species,
such as Surattha indentella (Kearfott) (Sorensen and Thompson 1984), Agapeta
zoegana (L.) (Story et al. 2001), Hydraecia immanis (Guenée) (Levine
1989), four saturniid and one sphingid species (Worth and Muller 1979), and
Earias insulana (Boisduval) (Yathom 1981), all of which showed significant
bias toward males at UV light traps. However, our data for some species
were quite different: species demonstrated sex-bias to varying degrees along
a spectrum from female-predominant to male-predominant (Fig. 1). It is
notable that bias in a particular direction or to a particular degree was not
family specific (Table 1); however, studies should be performed to compare
a broader range of species, especially species within a single genus, to determine
if degree of UV attraction is conserved in related taxa.
A published study with some parallel to ours is that of Persson (1976).
He reported that 48.0% of all collected noctuids were female. However,
he found that 13 noctuid species (of more than 300 surveyed) showed significant deviation from a 1:1 sex ratio, ranging from 68% males to 78%
females. The percentage of female noctuids in our study was lower at 40.1%,
432 Southeastern Naturalist Vol. 9, No. 3
but comparable to Persson’s results, we found a wide range of percentages
among individual species. Although no noctuid species in our study showed
significant bias toward females, females were predominant in 2 of the 8
species. Some important differences do exist between our studies: Persson’s
work was carried out in a humid subtropical environment, it used a mercury
bulb (with different UV and visible emission maxima than our bulb), and its
reported species did not overlap ours. However, our studies are in agreement
in that both report a wide range of sex ratios among captured species.
A number of factors might contribute to the observed sex-ratio variation
among our surveyed species. For some species, UV-trap collection may be
a consequence of the sex ratio of adult moths as they emerge from the pupa
stage, rather than a sex-specific attraction to the light. At fertilization, the
initial sex ratio of most moth species should be 1:1 since one sex is heterogametic
(De Prins and Saitoh 1999). That initial ratio could be offset by
variability in survival throughout embryonic and larval development as a result
of genetic and environmental factors. Studies of populations where adult
female moths are significantly more numerous than males have variously
attributed this to meiotic drive (Seiler 1920), to parthenogenesis (Lokki et al.
1975), and to bacterial infections selectively killing male larvae (Hurst 1993,
Hurst and Majerus 1993). It would be instructive to check the sex ratio of a
population of newly emerged adults in a species where our study has found
a significant skewing toward one sex at UV traps.
Even with a 1:1 proportion of the sexes upon emergence as adults, a
factor in producing an offset toward males taken in light traps could be the
relative flight activity of the two sexes. If males tend to spend a greater portion
of the night-time hours flying to forage for food or searching for females,
this behavior favors their perceiving and being attracted to a stationary UV
source. The concept of a diminished nightly flight time of females is consistent
with their having sex-specific energetic expense (egg production,
increased body mass when bearing eggs, and ovipositing activity). A parallel
phenomenon is seen in certain butterfly species where a 1:1 sex ratio occurs
in laboratory-reared populations, but a significant skewing toward males is
found in field-caught populations (Brussard and Ehrlich 1970). Those workers
concluded that greater flight activity (and thus visibility to collectors) by
males is the most likely cause of their being captured more frequently.
If light trapping occurs over less than the full flight period for a species, an
apparent skewed sex ratio may result from differential emergence times from
the pupal stage. For example, females of Panolis flammea typically eclose
before males (Leather and Barbour 1983), resulting in a population-level bias
toward females during the first part of the species’ flight period. As an exploration
of this phenomenon, we analyzed the temporal distribution of some
species whose sex-ratio bias might be masked when combining data from
the entire flight period. It will be recalled that a correlation analysis revealed
significant change from predominantly capturing males to predominantly
capturing females of Thioptera nigrofimbria (Fig. 2) and Tetanolita mynesalis
(Fig. 3), which is consistent with a hypothesis that the sex ratio has changed
over the two-month survey period. The resulting change in light-trap capture
2010 H.W. Garris and J.A. Snyder 433
counts could be further accentuated if unmated females increased their nightly
flight durations and/or ranges toward the end of the male flight period.
If a light-trap survey is limited to the last portion of a species’ flight
period, another possible factor coming into play is that one sex has a longer
average lifetime in the adult stage. In such a case, a skewed sex ratio would
be observed even if the adult population began with equal numbers. Once
again, the importance of surveying over the entire flight period is evident.
For several moth species, Persson (1976) found that females are most
abundant at a light trap during the first half of the night. This should have no
bearing on our study, since we collected specimens only after the entire night
had elapsed with the trap continuously in operation.
It could be posited that gravid females tend to fly closer to the ground because
of the greater mass imparted by their mature eggs, and therefore would
be less well represented in trap catches if the UV light source is considerably
above ground level. Placing multiple traps at significantly varying heights
above the substrate during identical sampling periods would test for this.
Finally, differences in visual perception and response may lead to one sex
being disproportionately represented in UV light-trap catches throughout the
flight period. This differential capture rate may occur by attracting a particular
sex from a greater distance or by eliciting a stronger phototactic response
at any distance. This potential factor could be determined in species where
appropriate numbers of newly eclosed adults can be obtained and tested in a
This study serves as a base set of observations for determining the extent
of male versus female sex bias in their attraction to light in the UV range.
Clearly, there are not always significantly more males represented in catches
at a UV light trap. Additional studies should be conducted to evaluate the relative
influence of female versus male flight periods, relative levels of activity,
and surviving emergent adult sex ratios of a variety of species to determine the
effect of each in producing a sex ratio at UV light traps. Understanding relative
sex ratios for species caught at UV light traps may serve to improve populationestimation
techniques of moths and other nocturnal insects as well as provide
information for the conservation or control of individual species.
The first author was supported by the Furman Advantage program. Equipment
was purchased through a grant from Furman University’s Research and Professional
Growth program. We thank Wade B. Worthen and two anonymous reviewers for valuable
comments and suggestions that improved the manuscript.
Brussard, P.F., and P.R. Ehrlich. 1970. The population structure of Erebia epipsodea
(Lepidoptera: Satyrinae). Ecology 51:119–129.
Cantelo, W.W., J.S. Smith, Jr., A.H. Baumhover, J.M. Stanley, and T.J. Henneberry.
1972. Suppression of an isolated population of the Tobacco Hornworm with
blacklight traps unbaited or baited with virgin female moths. Environmental
434 Southeastern Naturalist Vol. 9, No. 3
De Prins, J., and K. Saitoh. 2003. Karyology and sex determination. Pp.449–464,
In N.P. Kristensen (Ed.). Handbook of Zoology, Vol. IV: Part 36, Lepidoptera,
Moths, and Butterflies, and Vol. 2: Morphology, Physiology, and Development.
W. de Gruyter, Berlin, Germany and New York, NY. 564 pp.
Frank, K.D. 1988. Impact of outdoor lighting on moths: An assessment. Journal of
the Lepidopterists’ Society 42:63–93.
Gregg, P.C., G.P. Fitt, M. Coombs, and G.S. Henderson. 1994. Migrating moths collected
in tower-mounted light traps in northern New South Wales, Australia: Influence
of local and synoptic weather. Bulletin of Entomological Research 84:17–30.
Hassard, T.H. 1991. Understanding Biostatistics. St. Louis: Mosby-Year Book, Inc.,
St. Louis, MO. 292 pp.
Hurst, L.D. 1993. The incidences, mechanisms, and evolution of cytoplasmic sexratio
disorders in animals. Biological Reviews 68:121–193.
Hurst, G.D.D., and M.E.N. Majerus. 1993. Why do maternally inherited microorganisms
kill males? Heredity 71:81–95.
Leather, S.R., and D.A. Barbour. 1983. The effect of temperature on the emergence
of Pine Beauty Moth, Panolis flammea Schiff. (Lep., Noctuidae). Zeitschrift für
Angewandte Entomologie 96:445–448.
Levine, E. 1989. Forecasting Hydraecia immanis (Lepidoptera: Noctuidae) moth
phenology based on light-trap catches and degree-day accumulations. Journal of
Economic Entomology 82:433–438.
Lokki, J., E. Suomalainen, A. Saura, and P. Lankinen. 1975. Genetic polymorphism
and evolution in parthenogenetic animals. II. Diploid and polyploid Solenobia
triquetrella (Lepidoptera: Psychidae). Genetics 79:513–525.
Persson, B. 1976. Influence of weather and nocturnal illumination on the activity and
abundance of populations of noctuids (Lepidoptera) in south coastal Queensland.
Bulletin of Entomological Research 66:33–63.
Rice, W.R. 1989. Analyzing tables of statistical tests. Evolution 43:223–225.
Seiler, J. 1920. Geschlechtschromosomenuntersuchungen an Psychiden. I. Experimentelle
Beiinflussung der Geschlechts bestimmenden Reifeteilung bei Talaeporia
tubulosa Retz. Archiv für Zellforschung 15:249–268.
Sorensen, K.A., and H.E. Thompson. 1984. Light-trap response of the Buffalograss
Webworm, Surattha indentella Kearfott (Lepidoptera: Pyralidae), in Kansas.
Journal of the Kansas Entomological Society 57:719–722.
Steinbauer, M.J. 2003. Using ultra-violet light traps to monitor Autumn Gum Moth,
Mnesampela privata (Lepidoptera: Geometridae), in southeastern Australia. Australian
Story, J.M., W.R. Good, and L.J. White. 2001. Response of the Knapweed biocontrol
agent Agapeta zoegana L. (Lepidoptera: Cochylidae) to portable lights. Pan-
Pacific Entomologist 77:219–225.
Thomas, C.D. 1989. Limits and scope of light-trapping for studying moth population
dynamics. New Zealand Entomologist 12:89–90.
Thomas, A.W., and G.M. Thomas. 1994. Sampling strategies for estimating moth
species diversity using a light trap in a northeastern softwood forest. Journal of
the Lepidopterists’ Society 48:85–105.
United States Geological Survey (USGS). 2005. Orthoimagery at the USGS EROS.
Available online at http://eros.usgs.gov/website/Orthoimagery/index.php. Accessed
10 August 2009.
Worth, C.B., and J. Muller. 1979. Captures of large moths by an ultraviolet light trap.
Journal of the Lepidopterists’ Society 33:261–264.
Yathom, S. 1981. Sex ratio and mating status of Earias insulana females (Lepidoptera:
Noctuidae) collected from light traps in Israel. Israel Journal of Entomology