Optimization of traps for live trapping of Pine Wood Nematode vector Monochamus galloprovincialis

The pine sawyer beetle Monochamus galloprovincialis, a secondary pest of pines in Europe and North Africa, has become important as it was identified as the vector in Europe of Bursaphelenchus xylophilus, the causal agent of pine wilt disease (PWD). An effective trapping system is needed, not only for monitoring the insect vector but also for direct control of its population. Trapping may also provide key information on the nematode load carried by the beetles, allowing early detection of infections, provided that captured beetles remain alive within the trap. Highly effective attractants have been developed in recent years that are commonly used in combination with diverse standard trap designs. In this study, several trap designs were developed and compared to commercial standard models in order to determine which designs maximized the number of attracted insects actually caught and the proportion of them remaining alive. In total, 12 trap designs were evaluated in five field experiments carried out in France, Spain and Portugal. Teflon coating applied to the whole trap and extended, ventilated collecting cups resulted in a significant improvement of trap performance. These modifications led to significant increases of pine sawyer catches, up to 275%, when applied to multiple‐funnel or black cross‐vane traps, compared to standard designs. Furthermore, a significant proportion of the captured beetles remained alive within the trap. These findings have been used to develop new commercial traps (Econex Multifunnel‐12^® and Crosstrap^®; Econex, Murcia, Spain) available to forest managers. A model for insect survival within the trap was also fitted. Elapsed time between consecutive samplings, mean relative humidity and maximum radiation were the three most significant variables. Thus, traps should provide a suitable sample of live insects if sun exposure of the trap is minimized and a reasonable sampling schedule is implemented.     

Species characterization and responses of subcortical insects to trap‐logs and ethanol in a hardwood biomass plantation

相似文献   

Evaluation of traps of six different designs to capture pine sawyer beetles (Coleoptera: Cerambycidae)

1 Sawyer beetles (Coleoptera: Cerambycidae) and metallic wood‐boring beetles (Coleoptera: Buprestidae) attack recently killed or felled trees. They are attracted to potential hosts by volatiles and possibly by visual cues. We examined traps of six different designs to determine the role of silhouette and shape in the capture of pine sawyer beetles, Monochamus L. 2 Traps with a round black cylinder, a single vane (clear or black), or cross vanes (clear or black) all mounted above a pan were compared to an open pan. Each pan contained soapy water to retain insects. 3 Traps with a black silhouette were significantly more effective in capturing male and female Monochamus scutellatus (Say) than traps with or without clear vanes. For both sexes of M. mutator LeConte, pan traps were as effective as traps with a black silhouette. Silhouette was not important for the capture of male and female M. notatus (Drury). Trap catches for the six trap types were not significantly different for male Buprestis maculativentris Say. 4 A pan trap with a black silhouette is recommended for general use in capturing wood‐boring cerambycids and buprestids.     

Effect of aerosol surface lubricants on the abundance and richness of selected forest insects captured in multiple-funnel and panel traps

Survey and detection programs for native and exotic forest insects frequently rely on traps baited with odorants, which mediate the orientation of target taxa (e.g., the southern pine beetle, Dendroctonusfrontalis Zimmermann) toward a resource (e.g., host material, mates). The influence of trap design on the capture efficiency of baited traps has received far less empirical attention than odorants, despite concerns that intercept traps currently used operationally have poor capture efficiencies for some target taxa (e.g., large woodborers). Several studies have recently demonstrated that treating traps with a surface lubricant to make them "slippery" can increase their capture efficiency; however, previously tested products can be expensive and their application time-consuming. The purpose of this study was to evaluate the effect of alternate, easier to apply aerosol lubricants on trap capture efficiency of selected forest insects. Aerosol formulations of Teflon and silicone lubricants increased both panel and multiple-funnel trap capture efficiencies. Multiple-funnel traps treated with either aerosol lubricant captured significantly more Monochamus spp. and Acanthocinus obsoletus (Olivier) than untreated traps. Similarly, treated panel traps captured significantly more Xylotrechus sagittatus (Germar), Ips calligraphus (Germar), Pissodes nemorensis (Germar), Monochamus spp., A. obsoletus, Thanasimus dubius (F.), and Ibalia leucospoides (Hochenwarth) than untreated traps. This study demonstrates that treating multiple-funnel and panel traps with an aerosol dry film lubricant can increase their capture efficiencies for large woodborers (e.g., Cerambycidae) as well as bark beetles, a weevil, a woodwasp parasitoid and a bark beetle natural enemy (Coleoptera: Cleridae).     

Modified pheromone traps help reduce bycatch of bark‐beetle natural enemies

• 1 Thanasimus formicarius and Temnochila caerulea, two of the main predators of Ips sexdentatus, a well‐known forest pest in Southern Europe, are captured in high numbers when trapping I. sexdentatus as a result of the kairomonal effect of the lures used.
• 2 A preliminary field trial showed that predators could survive for at least 1 week within trapping containers, although predator mutilation and high predator death rates were observed.
• 3 Different modifications of conventional multiple funnel and slot traps with the objective of reducing natural enemy entrance into trap containers were bioassayed in field experiments conducted over four seasons. Based on the larger sizes of predators, different designs using welded wire‐mesh screens improved performance to different extents. Providing escape windows just above the screen on multiple funnel traps gave the most promising results, including when effect sizes among all tested designs were compared.
• 4 Thus, a simple modification of the lowest funnel of the multiple funnel traps would reduce the bycatch of T. formicarius and T. caerulea, hence improving the efficiency of trapping programmes by lowering the likely impact on natural populations of these predators.

     

An assessment of re-randomization methods in bark beetle (Scolytidae) trapping bioassays

1 Numerous studies have explored the role of semiochemicals in the behaviour of bark beetles (Scolytidae). 2 Multiple‐funnel traps are often used to elucidate these behavioural responses. Sufficient sample sizes are obtained by using large numbers of traps to which treatments are randomly assigned once, or by frequent collection of trap catches and subsequent re‐randomization of treatments. 3 Recently, there has been some debate about the potential for trap contamination to occur when semiochemical treatments (baits), and not trap‐treatment units (traps and baits), are re‐randomized among existing traps. Due to the volatility of many semiochemicals, small levels of contamination could potentially confound results. 4 A literature survey was conducted to determine the frequency of re‐randomizing semiochemical treatments (baits) vs. trap‐treatment units (traps and baits) in scolytid trapping bioassays. An experiment was then conducted to determine whether differences in the response of Dendroctonus brevicomis LeConte to attractant‐baited traps exist between the two methods. 5 The majority of papers examined reported use of a large number of fixed replicates (traps) rather than re‐randomization of treatments at frequent intervals. Seventy‐five percent of papers for which re‐randomization methods could be determined reported relocation of semiochemical treatments (baits) only. 6 No significant differences in trap catch were observed among multiple‐funnel traps aged with D. brevcomis baits (Phero Tech Inc., Canada) for 0, 30 and 90 days, suggesting that contamination did not influence the results. 7 It is concluded that re‐randomizing baits is a viable cost‐effective option to re‐randomizing trap and bait units.     

Response of garden chafer, Phyllopertha horticola, to plant volatiles: from screening to application

Field tests were performed on a golf course and in an apple orchard to screen synthetic plant volatiles with respect to their attractiveness for the garden chafer, Phyllopertha horticola L. (Coleoptera: Scarabaeidae), and to investigate the possible application of plant volatiles for garden chafer control. The chemicals tested were green leaf volatiles (GLV), terpenoids, and phenylpropanoids. Funnel traps baited with the GLV (Z)‐3‐hexen‐1‐ol, 1‐hexanol (Z)‐3‐hexenal, and hexanal captured more P. horticola than unbaited controls. Furthermore, traps baited with all tested floral terpenoids (i.e., geraniol, geranyl acetate, citronellol, linalool, and nerol) and phenylpropanoids (i.e., eugenol, anethol, isoeugenol, eugenyl acetate, and isoeugenyl acetate) captured more garden chafers than controls. Different dispenser types loaded once with a mixture of (Z)‐3‐hexen‐1‐ol (50%), geraniol (11.5%), eugenol (27%), and 2‐phenylethyl propionate (11.5%) attracted P. horticola over a whole flight season. A commercially available membrane dispenser had the best properties, combining the highest number of captured beetles with a low release rate. A simple modification of the trap design, i.e., a reduction of the funnel outlet diameter, significantly reduced the capture of beneficial non‐target insects (Apoidea), without influencing the number of captured garden chafers. A mass trapping experiment in the apple orchard revealed that the use of attractant traps significantly reduced the percentage of apples disfigured by feeding holes of adult garden chafers (control area: 18.9%, test area: 11.6%). The possible application of synthetic plant volatiles in mass trapping and monitoring approaches for garden chafer control is discussed.     

Considering species richness and rarity when selecting optimal survey traps: comparisons of semiochemical baited flight intercept traps for Cerambycidae in eastern North America

相似文献   

Comparison of reptile sampling in Iran with pitfall versus double-ended funnel traps

In north Khorasan province, Iran, we compared the effectiveness of two types of traps for collecting reptiles: funnel traps and pitfall traps. Three stations were monitored over three 10-day periods and in total 544 individuals (including 200 recaptures) belonging to 5 species and 4 families of lizards (Lacertidae) were collected. Funnel traps with 280 captures were more efficient than pitfall traps with 264 captures, but the differences between the two traps are not significant. Three species were captured most often in the three different stations respectively: station 1, Bunopus crassicauda (22% relative frequency); station 2, Eremias fasciata (29% relative frequency); and station 3, Trapelus agilis (32% relative frequency). Shannon species richness indices were higher for pitfall than funnel traps (i.e., 1.45 vs. 1.40), but the difference was not significant. Pitfall traps were better for capturing species that search widely for food, while the funnel traps preferentially captured species that climb plants, such as Trapelus agilis. We recommend using both types of trap to capture the various types of species in any one region.     

Analysis of single‐copy,nuclear microsatellite markers from flies collected on sticky traps

Sticky traps can provide large numbers of spatially referenced samples for use in molecular ecological studies of insects. However, the adhesives used on these traps, and the methods used to clean adhesive off trapped individuals, could potentially interfere with downstream molecular analyses. Specimens captured on sticky traps have been successfully used to analyse mitochondrial or multiple‐copy ribosomal DNA markers, but not single‐copy nuclear markers. Furthermore, the effects of trap adhesive and cleaning protocol on the success of molecular analyses have not been explored. Here, we examine the effects of trap adhesive, sample cleaning method, and sample storage condition on DNA concentration and purity, and on the ability to amplify single‐copy, nuclear microsatellite loci, using specimens of the western cherry fruit fly, Rhagoletis indifferens (Diptera: Tephritidae) captured on sticky traps in an orchard. We could extract DNA of high purity, and amplify microsatellite loci in multi‐plex polymerase chain reaction (PCR), under all combinations of treatments. However, DNA yield, DNA purity and the yield of PCR products were affected by treatment, with complex interactions among trap adhesive, sample cleaning method, and storage condition. Samples that were cleaned with acetone and stored dry had the highest DNA concentration. With respect to PCR amplification, samples cleaned with Histo‐clear produced much less product than those cleaned with acetone or not cleaned at all, whereas samples that were stored dry produced more PCR product than samples stored in ethanol. Insects captured on sticky traps can thus provide genetic data appropriate for molecular ecological analyses under a wide range of treatment conditions. However, potential interactions among adhesives, cleaning protocols and storage conditions suggest that any novel combination for treatment of samples from sticky traps should be tested on a small scale prior to collecting large numbers of samples for genetic studies.     

A sex pheromone‐baited trapping system for management of sweetpotato weevil,Cylas formicarius (Coleoptera: Brentidae)

A potent male attractant of sweetpotato weevil helps in monitoring and control of sweetpotato weevil in many production areas around the world. At present, it has not been used in Malaysia. Cost of the components of a trapping system is a major constraint in the adoption of male lure‐baited trapping by growers in Malaysia. Seven field trapping experiments were conducted from February 2013 to November 2015 as part of an effort to develop a simple, easy to construct, cost‐effective and efficient sex pheromone‐baited trap acceptable for use by farmers in Malaysia for monitoring and control of sweetpotato weevil (Cylas formicarius Fabricius). Overall, sweetpotato weevil trap catch was significantly affected by the number of windows in the trap, the killing agent used in the trap and the position of the trap relative to sweetpotato foliage, while trap size and trap colour did not significantly affect trap catch. Trap catch was best in plastic pole traps made from polyethylene terephthalate, with four window opening to facilitate weevil entry, with detergent solution as a killing agent and with the trap positioned from 0 to 40 cm above the crop canopy level. In a comparison study with commercial trap designs, sex pheromone‐baited plastic pole traps caught 60%–78% more weevils than were caught in sex pheromone‐baited delta traps, wing traps or unitraps. Optimization of trap characteristics is important for improving the performance of pheromone‐baited traps for use in population monitoring or mass‐trapping efforts to minimize crop damage by sweetpotato weevil infestation.     

The influence of trap type on evaluating population structure of the semifossorial and social rodent <Emphasis Type="Italic">Octodon degus</Emphasis>

Trap type may influence captures of individuals in different age-sex categories in small mammal studies, resulting in biased population and demographic information. We deployed 4 live trap types at burrow systems of the rodent, Octodon degus Molina, 1782, in central Chile to determine trap efficacy in capturing individuals of 6 demographic categories. We captured 2672 individuals in 17 709 trap days (15.1% trapping success). Tomahawks were the most efficient trap capturing half of individuals during both years, followed by mesh Sherman traps, large Sherman traps, and medium Sherman traps in 2005. All trap types equally sampled sexes. Large and medium Sherman traps provided similar demographic structure, where half of the individuals captured were pups; Tomahawk traps sampled more adults than pups. Relative captures of pups were similar across different trap types, suggesting that pups are equally sampled by each of the deployed trap types. Relative captures of adults were lower in Sherman traps, suggesting that this age class avoided solid-walled traps. For Octodon degus, the sole use of Tomahawk traps may produce sufficient, unbiased demographic data. Only 4 trap mortalities occurred (0.15%). Researchers may minimize trap mortality without compromising sufficient demographic sampling by trapping during peak animal activity.     

A carbon dioxide, heat and chemical lure trap for the bedbug, Cimex lectularius

Abstract A trap for the collection of bedbugs, Cimex lectularius Linnaeus (Hemiptera: Cimicidae), is described. The trap was baited with CO₂ (50–400 mL/min), heat (37.2–42.2 °C) and a chemical lure comprised of 33.0 μg proprionic acid, 0.33 μg butyric acid, 0.33 μg valeric acid, 100 μg octenol and 100 μg L‐lactic acid, impregnated into a gel. Laboratory studies, conducted in a square arena measuring 183 cm on each side, showed that traps with and without baits captured adult bedbugs, but traps with CO₂ emissions of 50–400 mL/min caught significantly (P < 0.05) more bedbugs than traps without CO_2. In an infested unoccupied apartment, traps with heat and with or without the chemical lure were tested without CO₂ on 29 trap‐days and with CO₂ on 9 trap‐days. The numbers of bedbugs captured were 656 and 5898 in traps without and with CO₂, respectively. The numbers of bedbugs of all development stages captured were significantly greater in traps with CO₂ (χ² = 15 942, d.f. = 1, P < 10^?9). A non‐parametric two‐way analysis of variance evaluation of six different traps with or without CO₂, heat or a chemical lure monitored over 19 trap‐days in an infested apartment showed that trap type was highly significant (n = 2833 bedbugs collected) (P < 10^?7). The trap with CO₂, heat and a chemical lure captured more bedbugs than the other traps, but only caught significantly more fourth and fifth instar nymphs than all other traps. Otherwise, the catches in this trap did not differ significantly from those caught by traps that contained CO₂ and heat only. The total numbers of bedbugs collected for each trapping date (pooling all six traps) followed an exponential decline over the trapping period. This type of trap, which caught bedbugs in unoccupied apartments with and without furniture, and in an occupied apartment, may have utility in studying the ecology of bedbugs, in detecting bedbug infestations and in reducing numbers of bites by trapping host‐seeking bedbugs.     

Comparison of sex pheromone traps for monitoring pink hibiscus mealybug (Hemiptera: Pseudococcidae)

The pink hibiscus mealybug, Maconellicoccus hirsutus (Green) (Hemiptera: Pseudococcidae), is a highly polyphagous pest that invaded Florida in 2002 and has recently been reported from several locations in Louisiana. Although identification of its sex pheromone in 2004 improved monitoring capabilities tremendously, the effectiveness and efficiency of different pheromone trap designs for capturing males has not been evaluated. We deployed green Delta, Pherocon IlB, Pherocon V, Jackson, and Storgard Thinline traps in Homestead, FL, and compared the number of male M. hirsutus captured per trap, the number captured per unit of trapping surface area, the amount of extraneous material captured, and the time taken to count trapped mealybugs. Pheromone-baited traps with larger trapping surfaces (green Delta, Pherocon IIB, and Pherocon V) captured more males per trap than those with smaller surfaces (Jackson and Storgard Thinline), and fewest males were captured by Storgard Thinline traps. However, Jackson traps captured as many or more males per square centimeter of trapping surface as those with larger surfaces, and the time required to count males in Jackson traps was significantly less than in green Delta, Pherocon IIB, and Pherocon V traps. Although all trap designs accumulated some debris and nontarget insects, it was rated as light to moderate for all designs. Based on our measures of effectiveness and efficiency, the Jackson trap is most suitable for monitoring M. hirsutus populations. Additionally, unlike the other traps evaluated, which must be replaced entirely or inspected in the field and then redeployed, only the sticky liners of Jackson traps require replacement, enhancing the efficiency of trap servicing.     

Efficacy of Fluon conditioning for capturing cerambycid beetles in different trap designs and persistence on panel traps over time

Fluon PTFE is a fluoropolymer dispersion applied as a surface conditioner to cross-vane panel traps to enhance trap efficiency for cerambycid beetles. We describe the results of three experiments to further optimize cerambycid traps of different designs and to test the effect of Fluon over time. We tested Fluon with Lindgren funnel and panel traps fitted with either wet or dry collection cups on catches of cerambycid beetles and how the effect of Fluon on panel traps persisted. Fluon-treated funnel traps with wet collection cups captured approximately 6x more beetles than the untreated funnel traps with wet collection cups. Untreated funnel traps with dry collection cups did not capture any beetles; however, Fluon-treated funnel traps with dry collection cups captured an average of four beetles per trap. Fluon-treated panel traps with wet collection cups captured approximately 9x more beetles than untreated panel traps with wet collection cups. Fluon-treated panel traps with dry collection cups captured approximately 11x more beetles than untreated panel traps with dry collection cups. The effect of Fluon on capturing cerambycid beetles did not decline after use in one or two field seasons. There was no significant difference in the number of beetles captured in freshly treated panel traps compared with traps that had been used for 1 or 2 yr. Fluon-treated traps captured nine species that were not captured in untreated traps. Conditioning both Lindgren funnel and panel traps with Fluon enhances the efficacy and sensitivity of traps deployed to detect exotic cerambycid species, or for monitoring threatened species at low population densities.     

Evaluation of traps used to monitor southern pine beetle aerial populations and sex ratios

1 Various kinds of traps have been employed to monitor and forecast population trends of the southern pine beetle (Dendroctonus frontalis Zimmermann; Coleoptera: Scolytidae), but their accuracy in assessing pine‐beetle abundance and sex ratio in the field has not been evaluated directly. 2 In this study, we used fluorescent powder to mark pine beetles emerging from six isolated infestations. We then compared estimates of total abundances and proportions of males emerging from within each infestation to the estimates from three types of traps: passive sticky traps (2, 5, 10 and 20 m away from the source of beetles), multi‐funnel traps baited with pine beetle attractants (100 m away) and pine trees baited with attractants (also 100 m away). 3 We found that the proportion of males captured in traps was significantly affected by the type of trap used. 4 Within an infestation, equal proportions of males and females were marked (0.53 ± 0.02 males; mean ± SE), but the proportions captured in trap trees and passive traps were more female biased (0.42 ± 0.03 and 0.46 ± 0.01 males, respectively). On the other hand, funnel traps provided an estimate of the proportion of males that was nearly identical to the proportion from within infestations (0.51 ± 0.03). 5 Numbers of marked beetles captured in traps were uncorrelated with the numbers of marked beetles emerging from the focal infestations. This suggests that traps positioned around an infestation may not be effective at estimating relative abundances of beetles within the infestation.     

Effects of pitfall trap size and sampling duration on collection of ground beetles (Coleoptera: Carabidae) in temperate forests

Pitfall trapping is a standard sampling method to compare the abundance or community structure of ground beetles. However, effects of sampling duration on biodiversity estimation of ground beetles according to different trap sizes have not been experimentally evaluated in temperate forests in Korea. Therefore, the objective of this study was to determine the interaction between trap sizes (SB, small‐sized bottle; MB, medium‐sized bottle; PC, plastic cup; PJ, plastic jar; PT, perforated type trap; FT, funnel‐type trap) and sampling duration (T1, 2 weeks × 2 sessions, 28 days; T2, 2 weeks × 4 sessions, 56 days; T3, 4 weeks × 2 sessions, 56 days) on estimation of ground beetle assemblages in Naejangsan National Park, a temperate forest in Korea. Funnel type and larger pitfall traps collected higher numbers of individuals and species than other trap sizes. Species composition of ground beetles was different by size of traps (SB, MB, and PC vs. PJ, PT, and FT). In particular, ground beetle composition in larger traps (PJ, PT, and FT) appeared to be influenced by environmental characteristics according to localities (e.g., soil characters and dominant tree species). These findings from our study support that pitfall trapping of ground beetles can be influenced by trap sizes per se as well as sampling durations and environmental characteristics. Thus, biodiversity monitoring in temperate forests should be conducted with long sampling duration (at least 28 days) using large‐sized traps (> 7.5 cm in diameter of trap mouthpart) considering expenses and study aims.     

An ordinal day model of spruce beetle trap capture phenology in northern Colorado

Spruce beetle populations (Dendroctonus rufipennis) (Coleoptera: Scolytinae) in Engelmann spruce (Picea engelmannii) stands were monitored at twelve sites for 6 years in northern Colorado using pheromone‐baited Lindgren funnel traps. During the trapping period, over 30,000 beetles were captured, and beetle abundance data were used to construct empirical models of trap capture phenology based on day of year, accumulation of thermal units and thermal day thresholds, with the goal of informing future regional trapping efforts and producing a simple model for use by forest managers. Ordinal day models outperformed thermal accumulation and thermal threshold models in both predictive power and parsimony. Mean date of earliest capture was Day 153 (June 2) and ranged from Day 126 to Day 161 (May 6–June 10), and mean date of final capture was Day 243 (August 31) and ranged from Day 220 to Day 286 (August 8–October 13). A two‐parameter logistic function was the most parsimonious of several ordinal day models, accounting for 81% of the variance in cumulative trap capture across all sites and dates. The model predicts 50% of trap captures to occur by Day 181 (June 30). This model has application as a decision support tool for forest ecosystem managers concerned with the timing of trap deployment or D. rufipennis mitigation treatments.     

Impact of trap design, windbreaks, and weather on captures of European corn borer (Lepidoptera: Crambidae) in pheromone-baited traps

Pheromone-baited traps are often used in ecological studies of the European corn borer, Ostrinia nubilalis (Hübner) (Lepidoptera: Crambidae). However, differences in trap captures may be confounded by trap design, trap location relative to a windbreak, and changes in local weather. The objectives of this experiment were, first, to examine differences in 0. nubilalis adult (moth) captures among the Intercept wing trap, the Intercept bucket/funnel UNI trap, and the Hartstack wire-mesh, 75-cm-diameter cone trap (large metal cone trap) as well as among three cone trap designs. Second, we examined the influence of the location of the large metal cone trap relative to a windbreak on the number of moths captured. Third, we examined the relationship between nightly mean air temperature, relative humidity, wind speed, precipitation, and the number of moths captured in large metal cone traps. The number of moths captured was significantly influenced by trap design, with large metal cone traps capturing the most moths. Wing and bucket traps were ineffective. Differences among trap captures were significant among trap locations relative to a windbreak. Under strong (>14 kph) or moderate (7 <14 kph) wind speeds, traps located leeward of the windbreak captured the most moths, but when wind speeds were light (<7 kph), traps not associated with windbreaks captured the most moths. The multiple regression model fitted to the relationship between number of moths captured per Julian date and nightly weather patterns was significant. Nightly mean air temperature was the most influential parameter in the model, and its relationship with moth capture was positive.     

An emergence trap for bark-dwelling insects, its efficiency and effects on temperature

• 1 A cloth cage with collecting bottles is described for trapping adult insects emerging from beneath tree bark in low densities.
• 2 The traps had little effect on temperature, although fluctuations tended to be slightly damped. They probably delayed emergence of adult codling moth, Cydia pomonella by a day or two only.
• 3 In a test of trap efficiency, all codling moths which developed to the adult stage inside the traps were recorded in the collecting bottles; a few larvae remained in diapause, and the proportion of these was greater in the traps than in an outdoor insectary, so that traps tend to provide a slight underestimate of adult numbers.