Next Article in Journal
A Review of the Genus Homidia (Collembola, Entomobryidae) in China Informed by COI DNA Barcoding, with the Description of Three New Species
Previous Article in Journal
Effect of Acheta domesticus Powder Incorporation on Nutritional Composition, Technological Properties, and Sensory Acceptance of Wheat Bread
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Insects
Volume 16
Issue 9
10.3390/insects16090973
insects-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Luthrodes pandava Larvae Can Distinguish Cycas Leaf Quality in Cafeteria Experiments

by
Thomas E. Marler
1,2
1
Philippine Native Plants Conservation Society, Inc., Ninoy Aquino Parks and Wildlife Center, Quezon City 1101, Philippines
2
Cycad Specialist Group, International Union for Conservation of Nature Species Survival Commission, 1196 Gland, Switzerland
Insects 2025, 16(9), 973; https://doi.org/10.3390/insects16090973
Submission received: 7 August 2025 / Revised: 12 September 2025 / Accepted: 15 September 2025 / Published: 17 September 2025
(This article belongs to the Section Insect Behavior and Pathology)
Browse Figures
Versions Notes

Simple Summary

The butterfly Luthrodes pandava is a widespread invasive species that produces larvae that feed on immature leaves of Cycas species. Since Cycas taxa are members of the group of plants known as cycads, and this plant group is the most threatened plant group that has been assessed, a greater understanding of the interactions between the host and the herbivore is needed. I evaluated the ability of L. pandava larvae to preferentially feed on leaves from high quality Cycas species. The larvae from wild populations of the butterfly obtained from Philippine and Thailand habitats that contain a native Cycas species were able to successfully discriminate. These larvae preferentially fed on leaflets from Cycas species that are considered to be high quality food. In contrast, larvae from invasion populations of the butterfly collected in urban settings fed on any Cycas leaflet offered as food. The beneficial behaviors exhibited by the wild butterfly populations appear to be lost over time when an urban butterfly population has access to numerous Cycas species for regeneration. The findings indicated that the geographic source of butterflies and feeding history influence feeding behavior.

Abstract

Cycads are being endangered by several anthropogenic threats, and invasions of non-native herbivores are among those threats. Various country invasions by the cycad blue butterfly Luthrodes pandava have threatened the local horticulture cycad trade and native cycad populations. Little is known about behaviors of the larval stage of this specialist herbivore, the life phase that causes damage to cycad leaf tissue. The objective here was to determine larval ability to discriminate among Cycas species that are known to exhibit contrasting susceptibility to L. pandava herbivory. Dual-choice cafeteria protocols were used whereby a single larvae was provisioned with one leaflet of minimally damaged species (low quality) and one leaflet of heavily damaged species (high quality). Consumption after an 8 h feeding cycle was measured as loss in leaflet area. Wild L. pandava populations derived from in situ Philippine and Thailand Cycas habitats preferentially fed on high quality leaflets. In contrast, invasive L. pandava populations from urban locations in Guam, Philippines, and Thailand exhibited no preferences in leaflet feeding choice. The findings indicated that larvae of this Lycaenidae specialist herbivore exhibit complex selection behaviors that depend on origin of the butterfly and feeding history. Discrimination capabilities of the in situ populations that evolved with a single native Cycas species demonstrated beneficial behaviors that appear to be lost over time by invasive populations that feed on numerous novel Cycas hosts within the invasive range.

1. Introduction

The global biodiversity crisis has emerged as a critical research agenda [1,2]. One approach has been to create groups of species to more fully understand extinction risks to the groups as a whole [3]. Cycads are members of an ancient group of spermatophytes, and some species have become commodities in the horticulture industry [4]. Cycads have emerged as the most threatened plant group that has been assessed globally [5,6,7,8]. Furthermore, sequential assessments have shown that the threat status to this group of plants is worsening with time [3].
Habitat loss and over-collection of wild plants comprise the greatest threats to cycads [5], but invasive species have been identified as one of the other threats that are directly related to human activities [9,10,11]. The Cycadaceae is a mono-generic family comprising 125 species [12], and two invasive arthropod herbivores have emerged as major threats to members of this family in recent years. The armored scale Aulacaspis yasumatsui Takagi can feed on numerous cycad species (Cycadophyta), but its ability to kill its host appears to be restricted to Cycas species [9]. The Luthrodes pandava Horsfield (Lepidoptera: Lycaenidae; formerly Chilades pandava) is a voracious leaf-feeding Lycaenidae butterfly that has a widespread native range [10,13,14]. Variation in the amount of L. pandava damage among co-mingled Cycas L. species is considerable, with the species which coevolved with a leaf-feeding butterfly revealing the least levels of herbivory [15].
Evidence-based conservation protocols are of paramount importance to mitigate the threats to cycad species survival [11]. Therefore, a greater understanding of the manner in which L. pandava threatens cycad plants and the defensive traits that the host plants utilize to confront L. pandava herbivory is needed. Moreover, for the endemic regions in which a native Cycas host co-exists with the native L. pandava, the ethical obligation to conserve both species presents a conundrum for decision-makers [15,16]. Very few ethology experiments have been conducted to more fully understand this bi-trophic relationship. When a L. pandava larvae is reared on leaflets from the most damaged Cycas species, a clear benefit has been demonstrated through oviposition performance of the consequent adults [17]. These authors reported that adult female L. pandava lifespan was increased 32% and lifetime egg production was increased more than 50% when individuals reared on leaves from heavily damaged Cycas species were compared with individuals reared from minimally damaged species. Similar influences on life history traits were recorded when L. pandava larvae diet was restricted to leaves from low quality Zamia species versus high quality Cycas species [18,19]. For the purposes of this study, the species that are heavily damaged by L. pandava feeding are considered high quality, and the species that are minimally damaged are considered low quality.
The host species that serves as larvae food clearly imposes an increase or decrease in adult performance of the herbivore. Therefore, discrimination capabilities would benefit the butterfly population in locations where more than one host species is available. Evaluating choices made by gravid L. pandava females indicated these individuals were able to discriminate among various Cycas species, preferentially ovipositioning on the leaves of high-quality species [20]. Interestingly, this behavior was restricted to a wild population of L. pandava which was derived from the endemic Cycas nongnoochiae K.D. Hill habitat. Females from an urban population of L. pandava of unknown origin and a history of feeding on numerous Cycas species were unable to discriminate among Cycas species, revealing no preferences when presented with high quality versus low quality leaves for ovipositioning. To my knowledge, the ability of L. pandava larvae to discriminate between high quality versus low quality Cycas leaves has not been studied to date.
Traditional dual-choice cafeteria feeding trials were employed here to determine the feeding behavior of L. pandava larvae when provisioned with leaflet tissue from one low quality Cycas species and one high quality Cycas species. The objective was to determine if individual larvae of this Cycas herbivore can discriminate food options. The findings are of importance for improving management decisions in botanic garden settings in which L. pandava has access to numerous host species grown in close proximity.

2. Materials and Methods

Dual-choice cafeteria trials were conducted in Guam, Philippines, and Thailand with L. pandava larvae derived from three in situ Cycas populations which had fed exclusively on their native Cycas host, and three urban populations of unknown origin and which had fed on several Cycas species located in the urban forest or garden community. The feeding trials were designed to provision one leaflet choice from a high quality Cycas species and one leaflet choice from low quality Cycas species, based on a common garden survey of L. pandava damage to 85 Cycas species [15]. The response variable calculated was amount of leaflet consumed from each leaflet during a 8 h feeding cycle. A total of 280 larvae derived from seven populations were exploited in this study.

2.1. Population Origins

Butterfly eggs and larvae were collected in Tak Fa, Thailand within the C. nongnoochiae endemic range on 18–19 June 2013. The eggs and larvae were transported to Nong Nooch Tropical Botanical Garden (NNTBG) in Chonburi, Thailand and fed C. nongnoochiae leaves until pupation. The Gardens’ extensive cycad collection was the source of the food. The resulting population of adults (Figure 1a) were fed honey on sponges and maintained in a communal cage ca. 1.6 m3 in volume. Production of larvae used in the cafeteria trials is described below.
The NNTBG’s Cycas germplasm supported a thriving L. pandava population of unknown origin (Figure 1b). This population had been feeding on numerous Cycas species for many years. Young C. revoluta leaflets containing numerous L. pandava eggs were collected from the Garden 13–16 July 2013 to produce the larvae as described below.
Butterfly larvae were collected in Mansalay, Mindoro, Philippines from a coastal Cycas edentata de Laub. population on 9 January 2019. The collections were made with permission from the owner of the copra farm that contained the plants. The trees on the farm were a component of a large C. edentata population that extended along the coastline in undisturbed habitats. The larvae were transported to a Cycas germplasm garden in Angeles City, Philippines and fed C. edentata leaflets until pupation. The resulting population of adults (Figure 2a) were fed honey on sponges and maintained in a communal cage ca. 1.6 m3 in volume. Production of larvae is described below.
The urban population of Cycas plants in the Angeles City urban forest supported a widespread L. pandava population of unknown origin (Figure 2b). This population had been feeding on the available Cycas bougainvilleana K.D. Hill, C. edentata, and C. revoluta. Expanding C. revoluta leaflets containing numerous L. pandava eggs were collected from C. revoluta plants 23–28 January 2019 to produce the larvae.
The invasive L. pandava population on the island of Guam (Figure 3a) was used for feeding trials in 2023. Origin of this population was most likely the Philippine Archipelago [13], and it had been feeding on urban landscape C. revoluta and in situ Cycas micronesica K.D. Hill trees since the 2003 invasion [21]. Expanding leaves with numerous L. pandava eggs were collected directly from C. revoluta in the urban forest on 22–27 January 2023 and transferred to an open-air laboratory setting at the University of Guam to produce the larvae for the cafeteria trials.
Butterfly larvae were collected in Silanguin Cove, Zambales, Philippines at the southern end of the known Cycas zambalensis Madulid & Agoo endemic range on 7 May 2024. The collections were made with permission from plants located in private beach resort landscapes. The larvae were transported to the Cycas germplasm garden in Angeles City, Philippines and fed C. zambalensis leaves until pupation. The resulting population of adults (Figure 3b) were fed honey on sponges and maintained in a communal cage ca. 1.6 m3 in volume for the production of larvae.
Expanding C. revoluta leaves containing numerous L. pandava eggs were collected 21–27 May 2024 from the urban population of plants in Angeles City. A population of L. pandava larvae derived from these eggs was reared using the same methods as the Zambales population.
For every population of caged adults described, partially expanded C. revoluta leaves were offered to the adult females for ovipositioning, and the resulting larvae were initially fed ad libitum with twice daily additions of fresh C. revoluta leaves. The timespan from egg to adult can be as short as 11–12 d, and each female adult is capable of producing 21–27 eggs per day [17]. Therefore, producing a sizable population of growing L. pandava larvae was a rapid endeavor. For every urban population in which eggs were collected, the resulting larvae were reared on C. revoluta leaf tissue at the same time as the wild populations.

2.2. Rearing and Handling Larvae

The prescribed age of the larvae and the window of time that feeding was allowed for the cafeteria trials were explored by preliminary studies. Tissue loss during feeding by small larvae can be minimal, as the 1st instar is less than 1.5 mm in length. Moreover, the smallest larvae feed on leaflet surfaces and the larger larvae feed on leaflet margins, so quantifying tissue loss following feeding is difficult for the small larvae. In contrast, leaflet tissue loss during feeding by 4th instar larvae nearing pupation is extensive and tissue loss is rapid. In order to exploit a median speed of herbivory, all cafeteria assays were conducted when larvae reached 6–7 mm in length during the 2nd instar. This size is about half of maximum size of a mature 4th instar larvae. This larvae length was achieved in 5–6 days for the urban populations and 6–7 days for the wild populations. Initial feeding trials were 24 h and 18 h in duration, but 100% of the leaflet tissue was consumed for each of these durations. Feeding trials of 12 h and 10 h enabled some remaining uneaten tissue, but more than half of the available tissue was consumed. Therefore, the potential for no-choice feeding for the latter portion of the window of time disallowed feeding for this length of time. The preliminary feeding times were used to determine that 8 h was adequate for all feeding trials because the 2nd instar larvae had consumed less than half of the available leaflet tissue within this duration. This protocol enabled each larvae to feed ad libitum while being presented with an active choice between leaflets of the two available Cycas species for the full duration of each 8 h trial.
Selection of the two Cycas species to exploit for the dual-choice trials was controlled by three factors. First, the larvae of this butterfly cannot consume mature leaflet tissue, so availability of actively expanding immature leaves is mandatory. The two species selected to provide food for each of the feeding trials was therefore restricted to locally available expanding leaves in the ex situ germplasm collections that were exploited for leaflet tissue. Second, the reported assessment of species susceptibility to L. pandava herbivory included 85 Cycas species [15]. This report included six species that exhibited the greatest level of butterfly damage, and eight species that exhibited the least level of damage. One species was selected from each of these two groups for each trial, ensuring the greatest level of contrast in level of herbivory between the two selected species. Third, in order to ensure that there was no influence of Cycas species in larvae feeding history on the outcomes, the two Cycas species offered in the feeding trials were novel substrates for the individual larvae used for the trials (food was restricted to C. revoluta prior to the trials) but also for the historical access to Cycas by each butterfly population.
Larvae were selected about 1800 HR on the day prior to each trial, and all leaf material was removed from the growth chambers to force an overnight fasting period. The feeding trials were initiated 0700–0800 HR the following morning. The feeding trial containers were plastic tubs with inside basal diameter of 8.5 cm and height of 6 cm (Figure 4a). The base of each tub was covered with a moistened double layer of coffee filter paper, and the leaflets and larvae were positioned gently in the center of the paper. Each dual-choice cafeteria feeding trial was conducted with a single 2nd instar larvae positioned between two leaflet sections (Figure 4b). The leaflets were immature and rapidly expanding when selected for the offered food. Each leaflet was shortened to a maximum of 8 cm by removing the basal portion with shears. The undisturbed apex of each leaflet was positioned adjacent to each larvae, ensuring the leaflet tissue in close proximity to the larvae at the start of the feeding trials did not contain any potential induced chemicals that may have resulted from the wound on the laminae.
The layout of the trial employed four initial arrangements for each butterfly population in order to account for directional bias in the navigation [22]. Two were positioned with the leaflets oriented east–west and the more resistant leaflet was placed on the south side of the larvae for half of these and on the north side of the larvae for the other half. Two were positioned with the leaflets oriented north–south and the more resistant leaflet was placed on the east side of the larvae for half of these and the west side of the larvae for the other half. An equal number of replications were ensured for each of the four arrangements. Moreover, in order to further counter directional bias, the tubs were gently rotated 90° every 2 h during each feeding cycle.
An initial digital photograph was secured (e.g., Figure 4b), then the larvae were directly observed until feeding was initiated. This time was recorded for each replication, then each tub was loosely covered. The larvae never attempted to navigate the sides of the vessels, and remained in contact with one or both of the leaflet sections for the duration of the 8 h feeding cycles. The edge feeding behavior resulted in the two partially eaten leaflet portions remaining fully intact (Figure 4c), and each was carefully extracted with tweezers and placed onto a fresh sheet of white paper to photograph a second time. This transfer was required because the abundant frass and a change in the filter paper color due to the release of chlorophyll and frass pigments during the feeding cycle. Deleting this background noise was required to ensure unambiguous color contrast between leaflet color and background color for the image analyses (described below).

2.3. Dual-Choice Feeding Decisions

The feeding trials with the endemic C. nongnoochiae butterfly population and the resident NNTBG population were conducted 19–24 July 2013. The tubs were placed on benches under concrete slabs that allowed diffuse light but blocked the sun. A shaded maximum/minimum thermometer was installed and data were recorded daily (Table 1). The photoperiod was 12.72 h and photosynthetic photon flux density (PPFD) was recorded by placing the quantum sensor (Skye SKP200, Skye Instruments, Llandrindod Wells, Powys, UK) inside the tubs for measurements. Two species of high quality host plants were selected as C. micronesica and Cycas seemanii A. Braun. (high quality). Two species of low quality host plants were selected as Cycas wadei Merr. and Cycas condaoensis K.D. Hill & S.L. Yang (low quality). These four Cycas species were novel substrates for the wild population of larvae. Ten trials were conducted for each of the four Cycas species combinations, for a total of 40 feeding trials for each of the two butterfly populations.
The feeding trials with the Mindoro C. edentata butterfly population and the urban Angeles City population were conducted in an ex situ cycad germplasm garden established by the University of Guam in Barangay Sapang Bato. The tubs were placed on benches under metal roofing that allowed scattered light. The photoperiod was 11.56 h, and the species selected for the feeding trials were the high quality C. micronesica and the low quality C. wadei. These species were novel food sources for the butterflies in the trials. Forty feeding trials were conducted with the two species choices for each of the two butterfly populations.
The feeding trials with the invasive Guam butterfly population were conducted on benches under the roof overhang of an open-air laboratory under ambient conditions. The photoperiod was 11.53 h and PPFD, and the high quality species was Cycas tansachana K.D. Hill & S.L. Yang and the low quality species was C. wadei. Both of these species were new to the naturalized butterfly population on the island. A cycad research garden maintained by the University of Guam was the source of these substrates. Forty feeding trials were conducted with the two species choices.
The feeding trials with the Zambales C. zambalensis butterfly population and the urban Angeles City population were conducted in the same nursery setting that was used for the 2019 trials. The photoperiod was 12.99 h, and the high quality leaflets were C. tansachana and the low quality leaflets were C. wadei. Both of these species were novel food materials for the butterflies in this feeding trial. Forty feeding trials were conducted with the two species choices for each of the two butterfly populations.

2.4. Image and Statistical Analysis

The 600 dpi digital images were evaluated using ImageJ version 1.53 [23] to calculate individual leaflet area initially and after feeding occurred. The feeding of late 2nd instar L. pandava larvae is typically edge feeding if the leaflet tissue is soft enough, so there were no feeding holes to evaluate in the area calculations. The precise area of the leaflets prior to the feeding trial differed slightly, so the response variable that was considered the most accurate for comparison among all of the trials was absolute tissue area loss due to feeding. This was calculated for each leaflet by subtracting the ending area from the initial area.
The initial leaflet area data conformed to parametric analyses because of the highly prescribed manner in which the leaflets were prepared, so a standard t-test was employed to confirm initial leaflet area was homogeneous. The ending leaflet area data did not meet parametric prerequisites for the wild butterfly populations. As a result, the nonparametric Mann–Whitney U test was employed to compare the two Cycas species in each trial. In contrast, the ending leaflet area for the feeding trials that used the urban butterfly populations met parametric analyses requirements, so a t-test was employed to compare herbivory of high quality versus low quality species. For the 2013 Thailand study, each of the four two-species combinations were analyzed separately (n = 10). For the other three studies, there were only two species included as food offerings within each trial, so the entire 40-replication data set was analyzed. Excel Version 2506 for Microsoft 365 was employed to conduct the t and U tests.
The feeding trials were not set up to subject the wild versus urban population response variables to statistical analysis. To gain more insight on the outcomes, the means ± standard deviation for feeding efficiency (time to begin feeding) were calculated for each wild and each urban population. Similarly, the total amount of herbivory during the 8 h feeding cycle was calculated for each replication by adding the total tissue lost from both leaflets. These data were also presented as means ± standard deviation.

3. Results

3.1. Thailand

The area of C. micronesica and C. wadei leaflets that were offered to the wild Thailand butterfly larvae did not differ (t = 0.094, p = 0.926), and the larvae were given access to 477 ± 16 mm2 (mean ± SD; n = 20) for each leaflet. Feeding by these larvae began after 1.7 ± 0.9 min. These larvae consumed some low-quality C. wadei leaflet tissue, but the amount was only 3% of the C. micronesica leaflet tissue that was consumed (Figure 5a). The urban Thailand butterfly larvae were offered leaflets from the same two Cycas species, and the initial leaflet area did not differ (t = 0.452, p = 0.657). Initial area was 484 ± 15 mm2 per leaflet. Feeding began after 0.9 ± 0.2 min. Contrary to the behavior of the wild population, these urban butterfly larvae consumed similar amounts of tissue from the two Cycas species (Figure 5b). The urban population revealed no ability to discriminate high versus low quality substrates. Total leaflet area consumed by the wild population was 261 ± 13 mm2, and total area consumed by the garden population was 282 ± 16 mm2.
The area of C. micronesica and C. condaoensis leaflets that were offered to the wild Thailand butterfly larvae did not differ (t = 0.079, p = 0.860), and the larvae were given access to 510 ± 13 mm2 for each leaflet. Feeding began after 1.8 ± 0.9 min. These larvae consumed a small amount of C. condaoensis leaflet tissue, but the amount was only 4% of the high-quality C. micronesica leaflet tissue that was consumed (Figure 5c). The urban Thailand butterfly larvae were offered leaflets from the same two Cycas species, and the initial leaflet area did not differ (t = 0.331, p = 0.744). Initial area was 502 ± 16 mm2 per leaflet. Feeding began after 0.8 ± 0.2 min. These urban butterfly larvae consumed similar amounts of tissue from the two Cycas species (Figure 5d). The wild population consumed 258 ± 16 mm2 of leaflet area, and the garden population consumed 291 ± 19 mm2 of leaflet area.
The area of C. seemanii and C. wadei leaflets that were offered to the wild Thailand butterfly larvae did not differ (t = 0.178, p = 0.861), and the larvae were given access to 487 ± 16 mm2 for each leaflet. Feeding began after 2.1 ± 0.9 min. The consumption of low-quality C. wadei leaflet tissue by these larvae was minimal and represented about 2% of the C. seemanii leaflet tissue that was consumed (Figure 5e). The urban Thailand butterfly larvae were offered leaflets from the same two Cycas species, and the initial leaflet area was similar (t = 0.694, p = 0.497). Initial area was 489 ± 13 mm2 per leaflet. Feeding began after 0.7 ± 0.2 min. Leaflet tissue from the two Cycas species was consumed in similar amounts by these urban butterfly larvae (Figure 5f). Total leaflet area consumed by the wild population was 249 ± 13 mm2, and total area consumed by the garden population was 285 ± 14 mm2.
The area of C. seemanii and C. condaoensis leaflets that were offered to the wild Thailand butterfly larvae did not differ (t = 0.440, p = 0.665), and the larvae were given access to 517 ± 11 mm2 for each leaflet. Feeding began after 2.2 ± 0.8 min. The consumption of low-quality C. condaoensis leaflet tissue by these larvae was minimal and represented about 3% of the C. seemanii leaflet tissue that was consumed (Figure 5g). The urban Thailand butterfly larvae were offered leaflets from the same two Cycas species, and the initial leaflet area was similar (t = 0.169, p = 0.868). Initial area was 509 ± 18 mm2 per leaflet. Feeding began after 0.8 ± 0.2 min. The urban butterfly larvae consumed similar amounts of tissue from the two Cycas species (Figure 5h). The wild population consumed 268 ± 11 mm2 of leaflet area, and the garden population consumed 290 ± 22 mm2 of leaflet area.

3.2. Philippines

The area of C. micronesica and C. wadei leaflets that were offered to the wild Mindoro butterfly larvae did not differ (t = 0.598, p = 0.552), and the larvae were given access to 515 ± 27 mm2 (mean ± SD; n = 80) for each leaflet. Feeding began after 2.1 ± 0.9 min. These larvae avoided the low-quality C. wadei leaflet tissue, with herbivory in the amount of about 7% of the C. micronesica leaflet tissue that was consumed (Figure 6a). The urban Angeles City butterfly larvae were offered leaflets from the same two Cycas species, and the initial leaflet area did not differ (t = 0.392, p = 0.696). Initial area was 499 ± 18 mm2 per leaflet. Feeding began after 0.9 ± 0.4 min. These urban butterfly larvae consumed similar amounts of tissue from the two Cycas species (Figure 6b). The wild population consumed 255 ± 35 mm2 of leaflet area, and the urban population consumed 282 ± 29 mm2 of leaflet area.
The area of C. tansachana and C. wadei leaflets that were offered to the wild Zambales butterfly larvae did not differ (t = 0.135, p = 0.893), and the larvae were given access to 461 ± 21 mm2 for each leaflet. Feeding began after 2.1 ± 0.9 min. These larvae avoided the low-quality C. wadei leaflet tissue, with herbivory in the amount of about 3% of the C. tansachana leaflet tissue that was consumed (Figure 6c). The urban Angeles City butterfly larvae were offered leaflets from the same two Cycas species, and the initial leaflet area did not differ (t = 0.135, p = 0.893), and initial area was 460 ± 18 mm2 per leaflet. Feeding began after 0.8 ± 0.4 min. These urban butterfly larvae consumed similar amounts of tissue from the two Cycas species (Figure 6d). Total leaflet area consumed by the wild population was 254 ± 33 mm2, and total area consumed by the urban population was 295 ± 23 mm2.

3.3. Guam

Larvae derived from the invasive population of L. pandava throughout the island of Guam were offered leaflets from high-quality Cycas tansachana and from low-quality Cycas wadei. The initial leaflet area did not differ (t = 0.332, p = 0.741), and the larvae were presented with 453 ± 19 mm2 for each of the leaflets. Feeding began after 0.7 ± 0.3 min for these larvae. These butterfly larvae consumed similar amounts of tissue from the two Cycas species during the dual-choice feeding cycle, revealing no ability to identify differences in food quality (Figure 7). Total leaflet area consumed by these urban butterfly larvae was 309 ± 29 mm2.

4. Discussion

The current global biodiversity crisis [1,2] has fueled the need to identify groups of plants that are threatened as a means of focusing conservation research efforts. The compelling global change factors present disproportionate levels of threats to rare plants with limited numbers or restricted endemic range [24,25]. This crisis is partly fueled by invasive species, and improvements in understanding the interactions between non-native arthropod herbivores and the new hosts that are found within their invasive range are needed as part of the efforts to combat the crisis. The relevance of these issues for cycad conservation includes the voracious leaf herbivore L. pandava which is capable of consuming immature leaf tissues from numerous Cycas and Zamia species [15,18,19]. The butterfly therefore represents an acute threat particularly for Cycas species that did not coevolve with a native butterfly herbivore [15]. This threat may be manifested when the butterfly invades new geographic regions with native Cycas species, or when the butterfly resides in an extensive cycad germplasm garden where numerous host species are growing together. The same phenomena may occur for the Zamia–Eumaeus bi-trophic herbivory system [26].
The results herein illuminated beneficial larval discrimination capabilities of three wild L. pandava populations that co-evolved with a single Cycas host species, showcasing behaviors that enabled preferential feeding on high-quality substrates when the choices were novel, first-time choices. However, this behavior was not evident in larvae from invasive L. pandava populations that had been afforded a multi-species history of access to numerous Cycas and Zamia hosts within a managed botanic garden in Thailand or an invaded urban forest setting in Guam and the Philippines.

4.1. Host and Herbivore

Inter-generic and inter-specific differences in herbivory are clearly evident in the group of plants known as cycads, but remain equivocal and not very well understood. Some factors that may define L. pandava herbivory diversity have been previously discussed [27]. Active selection of one host among several available hosts may be founded in differences in nutrient content and/or palatability. Active avoidance may be founded in constitutive or induced chemical defense. Active avoidance may also be a result of structural defense. Apparency differences [28] among host species may be causal of herbivory disparity as Cycas leaf expansion rates vary, and these rates may translate into timing of leaf maturity. The factors which govern herbivore–host relationships in cycad biology remain obscure, and the roles of endosymbionts and sequestered biochemicals as drivers of anti-herbivory behaviors in cycads are of particular interest and deserve more study [14,29,30,31,32,33,34,35,36,37].
In the Cycas–Luthrodes herbivory system, the decision by the herbivore to avoid low quality substrates and/or seek out high quality substrates is evident in adult female oviposition choices [17]. However, the immature stages of insects possess limited mobility and their sensory systems are constrained compared to the adult stage [38]. Feeding preferences and food discrimination ability of immature individuals often do not align with those of oviposition preferences of the female parents [39,40]. The new knowledge communicated herein reveals that L. pandava larvae can be endowed with the same selection abilities as the female adults, corroborating robust reviews indicating butterfly larvae do possess beneficial behaviors in selecting high quality versus low quality substrates [41].
The findings presented herein also revealed that the butterfly population being studied exerted profound influences on the larvae’s choices. This study included larvae from two subspecies of L. pandava. The populations from Thailand conform to the subspecies pandava from Southeast Asia, and the wild-collected populations from the Philippines are considered the separate subspecies vapanda [42]. Genetic signatures have been used to indicate the most likely source of the butterfly invasion of Guam is the Philippine archipelago [13]. Based on wing coloration, the urban population of L. pandava in the Philippines (Figure 2b) is more closely aligned with the Thailand population (Figure 1) than the wild-type in the Philippines (Figure 2a). The non-native C. revoluta is widely cultivated throughout most inhabited Philippine islands [43], and an inadvertent invasion of L. pandava vapanda within urban settings may have occurred in the past when C. revoluta plants were imported to support the nursery trade.
Despite substantial diversity of geographic origin and subspecies among the various feeding trials, every butterfly population obtained from in situ Cycas hosts exhibited unambiguous food selection behavior. Contrarily, every urban butterfly population that was studied exhibited no discrimination behaviors in food selection and consumed any Cycas leaflet that was offered ad libitum. Based on these results, the ability to discriminate during binary choices and feed primarily on high quality food was restricted to the wild populations.
The feeding efficiency determined as time lapsed until feeding was initiated was remarkably different for the wild type versus the urban type butterflies. In every paired experiment, the urban population initiated feeding with greater speed, averaging 0.8 min; and the wild population initiated feeding more slowly, averaging 2.0 min. A third difference between the wild and urban populations was measured as total leaflet area consumed during the 8 h feeding trials. In every case, the urban population (mean 288 mm2) consumed more than the wild population (mean 258 mm2), representing 12% greater daily consumption potential for the urban butterflies. Based on these results, the ability to exploit all available food sources and consume greater quantities of food per day would allow the urban populations to out-compete the wild populations at the larval stage.

4.2. Botanic Gardens

Ex situ cycad germplasm gardens are sometimes located nearby in situ populations of native congeneric cycad species. These managed gardens containing non-native species are artificially sympatric with the native congeneric species, some of which may be sexually compatible, and therefore create new biological interactions between the garden plants and the surrounding in situ plants. For example, the Montgomery Botanical Center curates an extensive ex situ Zamia collection [44] and is surrounded by in situ Zamia integrifolia L.f. populations. The habitat was relied upon for some of the initial studies that laid the foundation for our contemporary understanding of cycad pollination [45,46,47]. Similarly, the cycad germplasm gardens curated by the University of Guam are located amongst in situ C. micronesica habitats. This anthropogenic creation of close proximity between ex situ and in situ congenerics may cause consequential interactions through insects which travel between the garden and natural plants.
The permeability of botanic garden boundaries (e.g., [48]) influences herbivory in locations where a native cycad species resides. First, the non-native germplasm in botanic gardens may serve as anthropogenic brood sites of L. pandava, magnifying the neighborhood herbivory pressures imposed on the adjacent native Cycas host plants simply by increasing the butterfly population size. This same phenomenon may occur where non-native Zamia species in botanic gardens act as brood sites for Eumaeus atala Poey, magnifying the herbivory pressure on surrounding native Zamia plants. Second, the feeding behaviors of larvae on an in situ native host plant may originate from an egg oviposited by an adult that was reared on non-native botanic garden host plants. The results herein indicate these larvae from a garden-reared adult may cause greater herbivory damage than an oligophagous larvae originating from the native butterfly population. Third, an adult butterfly that was reared on high quality non-native host species inside the botanic garden may cause more in situ damage because of greater longevity and fecundity of these adults [17]. Finally, a more insidious threat may occur when the non-native cycad germplasm in a botanic garden becomes the source of genetic pollution of surrounding native cycad species due to the efflux of pollinator species carrying non-native but sexually compatible pollen to the native plants. The resulting hybrid zone surrounding the botanic garden perimeters would likely receive more diverse genetic pollution and expand in geographic range over time. Clearly, cycad germplasm curators are bound by an ethical responsibility to better understand how to protect the surrounding in situ native plants from the consequences of flying insects which move in and out of the managed ex situ plants.
One facet of garden management that cycad conservationists may exploit is the use of beneficial mixtures of host species to develop neighborhood diversity patterns that beneficially influence herbivore behavior. These protocols may clarify the role of plant genotypic diversity for modifying pest behaviors within the garden settings and decrease the impact that the garden has on the surrounding ecosystem [49,50,51,52]. A pragmatic outcome of this agenda may occur when insecticide applications are used in managed botanic gardens. The Cycas species which exhibit minimal L. pandava damage are clearly not in need of as much effort in applying expensive pesticides [15]. The costs of the pesticide program may therefore be lessened by grouping the heavily damaged Cycas species in close proximity as a means of reducing labor costs and increasing efficacy of the targeted pesticide applications.

4.3. Limited Focus on Relevant Concepts

The reasons for the more rapid larval growth and greater fecundity of adults when L. pandava are reared on the Cycas species that were considered to be high quality are not understood. This knowledge gap limits our ability to best define conservation protocols for host species in newly invaded territories and the reasons for differences in quality deserve further study.
Climate may be an important driver of L pandava herbivory in some locations, but the influences of climate change on L. pandava behavior has not been extensively discussed. Seasonal differences in L. pandava damage and have been reported in monsoon climates where increases in L. pandava herbivory occur during seasons with less rainfall [16,53]. Moreover, seasonal fluctuations of temperature may influence L. pandava performance [54]. These findings indicate future climate change may directly influence the threat levels of L. pandava on cycad conservation, yet what is known is not enough to make predictions about these phenomena. Seasonal difference in L. pandava population pressures in tropical regions such as Guam and the Philippines may be mediated by way of the timing and frequency of vegetative growth events which are required for L. pandava regeneration.
The issue of biological control has also not been developed for countering L. pandava damage to Cycas plants through management decisions. The potential for biological control of L. pandava has been discussed [55,56], and clearly needs to be afforded direct empirical research attention to create a sustainable approach to addressing the herbivory threat.
The topic of multiple cycad-specific arthropod herbivore invasions has also been inadequately studied. Indeed, Cycas plants are not host for single herbivore species, contrarily they may be required to withstand damage from multiple non-native herbivores representing contrasting guilds [57]. One recent study illuminated the relatively greater threat to C. micronesica imposed by A. yasumatsui when compared with several other herbivores including L. pandava [58]. These outcomes were evident in Guam Island where the A. yasumatsui invasion preceded the L. pandava invasion, but equally evident in Rota Island where the L. pandava invasion preceded the A. yasumatsui invasion. The butterfly and armored scale populations appear to be in direct concurrent competition, as the damage by L. pandava increased and decreased in tandem with decreased and increased A. yasumatsui damage [57]. Developing the best management protocols to combat these and other unique interactions among multiple invasive cycad pests will require more research.
Finally, continuing expansion of the invasive range of this devastating cycad pest calls for greater focus on inspection and quarantine protocols in all state territories with a native Cycas species. The first record of a new country invasion was reported in 1992 when the island of Okinawa was reported as invaded [59]. The most recent record of invasion was 2025 when regenerating L. pandava populations were reported in Australia [60]. These new invasions are particularly threatening to range-limited endemic Cycas species that have never been challenged by an alien leaf herbivore. However, the threats to native biota may be even more magnified for Cycas populations that support a native leaf herbivore, such as the Australian Cycad Blue Theclinesthes onycha Hewitson [61]. In these situations, the threat of extinction is caused by L. pandava is not restricted to the new Cycas host species, it extends to the native herbivore which is forced into direct competition with the invasive species, as the native butterfly relies on the Cycas host for regeneration.

5. Conclusions

The threat to cycad conservation caused by the cycad leaf herbivore L. pandava has increased in recent years due to new territory invasions. The results from dual choice cafeteria trials indicated that L. pandava larvae derived from in situ Cycas habitats exhibited beneficial discriminatory behaviors when challenged with novel food choices. However, these selection behaviors were not evident for larvae from a resident L. pandava population which was provisioned with larval food from numerous Cycas species that were not available within the butterfly’s native habitat. The findings indicate that Cycas plants within botanic gardens with resident L. pandava are threatened by an herbivore that is more voracious than the corresponding wild type herbivore. Moreover, when a botanic garden is located within the natural range of a native Cycas species, or when an insular Cycas population experiences a new L. pandava invasion, the threats to the native plants are magnified by the herbivore’s ability to become less discerning, more efficient, and more voracious than the wild type butterflies.

Funding

This research was funded in part by the United States Department of Agriculture McIntire Stennis project number 2013-32100-06057.

Data Availability Statement

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding author.

Acknowledgments

I thank Kampon Tansacha for logistical support, nursery bench space, and access to the germplasm collection at the Nong Nooch Tropical Botanical Garden. I thank Anders Lindström and Paris Marler for help in maintaining the Thailand butterfly population. Gil Cruz and Frankie Matanane are thanked for help in maintaining the Guam ex situ germplasm sites. Gerard Sigua is thanked for logistical support in maintaining the Philippine germplasm garden.

Conflicts of Interest

Author Thomas Marler was a volunteer for the non-government organization Philippine Native Plants Conservation Society.

Abbreviations

The following abbreviations are used in this manuscript:
NNTBGNong Nooch Tropical Botanical Garden

References

  1. Affinito, F.; Williams, J.M.; Campbell, J.E.; Londono, M.C.; Gonzalez, A. Progress in developing and operationalizing the Monitoring Framework of the Global Biodiversity Framework. Nat. Ecol. Evol. 2024, 8, 2163–2171. [Google Scholar] [CrossRef]
  2. McGowan, P.J.; Hutchinson, A.; Brooks, T.M.; Elliott, W.; Hoffmann, M.; Mair, L.; McDougall, A.; Raimondo, D.C.; Butchart, S.H. Understanding and achieving species elements in the Kunming–Montreal Global Biodiversity Framework. Bioscience 2024, 74, 614–623. [Google Scholar] [CrossRef]
  3. Butchart, S.H.M.; Akçakaya, H.R.; Berryman, A.J.; Brooks, T.M.; Burfield, I.J.; Chanson, J.; Dias, M.P.; Donaldson, J.S.; Hermes, C.; Hilton-Taylor, C.; et al. Measuring trends in extinction risk: A review of two decades of development and application of the Red List Index. Phil. Trans. R. Soc. B 2025, 380, 20230206. [Google Scholar] [CrossRef]
  4. Norstog, K.J.; Nicholls, T.J. The Biology of the Cycads; Cornell University Press: Ithaca, NY, USA, 1997. [Google Scholar]
  5. Donaldson, J.S. (Ed.) Cycads: Status Survey and Conservation Action Plan; IUCN/SSC Cycad Specialist Group: Cambridge, UK, 2003. [Google Scholar]
  6. Brummitt, N.A.; Bachman, S.P.; Griffiths-Lee, J.; Lutz, M.; Moat, J.F.; Farjon, A.; Donaldson, J.S.; Hilton-Taylor, C.; Meagher, T.R.; Albuquerque, S.; et al. Green plants in the red: A baseline global assessment for the IUCN sampled red list index for plants. PLoS ONE 2015, 10, e0135152. [Google Scholar] [CrossRef]
  7. Fragniere, Y.; Betrisey, S.; Cardinaux, L.; Stoffel, M.; Kozlowski, G. Fighting their last stand? A global analysis of the distribution and conservation status of gymnosperms. J. Biogeogr. 2015, 42, 809–820. [Google Scholar] [CrossRef]
  8. Yessoufou, K.; Daru, B.H.; Tafirei, R.; Elansary, H.O.; Rampedi, I. Integrating biogeography, threat and evolutionary data to explore extinction crisis in the taxonomic group of cycads. Ecol. Evol. 2017, 7, 2735–2746. [Google Scholar] [CrossRef]
  9. Tang, W.; Donaldson, J.; Haynes, J.; Terry, I. International Union for Conservation of Nature Cycad Specialist Group Report and Recommendations on Cycad Aulacaspis Scale, Aulacaspis yasumatsui Takagi (Hemiptera: Diaspididae); IUCN: Gland, Switzerland, 2005. [Google Scholar]
  10. Tang, W.; Oberpreiler, R. Pest Alert: Insect Pests of Cycads; IUCN/SSC Cycad Specialist Group, IUCN: Gland, Switzerland, 2006. [Google Scholar]
  11. Mankga, L.T.; Yessoufou, K. Factors driving the global decline of cycad diversity. AoB PLANTS 2017, 9, plx022. [Google Scholar] [CrossRef]
  12. Calonje, M.; Stevenson, D.W.; Osborne, R. The World List of Cycads. 2013–2025. Available online: http://www.cycadlist.org (accessed on 6 August 2025).
  13. Wu, L.-W.; Yen, S.-H.; Lees, D.C.; Hsu, Y.-F. Elucidating genetic signatures of native and introduced populations of the Cycad Blue, Chilades pandava to Taiwan: A threat both to Sago Palm and to native Cycas populations worldwide. Biol. Invasions 2010, 12, 2649–2669. [Google Scholar] [CrossRef]
  14. Whitaker, M.R.; Salzman, S. Ecology and evolution of cycad-feeding Lepidoptera. Ecol. Lett. 2020, 23, 1862–1877. [Google Scholar] [CrossRef]
  15. Marler, T.E.; Lindström, A.J.; Terry, L.I. Chilades pandava damage among 85 Cycas species in a common garden setting. HortScience 2012, 47, 1832–1836. [Google Scholar] [CrossRef]
  16. Purti, N.; Singh, L.J.; Pandey, A.K. New hosts for the cycad blue butterfly, Luthrodes pandava, Horsfield (Lepidoptera: Lycaenidae) in an island ecosystem. Feddes Repert. 2022, 133, 234–243. [Google Scholar] [CrossRef]
  17. Marler, T.E.; Lindström, A.J.; Marler, P.N. Diversity in Cycas (Cycadales: Cycadaceae) species offered as larval food influences fecundity of Chilades pandava (Lepidoptera: Lycaenidae) adults. Int. J. Insect Sci. 2017, 9, 1–7. [Google Scholar] [CrossRef] [PubMed]
  18. Hassan, G.M.; Batt, M.A.; El-Aassar, M.R. Discriminatory of some morphometry and biological aspects of the Cycad blue butterfly, Chilades pandava reared on Cycas and Zamia ornamental palms. Braz. J. Biol. 2024, 84, e262610. [Google Scholar] [CrossRef]
  19. Purti, N.; Singh, L.J.; Pandey, A.K. Life history traits of cycad blue butterfly, Luthrodes pandava Horsfield (Lepidoptera: Lycaenidae) on cycad hosts in an Island ecosystem. Feddes Repert. 2024, 135, 44–58. [Google Scholar] [CrossRef]
  20. Marler, T.E.; Lindström, A.J.; Marler, P.N. Chilades pandava mothers discriminate among Cycas species during choice tests, but only in an endemic naïve population. Plant Signal. Behav. 2016, 11, e1208879. [Google Scholar] [CrossRef]
  21. Moore, A.; Marler, T.; Miller, R.H.; Muniappan, R. Biological control of cycad aulacaspis scale on Guam. Cycad Newsl. 2005, 28, 6–8. [Google Scholar]
  22. Roberts, J.M.; Clunie, B.J.; Leather, S.R.; Harris, W.E.; Pope, T.W. Scents and sensibility: Best practice in insect olfactometer bioassays. Entomol. Exp. Appl. 2023, 171, 808–820. [Google Scholar] [CrossRef]
  23. Schneider, C.A.; Rasband, W.S.; Eliceiri, K.W. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 2012, 9, 671–675. [Google Scholar] [CrossRef]
  24. Thuiller, W.; Albert, C.; Araújo, M.B.; Berry, P.M.; Cabeza, M.; Guisan, A.; Hickler, T.; Midgley, G.F.; Paterson, J.; Schurr, F.M.; et al. Predicting global change impacts on plant species’ distributions: Future challenges. Perspect. Plant Ecol. Evol. Syst. 2008, 9, 137–152. [Google Scholar] [CrossRef]
  25. Enquist, B.J.; Feng, X.; Boyle, B.; Maitner, B.; Newman, E.A.; Jørgensen, P.M.; Roehrdanz, P.R.; Thiers, B.M.; Burger, J.R.; Corlett, R.T. The commonness of rarity: Global and future distribution of rarity across land plants. Sci. Adv. 2019, 5, eaaz0414. [Google Scholar] [CrossRef]
  26. Sierra-Botero, L.; Calonje, M.; Robbins, R.K.; Rosser, N.; Pierce, N.E.; López-Gallego, C.; Valencia-Montoya, W.A. Cycad phylogeny predicts host plant use of Eumaeus butterflies. Ecol. Evol. 2023, 13, e09978. [Google Scholar] [CrossRef]
  27. Marler, T.E.; Lindström, A.; Terry, L.I. Information-based or resource-based systems may mediate Cycas-herbivore interactions. Plant Signal. Behav. 2012, 7, 760–762. [Google Scholar] [CrossRef][Green Version]
  28. Root, R.B. Organization of a plant-arthropod association in simple and diverse habitats: The fauna of collards (Brassica oleracea). Ecol. Monogr. 1973, 43, 95–124. [Google Scholar] [CrossRef]
  29. Schneider, D.; Wink, M.; Sporer, F.; Lounibos, P. Cycads: Their evolution, toxins, herbivores and insect pollinators. Naturwissenschaften 2002, 89, 281–294. [Google Scholar] [CrossRef]
  30. Joshi, Y.; Bansal, P.; Yadav, A.L. Cycadales’ defense against insect and mammalian herbivory: Do endophytic fungi have to say something?—A hypothesis. Fungal Biol. Rev. 2024, 50, 100393. [Google Scholar] [CrossRef]
  31. Prado, A. Chemical Ecology of Azoxyglycosides: Plant-Insect Interactions Between Neotropical Cycads (Zamiaceae) and Leaf Herbivores. Ph.D. Dissertation, McGill University, Montreal, QC, Canada, 2014. [Google Scholar]
  32. Salzman, S. The Ecology and Evolution of Cycads and Their symbionts. Ph.D. Dissertation, Harvard University, Cambridge, MA, USA, 2019. [Google Scholar]
  33. Whitaker, M.R.L.; Gilliéron, F.; Skirgaila, C.; Mescher, M.C.; De Moraes, C.M. Experimental evidence challenges the presumed defensive function of a “slow toxin” in cycads. Sci. Rep. 2022, 12, 6013. [Google Scholar] [CrossRef]
  34. Whitaker, M.R.L.; Banack, S.A.; Mescher, M.C.; Cox, P.A.; De Moraes, C.M. BMAA in cycad-feeding Lepidoptera: Defensive sequestration or bioaccumulation? Front. Ecol. Evol. 2023, 11, 1114636. [Google Scholar] [CrossRef]
  35. Du, Q.; Xing, N.; Guo, S.; Li, R.; Meng, X.; Wang, S. Cycads: A comprehensive review of its botany, traditional uses, phytochemistry, pharmacology and toxicology. Phytochem. 2024, 220, 114001. [Google Scholar] [CrossRef]
  36. Osborne, R.; Stevenson, D.W.; Ferreira, D. Phytochemistry of the Cycadales. Bot. Rev. 2025, 24, 1–65. [Google Scholar] [CrossRef]
  37. Salzman, S.; Bustos-Díaz, E.D.; Whitaker, M.R.L.; Sierra, A.M.; Cibrián-Jaramillo, A.; Barona-Gómez, F.; Villarreal Aguilar, J.C. Chemical ecology of symbioses in cycads, an ancient plant lineage. New Phytol. 2025, 246, 1494–1504. [Google Scholar] [CrossRef]
  38. Bernays, E.A.; Chapman, R.F. Host-Plant Selection by Phytophagous Insects; Chapman and Hall: London, UK, 1994. [Google Scholar]
  39. Soler, R.; Pineda, A.; Li, Y.; Ponzio, C.; van Loon, J.J.A.; Weldegergis, B.T.; Dicke, M. Neonates know better than their mothers when selecting a host plant. Oikos 2012, 121, 1923–1934. [Google Scholar] [CrossRef]
  40. Gamberale-Stille, G.; Söderlind, L.; Janz, N.; Nylin, S. Host plant choice in the comma butterfly-larval choosiness may ameliorate effects of indiscriminate oviposition. Insect Sci. 2014, 21, 499–506. [Google Scholar] [CrossRef]
  41. Homem-de-Carvalho, D.; Vitor Caetano Souza, J.; Rodrigues, D. A review and meta-analysis of host-plant selection by lepidopteran larvae: No evidence of diet breadth constraints on larval decision-making. Oikos 2025, 2025, e11333. [Google Scholar] [CrossRef]
  42. Hsu, Y.F. Systematic position and description of Chilades peripatria sp. nov. (Lepidoptera: Lycaenidae). Bull. Inst. Zool. Acad. Sin. 1989, 28, 55–62. [Google Scholar]
  43. Madulid, D.A.; Agoo, E.M.G. Taxonomy and conservation of Philippine Cycads. Blumea 2009, 54, 99–102. [Google Scholar] [CrossRef][Green Version]
  44. Montgomery Botanical Center. Introduction to Cycads. Available online: https://www.montgomerybotanical.org/about/cycads/ (accessed on 12 September 2025).[Green Version]
  45. Tang, W. Heat production in cycad cones. Bot. Gaz. 1987, 148, 165–174. [Google Scholar] [CrossRef]
  46. Tang, W. Insect pollination in the cycad Zamia pumila (Zamiaceae). Amer. J. Bot. 1987, 74, 90–99. [Google Scholar] [CrossRef]
  47. Tang, W. Reproduction in the cycad Zamia pumila in a fire-climax habitat: An eight-year study. Bull. Torrey Bot. Club 1990, 117, 368–374. [Google Scholar] [CrossRef]
  48. Meinard, Y.; Georges, J.-Y. Frontiers of conservation. Conserv. Biol. 2025, 39, e14432. [Google Scholar] [CrossRef]
  49. Hambäck, P.A.; Inouye, B.D.; Andersson, P.; Underwood, N. Effects of plant neighborhoods on plant–herbivore interactions: Resource dilution and associational effects. Ecology 2014, 95, 1370–1383. [Google Scholar] [CrossRef] [PubMed]
  50. Stastny, M.; Agrawal, A.A. Love thy neighbor? reciprocal impacts between plant community structure and insect herbivory in co-occurring Asteraceae. Ecology 2014, 95, 2904–2914. [Google Scholar] [CrossRef]
  51. Grettenberger, I.M.; Tooker, J.F. Inter-varietal interactions among plants in genotypically diverse mixtures tend to decrease herbivore performance. Oecologia 2016, 182, 189–202. [Google Scholar] [CrossRef] [PubMed]
  52. Hahn, P.G.; Orrock, J.L. Neighbor palatability generates associational effects by altering herbivore foraging behavior. Ecology 2016, 97, 2103–2111. [Google Scholar] [CrossRef]
  53. Tiple, A.; Agashe, D.; Khurad, A.M.; Kunte, K. Population dynamics and seasonal polyphenism of Chilades pandava butterfly (Lycaenidae) in central India. Curr. Sci. 2009, 97, 1774–1779. [Google Scholar]
  54. Ravuiwasa, K.T.; Tan, C.-W.; Hwang, S.-Y. Temperature-dependent demography of Chilades pandava peripatria (Lepidoptera: Lycaenidae). J. Econ. Entomol. 2011, 104, 1525–1533. [Google Scholar] [CrossRef]
  55. Chang, Y.C. Morphology, life history and damage of Cycas blue butterfly Chilades pandava as well as pathogencity of entomogenous fungus to its larva. Bull. Taiwan For. Res. Inst. New Ser. 1989, 4, 43–50. [Google Scholar]
  56. El-Aassar, M.R.; Batt, M.A.; Hassan, G.M. The occurrence of Chilades pandava infested Zamia palms and its control by some biological and safe compounds. Menoufia J. Plant Prot. 2024, 9, 189–197. [Google Scholar] [CrossRef]
  57. Marler, T.E. Temporal variations in leaf miner, butterfly, and stem borer infestations of Cycas micronesica in relation to Aulacaspis yasumatsui incidence. HortScience 2013, 48, 1334–1338. [Google Scholar] [CrossRef]
  58. Marler, T.E.; Cruz, G.N. Insular Cycas micronesica habitats respond similarly to Aulacaspis yasumatsui invasion, regardless of co-occurring consumers. Forests 2024, 15, 22. [Google Scholar] [CrossRef]
  59. Mitsuhashi, W. A new record of Chilades pandava (Lepidoptera, Lycaenidae) from Japan. Choken Field 1992, 81, 8–9. (In Japanese) [Google Scholar]
  60. Lambkin, T.; Mclaren, R.G. First confirmed breeding population of the cycad blue butterfly, Luthrodes pandava (Horsfield) (Lepidoptera: Lycaenidae), from Australia. Austral. Entomol. 2025, 52, 83–89. [Google Scholar]
  61. Braby, M.F. The Complete Field Guide to Butterflies of Australia; CSIRO Publishing: Collingwood, Australia, 2004. [Google Scholar]
Insects 16 00973 g001
Figure 1. Luthrodes pandava adult female butterflies in Thailand. (a) Derived from Tak Fa from an endemic Cycas nongnoochiae habitat; (b) derived from Nong Nooch Tropical Botanical Garden.
Figure 1. Luthrodes pandava adult female butterflies in Thailand. (a) Derived from Tak Fa from an endemic Cycas nongnoochiae habitat; (b) derived from Nong Nooch Tropical Botanical Garden.
Insects 16 00973 g001
Insects 16 00973 g002
Figure 2. Luthrodes pandava adult female butterflies in the Philippines. (a) Derived from Mansalay, Mindoro from a Cycas edentata habitat; (b) derived from Angeles City, Pampanga from Cycas plants in the urban forest.
Figure 2. Luthrodes pandava adult female butterflies in the Philippines. (a) Derived from Mansalay, Mindoro from a Cycas edentata habitat; (b) derived from Angeles City, Pampanga from Cycas plants in the urban forest.
Insects 16 00973 g002
Insects 16 00973 g003
Figure 3. Morphology of Luthrodes pandava from two islands. (a) Invasive butterfly adult on Guam feeding on the invasive Bidens pilosa; (b) native butterfly in southern Zambales, Philippines.
Figure 3. Morphology of Luthrodes pandava from two islands. (a) Invasive butterfly adult on Guam feeding on the invasive Bidens pilosa; (b) native butterfly in southern Zambales, Philippines.
Insects 16 00973 g003
Insects 16 00973 g004
Figure 4. Dual-choice cafeteria feeding trials with 2nd instar Luthrodes pandava larvae and partially expanded leaflets from two Cycas species. (a) Vessels used for the 8 h feeding trials; (b) initial layout and positioning of the larvae with high-quality Cycas micronesica leaflet (left) and low-quality Cycas wadei leaflet (right); (c) example of appearance of the partially consumed leaflet tissue after 8 h feeding cycle. Markers on scale are 1 mm.
Figure 4. Dual-choice cafeteria feeding trials with 2nd instar Luthrodes pandava larvae and partially expanded leaflets from two Cycas species. (a) Vessels used for the 8 h feeding trials; (b) initial layout and positioning of the larvae with high-quality Cycas micronesica leaflet (left) and low-quality Cycas wadei leaflet (right); (c) example of appearance of the partially consumed leaflet tissue after 8 h feeding cycle. Markers on scale are 1 mm.
Insects 16 00973 g004
Insects 16 00973 g005
Figure 5. The area of leaflet tissue consumed by Thailand Luthrodes pandava larvae during an 8 h feeding cycle. Yellow bars depict high quality species, blue bars depict low quality species. (a,c,e,g) Wild population collected from Tak Fa; (b,d,f,h) urban population collected from Nong Nooch Tropical Botanical Garden Mean ± SD, n = 10. Bars within each panel with same letters do not significantly differ.
Figure 5. The area of leaflet tissue consumed by Thailand Luthrodes pandava larvae during an 8 h feeding cycle. Yellow bars depict high quality species, blue bars depict low quality species. (a,c,e,g) Wild population collected from Tak Fa; (b,d,f,h) urban population collected from Nong Nooch Tropical Botanical Garden Mean ± SD, n = 10. Bars within each panel with same letters do not significantly differ.
Insects 16 00973 g005
Insects 16 00973 g006
Figure 6. The area of leaflet tissue consumed by Philippine Luthrodes pandava larvae during an 8 h feeding cycle. Yellow bars depict high quality species, blue bars depict low quality species. (a) Wild population collected from Mindoro; (b) urban population collected from Angeles City; (c) wild population collected from Zambales; (d) urban population collected from Angeles City. Mean ± SD, n = 40. Bars within each panel with same letters do not significantly differ.
Figure 6. The area of leaflet tissue consumed by Philippine Luthrodes pandava larvae during an 8 h feeding cycle. Yellow bars depict high quality species, blue bars depict low quality species. (a) Wild population collected from Mindoro; (b) urban population collected from Angeles City; (c) wild population collected from Zambales; (d) urban population collected from Angeles City. Mean ± SD, n = 40. Bars within each panel with same letters do not significantly differ.
Insects 16 00973 g006
Insects 16 00973 g007
Figure 7. The area of leaflet tissue consumed by larvae derived from the invasive population of Luthrodes pandava throughout the island of Guam. Yellow bar depicts high quality species, blue bar depicts low quality species. Larvae were provided access to two Cycas species for 8 h. Mean ± SD, n = 40. Bars with same letters do not significantly differ.
Figure 7. The area of leaflet tissue consumed by larvae derived from the invasive population of Luthrodes pandava throughout the island of Guam. Yellow bar depicts high quality species, blue bar depicts low quality species. Larvae were provided access to two Cycas species for 8 h. Mean ± SD, n = 40. Bars with same letters do not significantly differ.
Insects 16 00973 g007
Table 1. Experimental conditions for Luthrodes pandava larvae feeding trials using Cycas hosts in Guam, Philippines, and Thailand.
Table 1. Experimental conditions for Luthrodes pandava larvae feeding trials using Cycas hosts in Guam, Philippines, and Thailand.
TrialDatesMax Temp (°C) Min Temp (°C)PPFD
(µmol·m−2·s−1)
Thailand
C. nongnoochiae		19–24 July
2013		32.926.624–32
Philippine
C. edentata		30 January–5 February 201928.722.333–40
Guam
Population		29 January–4 February 202330.126.120–24
Philippine
C. zambalensis		28 May–5 June
2024		33.326.126–31
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Marler, T.E. Luthrodes pandava Larvae Can Distinguish Cycas Leaf Quality in Cafeteria Experiments. Insects 2025, 16, 973. https://doi.org/10.3390/insects16090973

AMA Style

Marler TE. Luthrodes pandava Larvae Can Distinguish Cycas Leaf Quality in Cafeteria Experiments. Insects. 2025; 16(9):973. https://doi.org/10.3390/insects16090973

Chicago/Turabian Style

Marler, Thomas E. 2025. "Luthrodes pandava Larvae Can Distinguish Cycas Leaf Quality in Cafeteria Experiments" Insects 16, no. 9: 973. https://doi.org/10.3390/insects16090973

APA Style

Marler, T. E. (2025). Luthrodes pandava Larvae Can Distinguish Cycas Leaf Quality in Cafeteria Experiments. Insects, 16(9), 973. https://doi.org/10.3390/insects16090973

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop