The effects of fluctuating temperatures on degree-day development and life history parameters of Pseudacysta perseae (Hemiptera: Tingidae)

Abstract

Pseudacysta perseae (Heidemann 1908) (Hemiptera: Tingidae) is a foliar pest of avocados. The effects of 6 fluctuating temperature regimens, which averaged 15, 20, 25, 30, 32, and 35 °C over a 24-h period, on the developmental and reproductive biology of P. perseae were investigated. Selected temperature cycles are representative of avocado production regions in California (US). Fluctuating temperature regimens had significant effects on P. perseae development times, fecundity, fertility, longevity, and survivorship rates. One linear model (Ordinary Linear) and 7 nonlinear regression functions (Beta, Brière-2, Lactin-2, Lobry–Rosso–Flandrois, Performance-2, Ratkowsky, and Weibull) were utilized to investigate the correlation between fluctuating temperature profiles and P. perseae development times. The Beta and Weibull models failed to converge. Model parameters, T_min, T_opt, and T_max, were estimated as 1.72 to 9.78 °C, 31.04 to 31.57 °C, and 34.05 to 39.38 °C, respectively. The thermal requirement for development, K, was estimated as 476.19 degree-days. At 32 °C, P. perseae females exhibited 4 egg-laying peaks around days 11, 35, 54, and 63 of life. A maximum daily average of eggs laid (i.e., fecundity) was 6.07 on day 35 and the average daily egg-laying rate was 3.08 eggs over a 69-day span. The maximum proportion of eggs that hatched (i.e., fertility) was 0.49 on day 31, and the average daily proportion of hatched eggs was 0.10. This study confirmed that P. perseae passes through 4 nymphal instars, not 5 as previously reported. In addition, sexual dimorphism with respect to the coloration of the fourth antennal segment is documented and is substantially darker in adult males.

Graphical Abstract

Graphical Abstract.

Introduction

Pseudacysta perseae (Heidemann 1908) (Hemiptera: Tingidae: Tingini), known locally as avocado lace bug, has a host range that is restricted to plant species in the family Lauraceae, which includes avocados (Persea americana Miller) (Hoddle 2004). The presumed native range of P. perseae is thought to incorporate parts of the southeastern United States (e.g., Florida, Georgia, and Texas), the Caribbean, México, and possibly Central and South America (Heidemann 1908, Mead and Peña 1991). Pseudacysta perseae was first found in California (US) in 2004, infesting residential trees, primarily non-Hass avocado varieties, in southern San Diego County, and was considered a minor pest problem as it failed to spread into commercial avocado production areas which are dominated by the Hass variety (Hoddle et al. 2005, Humeres et al. 2009a, b). In California, the Hass cultivar accounts for ~95% of the crop and commercially grown avocados were valued at $458 million (US) in 2022 (CAC 2023).

Subsequent to the original 2004 detection in California, P. perseae has undergone extensive local and global range expansion. In 2017, P. perseae outbreaks were reported for the first time in commercial Hass avocado orchards in northern San Diego and Riverside Counties in California. This pest subsequently spread to Los Angeles (2019), Orange (2022), and Santa Barbara (2023) Counties in California (Hoddle 2022a, b). In 2014, P. perseae was found on the island of Madeira, an autonomous region of Portugal located off the northwestern coast of Africa (EPPO 2015). In 2019, P. perseae were found infesting avocado groves on the Hawaiian Islands (US) of O’ahu, Maui, Hawaiʻi, and Kaui (Matsunaga and Silva 2022). Pseudacysta perseae was detected on avocados in 2022 in Gabon, Africa (Poligui et al. 2023). With respect to California, molecular analyses indicated that the original invading population detected in 2004 and identified as haplotype A, likely originated from Nayarit in México (Rugman-Jones et al. 2012). Subsequent analyses using P. perseae specimens collected post-2017 from newly invaded regions of California and Hawaiʻi indicated that a second mitochondrial haplotype, G, was prevalent. This finding indicated that California had experienced a second invasion event, possibly from Florida (US), and suggested that California, given its closer proximity to Hawaiʻi when compared to Florida, may have acted as a bridgehead for the Hawaiʻian invasion (Dadlani et al. 2025).

Female P. perseae lay eggs on the underside of leaves and eggs are often covered with a protective black “tar-like” secretion (Hoddle et al. 2005). Adult and nymphal P. perseae feed on the undersides of mature leaves. It is widely assumed that P. perseae has 5 instars (Heidemann 1908, Morales Romero et al. 2002, Peña et al. 2012, Guidoti et al. 2015), but this assumption lacks adequate documentation. Economic damage to avocado leaves results from feeding injury caused by adults and nymphs which often manifests as large necrotic islands in the central regions of leaves. Damage may be amplified by pathogenic fungi, such as Colletotrichum spp., that infest leaves through feeding wounds (Mead and Peña 1991). Consequently, high-density P. perseae infestations cause economic damage by injuring foliage which results in reduced photosynthetic capacity, and premature abscission of heavily damaged leaves exposes immature fruit to sun, which can result in fruit loss due to sunburn. Economic losses are further increased when applications of insecticides are made to reduce high population densities to less damaging levels (Peña et al. 2012, Hoddle et al. 2005, Humeres et al. 2009b, Byrne et al. 2010).

Abiotic environmental factors, particularly temperature, influence the development, behavior, reproduction, mortality, phenology, and population dynamics of insects (Pedigo 1989, Hallman and Denlinger 1998, Neven 2000, Nelson et al. 2013). Insects are poikilothermic and important life history functions, like metabolic rates, are affected by ambient temperature (Neven 2000). Limited information is available on the effects of temperature on P. perseae developmental and reproductive biology. In the Dominican Republic, P. perseae development from egg to adult on Hass avocado leaves required 22 days at an average outdoor temperature of 26 ± 2 °C (Antun 1991). In the laboratory, the development times of immature P. perseae were quantified across 5 (i.e., 20, 22, 25, 28, and 30 °C) temperatures that remained constant over a 24-h period (Morales Romero et al. 2000). Under these static temperature conditions, the life cycle of P. perseae from egg to adult spanned 21 (at 30 °C) to 42 (20 °C) days (Morales Romero et al. 2000). The minimum temperature above which development occurred under these conditions was identified as ~9 °C (Morales Romero et al. 2000). However, the optimal and upper-development threshold temperatures, above which insects are unable to develop, were not quantified by Morales Romero et al. (2000).

Insects can tolerate a wide range of temperatures, and under field conditions, over any given 24-h period, insects may be exposed to temperatures that can positively or negatively influence survivorship, development, and reproductive rates (Hallman and Denlinger 1998, Hoddle et al. 2023). With respect to avocado orchards, prevailing ambient outdoor temperatures are not constant and P. perseae encounter temperature cycles that fluctuate between daily highs and nighttime lows over any given 24-h period. Consequently, fluctuating temperature variations may affect P. perseae development times, survivorship rates, and fecundity when compared to similarly collected data obtained under constant temperature regimes (McCalla et al. 2019, Milosavljević et al. 2019, 2020, Hoddle et al. 2023). To date, no studies have investigated the effects of fluctuating temperature cycles which are characteristic of avocado production regions in California on the life history traits (e.g., development times) of P. perseae. To address this shortcoming, this study investigated the effects of 6 fluctuating temperature profiles that are representative of climatic conditions across avocado production areas in southern California on selected life history parameters of P. perseae. The intent of this work was to improve understanding of the effects of temperature on P. perseae developmental and reproductive biology and to provide the necessary baseline information for the development of degree-day models. Results of work reported here on the influence of fluctuating temperatures on P. perseae development enabled the identification of lower, optimal, and upper thermal tolerances and degree-day accumulation needed to complete development and associated temperature effects on reproductive biology.

Materials and Methods

Source of Experimental Pseudacysta perseae

Pseudacysta perseae adults used for experiments were collected from Hass avocado orchards in northern San Diego County, California. Adult males and females were confined in Munger cells (Munger 1942) on Hass avocado leaves upon which they fed and oviposited. Leaves were collected from unsprayed Hass avocado trees on the UC Riverside campus and were thoroughly washed and dried before use. Munger cells were maintained in a climate-controlled cabinet at a fluctuating temperature that averaged 30 °C, 60 ± 5% relative humidity, and a 14:10 h light: dark photoperiod. The progeny of field-collected adult P. perseae maintained under these controlled laboratory conditions were used in experiments detailed below.

Experimental Rearing Set-Up and Temperature Cabinet Programing

Pseudacysta perseae reared on Hass avocado leaves were maintained in an observation arena that was constructed using a modified Munger cell set-up (see Supplementary Fig. 1 for the set up of these experimental arenas). To create the observation arena, earthquake putty (QuakeHold Museum Putty, Ready America, San Marcos, CA, USA) was used to secure a Hass avocado leaf to a central Munger cell plate (7.5 cm wide × 10.5 cm long) made of plexiglass, with a central 3 cm diameter hole and two 6 mm diameter aeration holes drilled into the wall of the arena. Aeration holes were covered with fine metal mesh to prevent insect escape. The central 3 cm hole allowed access to the avocado leaf on which P. perseae were placed. This central portion of the Munger cell with P. perseae located on the exposed avocado leaf was sealed with a clear plexiglass cover (7.5 × 10.5 cm) to prevent escape. Munger cells attached to avocado leaves with P. perseae nymphs were placed on water-saturated foam pads contained within a stainless steel pan (25 cm × 22 cm × 4 cm) to retain water. Assays with egg-bearing leaves were left uncovered to prevent condensation and excessive fungus growth. As leaves deteriorated, P. perseae nymphs and adults were transferred using a 0.5 mm camel-hair paintbrush to new Munger cells with fresh Hass avocado leaves.

Temperature-driven development studies were conducted at 6 fluctuating temperature profiles that averaged 15, 20, 25, 30, 32, and 35 °C over a 24-h period. To achieve the target average temperatures of 15, 20, 25, and 30 °C, the oscillating temperature cycle used to program climate-controlled cabinets (Darwin Chambers, St. Louis, MO, USA) was modeled using an average of 5 years (1 January 2017 to 31 December 2021) of hourly daily temperature data from the California Irrigation Management Information System (CIMIS) weather station, Escondido SPV #153 in San Diego County, California, USA. For average temperatures of 32 and 35 °C, daily temperature data were obtained from the CIMIS weather station, Borrego Springs #207 in the Imperial/Coachella Valley region in San Diego County for the same 5-year period. These temperature data were used to program the climate-controlled cabinets with incremental hourly temperature changes over a 24-h period to produce each target mean temperature (Table 1). Humidity in cabinets was maintained at 60 ± 5% RH which was the average relative humidity recorded in San Diego County (CIMIS 2021). Photoperiod was set at 14:10 h (L:D) with a light intensity of 100 μE m⁻² s⁻¹ for every experimental temperature profile. To ensure the desired experimental temperature profiles and relative humidity were achieved, HOBO Pro V2 Temperature/RH loggers (Onset Computer, Pocasset, Massachusetts, USA) were programmed to record temperature and humidity at 30-min intervals inside the climate-controlled cabinets throughout the duration of the experiment. These data were used to confirm temperature cabinets were performing the programmed temperature cycles.

Table 1.

Stepwise hourly temperature ramping in temperature cabinets used for fluctuating temperature regimens for assessing the effects of temperature on Pseudacysta perseae development and reproductive biology

Hour	Mean temperature (°C)						Photoperiod
	15	20	25	30	32	35
0100	11.4	15.3	19.5	25.2	23.9	28.4	Dark
0200	11.1	15	19.2	24.0	23.2	27.7
0300	10.7	14.8	18.8	24.0	22.3	26.8
0400	10.3	14.5	18.4	23.3	21.5	26.0
0500	10.0	14.2	18.0	22.7	21.5	26.2
0600	9.8	14.3	17.8	23.0	22.7	26.6	Light
0700	10.2	15.4	19.2	24.9	27.7	30.0
0800	12.0	17.6	21.5	27.2	32.1	33.2
0900	14.5	20.4	24.6	31.1	34.6	35.2
1000	17.2	23.1	28.0	33.2	36.1	36.8
1100	19.4	25.6	30.6	35.6	37.3	38.1
1200	21.1	27.1	32.8	36.4	38.3	39.1
1300	21.7	27.8	33.9	37.4	39.2	40.2
1400	21.8	27.7	33.6	38.0	39.9	41.0
1500	21.3	27.1	32.6	37.2	40.3	41.6
1600	20.5	26.1	31.5	34.6	40.4	42.0
1700	19.1	24.7	30.0	33.7	39.9	41.9
1800	17.2	22.4	28.2	32.0	39.2	41.2
1900	15.5	20.1	26.0	31.0	37.0	39.2
2000	14.3	18.6	23.8	30.0	34.8	37.3	Dark
2100	13.7	17.4	22.9	29.3	32.4	35.6
2200	13.1	16.6	22.0	27.0	30.6	34.1
2300	12.6	16.2	21.3	26.0	28.3	32.1
2400	11.9	15.8	20.4	25.0	26.1	30.2
Total steps	24	24	24	24	24	24

Quantification of Number of Nymphal Instars and Photo-Documentation of P. perseae Life Stages

It has been widely assumed that P. perseae has 5 nymphal instars (Heidemann 1908, Morales Romero et al. 2002, Peña et al. 2012, Guidoti et al. 2015), but there is no robust documentation supporting this assumption. To confirm or refute the assumption that P. perseae has 5 instars, the number of P. perseae nymphal instars was quantified at 30 °C. The transition from first instar to second instar, second instar to third instar, and third instar to fourth instar was monitored for 20 specimens under a microscope every 12 to 14 h. A clear plastic 24-well (16 mm in diameter; 3.5 ml volume) cell culture plate (12 cm × 18 cm × 1.2 cm with lid) (Falcon 351147, Corning Life Sciences, Union City, CA) was utilized for these studies. Circular Hass avocado leaf disks, excised from whole leaves with a 16 mm cork borer were placed on the floor of each well with the lower leaf side facing up. Each well (n = 24) was inoculated with 1 first instar nymph and leaf disks were replaced every 3 days. Observation of an ecdysed exoskeleton indicated when a specimen had advanced to the next instar.

To visually document the life stages and sexual dimorphism of adult P. perseae, live eggs, nymphs (instars 1 to 4), and adult males and females, were digitally photographed in the laboratory at 2× magnification using a digital camera (Canon 90d DSLR; Macro Twin Lite MT-24EX with the Canon MP-E 65 mm lens). Digital photographs were edited using Adobe Photoshop and annotated on Adobe Illustrator (Adobe Systems Inc., San Jose, CA). The most extensive description of P. perseae life stages is provided by Heidemann (1908) and this digital imaging work was performed to supplement these written descriptions and line drawings.

Temperature-Driven Development Experiments

For each experimental fluctuating temperature regimen (Table 1), a total of ~30 field-collected adult male and female P. perseae adults (~1 to 2 weeks of age) were set up in Munger cells and left to mate and oviposit on Hass avocado leaves. Each day adults were transferred to new leaves and freshly oviposited eggs were maintained on labeled leaves that were kept on water-saturated foam pads held in stainless steel pans and observed daily for nymph eclosion under a microscope at 10× to 40× magnification. Leaves were discarded once no further nymphs emerged for a continuous period of 5 days. Emerged P. perseae nymphs (referred to as G₀) were placed individually into Munger cells with Hass avocado leaves and development was monitored daily until adulthood. The duration of nymphal development was recorded for each instar, which was confirmed by the presence of discarded exoskeletons. Nymphs that died prematurely due to unnatural causes (e.g., trapped in a water droplet or adhered to quake putty) were excluded from data analyses. Adults were sexed under a dissecting microscope, and mating pairs consisting of 1 male and 1 female were established to measure the lifetime fecundity and longevity of adults. For each experimental temperature, 10 to 15 mating pairs were set up and followed until death. In the event of adult mortality, a new male or female was introduced as a replacement. This arrangement ensured that both males and females had the chance to mate with alternative individuals throughout their entire natural lifespan. Replacement adults were not used for data analyses and adult longevity was assessed every 24 h.

To ensure adequate sample sizes for development studies conducted at 15 and 35 °C, P. perseae eggs (i.e., G₁) laid at, or first instar nymphs (< 12 h of age) that emerged at 32 °C, were transferred to the 15 and 35 °C temperature cabinets. This departure from the preceding experimental protocol described above was necessary to obtain nymphs for development studies as oviposition either did not occur (15 °C) or nymphs failed to emerge from oviposited eggs (35 °C) at these lower- and upper-temperature extremes.

Daily and Lifetime Fecundity and Fertility Estimates

Female fecundity (i.e., number of eggs laid) was recorded daily at a fluctuating average temperature of 32 °C until female death from natural causes occurred. This temperature was selected for daily and lifetime fecundity studies as it was the optimal temperature regimen for the development of P. perseae (see Results section). Adults utilized for this fecundity experiment were reared to adulthood at 32 °C. Twenty-three mating male–female pairs (< 24 h of age) that emerged at 32 °C were set up in 23 Munger cells with Hass avocado leaves as the feeding and oviposition substrate. The daily number of eggs laid by each female was recorded from the first day of adulthood until the day of death. A digital photograph of the observation arena enclosing the avocado leaf was captured using a microscope-mounted camera at 2.5× magnification. Labeled (i.e., mating pair identifier and date) digital images were used for mapping of eggs and determining developmental fate (i.e., hatching status). In the event of male mortality, a new male was introduced as a replacement, ensuring that females had the opportunity to continuously mate with males throughout their entire lifespan. Hass avocado leaves in Munger cells with eggs were removed, labeled, and replaced with fresh leaves every 6 to 8 days. Labeled leaves removed from Munger cells were cross-referenced to their corresponding digital maps and eggs were examined daily under a microscope at 10× to 40× magnification for nymph emergence. Nymph emergence from individual eggs, identified by an open operculum, laid on specific days was determined from digital photos with mapped eggs. Daily observations of Hass leaves with eggs continued for 15 days from the date of the last egg that hatched. Average daily egg oviposition rates, lifetime fecundity (i.e., total number of eggs laid), egg hatch rates (i.e., fertility), and longevity in days for each experimental female were determined.

Statistical Analyses of Preimaginal Developmental, Adult Longevity Times, and Female Fertility Data Across Fluctuating Temperature Regimens

SAS (SAS Statistical Analysis Software version 9.4, 2013) was used for all statistical analyses reported here. Generalized linear mixed models (GLMMs) and the PROC GLIMMIX procedure were used to evaluate the correlation between temperature and development duration times in G₁P. perseae offspring. All models included fixed effects such as the fluctuating temperature profile (15 to 35 °C), sex, and their interaction. Separate models were created for each developmental stage, encompassing P. perseae eggs, first, second, third, and fourth instar nymphs, along with combined male and female egg-to-adult development and adult longevity. Only data from individuals who reached adulthood and sexed as males or females were included in analyses for examined variables. Individuals who did not reach adulthood were omitted from analyses. A nested structure was included in all models to address potential statistical dependencies, this approach incorporated G₁P. perseae within the G₀ parent identity and temperature profile (Hoddle et al., 2023). Poisson distributions were used across all models, as determined by the variances of the response variables.

For the 35 °C data, a substantial proportion (0.97) of eggs laid at this temperature failed to hatch, and nymphs that emerged typically died shortly after eclosing. Consequently, a subgroup of first instars, which comprised a total of 37 nymphs, that emerged from eggs laid at 32 °C were moved to 35 °C (see above for details) for data collection on development times at 35 °C for each stage, overall development times, and adult longevity. To justify this approach, egg hatching times for individuals that hatched at 35 °C but subsequently died (hereafter referred to as 35-to-35 °C hatching times) were initially compared with both egg hatching times for individuals that hatched at 32 °C and were reared to adulthood at 32 °C (hereafter referred to as 32-to-32 °C hatching times) and egg hatching times for individuals that hatched at 32 °C and were transferred to 35 °C (hereafter referred to as 32-to-35 °C hatching times). Utilizing GLMMs and the PROC GLIMMIX procedure in SAS (SAS version 9.4, 2013), the analysis incorporated a fluctuating temperature profile (32-to-32 °C, 32-to-35 °C, and 35-to-35 °C) as a fixed variable, with egg hatching times in days as the dependent variable. A nested structure was incorporated into the model to address potential statistical dependencies, involving G₁P. perseae eggs within the G₀ parent identity and temperature profile (Hoddle et al., 2023). Hatching times were modeled with a Poisson distribution as dictated by the variance of the response variable. No significant differences in egg hatching times were observed among 32-to-32 °C, 32-to-35 °C, and 35-to-35 °C conditions (see Results section below). Consequently, for the comprehensive models covering temperatures from 15 to 35 °C, calculations for 32 °C utilized 32-to-32 °C egg hatching times, and 35 °C utilized 32-to-35 °C egg hatching times.

Similarly, GLMMs and the PROC GLIMMIX procedure in SAS (SAS version 9.4, 2013) were used to evaluate the effect of varying temperatures on both the G₂ eggs laid by G₁ female P. perseae and the resulting percentage of hatched eggs. Considering the fluctuating temperature profile (excluding 35 °C, where no eggs were laid and no hatch occurred) as a fixed effect in all models, G₂ egg counts followed a negative binomial distribution, while the resulting percentage of hatched eggs was modeled using a binomial distribution. In addition, the model for egg counts incorporated G₁ female longevity and the interaction of egg counts with temperature as covariates to account for potential influences on the outcomes. Conversely, the model for the resulting percentage of hatched eggs omitted these covariates and focused solely on the fixed effect of temperature. A nested structure was integrated into all models to address potential statistical dependencies, incorporating G₁P. perseae females within the G₀ parent identity and temperature profile (Hoddle et al., 2023).

In all GLMMs utilized, temperature was categorized instead of treated as a continuous variable due to its highly nonlinear relationship with each analyzed variable (Streiner 2002, Pasta, 2009, McCalla et al. 2019, Milosavljević et al. 2019). Significant main effects were verified through pairwise comparisons using the least-squared means option (utilizing SAS GLIMMIX procedure and LSMEANS statement [SAS version 9.4, 2013]), adjusting for multiple comparisons using the Tukey–Kramer method at a significance level below 0.05. Importantly, no significant differences in egg-to-adult development times were observed between G₁ females and G₁ males (refer to Tables 2 and 3 in the Results section), and the values for G₁ males and G₁ females were combined for subsequent model fitting.

Table 2.

Results from generalized linear mixed models analyzing the effects of sex, temperature (characterized by fluctuating temperature regimens averaging 15, 20, 25, 30, 32, or 35 °C, over a 24-h period), and their interactions on the development times of first-generation (G₁) Pseudacysta perseae eggs (A), along with first, second, third, and fourth instar larvae (B–E), combined egg-to-adult transition (F), and adult longevity (G) when subjected to fluctuating temperature regimes

(A) Eggs (development)	Num df	Den df	F	P
Sex (S)	1	146	0.06	0.8082
Temperature (T)	5	146	63.80	<0.0001*
S × T	5	146	0.11	0.9897
(B) First instar larvae (development)	Num df	Den df	F	P
Sex (S)	1	146	0.18	0.6700
Temperature (T)	5	146	82.48	<0.0001*
S × T	5	146	0.05	0.9983
(C) Second instar larvae (development)	Num df	Den df	F	P
Sex (S)	1	146	0.01	0.9850
Temperature (T)	5	146	66.77	<0.0001*
S × T	5	146	0.18	0.9700
(D) Third instar larvae (development)	Num df	Den df	F	P
Sex (S)	1	146	0.03	0.8723
Temperature (T)	5	146	74.62	<0.0001*
S × T	5	146	0.11	0.9899
(E) Fourth instar larvae (development)	Num df	Den df	F	P
Sex (S)	1	146	0.02	0.8765
Temperature (T)	5	146	52.88	<0.0001*
S × T	5	146	0.14	0.9826
(F) Egg-to-adult (development)	Num df	Den df	F	P
Sex (S)	1	146	0.04	0.8388
Temperature (T)	5	146	324.32	<0.0001*
S × T	5	146	0.09	0.9932
(G) Adults (longevity)	Num df	Den df	F	P
Sex (S)	1	143	0.01	0.9796
Temperature (T)	5	143	89.67	<0.0001*
S × T	5	143	0.41	0.8420

^*Indicates significance at a level below 0.05.

Table 3.

Mean development times (mean days ± SE) for first-generation (G₁) Pseudacysta perseae eggs, along with first, second, third, and fourth instar larvae, combined egg-to-adult transition, and adult longevity, were observed under 6 fluctuating temperature regimes that averaged 15, 20, 25, 30, 32, or 35 °C, over a 24-h period

Temp. (°C)	Average duration times (mean days ± SE)
	Eggs		First instar larvae		Second instar larvae		Third instar larvae		Fourth instar larvae		Eggs-to-adults		Adults
	Female	Male	Female	Male	Female	Male	Female	Male	Female	Male	Female	Male	Female	Male
15	24.08 ± 0.56a [13]	24.73 ± 0.44a [15]	12.38 ± 0.38a [13]	12.73 ± 0.38a [15]	10.61 ± 0.31a [13]	11.07 ± 0.37a [15]	11.31 ± 0.28a [13]	11.41 ± 0.29a [15]	12.61 ± 0.46a [13]	12.21 ± 0.38a [15]	71.07 ± 0.97a [13]	72.13 ± 0.72a [15]	69.61 ± 4.71a [13]	64.87 ± 2.96a [15]
20	20.08 ± 0.33b [13]	19.27 ± 0.33b [11]	11.38 ± 0.38a [13]	12.09 ± 0.37a [11]	10.92 ± 0.33a [13]	11.27 ± 0.49a [11]	12.15 ± 0.46a [13]	12.09 ± 0.46a [11]	10.08 ± 0.33a [13]	10.01 ± 0.49a [11]	64.61 ± 0.62b [13]	64.73 ± 0.88b [11]	39.54 ± 2.03b [13]	42.54 ± 3.14b [11]
25	11.06 ± 0.17c [16]	10.81 ± 0.21c [16]	2.94 ± 0.17b [16]	3.31 ± 0.15b [16]	3.63 ± 0.18b [16]	3.69 ± 0.18b [16]	3.69 ± 0.27b [16]	3.56 ± 0.16b [16]	5.19 ± 0.19b [16]	5.06 ± 0.14b [16]	26.51 ± 0.24c [16]	26.44 ± 0.22c [16]	26.56 ± 1.44c [16]	29.51 ± 1.26c [16]
30	11.38 ± 0.22c [16]	11.53 ± 0.24c [15]	3.51 ± 0.16b [16]	3.61 ± 0.19b [15]	2.75 ± 0.14b [16]	3.13 ± 0.21b [15]	2.94 ± 0.14b [16]	3.07 ± 0.21b [15]	3.56 ± 0.18b [16]	4.02 ± 0.29b [15]	24.13 ± 0.41c [16]	25.33 ± 0.47c [15]	27.38 ± 2.81c [13]	28.41 ± 3.21c [15]
32	9.93 ± 0.27c [15]	10.29 ± 0.22c [17]	2.27 ± 0.12b [15]	2.29 ± 0.14b [17]	2.81 ± 0.11b [15]	2.47 ± 0.17b [17]	2.53 ± 0.13b [15]	2.76 ± 0.18b [17]	3.73 ± 0.12b [15]	3.41 ± 0.21b [17]	21.27 ± 0.27d [15]	21.23 ± 0.35d [17]	22.87 ± 2.52c [15]	23.35 ± 1.14c [17]
35	10.01 ± 0.36c [6]	10.61 ± 0.61c [5]	3.51 ± 0.43b [6]	3.41 ± 0.51b [5]	3.33 ± 0.21b [6]	3.01 ± 0.32b [5]	3.83 ± 0.31b [6]	3.21 ± 0.21b [5]	4.33 ± 0.21b [6]	4.21 ± 0.21b [5]	25.01 ± 0.58c. [6]	24.41 ± 0.81c [5]	5.51 ± 1.99d [6]	4.61 ± 1.47d [5]
P	< 0.0001		< 0.0001		< 0.0001		< 0.0001		< 0.0001		< 0.0001		< 0.0001

Means sharing the same column and marked with the same letter were not found to be significantly different at α = 0.05 (LSMEANS), number within square brackets indicates the count of individuals. Numbers in square brackets [] indicate the number of individuals in study.

Fitting of Models to Temperature-Driven Pseudacysta perseae Development Data

One linear model (Ordinary Linear [Campbell et al., 1974]) and 7 nonlinear regression functions (Beta [Yin et al. 2003, Auzanneau et al. 2011, Shi et al. 2015], Brière-2 [Brière et al. 1999], Lactin-2 [Logan et al. 1976, Lactin et al. 1995], Lobry–Rosso–Flandrois [Lobry et al. 1991, Rosso et al. 1993], Performance-2 [Shi et al. 2011, Wang et al. 2013], Ratkowsky [Ratkowsky et al. 1983], and Weibull [Angilletta 2006]) were utilized to investigate the correlation between fluctuating temperature profiles and P. perseae development times (refer to McCalla et al. 2019, Milosavljević et al. 2019, and Table 5 for model equations). All models were fitted to data spanning the temperature range 15 to 35 °C, using egg-to-adult development rates (1/d), which are the reciprocals of mean egg-to-adult development times (days⁻¹) for G₁P. perseae (i.e., combined male and female development data). Among the 7 nonlinear models examined, the Beta and Weibull models failed to converge and were therefore unable to fit to G₁P. perseae development rates. Consequently, the Beta and Weibull models were excluded from analyses.

Table 5.

Mathematical models, parameter estimates, and goodness-of-fit metrics for 6 performance functions describing the relationship between temperature and development rates (D_r) of first-generation (G₁) Pseudacysta perseae (consolidating male and female data) reared under 6 fluctuating temperature profiles, averaging 15, 20, 25, 30, 32, or 35 °C, over a 24-h period

Model	Model equation†	Parameter	Parameter estimate	Reference
Ordinary Linear‡	$D_r=a+bT$	a b K (degree-days)‡ Tmin‡ R2adj	−0.0195 0.0021 476.19 9.29 0.9109	Campbell et al. (1974)
Lactin-2 (Logan-Lactin)	$D_r=\lambda +e^{\rho T}-e^{(\rho T_u-(T_u-T)/\delta T)}$	λ ρ δ Tmin* Tu (Tmax*) Topt* RSS	−1.0217 0.0022 2.1158 9.77 37.05 31.57 0.000019	Logan et al. (1976) Lactin et al. (1995)
Brière-2	$D_r=aT(T-T_{min}){(T_{max}-T)}^{1/b}$	a b T min T max T opt * RSS	0.000046 5.346 4.94 34.05 31.39 0.000032	Brière et al. (1999)
Lobry–Rosso–Flandrois (LRF)	$D_r={\mu }_{opt}\frac{(T-T_{max}){(T-T_{min})}^2}{(T_{opt}-T_{min})[(T_{opt}-T_{min})(T-T_{opt})-(T_{opt}-T_{max})(T_{opt}+T_{min}-2T)]}$	μ opt T min T max T opt RSS	0.0455 2.94 37.14 31.51 0.000016	Lobry et al. (1991) Rosso et al. (1993)
Performance-2	$D_r=b(T-T_{min})(1-e^{c(T-T_{max})})$	T min T max T opt * b c RSS	9.87 36.09 31.04 0.0023 0.4435 0.000019	Shi et al. (2011) Wang et al. (2013)
Ratkowsky	$\begin{array}{lll}& {\displaystyle \sqrt{D_r}=b(T-T_{min})(1-e^{c(T-T_{max})});}& {\displaystyle D_r={\left[b(T-T_{min})(1-e^{c(T-T_{max})})\right]}^2}\end{array}$	b c T min T max T opt * RSS	0.0082 0.2596 1.72 39.38 31.11 0.000015	Ratkowsky et al. (1983)

^†In every model, T represents the temperature in degrees Celsius, and D_r signifies the development rate at temperature T; in the Ratkowsky, Lactin-2, and Performance-2 models, e signifies the base of the natural logarithms; other symbols represent specific model parameters, and a comprehensive description of models and parameters can be found in the respective references.

^‡Linear regression was used to calculate the theoretical minimum developmental threshold (T_min = − a/b, with a representing the development rate at T = 0 °C and b as the slope), along with the thermal constant or degree-days required for development completion (K = 1/b); temperatures at which development rate of Pseudacysta perseae larvae deviated from rectilinearity were omitted from analyses (i.e., development rate at 35 °C).

^*In all nonlinear models, with the exception of the LRF model, the optimal development point (T_opt) was derived from the peak of the development curve (where D_r = max); in the Lactin-2 model, the theoretical minimum and maximum developmental thresholds (T_min and T_max) were determined at the points where the curve intersected the temperature axis (where D_r = 0).

Linear regression (Campbell et al., 1974) was conducted using the PROC REG procedure in SAS to determine the degree-days necessary for egg-to-adult development of G₁P. perseae (i.e., K, used combined male and female data) and the theoretical lower development threshold (i.e., T_min) under fluctuating temperature profiles. Parameter K was calculated as the reciprocal of the line slope, and T_min was obtained by solving for y = 0 for the fitted regression equation (Campbell et al. 1974). Linear model goodness of fit was assessed using R-squared adjusted (R²_adj), where an R²_adj > 0.9 indicated a satisfactory fit to the data (see Table 5; [Campbell et al. 1974]). Cook’s D metric was utilized to identify and exclude influential outliers from the analysis. Outliers were categorized as observations with D > 4/number of observations as being overly influential (Bollen and Jackman 1990).

The PROC NLIN procedure in SAS (SAS version 9.4 2013) was applied to G₁P. perseae egg-to-adult development rates to parameterize nonlinear models (Shi and Ge, 2010). All 5 tested nonlinear models featured 4 parameters, which resulted in identical degrees of freedom (df) (Mirhosseini et al. 2017, Ratkowsky and Reddy 2017). Consequently, the assessment of goodness of fit for the nonlinear models relied on the residual sum of squares (RSS), expressed as:

$${\displaystyle \begin{array}{l}{\displaystyle RSS={\displaystyle \sum _{i=1}^n}{(y_i-{{\displaystyle \widehat{y}}}_i)}^2}\end{array}}$$

where, n denotes the number of observations, and y_i and ŷ_i represent observed and expected development rates at ith temperature, respectively. Nonlinear models with the smallest RSS values were indicative of a superior fit to datasets (Shi et al. 2015, Ratkowsky and Reddy 2017, McCalla et al. 2019, Milosavljević et al. 2019, 2020). R²_adj was not employed to evaluate the goodness of fit for nonlinear models as it inaccurately characterizes the validity of a nonlinear fit to data (Spiess and Neumeyer 2010).

All temperature-based development models were fitted by considering the development rate as the response variable which satisfied homogeneity assumptions (Ratkowsky 2004). No additional transformations were required for the Brière-2, Lactin-2, LRF, and Performance-2 expressions, and each model was subjected to least squares estimation in its original form (Ratkowsky 2004). In the case of the square-root model (i.e., the Ratkowsky model), both sides of the equation were squared, ensuring that the left-hand side represented the development rate and not the square root of the rate (Shi and Ge 2010). All model outputs were visualized in SigmaPlot (SigmaPlot Version 12.3, 2013).

Results

Effects of Temperature on G₁P. perseae Development Times

No significant differences were found in egg hatching times among 32-to-32 °C, 32-to-35 °C, and 35-to-35 °C rearing conditions (F = 0.98; df = 2,47; P = 0.39). Consequently, for the models fitted to development times for G₁P. perseae eggs, nymphal instars, and combined egg-to-adult development (this section), and adult longevity (see section below) at temperatures ranging 15 to 35 °C, calculations for 32 °C used 32-to-32 °C egg hatching times, and calculations for 35 °C used 32-to-35 °C egg hatching times, as outlined in the statistical analysis section above.

Consequently, mean development times for G₁P. perseae eggs, nymphal instars, and egg-to-adult development varied with temperature (temperature effects: P < 0.0001 for all response variables; Tables 2 and 3), and no significant differences based on sex were detected (sex effects: P > 0.67 for all response variables; Table 2). In addition, the interaction between sex and temperature had no significant effect (sex × temperature interaction effects: P > 0.84 for all response variables; Table 2) on the duration of G₁P. perseae eggs, nymphal instars, and combined egg-to-adult development time.

Effects of Temperature on G₁P. perseae Longevity Times and Female Fecundity and Fertility Rates

G₁P. perseae males and females exhibited different longevity times with exposure to fluctuating temperature regimens (temperature effect; Tables 2 and 3), and the effect of sex was not significant (sex effect; Table 2). The interaction between sex and temperature was not significant (sex × temperature effect; Table 2). Longevity for P. perseae individuals was shortest at 35 °C and longest at 15 °C, followed by 20 °C (Table 3). Mean G₂ egg counts for G₁P. perseae females varied significantly with temperature (F = 3.62; df = 4, 61; P = 0.01), reaching the highest numbers at 25, 30, and 32 °C (Table 4). In addition, G₂ egg counts increased with G₁ female longevity (female longevity effect: estimate: 0.075; t = 8.41; df = 1, 53; P < 0.0001), and these effects remained consistent across different temperature profiles (interaction between sex and female longevity effect: F = 11.44; df = 4, 58; P < 0.0001) (Table 4). Similar to G₂ egg counts, the resulting percentage of hatched G₂ eggs varied across different temperature conditions (F = 4.41; df = 4, 59; P = 0.0035), with the highest values observed over the temperature range 20 to 32 °C (Table 4).

Table 4.

Mean fertility rates, represented by the numbers of second-generation (G₂) Pseudacysta perseae eggs laid (± SE) and the percentage of G₂Pseudacysta perseae nymphs hatched (± SE), were examined for first-generation (G₁) female Pseudacysta perseae reared under 6 fluctuating temperature profiles (i.e., 15, 20, 25, 30, 32, or 35 °C)

	Mean fertility rate per G1 female
Temperature (°C)	Mean no. G2 eggs laid ± SE	Mean % G2 eggs hatched ± SE
15	15.69 ± 2.78c	4.45 ± 1.93b
	[13] (204)	[13] {11}
20	27.15 ± 1.61b	22.65 ± 2.98a
	[13] (353)	[13] {75}
25	53.87 ± 3.84a	28.26 ± 1.39a
	[16] (862)	[16] {239}
30	53.12 ± 10.27a	21.59 ± 2.86a
	[16] (850)	[15] {202}
32	46.86 ± 8.39a	24.53 ± 2.87a
	[15] (656)	[14] {155}
35	No eggs were laid	No egg hatch occurred
	[6] (0)	[0] (0)
P	< 0.0001	0.0035

Means within the same column, denoted by the same letter, exhibited no significant differences at α = 0.05 (LSMEANS). The number in square brackets [] denotes the quantity of evaluated G₁ female P. perseae at each temperature profile, while the number in round brackets () indicates the total count of G₂P. perseae eggs laid at each temperature profile; and the number in curly brackets {} represents the total number of resulting G₂P. perseae eggs hatched at each temperature profile. No eggs were laid, and no hatching occurred at 35 °C; hence, the 35 °C temperature profile was excluded from both analyses.

At the optimal developmental temperature, 32 °C, female P. perseae exhibited 4 egg-laying peaks around days 11, 35, 54, and 63 of life (Fig. 1). Eggs were laid daily after day 3 of adult life and the average daily egg-laying rate was 3.08 over a 69-day span (Fig. 1). A maximum daily average of eggs laid (i.e., fecundity) was 6.07 eggs laid on day 35 (Fig. 1). Daily fertility, measured as the number of eggs that hatched, indicated that P. perseae eggs begin to hatch reliably from the seventh day of female adulthood day (Fig. 1). The average daily proportion of hatched eggs was 0.10, and the maximum proportion of eggs that hatched was 0.49 on day 31. Eggs failed to hatch on 0.30 days that eggs were oviposited (Fig. 1).

Fig. 1.

Daily fecundity (i.e., mean number of eggs laid per day) and fertility (i.e., proportion of eggs hatched on day of laying) of Pseudacysta perseae females held at 32 °C (fluctuating).

Fitting Models to Temperature-Driven Development Data

The Ordinary Linear model demonstrated a strong fit to data for G₁P. perseae, with an R²_adj value exceeding 0.91 (Table 5). For G₁P. perseae, the linear regression equation predicted a T_min (i.e., lower development threshold) of 9.29 °C and a thermal requirement for development completion (i.e., K) of 476.19 degree-days above this minimum threshold (Table 5; Fig. 2a). The 5 nonlinear models that converged exhibited good fits, with RSS values below 0.0001 for G₁P. perseae (Table 5). Estimations of T_opt (i.e., estimated optimal temperature for development) were consistent among models for G₁P. perseae, ranging from 31.04 to 31.57 °C (Table 5; Fig. 2b–f). However, there was notable divergence in model predictions for T_min (i.e., minimum temperature for development), with estimates ranging from 1.72 to 9.78 °C for G₁P. perseae (Table 5; Fig. 2b–f). A smaller variance in model predictions was observed for T_max (i.e., upper-temperature limit for development), with a predicted range spanning 34.05 to 39.38 °C for G₁P. perseae (Table 5; Fig. 2b–f).

Fig. 2.

Projected rates of total development for G₁ (a–f) Pseudacysta perseae (males and females combined) were assessed across 6 fluctuating temperature profiles (15, 20, 25, 30, 32, or 35 °C) using a) Linear, b) Brière-2, c) Lactin-2, d) LRF, e) Performance-2, and f) Ratkowsky models. In all charts, the ordinate represents the development rate (1/D, in days⁻¹), while the abscissa indicates the temperature (°C). Symbols denote observed data in all charts. For the linear regression model (a), the final data point (i.e., 35 °C) for Pseudacysta perseae development times was excluded due to significant deviation from a straight line.

Quantification of Number of Nymphal Instars and Photo-Documentation of P. perseae Life Stages

Post egg eclosion, the number of instars P. perseae nymphs pass through, confirmed by detection of ecdysed exoskeletons, to reach adulthood, is 4 and not 5, as commonly assumed (e.g., Morales Romero et al. 2002). In addition, with respect to Blatchley’s description (Blatchley 1926) of the fourth antennal segment in adults, this terminal distal segment is not uniformly dark for males and females. The coloration of the fourth antennal segment exhibits sexual dimorphism, with the segment in males being distinctly dark- and light-colored for females (Supplementary Fig. S2).

Discussion

The relationship between temperature and life cycle characteristics such as reproductive capacity, development, and mortality rates strongly influences the impacts of pest species in terms of economic and ecological damage, and speed and patterns of spread (Rebaudo and Rabhi, 2018). To better understand the effects of temperature on the potential pestiferousness of P. perseae, fluctuating temperature cycles that ranged from 15 to 35 °C over a 24-h period aimed to encapsulate the breadth of thermal environments that P. perseae might encounter in avocado orchards in California. The findings of this study provide important insights into the developmental biology and thermal tolerance of P. perseae which are essential for understanding how an important abiotic factor, like temperature, drives the population dynamics of P. perseae across different avocado growing climates in California (e.g., cool coastal vs. hotter interior growing areas) and elsewhere (e.g., Hawaiʻi, México, and Gabon). Improved understanding of abiotic factors, like temperature, that underlie population growth can assist with the development of regionalized integrated pest management (IPM) strategies for this pest. These temperature-based management programs can be custom-designed to specific growing areas to target specific life stages or to assess when population build-up and associated economic damage is likely to be highest.

The variation in adult longevity times and female fertility (i.e., egg hatching) rates with temperature further indicates how thermal conditions influence the life history traits of P. perseae. Notably, longevity decreases significantly as temperature increases over the range of 15 to 35 °C. In addition, fecundity (i.e., total number of eggs laid) peaks over 25 to 32 °C, suggesting a trade-off between lifespan and reproductive success. This finding indicates that while higher temperatures may decrease average longevity, warmer temperatures can positively influence reproductive rates, which may potentially result in more severe infestations when ambient environmental conditions fall within this optimal 25 to 32 °C temperature range. This finding has important implications for pest management, as monitoring, and possibly control strategies, may need to be intensified under these prevailing temperature conditions to mitigate the proliferation of P. perseae populations which could cause economic damage.

The high mortality rates observed among eggs, nymphs, and adults of P. perseae at the upper threshold temperature that averaged 35 °C over a 24-h period, underscores the negative impact of upper-level temperature extremes on the survivorship rates and fitness of this pest. The negative effects of high temperatures on P. perseae are important, as it indicates that elevated temperatures, akin to potential climate warming scenarios, seasonal temperature peaks, or intense heat waves of short duration, such as Santa Ana wind events in California, may have significant and deleterious impacts on P. perseae populations. In addition, as climate change increases average temperature levels in southern California (Pathak et al., 2018), the effects of heat on P. perseae (and avocado production in general) may become more pronounced, and extreme heat events could naturally limit the geographic spread and population growth of P. perseae.

The identification of a multi-peak fecundity and fertility pattern in P. perseae at the optimal developmental temperature of 32 °C highlights the necessity for continuous monitoring and developing biologically based timed interventions to curb population surges. The occurrence of up to 4 fecundity and fertility peaks suggests that P. perseae females can maintain substantial reproductive momentum as they age. This could cause populations to rebound unexpectedly after observed population surges, which necessitates the need for regular population monitoring at times when temperatures favor P. perseae development and reproduction.

In comparing the developmental parameters of P. perseae across previously conducted temperature-development studies (Morales Romero et al. 2000 and Humeres unpublished data 2008 [see Supplementary Table S1 for linear regression analyses of these data sets]), this study, conducted under fluctuating temperature regimes that are representative of climatic conditions across infested areas in southern California, enables important comparisons to the findings of Morales Romero et al. (2000) and Humeres (2008, unpublished data), who conducted development studies under constant temperature conditions (Supplementary Table S1). The T_min values observed in this study (linear regression, 9.29 °C; Lactin-2, 9.77 °C; and Performance-2, 9.87 °C) and Morales Romero et al. (2000) (9.29 °C) are lower than those derived from analyses of Humeres’ unpublished data (2008) (11.32 °C). This outcome may be attributable to the narrower temperature ranges (i.e., 28 to 33 °C) used in experiments conducted by Humeres (2008, unpublished data). The higher K value of this study (i.e., 476.19 vs. 434.78 [Morales Romero et al. 2000] and 217.33 [Humeres 2008 unpublished] [Supplementary Table S1]) in this study suggests that under average fluctuating temperature conditions used in these studies, P. perseae may require increased cumulative temperature exposure above T_min to complete development. However, there is one important caveat that needs consideration; this being the P. perseae haplotype used for temperature-driven development studies. Haplotype G was used in this study and by Morales Romero et al. (2000) whose research was conducted with P. perseae populations in Cuba (haplotype G is the dominant haplotype in the Caribbean [Rugman-Jones et al. 2012; Dadlani et al. 2025]). In comparison, studies by Humeres (2008, unpublished) were conducted with haplotype A prior to the invasion of California by haplotype G in 2017 (Dadlani et al. 2025). Genetic differences as identified by haplotype designation (i.e., A [Mexican origin] vs. G [widespread and common in the Caribbean and Florida {Dadlani et al. 2025}]) may introduce a confounding genetic variable that could influence measured development parameters across a range of temperatures.

This study confirmed that P. perseae has 4 nymphal instars (i.e., 4 molts were observed prior to nymphs reaching adulthood) and not 5 as stated in other studies (Heidemann 1908, Morales Romero et al. 2002, Peña et al. 2012, Guidoti et al. 2015). Owing to the ambiguous nature of experimental conditions that investigated and described P. perseae instar development (Morales Romero et al. 2002) it is unclear how transitions between stadia were recognized and confirmed. In this study, the transition between instars was closely monitored for molted exoskeletons which confirmed ecdysis had occurred and transition to the next instar had happened. Consequently, the confirmation of 4 instars in P. perseae deviates from the typically reported pattern of 5 instars that is typical of other species of Tingidae (Heidemann 1908, Nair and Braman, 2012, Guidoti et al., 2015). A review of literature for decreased number of instars identified 2 other tingid species that have been documented and confirmed as having 4 nymphal stages, Stephanitis rhododendri (Horváth) (Tinigidae: Tingini) and Tanybyrsa cumberi (Drake) (Tinigidae: Tingini) that feed on Rhododendron spp. (Ericaceae) and Astelia banksia (Liliaceae), respectively (May 1977, Štys and Davidová 1989).

Heidemann (1908) speculated that the last instar of P. perseae was the fifth instar, mirroring the number of instars observed in other tingid species described at the time. This assumption by Heidemann (1908) likely influenced Morales Romero et al. (2002) who reported that P. perseae has 5 instars. Morales Romero et al. (2002) provide a dichotomous key for identifying 5 instars of P. perseae, utilizing head diameter, body length, and coloration. A close examination of the descriptions in Morales Romero et al. (2002) on how first and second instars were separated indicates that what was considered 2 separate stages is, in fact, just 1, the first instar, which exhibits 2 color morphs, a very pale brown color upon eclosion from the egg (first instar according to Morales Romero et al. 2002) which then transitions to a darker dull red color (second instar according to Morales Romero et al. 2002), which is likely due to post-emergence melanization in the first instar. In comparison to Morales Romero et al. (2002) who used color change from light to dark in the first instar nymph to indicate transition to the second instar, this study tracked exoskeleton shedding instead and determined that observed color changes in first instar P. perseae does not correspond with ecdysis to the second instar. This finding challenges the 5-instar dogma by clearly demonstrating that P. perseae has only 4 nymphal instars.

In addition, this study identified a new sexual dimorphism trait that distinguishes male and female P. perseae. Contrary to Blatchley’s description (Blatchley 1926) of adult P. perseae antennae as having a “blackish apical half of the fourth antennal segment” (Blatchley 1926, Mead and Peña 2016), findings reported here indicate that females can be distinguished by having a light-colored apical half on the fourth antennal segment, while males display a dark, almost black, colored terminal segment. This finding simplifies sex determination when field sampling.

Supplementary Material

Supplementary material is available at Annals of the Entomological Society of America online.

Author contributions

Lakshmi Dadlani (Data curation, Investigation, Writing—original draft [equal]), Ivan Milosavljevic (Formal analysis, Writing—review & editing [equal]), and Mark Hoddle (Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing—original draft, Writing—review & editing [equal])

Funding

This work was supported, in part, by the California Avocado Commission, research agreement number CAC-65131-00-000, ‘Phenology and Ecology of Avocado Lace Bug in Southern California’ awarded to MSH.

Conflicts of interest. The authors have no relevant financial or non-financial interests to disclose.