Beetle eggs, larvae, and adults used in these experiments were drawn from a CRB colony that had been established using adults collected from traps deployed in the field. These field-caught specimens were quarantined for at least 7 days within the Hawai’i Department of Agriculture Plant Quarantine facilities to ensure that they harbored no obvious pests or pathogens then transferred to the University of Hawai’i Manoa Arthropod Containment Laboratory (UH-ACL). Adult beetles collected from the field (founders) were placed in boxes (26 cm × 19 cm × 17 cm) containing a mulch mixture and allowed to mate and lay eggs.

Initially, a 50:50 (v/v) mixture of the commercial compost Menehune MAGIC™ (Hawaiian Earth Products, Kapolei, HI, USA) and coconut coir peat (Pacific Grower Supplies, Hilo, HI, USA) was used as the medium for all life stages and the food source for larvae. In 2017, the mulch used in the colony transitioned to green waste that is pruned from campus plants and processed by the University of Hawaii's Landscaping Services. The Landscape Services mulch was sieved through either a 1.6-mm or a 6.4-mm mesh to remove large pieces of material, creating either “fine” or “coarse” mulch, respectively. The coarse mulch was used in the adult boxes.

The F1 generation was reared to the adult stage. As eggs were collected from the founder boxes with adults, they were placed in individual containers (26 cm × 19 cm × 17 cm) containing mulch. The larvae then fed upon the mulch, pupated, and emerged as adults, which were fed as described below. This new generation of CRB adults was allowed to mate. After the successful production of an F2 generation, adults of the founder, F1, and F2 generations were commingled. Up to 15 adult CRBs with a female-to-male ratio of approximately 2:1 were housed in plastic containers (43 cm × 28 cm × 17 cm) containing coarse mulch. Eggs were collected weekly and transferred to plastic containers containing fine mulch, up to 300 eggs/container. After reaching the late second or early third instar, larvae were transferred to individual 1-L bottles containing coarse mulch until pupation and imago emergence. Rearing took place in I-41LLVL Biological Incubators (Percival Scientific, Perry, IA, USA) at temperatures between 28°C and 30°C, and the medium was exchanged as needed. Humidity was provided by placing buckets of water within the incubator. Adult beetles were fed a solid artificial diet once per week, amounting to approximately 13 g/beetle/week. The diet consisted of 33 g of sugar and 26 g of Muscle Milk vanilla-flavored 100% Whey Protein Isolate and Concentrate (Cytosport, Walnut Creek, CA, USA) dissolved in 1 L of boiling water, and solidified with 15 g of agar (Fisher BioReagents, Pittsburgh, PA, USA). All specimens used in this study came from offspring produced by the commingled founder, F1, and F2 adult beetles, and all experimentation was conducted within the UH-ACL.

Approximately 500 g of biosolids collected from a single location within the CRB infestation zone was thoroughly examined to ensure that no CRB life stages were present. These biosolids were then submitted to the University of Hawai’i Agriculture Diagnostic Service Center to be analyzed for pH, electroconductivity (EC), and concentration of phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), and sodium (Na) (6). The resulting values were compared with the range of values for soils in Hawai’i (7, 8).

To evaluate the ability of CRBs to survive and colonize biosolids, eggs collected from the colony the day prior were placed individually in 111-cc translucent plastic containers with a plastic lid with four holes pierced in the top for air exchange (Dart Container Corporation, Mason, MI, USA). Each cup contained 100% biosolids, a mixture of fine mulch and biosolids (75%, 50%, or 25% biosolids, v/v), or 100% fine mulch (control treatment), as described for colony rearing. Each container held approximately 100 cc of one of the biosolid mixtures, and a single egg was placed on the surface and lightly covered with the mulch.

Ten eggs were evaluated for each of the four mixtures and the control. Eggs were initially checked after 2 weeks, and weekly thereafter, for hatching. If the egg hatched, the larva was monitored weekly for its mortality, instar stage, and weight. After 5 weeks, the surviving larvae were transferred to larger (≈ 575 cc) Magenta™ GA-7 vessels (Sigma-Aldrich, St. Lois, MO, USA) with perforated lids, containing freshly made mulch at the same biosolids to fine mulch ratio. Monitoring for larval survival, instar, and weight continued until all surviving larvae for the control treatment reached the third instar. If a larva was dead, it was noted as “dead” and no further records were maintained for that individual. Unhatched eggs were returned to the container for continued observation until the egg hatched or the egg was not found for 2 consecutive weeks.

If an egg or larva was not found, mulch was kept for another week of observation in case the egg or first instar was missed. If nothing was found for 2 consecutive weeks of observation, that egg/larvae was determined to be “dead”. If larvae were found alive after 1 week of missed observation, the life stage and weight were recorded and monitoring continued. If a larva was found after 1 week of missed observation and it was dead, it was noted as “dead” and no further records were maintained for that individual. The procedure was replicated three times.

Survival curves were created and plotted for overall larval survival in each media treatment using the Surv function from the Survival package (9) in R (10). Time was reported in the number of weeks from when eggs entered each medium until larval death or the end of the observation period. A pairwise comparison of survival curves by log-rank tests was used to determine the specific significance between treatments.

To evaluate the survival of CRBs in mulch containing incremental concentrations of common metal salts, the Menehune MAGIC™ potting mix was spread out and dehydrated in aluminum pans for at least 1 day in an oven at 120°C. Single preliminary trials were conducted with salt solutions at 0.25 M, 0.5 M, 0.75 M, and 1 M concentrations. Deionized water (dH₂O) served as a control treatment. Each cubic centimeter of dried mulch was rehydrated with 0.46 mL of salt solution or dH₂O and placed in a Magenta™ GA-7 vessel or 32-oz round culture vessel (PhytoTechnology Laboratories, Lenexa, KS, USA) with a perforated lid. Sodium chloride (NaCl, m.w. 58.44), potassium chloride (KCl, m.w. 74.55), magnesium sulfate (MgSO₄, m.w. 120.37), calcium sulfate (CaSO₄·2H₂O, m.w. 172.17), magnesium chloride (MgCl₂·6H₂O, m.w. 203.3), and calcium chloride (CaCl₂·2H₂O, m.w. 147.02) were the metal salts selected for evaluation. To evaluate these treatments, first-instar larvae and second-instar larvae were placed in separate communal containers and incubated at 30°C. Each treatment had five of each instar stages. Larvae were evaluated after 1 week for survival. Those larvae that disintegrated and whose remains were not recovered were considered dead. Three replicates were performed for each instar stage and salt concentration.

The results from the preliminary trials were used to predict a range of concentrations of the various salts that might be lethal to the first and second instars of CRBs. These subsequent trials followed the same methodology as the preliminary trials, but the range of concentrations of salts (in 0.1M increments) varied depending on the larval stage and salt used. After the initial test with MgSO₄, subsequent tests were conducted using a commercially available Epsom salt (MgSO₄·7H₂O; Gro Tec, Inc., Madison, GA, USA) based on the application rate on the package. Treatments of 1× (0.13 M), 2× (0.26 M), and 5× (0.64 M) of the recommended application rate of Epsom salt was used to test both the first and second instars.

The survival data for the secondary trials were analyzed using R software (10) with the dplyr package (11). Univariate ANOVAs were conducted for each life stage (first or second instar) and for each salt (sodium chloride, potassium chloride, Epsom salt/magnesium sulfate, calcium sulfate, magnesium chloride, and calcium chloride) tested. Analyses of instar stages across salt treatments were not combined because different salt concentrations were used for the first and second instars. For each ANOVA, a Tukey honestly significant difference (HSD) test was conducted to determine differences in larval (first or second instar) survival at different concentrations of the salt tested.

The effects of acute and chronic exposure of CRB first and second instars to a commercial formulation of Epsom salt (MgSO₄·7H₂O) intended for horticultural applications (Gro Tec, Inc.) were evaluated. Replicated acute exposure trials were conducted as described above, except that the Menehune MAGIC™ potting mix was replaced with the fine mulch sourced from UH landscaping, as used in the colony maintenance. The mulch was rehydrated using Epsom salt concentrations at 1×, 2×, and 5× the manufacturer’s recommended application rate for potted palms (0.5 cup/gallon of water), which was estimated to be 0.13, 0.26, and 0.64 M, respectively. The control treatment consisted of rehydration with dH₂O. For each treatment, the percentage that survived to each larval stage was analyzed by ANOVA and the Tukey HSD test, as described above.

For chronic exposure, control and Epsom salt-treated fine mulch was placed into the 111-cc translucent plastic containers described above. An individual CRB egg was placed in the center of each container and covered with approximately 5 mm of fine mulch, as described for colony rearing. The containers were maintained at 30°C and weekly observations began after 2 weeks. If larvae were present, their life stage and weight were recorded, then the larvae were returned to their containers. If the eggs were still present and appeared vital, they were placed back into the cup for further observation. If the eggs were missing or in poor condition, or there was no evidence of larvae following 2 weeks of observations, the egg was considered dead and no further records were kept for that specimen. All cups with samples were returned to 30°C and checked weekly for life stage and weight. After 5 weeks, the mulch was replaced with fresh mulch. After 10 weeks, larvae were moved into Magenta™ GA-7 vessels with fresh mulch to accommodate larval growth. Mulch was replaced after 13 weeks, and again after 19 weeks if there was no evidence of pupation. Once pupation was initiated, weekly measurements ceased, due to the fragility of this life stage, but the containers were monitored for emergence. Once adults emerged, measurements of beetle length, elytron length, width, height, and weight were recorded. Three independent replicates were performed concurrently, with 10 eggs per treatment for each replicate.

As described for the biosolid trials, survival curves were created and plotted for overall larval survival. Curves were created for each using the Surv function from the Survival package (9) in R (10). Time was reported in the number of weeks from when eggs entered each medium until larval death or the end of the observation period. A pairwise comparison of survival curves by log-rank tests was used to determine the specific significance between treatments.

To determine if the presence of Epsom salt had an impact on the oviposition behavior of female CRBs, choice tests were conducted with specimens caught from the environment and maintained in the colony at the UH-ACL. These tests were performed under laboratory conditions in a wooden chamber with an arena composed of three compartments. One compartment contained sand and served as a putatively undesirable substrate for oviposition. The other two compartments contained green waste that had been passed through a 13-mm screen. The green waste in one compartment was dehydrated as described above and rehydrated with dH₂O. The other compartment contained green waste that was rehydrated with either 2× or 5× Epsom salt solution. The contents of all three compartments were flush with the dividers, with no surface obstacles between treatments. A single female was placed in the sand compartment, the lid of the chamber was closed and secured, and the chamber was stored at 30°C. The experiment concluded after 1 week when the compartments were excavated, and the number of eggs recovered from each compartment was recorded. A different female and fresh mulch and sand were added to each compartment at the beginning of each replication. A total of 12 replicates were performed for each Epsom salt concentration (2× and 5×).

For each choice experiment, the number of eggs laid per beetle in each substrate was compared using a non-parametric Wilcoxon rank-sum test (JMP Pro 13). If no eggs were found in any substrate, this category was not included in the analysis. The oviposition pattern for each beetle was scored and analyzed using categorical data analysis techniques. A single female beetle was scored as laying her eggs in mulch with Epsom salt added, mulch without salt added, sand, both mulches, or none. The null hypothesis, that oviposition would be equally distributed across substrates, was tested using a chi-squared goodness-of-fit test (JMP Pro 13). Where appropriate, a subdivided chi-squared goodness-of-fit test (12) was subsequently conducted to determine if the distribution of ovipositing was the same in both mulch substrates within the same trial, or if one of the two mulch substrate options was preferred. The sand category was not included in any subsequent analysis, as no beetles were ever found to oviposit in that substrate.

An analysis of the biosolid material within the CRB infestation zone and comparison with the expected values for heavy agricultural soils in Hawai’i revealed that the biosolid pH, P, and Ca values were similar to those expected, whereas the EC, K, Mg, and Na values were very highly elevated (Table 1). These elevated values suggest that the biosolid material would not support the growth of most agricultural crops (7).

Treatments with 25% biosolids had higher survival probabilities than treatments with 50% or more biosolids (Figure 1). Survival probabilities for treatments with 25% biosolids appeared comparable to survival probability with the control treatment. In addition, survival probability was similar for the 75% and 100% biosolid treatments. The pairwise comparison of the survival curves confirmed that there was no significant difference in the survival of larvae between the 25% biosolid medium and the control (p = 0.97). Additionally, the survival probabilities did not differ significantly between the 75% and 100% biosolid treatments (p = 0.42). Larval survival in the treatment with 50% biosolids was significantly different from survival with the control treatment (p = 2.31e–3) and from survival with the 25% (p = 1.159e–2), 75% (p = 4.4e–4), and 100% (p = 1.17e–3) biosolid treatments.

The survival of first- and second-instar CRB larvae was evaluated after 7 days of exposure to mulch containing incremental concentrations of six different salts: NaCl, KCl, MgSO4, CaSO₄·2H₂O, MgCl₂·6H₂O, and CaCl₂·2H₂O. All salts significantly reduced the survival of these larvae as their concentrations in the mulch increased (Figures 2A–D and data not shown). Survival rates were typically higher for second-instar larvae than for first-instar larvae. Mulch rehydrated with solutions greater than 0.5 M in concentration, regardless of the salt used, typically resulted in no survival of either first- or second-instar larvae after 7 days of exposure.

A commercially available Epsom salt intended for horticultural applications was evaluated for its impact on CRB survival following acute exposure. At 1×, 2×, and 5× the recommended application rate following a 7-day (acute) exposure, the survival of first- and second-instar larvae decreased as the amount of Epsom salt in the mulch increased, with no first-instar larvae surviving the 5× treatment (Figure 2E). Following chronic exposure, overall larvae survival probabilities were higher for the control, 1×, and 2× treatments than for the 5× treatment. They also did not appear to be significantly different from each other (Figure 3). This was confirmed with the pairwise log-rank test. The 5× treatment was significantly different from all other treatments (control: p = 3.0e–11; 1×: p = 8.2e–6; 2×: p = 4.0e–8). The only other significant difference between treatments was between the 1× and 2× treatments (p = 0.049).

The development of first- and second-instar CRB larvae (but not third-instar larvae or pupae) was delayed when chronically exposed to mulch containing 2× the recommended rate of Epsom salt (Figure 4A). Furthermore, the average weight of the larvae decreased with increasing levels of Epsom salt in the mulch (Figure 4B). The body weight and length of newly emerged adult CRBs were also reduced with increasing Epsom salt levels (Figures 4C, D). Similar to acute exposure experiments, no larvae survived in mulch supplemented with 5× the recommended rate of Epsom salt.

Female CRBs laid on average ( ± SE) 29 ± 9 eggs when presented with the choices of mulch amended with 2× Epsom salt, unamended mulch, and sand. Although slightly more eggs were laid by larvae receiving the 2× Epsom salt treatment than by larvae subjected to the control treatment, the difference was not statistically different (Table 2). No beetles laid eggs in the sand. Three of the 12 beetles oviposited in both 2× Epsom salt and control treatments, eight chose either the 2× Epsom salt (n = 3) or control (n = 5) treatments, and one beetle did not oviposit. The numbers of females that laid eggs in either the 2× Epsom salt or control treatment, or both, within the same trial were equally distributed (χ² = 0.73, df = 2, p = 0.6951), and there was no significant preference for the 2× Epsom salt or control treatment (χ² = 0.50, df = 1, p = 0.4795).

Female CRBs laid on average ( ± SE) 38 ± 6 eggs when presented with the choice of 5× Epsom salt-amended mulch, unamended mulch, or sand. Significantly more eggs were laid in the control treatment than in the 5× Epsom salt treatment (Table 3). No beetles laid eggs in the sand. Three of the 12 beetles oviposited in both the 5× Epsom and control mulch, one oviposited in the 5× Epsom salt treatment, seven oviposited in the control mulch, and one did not oviposit. The numbers of females that laid eggs in either the 5× Epsom salt or control treatment, or both, or neither, within the same trial were not equally distributed (χ² = 8.00, df = 3, p = 0.0460), and there was a significant preference for the control treatment over the 5× Epsom salt treatment and sand (χ² = 4.50, df = 1, p = 0.0339).