Oviposition Behavior of the Bed Bug, Cimex lectularius (Hemiptera: Cimicidae)

Female bed bugs, Cimex lectularius, in strains from three widely separated geographic regions were evaluated for diel and daily periodicity in oviposition, as well as for the duration of oviposition per se and apparent egg marking behavior. Females that were mated and fed for the first time and laying their first clutch of eggs and females fed a second time before laying their second clutch of eggs both oviposited preferentially in the scotophase of a 14:10 L:D cycle. Oviposition began before the photophase ended, indicating it was not induced by the onset of darkness and suggesting that oviposition may be governed by a circadian rhythm. First-clutch females began to oviposit closer to the onset of darkness and 1–2 days later than second-clutch females, respectively, suggesting that first-clutch females were more strictly entrained to the L:D cycle and were still undergoing reproductive maturation. In all cases, oviposition of one clutch was finished by day 10, but the frequency distributions of eggs laid daily by first-clutch and second-clutch females, as well as those of females from the three different strains, were all significantly different. The acts of oviposition and subsequent apparent egg marking behavior (a novel observation) lasted mean durations of 4.7 s and 24.1 s, respectively. During apparent egg marking, a female would repeatedly touch the abdominal cuticle in the region of the scent glands with her metathoracic tarsi and then touch the newly laid egg while concurrently waving the tip of her abdomen back and forth across the surface of the egg.

Keywords: Bed bug; Hemiptera; Cimicidae; Oviposition; Behavior

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s10905-023-09830-x.

Insect eggs are immobile and at risk of predation, parasitism, microbial infection, and desiccation (Kaltenpoth et al. [21]; Deas and Hunter [6]; Newcombe et al. [27]). Oviposition behavior by females plays an important role in offspring success (Van Dyck and Regniers [35]; Guidobaldi and Guerenstein [13]). Females select oviposition sites that reduce exposure of eggs and newly eclosed first instars to natural enemies and provide food resources for their progeny (Gullan and Cranston [14]; Van Dyck and Regniers [35]). Alternatively, or in addition, many female insects utilize chemical defenses to protect their eggs (Newcombe et al. [27]). Defensive compounds may be synthesized by the female de novo, or they can be sequestered through the host diet of the parents, usually the female (Blum and Hilker [3]; Newcombe et al. [27]). Females often apply microbial or chemical defensive secretions to the egg surface or environment (Kaltenpoth et al. [21]; Matsuura et al. [24]; Hosoe et al. [16]), but they may also be incorporated into the chorion or internal contents of developing eggs (Eisner et al. [7]; Blum and Hilker [3]; Newcombe et al. [27]).

The ovipositional behavior of hematophagous insects has been studied extensively due to the risk they pose to public health (Ampleford and Davey [1]; Bentley and Day [2]; Lorenzo and Lazarri [22]; Feliciangeli [9]). For bed bugs, Cimex lectularius (Hemiptera: Cimicidae), Johnson ([20]) elucidated the duration between blood feeding and oviposition and the duration in days that females oviposit after feeding and mating. Our objectives were to further characterize oviposition in the bed bug, including the diel and daily oviposition patterns exhibited by female bed bugs of three geographically distinct North American populations (hereafter referred to as strains) after their first and second blood meal, and to determine the time taken to complete deposition of one egg. In addition, we observed apparent egg marking behavior for the first time in C. lectularius.

Bed bugs were obtained for laboratory studies conducted in 2016 as needed from three permanent laboratory colonies established from collections of wild insects: CIN-1, Cincinnati, OH 2005; NY-1, New York, NY 2007; LEX-8, Lexington, KY 2012. Bugs were housed in an incubator (Percival Scientific, Perry, IA) at 27 °C, 70% RH, and 14:10 L:D. Weekly blood meals (Montes et al. [25]) comprised defibrinated rabbit blood (Quad Five, Rygate, MT or Hemostat, Dixon, CA) pipetted into glass mosquito feeders (Kimble Chase Custom Glass Shop, Vineland, NJ) heated to 39 °C with a circulating water bath. Bed bugs in 59 mL plastic jars (Consolidated Plastics, Stow, OH) fed by inserting their proboscis through fine mesh organza covering the jars and penetrating a layer of Parafilm® (Bemis Co., Neenah, WI) to reach the blood.

Time-lapse photography was used to document the periodicity of oviposition by 20 females from each strain that had mated and taken their first blood meal 24 h before the experiment, and 20 females from each strain that had mated, fed and produced their first clutch of eggs seven days before the experiment. These females were provided a second blood meal 24 h before the experiment. A Cannon 60D camera (Canon USA, Melville, NY) was fitted with a 100 mm macro lens, and a Kodak No. 29 Red Wratten gel filter (Eastman Kodak Co., Rochester, NY) was suspended over the flash to prevent disturbing bed bugs during the scotophase. The camera was positioned 45 cm from four groups of 10 females each (two groups fed for the first time and two groups fed for the second time) held separately in four wells of a six-well plastic plate (VWR, Radnor, PA). The wells were lined with Teflon to prevent escape and movement on the walls and floor. Inside the well, the insects rested on a 3.5 × 1.75 cm piece of blotting paper pressed firmly into the bottom of the cell to minimize the chance of oviposition on the underside. The apparatus was held in an incubator at 27 °C, 70% RH, and 14:10 L:D. One picture was taken at the start of each hour, beginning 24 h post blood meal at the first hour of the 10-h scotophase and ending after the last hour of the photophase on day 11.

Each photograph was analyzed using Adobe Photoshop (Adobe, San Jose, CA). The cropping feature was utilized to isolate one cell of the six-well plate at a time, and the align tool was used to position the image so that eggs on the blotting paper were visible. The photos were stacked in time sequence, and a transparent sheet was mounted to the monitor of a desktop computer so that eggs deposited each hour could be marked and labeled with the day and time of egg laying.

An infrared-sensitive camcorder (DNV16HDZ, Bell and Howell, New York, NY) and an infrared illuminator were used to record videos during the scotophase to capture oviposition-associated behavior. Videos of 17 ovipositing females were recorded for 30–60 min during the scotophase. All behavior that appeared to be related to oviposition was noted, along with its duration. A sample video demonstrating observed behavior(s) is provided in the Supplementary Materials.

Data were pooled for each group of 20 female bed bugs of all three strains and analyzed separately using Chi-square tests (α = 0.05). Numbers of eggs laid during the scotophase and the photophase were compared separately for first-clutch and second-clutch females of each strain against expected 50:50 distributions. Frequency distributions of eggs laid during each hour of the 24-h L:D cycle and for each day of the 10-day oviposition duration were compared between first-clutch and second-clutch females using contingency tables. Rows with totals of 0 or 1 were deleted to avoid low expected values. Frequency distributions comparing oviposition by females of all three strains over nine days of the 10-day oviposition duration were also compared using contingency tables, with separate analyses comparing first-clutch and second-clutch females within each strain, and among the three strains for first-clutch females, second-clutch females and both groups together. The LEX-8 strain comparison of first-clutch and second-clutch females on day 1 was deleted because no eggs were laid. For all three colonies, day 10 was deleted because only two eggs were laid. The mean numbers of eggs laid by first-clutch females and second-clutch females among the three colonies were compared using a paired t-test. Data on the duration of the act of oviposition and an apparent egg-marking behavior were plotted as frequency distributions.

Females from the three strains laid strikingly different numbers of eggs both within and between treatments, ranging (for example) from 2.8 per female for first-clutch LEX-8 females to 10.8 per female for first-clutch CIN-1 females (Table 1). When females were grouped by the first or second clutch, the 60 first-clutch females (20 from each strain) laid a mean (± SE) of 124.0 ± 47.7 eggs (2.1 per female) over the 10 days, and the 60 s-clutch females laid a mean of 183.0 ± 40.8 eggs (3.1 per female) over the same duration. A paired t-test revealed no significant differences (t = 2.8892, df = 4, P = 0.1018) between the means. Additionally, the numbers of eggs laid may be slightly underestimated; even though care was taken to prevent females from reaching the underside of the blotting paper oviposition substrate, some females may have found a crevice to reach the underside. Any eggs laid on the reverse side would not have been detected because the camera could only view eggs laid on one side of the oviposition substrate.

Table 1 Comparison of numbers of eggs laid and percentages of eggs laid in the scotophase and the photophase by first clutch and second clutch female bed bugs from three strains. There were 20 females evaluated for each of the six groups

Females from all three colonies laid significantly more eggs in the scotophase than in the photophase. First-clutch and second-clutch females laid means of 89.1% and 80.7% of their eggs, respectively, in the scotophase. The predominance of oviposition during the early hours of the scotophase by females from all three strains is evident (Fig. 1). There was an apparent tendency for second-clutch females to begin oviposition more frequently at the end of the photophase and early in the scotophase than first-clutch females. Still, the frequency distributions of eggs laid per hour between the two groups were significantly different only for CIN-1 (χ2 = 55.6386, df = 17, P < 0.0001) (Fig. 1A) and not for NY-1 (χ2 = 26.7933, df = 19, P = 0.1096) or LEX-8 (χ2 = 11.9259, df = 10, P = 0.32900) (Fig. 1B, C). An anomaly occurred for CIN-1, wherein first-clutch females laid more eggs than second-clutch females in the first hour of the scotophase, but this was offset by higher numbers of eggs laid by second-clutch females in the last two hours of the photophase (Fig. 1A).

Graph: Fig. 1Diel patterns of oviposition by 20 female bed bugs from each of three different strains, A. CIN-1, B. NY-1 and C. LEX-8 over a duration of 10 days under 14:10 L:D. First-clutch females had mated and fed for the first time 24 h before the experiment began and they laid their first clutch of eggs. Second-clutch females had mated, fed and produced their first clutch of eggs seven days prior to the experiment and had taken a second blood meal 24 h before the experiment. The scotophase ran from 1–10 h and the photophase ran from 11–24 h

The frequency distributions of eggs laid by first-clutch and second-clutch females over the 10-day oviposition duration (Fig. 2) were all significantly different (CIN-1, χ2 = 56.9655, df = 8, P < 0.0001; NY-1, χ2 = 151.8624, df = 8, P < 0.0001; LEX-8, χ2 = 14.5025, df = 7, P = 0.0429). Similarly, the frequency distributions among females from the three strains were significantly different for first-clutch females (χ2 = 85.7494, df = 16, P < 0.0001), second-clutch females (χ2 = 170.6020, df = 16, P < 0.0001), and both together (χ2 = 182.0825, df = 16, P < 0.0001). The patterns over time for CIN-1 and NY-1 were similar (Fig. 2A, B), with second-clutch females tending to lay eggs in much higher numbers than first-clutch females on days 1–3, and for the pattern to be reversed at a later time (days 7–10 for CIN-1, and days 5–7 for NY-1). For LEX-8, however, second-clutch females laid more than twice the number of eggs as first-clutch females on days 2–3, and 6–9, and in no case did first-clutch females lay more eggs than second-clutch females (Fig. 2C).

Graph: Fig. 2Daily patterns of oviposition by 20 female bed bugs from each of three strains, A. CIN-1, B. NY-1 and C. LEX-8, over 10 days under 14:10 L:D. First-clutch females had mated and fed for the first time 24 h before the experiment began and they laid their first clutch of eggs. Second-clutch females had mated, fed and produced their first clutch of eggs seven days prior to the experiment and had taken a second blood meal 24 h before the experiment. The scotophase ran from 1–10 h and the photophase ran from 11–24 h

Before oviposition, females exhibited heightened movement while curving their abdominal tip downward. The act of oviposition by 17 NY-1 female bed bugs was rapid, taking a mean duration of 4.65 ± 0.69 s from the time that a female touched the substrate with her abdominal tip to the time that an egg had fully emerged from her body, with a range of 2–12 s (Fig. 3A).

Graph: Fig. 3Frequency distributions for duration of oviposition by 17 bed bug females (A) and apparent egg marking behavior (B) by the same females immediately after oviposition

Following oviposition, all 17 females immediately engaged in apparent egg marking behavior that took a mean duration of 24.06 ± 1.53 s to complete, with a range of 17–41 s (Fig. 3B). Females repeatedly touched their ventral abdominal cuticular surface with their metathoracic tarsi, and then touched the egg with the same tarsi. Concurrently, they rapidly moved their abdominal tip back and forth over the recently laid egg.

We have demonstrated that: female bed bugs oviposit primarily during the scotophase, females that are laying a second clutch of eggs tend to oviposit at the end of the photophase and early in the scotophase at higher levels than females laying their first clutch of eggs, females spend a mean of 4.7 s in the act of oviposition, and that females engage in apparent egg-marking behavior with a mean duration of 24.1 s. Egg-marking behavior has not been previously described in bed bugs.

Commencement of oviposition late in the photophase by females from all three colonies (Fig. 1) indicates that unlike locomotor activity (Romero et al. [31]), oviposition is not induced by the onset of darkness. The results suggest a circadian rhythm (Cury et al. [5]), as demonstrated for locomotor activity in bed bugs (Romero et al. [31]). Whereas induction of locomotor activity by unfed bed bugs (Romero et al. [31]) is a prelude to foraging for a blood meal under cover of darkness, fed and mated females will likely remain in an aggregation harborage where the commencement of oviposition during the photophase would pose minimal danger.

Although females from each strain that fed for a second time laid numerically more eggs in their second clutch than females that fed, mated, and laid their first clutch of eggs (Table 1), there was no significant difference between the size of first and second clutches among females from the three strains. However, more replication and testing of additional strains should be pursued in future studies, given the descriptive differences observed here. Additionally, because of the deleterious physiological effects of repeated traumatic insemination for females (Stutt and Siva-Jothy [34]; Polanco et al. [29]), we hypothesize that with further mating, subsequent egg clutches would continue to gradually decline for mated females.

The daily frequency distributions of eggs laid by females from the three strains (Fig. 2) are in agreement with the 10-day duration of oviposition after feeding found by Polanco et al. ([29]), which in turn is correlated with almost complete digestion of the blood meal (Saveer et al. [32]). However, the differences for first-clutch females, second-clutch females, and both groups among females from all three strains are more pronounced (Fig. 2) than the differences found among three strains of females by Polanco et al. ([29]). The first oviposition by females from CIN-1 and NY-1 strains on the second day after mating and feeding (day 1 in Fig. 2 is the second day after mating and feeding) is in agreement with the timing of the first oviposition found by Johnson ([20]) and How and Lee ([17]), but first oviposition by LEX-8 females on the third day after mating and feeding is the same as found for the three strains tested by Polanco et al. ([29]). The delay of first-clutch females in reaching their maximal oviposition rate compared to second clutch females suggests that first-clutch females are still developing reproductive capacity in the first few days after their initial adult feeding and mating. Similarly, eggs laid soon after a female has fed take longer to hatch than eggs laid later, suggesting that following fertilization within the ovaries [a trait unique to bed bugs (Evison et al. [8])], embryonic development within the ovaries is also not finished by the time the first eggs are laid (Johnson [19]).

In an experiment to determine resistance to combinations of pyrethroid and neonicotinoid insecticides, Gordon et al. ([10]) tested the same strains of CIN-1 and NY-1 bed bugs as used in our study and field-collected LEX-8 that were used to establish a laboratory colony. During several generations in the laboratory, in the absence of insecticide-driven selection pressure, CIN-1 and NY-1 bugs had evolved away from resistance. After 14 days post-exposure to Temprid® (Bayer Crop Science, Research Triangle Park, NC), NY-1 and CIN-1 bed bugs experienced 97.5% and 100% mortality, respectively, while none of the 60 LEX-8 bugs tested died. Studies with bed bugs (Gordon et al. [11]) and other insects such as fruit flies (Homem et al. [15]) or aphids (Jackson et al. [18]) have demonstrated negative consequences on reproductive capacity due to selection for insecticide resistance. We hypothesize that the tradeoff of developing enduring insecticide resistance traits in the LEX-8 strain may have resulted in reduced fecundity (Table 1).

Following observation of apparent egg marking behavior (Fig. 3), an attempt was made to capture the volatiles deposited by females on newly laid single eggs using solid-phase microextraction and coupled gas chromatography-mass spectroscopy (GC–MS) (Crawley [4]), but no volatile chemicals were detected. Subsequently, Zhang et al. ([36]) aerated groups of approximately 50 newly-laid bed bug eggs until hatching, analyzed the captured volatiles using GC–MS and found eucalyptol (aka 1,8-cineole) (1,3,3-trimethyl-2-oxabicyclo[2.2.2]octane) to be the only significant compound identified. (E)-2-Hexenal and (E)-2- octenal, which serve as aggregation pheromone components at low concentrations and as alarm pheromones at high levels (Gries et al. [12]), were not present in captured egg volatiles and eucalyptol was not present in the captured volatiles from containers holding adults, nymphs and feces that emitted (E)-2-hexenal and (E)-2-octenal. It is unknown how and where bed bugs produce eucalyptol, but another hemipteran, the cotton seed bug, Oxycarenus hyalinipennis (Costa) (Hemiptera: Lygaeidae) produces it in the metathoracic scent glands (Olagbemiro and Staddon [28]).

Insects may mark their eggs with repellent or antimicrobial compounds that protect them from predation, parasitism, or infection (Kaltenpoth et al. [21]; Matsuura et al. [24]; Hosoe et al. [16]; Newcombe et al. [27]). Prokopy ([30]) cites several instances in which adult insects mark their eggs or the oviposition site with a repellent pheromone that reserves a scarce food resource or habitat for the immature insect that ecloses from the egg. Eucalyptol is found in many plants and insects and may serve as an insect repellent or attractant [eucalyptol (pherobase.com)]. However, against Rhodnius prolixus nymphs, eucalyptol was found to be 3.5 × less toxic than Vapona (an organophosphate insecticide) and was not repellent (Sfara et al. [33]). Because bed bug eggs bearing eucalyptol are found in protective aggregations, and eucalyptol may enhance an aggregation's protective function through its antimicrobial activity (Mączka et al. [23]), it is more likely that eucalyptol would serve as an aggregation rather than a repellent pheromone for bed bugs. Further investigation is needed to determine if eucalyptol, or other compounds, are transferred to bed bug eggs during apparent egg-marking behavior and the biological function of the hypothesized egg-marking pheromone.

We thank Scott Bessin for technical assistance, primarily with the analysis of photographs. We thank Josiah Ritchey for verification of statistical analyses and approaches.

Sydney Crawley: conceptualization, methodology, investigation, data curation, formal analysis, writing, and editing. Kenneth Haynes: conceptualization, methodology, formal analysis, supervision, review and editing. John Borden: formal analysis, writing, review and editing.

This work was supported by the USDA National Institute of Food and Agriculture as an approved Hatch project of the Kentucky Agricultural Experiment Station.

Data are on file with the senior author and are available on request.

Not applicable.

The research reported herein did not require ethical approval.

All authors of this manuscript have read the pre-submission version, agree that it be submitted for publication in the Journal of Insect Behavior, and assert that neither the manuscript nor any of its components have been submitted to, accepted by or published in another journal.

The authors declare no competing interests.

The authors have no conflicts of interest that pertain to this paper.

Below is the link to the electronic supplementary material.

Graph: Supplementary file1 (WMV 7709 KB)

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.