Relative Humidity Effects on the Life History of Hermetia illucens (Diptera: Stratiomyidae) Author(s): L. A. Holmes, S. L. Vanlaerhoven, and J. K. Tomberlin Source: Environmental Entomology, 41(4):971-978. 2012. Published By: Entomological Society of America DOI: http://dx.doi.org/10.1603/EN12054 URL: http://www.bioone.org/doi/full/10.1603/EN12054

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. PHYSIOLOGICAL ECOLOGY Relative Humidity Effects on the Life History of Hermetia illucens (Diptera: Stratiomyidae)

1,2 1 3 L. A. HOLMES, S. L. VANLAERHOVEN, AND J. K. TOMBERLIN

Environ. Entomol. 41(4): 971Ð978 (2012); DOI: http://dx.doi.org/10.1603/EN12054 ABSTRACT Black soldier ßies, Hermetia illucens (L.) (Diptera: Stratiomyidae) are of particular interest for their application in waste management such as reducing manure accumulations in conÞned feeding operations. Determining black soldier ßy development time as a result of climatic variations will allow for optimizing their utilization as a waste management agent at landÞll sites and conÞned feeding operations. To implement a black soldier ßy waste management program in Canada, where seasonal variability does not support H. illucens development on a year round basis, determining maximum and minimum abiotic thresholds to sustain larval development is important. In Canadian winters, maintaining greenhouse temperatures necessary for black soldier ßy development results in low relative humidity that could impact their development. The objective of this study was to determine relative humidity thresholds on egg eclosion and adult emergence. Egg eclosion success was measured at 25, 40, 50, 60, and 70% relative humidities and adult emergence success was measured at 25, 40, and 70% RH. Egg eclosion and adult emergence success increased with increasing relative humidities, while development time decreased with rising relative humidities.

KEY WORDS egg desiccation hypothesis, egg clustering, waste management, egg eclosion, adult emergence

The black soldier ßy, Hermetia illucens (L.) (Diptera: Sheppard, 2002) and the effects of temperature and Stratiomyidae) is a temperate and tropical species that diet on development (Sheppard et al. 1994; Tomberlin develops in decomposing organic material with three et al. 2002, 2009) were conducted with a goal of main- generations per year in the southeastern United States taining a self-sustaining colony for year-round waste (Tomberlin et al. 2002). Females mate once with one conversion. oviposition event in their lifetime and mated females A neglected aspect of black soldier ßy life history is selectively oviposit 320Ð620 eggs in dry crevices near the effect of relative humidity (RH) on black soldier a moist food resource ϳ2 d after successful copulation ßy development and survivorship. Upon egg eclosion, (Tomberlin et al. 2002). Hatched neonate larvae neonate larvae feed on moist decomposing organic quickly make their way to the resource (Booth and matter until development to the postfeeding stage at Sheppard 1984). Previous life history studies on the which time they leave their food resource in search of black soldier ßy were primarily limited to their use in a place to pupate and complete metamorphosis, this waste management. Black soldier ßies efÞciently con- natural life-history of the black soldier ßy occurs in vert organic waste and therefore have been used to conÞned animal feeding operations in the southeast- reduce waste in large conÞned animal feeding oper- ern United States (Sheppard et al. 1994). As a result, ations (Sheppard et al. 2002) as well as to generate successful manure management systems with the supplemental livestock feed (Booram et al. 1977). black soldier ßy were in place in poultry feeding Therefore, developmental studies including mating houses in 1994 without the use of separate facilities or behaviors (Booth and Sheppard 1984, Tomberlin and special equipment to control the ambient environ- ment (Sheppard et al. 1994). However, unlike the southeastern United States, indoor heating systems are On behalf of the University of Windsor in Windsor, Ontario, Can- ada and Texas A&M University in College Station, TX, all experi- necessary for greenhouses in regions experiencing mental methodologies in this research were performed in compliance cold weather resulting in low RH conditions. with all legalities in both Canada and the United States of America. Because low RH increases mortality in many in- The authors on this research have no conßict of interest with the sects, particularly during the egg stage (Sabelis 1985, funding agencies; however, L.A.H. was employed as a teaching as- sistant by the University of Windsor throughout the duration of this as cited in Schausberger 1998), this could impede study. implementation of a year-round black soldier ßy waste 1 Department of Biology, University of Windsor, 401 Sunset Ave- management program in Canada and other locations nue, Windsor, Ontario, Canada, N9B 3P4. with a cold season. The objective of this study was to 2 Corresponding author, e-mail: [email protected]. 3 Department of Entomology, Texas A&M University, 2475 TAMU, determine the effects of RH on black soldier ßy de- College Station, TX, 77843-2475. velopment, including egg eclosion and successful

0046-225X/12/0971Ð0978$04.00/0 ᭧ 2012 Entomological Society of America 972 ENVIRONMENTAL ENTOMOLOGY Vol. 41, no. 4 adult emergence, with a goal of deÞning optimal con- A Hobo U12Ð012 data logger (Onset Computer Co- ditions for black soldier ßy production in cold-climate operation, Pocasset, MA) was placed in each growth agricultural facilities. chamber on the top shelf next to the tray of egg clusters. The data logger recorded RH, temperature and light intensity every 15 min. Eggs were monitored Materials and Methods every 12 h for successful egg eclosion. Two days after Experiment: Egg Eclosion. Source of Eggs. Eggs egg eclosion, identiÞed by observing neonatal larvae were collected from a H. illucens laboratory colony in each cup, egg clusters were stored at Ϫ25ЊC until housed in a cage (1.8 ϫ 1.8 ϫ 1.8 m and 1.5 mm mesh the unhatched eggs were counted. Time to egg eclo- screen) maintained in a greenhouse, outside the Fo- sion and per cent successful egg eclosion was recorded rensic Laboratory for Investigative Entomological Sci- for each replicate within each RH treatment. ences (F.L.I.E.S.) Facility located at Texas A&M Uni- Experiment: Adult Emergence. Source of Larvae. versity in College Station, TX. The laboratory colony Eggs were collected using the same methods as the egg originated from eggs used in Dr. Craig SheppardÕs eclosion experiment described previously. Each cup colony in Tifton, GA. Eggs for each replicate across contained two egg clusters (egg contribution from two RH treatments were collected withina4hwindow. individual females to ensure the necessary survivor- Eggs were collected in a three layer, 3 ϫ 5 cm cor- ship required to proceed with adult emergence ex- rugated cardboard, held together with Elmer white periment) and cups were divided among three RH glue with 3 ϫ 4 mm ßutes used as an oviposition treatments (25, 40, and 70%). Because of limited substrate, taped 5 cm above the oviposition substrate growth chamber availability, all replicates for all treat- with the ßutes perpendicular to the substrate. Ovipo- ments were placed in a single walk-in growth chamber sition attractant was composed of moist-to-liqueÞed maintained at 70% RH, 25ЊC and a photoperiod of 14:10 Gainesville diet (5:3:2 hand mixture of wheat bran, (L:D) h until the postfeeding stage of development. alfalfa, and corn meal, respectively), (Producers Co- Upon egg eclosion, larvae were fed 10 g of dry Gaines- operative Association, Bryan, TX), developed for rear- ville diet (Tomberlin et al. 2002, Hogsette 1992) mixed ing house ßies Musca domestica (L.) (Diptera: Mus- with 18 ml of water aliquots (Ϸ70% moisture), ad cidae) (Tomberlin et al. 2002, Hogsette 1992). libitum until larvae reached the postfeeding stage of Treatments. Each egg-containing corrugated card- development, indicated by their change in color from board ßute was dissected allowing the removal of the cream to black. egg clusters present. While maintaining egg cluster Treatments. To determine the effect RH has on size and form, all egg clusters were combined and then adult emergence, all larvae from all three RH treat- distributed into 30 ml clear plastic cups (Dart P100, ments (25, 40, and 70%) were reared in the same Dart Container Corporation, Mason, MI). Each cup environmental conditions until growth to the post- contained one egg cluster (egg contribution from one feeding stage of development. Neonatal larvae live individual female) composed of 1520 Ϯ 43 eggs. Cups and feed on moist decomposing organic matter. While with eggs were placed into a 30-well clear plastic tray developing in this moist environment, feeding larvae (Bio-Serv, Frenchtown, NJ) and assigned a RH treat- are not readily exposed to the ambient RH. However, ment. RH treatments in this study were chosen based upon completion of the feeding stage, postfeeding on Clark and FaethÕs design (Clark and Faeth 1998) larvae leave the moist environment in search of a safe and preliminary experiments on adjusting the RH lev- place to pupate, however, by leaving the food re- els in the growth chambers (Model I-36LLVLC8, Per- source; the postfeeding larvae are readily exposed to cival ScientiÞc Inc., Perry, IA). It was evident that the ambient environment, where RH may have a signiÞ- growth chambers could not maintain humidity levels cant impact on pupation and adult emergence. As a below 25%; therefore, 25% RH was selected to test for result the effect of RH on adult emergence was con- the lower RH threshold and 70% RH was selected as ducted on postfeeding larvae. Thirty postfeeding lar- the control as a result of successful development in vae from each replicate per RH treatment were re- Tomberlin et al. (2002). Later, a continuum of 40, 50, moved from their food in the walk-in growth chamber and 60% RH effects on egg eclosion was also examined. and placed individually into 30 ml clear plastic cups, This experiment was completed in three stages. without lids, for maximum RH exposure. Cups were First, 10 replicates of 25 and 70% RH effects on egg placed in a 30-well tray and placed into their respec- eclosion were simultaneously compared in two sepa- tive growth chambers (Percival ScientiÞc Inc.) cor- rate growth chambers. Second, 10 additional repli- responding to their respective relative humidities (25, cates of 25 and 70% RH treatment were compared by 40, or 70%) and 25ЊC and a photoperiod of 14:10 (L:D) which the growth chambers were switched for each h. The temperature and photoperiod used for this RH so that each treatment experienced each chamber. portion of the study was as a result of shared growth Finally, 20 replicates of 40, 50, and 60% RH effects on chamber space. The walk-in growth chamber used to egg eclosion were compared in three separate growth rear the black soldier ßy eggs to the postfeeding stage chambers. Regardless of RH treatment, each growth of development housed other research projects and chamber maintained a temperature of 27ЊC with a therefore 25ЊC and a photoperiod of 14:10 (L:D) h was photoperiod of 12:12 (L:D) h. Photoperiod was cho- used instead of the conditions used to maintain the sen to mimic the rearing conditions used to maintain parent stock colonies. To maintain environmental the parent stock colony. consistency, the temperature and photoperiod used to August 2012 HOLMES ET AL.: RELATIVE HUMIDITY EFFECTS ON H. illucens 973

-rear the black soldier ßy from eggs to the postfeeding Table 1. Mean (؎ SE) time to egg eclosion and per cent suc stage of development were also used in the treatment cessful egg eclosion in each treatment, 25, 40, 50, 60, and 70% RH growth chambers. Postfeeding larvae were observed every 12 h to Time to egg Successful RH (%) determine time of pupation for each larva. Individual eclosion (hours) egg eclosion (%) postfeeding larvae were kept in their original cup Experiment 1a 25 GC Ib 124.43 Ϯ 1.85A 8.44 Ϯ 0.75a throughout the duration of the experiment. Upon pu- GC II 138.00 Ϯ 0.00B 5.44 Ϯ 1.40a pation, pupae remained in their cups without lids for 70 GC I 87.63 Ϯ 2.70C 64.67 Ϯ 7.69b GC II 80.78 Ϯ 3.75C 86.22 Ϯ 2.91b 5 d after pupation, to maximize RH exposure before Experiment 2 40 90.11 Ϯ 0.35A 19.86 Ϯ 1.27a adult emergence. Six-day-old pupae were capped with 50 87.84 Ϯ 1.30B 38.00 Ϯ 2.29b clear plastic lids (100PCL25 Dart Container Corpo- 60 71.21 Ϯ 0.37C 72.74 Ϯ 2.46c ration), in anticipation of adult emergence and Means for time to egg eclosion and percent of successful egg pierced once with a 29 gauge syringe to allow con- eclosion within the same exp followed by different letters are signif- tinuous RH exposure. Pupae were monitored every icantly different (P Ͻ 0.05, Wilcoxon/KruskalÐWallis, SAS JMP 8.0.1). 12 h for successful adult emergence. By keeping each a The 25% and 70% were analyzed separately from 40%, 50%, and 60%. postfeeding larvae in their original cups until adult b GC ϭ growth chamber, for purpose of depicting chamber effect. emergence, the length of the postfeeding and pupal stages of development, successful adult emergence and adult longevity were recorded for each larva. Results Dissections were completed on all pupae that did not Experiment: Egg Eclosion. Lower RH Threshold: 25 emerge in the 25 and 40% RH treatments to observe and 70%. Regardless of switching the growth cham- any indication of desiccation. Hobo units were used to bers, eggs subjected to 25% RH had higher mortality record temperature, RH, and light intensity as previ- (Table 1; ␹2 ϭ 27.8882, df ϭ 1, P Ͻ 0.001) and slower ously described. development (Table 1; ␹2 ϭ 28.7834, df ϭ 1, P Ͻ 0.001) Statistics. Experiment: Egg Eclosion. All statistics than eggs in 70% RH. After repeating the experiment were computed using SAS JMP 8.0.1 statistical soft- and switching growth chambers for each treatment, ware (SAS Institute Inc., Cary, NC). The data failed to there was a difference with respect to time to egg meet the assumptions of normality (homogeneity of eclosion for the 25% RH treatment (␹2 ϭ 12.5057; df ϭ variance and normality goodness-of-Þt) despite data 1; P ϭ 0.0004; DunnÐSidak adjusted ␣ ϭ 0.0253). Time transformation attempts (Box Cox, arcsine and Log10 to egg eclosion for the 25% RH treatments in chamber transformations). Thus, a nonparametric one-way 1 was delayed 10.3% compared with chamber 2. In Wilcoxon/KruskalÐWallis (Ranked Sums) test was contrast, there was no difference with respect to time used for both analyses of time to egg eclosion and the to egg eclosion for the 70% RH treatment when percent successful egg eclosion. Pairwise comparisons switching the growth chambers (␹2 ϭ 0.1330; df ϭ 1; using Wilcoxon/MannÐWhitney U tests were used on P ϭ 0.7154). There was no differences observed when signiÞcant results. Alpha values were adjusted using switching the growth chambers with respect to mean the DunnÐSidak procedure to correct for type I error (ϮSE) percent successful egg eclosion for 25 and 70% ␹2 ϭ ϭ ϭ ␹2 ϭ as a result of multiple comparisons (Quinn and Ke- RH ( 3.9153, df 1, P 0.0478 and 3.9637, ϭ ϭ ough 2002). A one way analysis of variance (ANOVA) df 1, P 0.0465, respectively; DunnÐSidak adjusted ␣ ϭ was used to compare the number of eggs in each egg 0.0253). The number of individual eggs in each egg cluster (contribution by one female) was not different cluster across treatments and replicates. ϭ ϭ Experiment: Adult Emergence. The data failed to across treatments (F1,36 0.1435; P 0.7070). meet the assumptions of normality (homogeneity of Hobo data loggers recorded the temperature, RH and light intensity every 15 min. The mean (ϮSE) variance and normality goodness-of-Þt) despite data temperature, RH and light intensity in the 25% RH transformation attempts (Arcsine and Box Cox trans- treatment for chambers 1 and 2 were 28.45 Ϯ 0.02ЊC, formations). Therefore, a nonparametric one-way 24.42 Ϯ 0.01% RH, 2698.07 Ϯ 41.25 Lux and 28.79 Ϯ Wilcoxon/KruskalÐWallis test with Wilcoxon/MannÐ 0.02ЊC, 24.12 Ϯ 0.01% RH, 3001.29 Ϯ 68.32 Lux, re- Whitney U pairwise comparisons tests were used to spectively. The mean (ϮSE) temperature in the 70% determine postfeeding and pupal mortalities and the RH treatment for chambers 1 and 2 were 27.63 Ϯ percent successful adult emergences with respect to 0.02ЊC, 70.19 Ϯ 0.01% RH, 3127.42 Ϯ 43.92 Lux and treatment. Alpha values were adjusted using the 27.59 Ϯ 0.04ЊC, 69.17 Ϯ 0.01% RH, 3120.02 Ϯ 123.23 DunnÐSidak procedure (Quinn and Keough, 2002, p. Lux, respectively. 48). A one way ANOVA was used to compare the RH Continuum: 40, 50, and 60%. Eggs subjected to number of eggs in each egg cluster across RH and 40% RH had higher mortality and slower development replicates. than eggs in 50% and 60% RH (Table 1, ␹2 ϭ 50.2911, A full factorial Cox Proportional Hazard test (model df ϭ 2, P Ͻ 0.001 and ␹2 ϭ 43.4409, df ϭ 2, P Ͻ 0.001, effects: sex, (RH) and sex*RH) was used to compare respectively). Additionally, eggs in the 50% RH treat- survivorship differences with respect to the length of ment also had higher mortality and slower develop- the postfeeding and pupal stages of development and ment than eggs in 60% RH (Table 1, ␹2 ϭ 28.7371, df ϭ adult longevity between relative humidities. 1, P Ͻ 0.001 and ␹2 ϭ 24.1606, df ϭ 1, P Ͻ 0.001, 974 ENVIRONMENTAL ENTOMOLOGY Vol. 41, no. 4

Table 2. RH effects on length of development, mortality, adult emergence, and adult longevity

Length of development stage (d) Mortality (%) Successful adult Adult RH (%) Postfeeding Pupal Postfeeding Pupal emergence (%) longevity (d) 25 10.36 Ϯ 0.09A 8.92 Ϯ 0.15A 62.00 Ϯ 0.06a 65.00 Ϯ 5.9a 16.00 Ϯ 3.6A 5.17 Ϯ 0.14a 40 9.72 Ϯ 0.08B 8.57 Ϯ 0.06B 26.00 Ϯ 0.05b 23.00 Ϯ 4.2b 59.00 Ϯ 5.9B 6.68 Ϯ 0.10b 70 9.48 Ϯ 0.09C 8.41 Ϯ 0.06C 3.00 Ϯ 0.01c 2.00 Ϯ 0.5c 93.00 Ϯ 1.2C 7.94 Ϯ 0.09c

Means for development times, mortalities, successful adult emergence, and adult longevity followed by different letters within the same column are signiÞcantly different (P Ͻ 0.05, Wilcoxon/KruskalÐWallis, Cox Proportional Hazard, SAS JMP 8.0.1). respectively). The number of individual eggs in each The sex ratio was 76.74, 67.60, and 71.98% male in the egg cluster (contribution by one female) was not 25, 40, and 70% RH treatments, respectively. The pro- ϭ ϭ different across treatments (F2,57 2.5797; P portion of males was higher than females for each RH 0.0846). (␹2 ϭ 165.9885, df ϭ 1, P Ͻ 0.0001); however, RH was Hobo data loggers recorded the temperature, RH not a factor on the sex ratio (␹2 ϭ 3.441, df ϭ 2, P ϭ and light intensity every 15 min. The mean (ϮSE) 0.1790). The mean (ϮSE) longevity for males and temperature, RH and light intensity for the 40, 50, and females was 5.08 Ϯ 0.18 d and 5.3 Ϯ 0.21, 6.74 Ϯ 0.13, 60% RH treatments were, 27.36 Ϯ 0.01ЊC, 35.79 Ϯ and 6.51 Ϯ 0.16 d and 8.00 Ϯ 0.11 and 7.77 Ϯ 0.14 d, 0.03% RH, 2302 Ϯ 56.34 Lux, 27.82 Ϯ 0.02ЊC, 48.89 Ϯ respectively in the 25, 40, and 70% RH treatments, 0.07% RH, 2726 Ϯ 66.80 Lux and 28.81 Ϯ 0.02ЊC, respectively. Sex did not have an effect on the timing 58.59 Ϯ 0.07% RH, 3736 Ϯ 91.55 Lux, respectively. of adult emergence, (␹2 ϭ 2.7546, df ϭ 1, P ϭ 0.0970), Experiment: Adult Emergence. Postfeeding larval nor adult longevity (␹2 ϭ 0.5388, df ϭ 1, P Ͻ 0.4629). and pupal mortalities differed between RH (Table 2, There was no effect of sex and the interaction of RH ␹2 ϭ 37.0437, df ϭ 2, P Ͻ 0.001 and ␹2 ϭ 42.2135, df ϭ and sex on adult longevity (␹2 ϭ 1.5659, df ϭ 1, P ϭ 2 2, P Ͻ 0.001, respectively). Postfeeding larvae and 0.2108 and ␹ ϭ 0.3779, df ϭ 2, P Ͻ 0.8278, respec- pupae in 25% RH had higher mortality than those in tively). RH independently affected adult longevity 2 the 40% (␹2 ϭ 14.7220, df ϭ 1, P Ͻ 0.001 and ␹2 ϭ (Fig. 1c, ␹ ϭ 103.6544, df ϭ 2, P Ͻ 0.001), such that 16.8152, df ϭ 1, P Ͻ 0.001, respectively) and 70% RH adults in 25% RH survived fewer days than those in 40 (␹2 ϭ 26.7192, df ϭ 1, P Ͻ 0.001 and ␹2 ϭ 27.1328, df ϭ and 70% RH, and the adults in 40% RH survived fewer 1, P Ͻ 0.001, respectively) and those in 40% RH had days than those in 70% RH. higher mortality than those in 70% RH (␹2 ϭ 17.4448, Hobo data loggers recorded the temperature, RH df ϭ 1, P Ͻ 0.001 and ␹2 ϭ 24.8949, df ϭ 1, P Ͻ 0.001, and light intensity every 15 min upon being placed in respectively). Successful adult emergence differed their treatment growth chambers at the postfeeding Ϯ between the 25, 40, and 70% RH treatments (Table 2, stage of development. The mean ( SE) tempera- ␹2 ϭ 41.2807, df ϭ 2, P Ͻ 0.001). Pupae in 25% had ture, RH and light intensity for the 25, 40, and 70% Ϯ Њ Ϯ lower successful adult emergences than those in the 40 RH treatments were, 26.23 0.01 C, 30.04 0.11% Ϯ Ϯ Њ Ϯ and 70% RH treatments (␹2 ϭ 18.4713, df ϭ 1, P Ͻ 0.001 RH, 3034.67 25.71 Lux, 26.23 0.02 C, 41.32 Ϯ Ϯ Њ and ␹2 ϭ 26.4919, df ϭ 1, P Ͻ 0.001, respectively) and 0.06% RH, 3642.31 29.19 Lux and 26.96 0.02 C, Ϯ Ϯ pupae in the 40% RH had lower successful adult emer- 69.53 0.07% RH, 3665.63 29.94 Lux, respectively. gences than those in the 70% RH (␹2 ϭ 21.3952, df ϭ 1, P Ͻ 0.001). There was no effect of sex and the interaction of RH Discussion and sex on time spent in the postfeeding stage of Our predictions are conclusive; RH has signiÞcant 2 development (␹ ϭ 1.5194, df ϭ 1, P ϭ 0.2177 and effects on successful black soldier ßy egg eclosion and 2 ␹ ϭ 1.1851, df ϭ 2, P ϭ 0.5529, respectively). RH adult emergence. In low RH environments, water loss independently affected the time spent in the post- through the egg and pupal membranes can be detri- feeding stage of development (Fig. 1a, ␹2 ϭ 16.4343, mental to the survivorship of holometabolous insects, df ϭ 2, P Ͻ 0.001), such that a greater proportion of resulting in desiccation (Wigglesworth 1984). We be- larvae in the 25% RH treatment spent more time in the lieve the deleterious effects 25% RH had on egg eclo- postfeeding stage of development than larvae in other sion success was attributed to desiccation. Stamp relative humidities. Similarly, there was no effect of (1980) was the Þrst to coin the egg desiccation hy- sex and the interaction of RH and sex on the time spent pothesis, an oviposition strategy used in Lepidoptera in the pupal stage of development (␹2 ϭ 3.0472, df ϭ to prevent egg desiccation. Clark and Faeth (1997, 1, P ϭ 0.0809 and ␹2 ϭ 1.6552, df ϭ 2, P ϭ 0.4371, 1998) followed up on StampÕs hypothesis with refer- respectively). RH independently affected the time ence to egg cluster size and composition, where they spent in the pupal stage of development (Fig. 1b, ␹2 ϭ observed eggs on the surface of egg clusters desic- 10.5336, df ϭ 2, P Ͻ 0.005), such that a greater pro- cated, while eggs within the egg cluster did not. We portion of larvae in the 25% RH treatment spent more observed 6.84% successful egg eclosion at 25% RH and time in the pupal stage of development than larvae in similar to Clark and Faeth (1998), the eggs located on other relative humidities. the surface of egg clusters appeared desiccated, while August 2012 HOLMES ET AL.: RELATIVE HUMIDITY EFFECTS ON H. illucens 975

Fig. 1. Proportion of survivorship at each stage of development over time. (a) The effect of RH on time spent in the postfeeding stage of development was different across treatments. (b) The effect of RH on time spent in the pupal stage of development was different across treatments. (c) The effect of RH on adult longevity was different across treatments. (P Ͻ 0.05, Wilcoxon/KruskalÐWallis, Cox Proportional Hazard, SAS JMP 8.0.1). eggs located within the egg cluster had successfully tion under different RH conditions would provide eclosed (unpublished data). Black soldier ßy lay their further insight. eggs in clusters composed of multiple layers, taking on Furthermore, time to egg eclosion was inversely a three dimensional form of the crevice in which they proportional to successful egg eclosion at low relative were oviposited in; however, when no oviposition humidities compared with high relative humidities. substrate is provided, females will lay spherical clus- Eggs in 25% RH took between 1.5 and 2.5d longer to ters composed of multiple layers which are loosely eclose than eggs in 70% RH. Why low RH conditions packed compared with using conÞned crevices (un- delays egg eclosion is not known (Smith 1993; Guar- published data). Therefore, egg cluster shape, size, neri et al. 2002, as cited in Han et al. 2008); however, and layer composition is manipulated by either the Willmer (1982) suggested longer development times oviposition substrate provided or the oviposition site are a result of an increase in energy costs for conserv- selected by the female. In agreement with Clark and ing water because of transpiration. When insects take Faeth (1997, 1998), egg clustering may be an ovipo- in oxygen through their spiracles, they concomitantly sition strategy for the black soldier ßy; however, con- lose water (Wigglesworth 1984). Unlike juveniles and ducting choice experiments on oviposition site selec- adults who have the physiology to regulate water loss 976 ENVIRONMENTAL ENTOMOLOGY Vol. 41, no. 4 trade-offs with respiration, eggs are often immobilized time in their tolerance of RH variation. For the pur- and without mechanisms of controlling water loss dur- pose of our objective, although the lower RH thresh- ing respiration, a requirement of embryo development old could not be determined for egg eclosion or adult (Zrubek and Woods 2006). The rate of oxygen con- emergence, mortality because of desiccation provides ductance through the egg cuticle is very complex sufÞcient evidence for using higher relative humidi- (Woods et al. 2005) and the relationship between ties (Ն50% RH) in colony maintenance. Although, oxygen conductance and ambient relative humidity higher successful egg eclosion was evident in the 50, and its effect on egg development is unclear. 60, and 70% RH treatments, demonstrating that egg Interestingly, eggs in the 60% RH treatment eclosed eclosion and adult emergence success increases with before eggs in the 70% RH treatment with a compa- RH, there was successful egg eclosion and adult emer- rable eclosion success; however, because the mean gence in the lower RH treatments, leading us to be- temperature in the 60% RH growth chamber was re- lieve that developing a low RH strain of black soldier corded a degree higher than in the mean temperature ßy may be possible. in the 70% RH growth chamber, we cannot attribute Our results conÞrm desiccation is highly evident on this difference solely on the effects of RH. Insect postfeeding larvae and pupae under low relative hu- development is directly proportional to temperature midities with only 16% adult emergence at 25% RH. with lower temperatures slowing the rate of develop- Desiccation was relatively equally evident in both ment and higher temperatures increasing the rate of postfeeding larvae and pupae at 25% RH. Desiccated development in many insects (Gullan and Cranston postfeeding larvae were identiÞed by their brittle, 2010). Tomberlin et al. (2009) demonstrated that truncated, and ßattened appearance. Desiccated pu- black soldier ßy development at 27ЊC was delayed by pae were identiÞed at the end of the experiment; 1 mo 11% when compared with development at 30ЊC. It after the pupae in the 70% RH emerged, when emer- would be necessary to run a replicated study with the gence did not occur. Desiccated pupae in the 25 and same experimental design, introducing growth cham- 40% RH treatments were dissected to assess evidence ber as an added treatment to decipher if a growth of dehydration. High proportions of hollowed cavities chamber effect with respect to temperature is evident. within the puparium were evident and fat bodies were Because eggs successfully eclosed at 25% RH, we completely dehydrated. However, the 7% of pupae in cannot yet decipher the lower RH threshold for black the 70% RH treatment that did not emerge were not soldier ßy egg eclosion. To determine the lower RH dissected for evidence of desiccation, because this threshold, alternative measures in achieving lower RH mortality rate was similar to the mortality rate in conditions in the growth chambers, possibly using Tomberlin et al. (2009) under similar conditions. saturated salt solutions (Pappas, Broufas, and Koveos An insectÕs metabolic response to desiccation in- 2008, Clark and Faeth 1998) would be necessary be- volves an up regulation of the desiccation stress pro- cause the growth chambers used cannot maintain rel- tein 28 (dsp28) and juvenile hormone (JH) (Kroeker ative humidities below 25% under normal automated and Walker 1991a). Kroeker and Walker (1991b) stud- programming (unpublished data). Furthermore, eggs ied the correlation of up regulated dsp28 and JH in the developing at 25% RH in chamber 1 eclosed 12 h common mealworm Tenebrio molitor (L.) (Co- before those developing in chamber 2. The mean tem- leoptera: Tenebrionidae). They found that JH and perature and RH only differed by 0.3ЊC and 0.3% RH, dsp28 are up regulated during ecdysis, down regulated respectively, between the two growth chambers; how- just before pupation and not up regulated again until ever, the mean light intensity in chamber 2 was 10.6% adult reproduction (Kroeker and Walker 1991b). higher than in chamber 1. In Drosophila melanogaster Should this mechanism be similar in the black soldier (Meign) (Diptera: Drosophilidae), increased light in- ßy, a desiccation response to low RH may involve an tensity slows development (Bruins et al. 1991), there- up regulation of dsp28 and JH during the postfeeding fore, the delay in development in chamber 2 may be stage of development, down regulation in the pupal attributed to the increased light intensity. stage and up regulation again in the adult stage during Because black soldier ßies are found more com- mating and reproduction. This would explain why monly in tropical and subtropical humid environments postfeeding larvae were able to successfully pupate; (Booth and Sheppard 1984), they probably face little however, inevitably succumb to desiccation in the selection for tolerating low RH. However, because pupal stage of development because of a down regu- some eclosion success was evident at lower relative lation of dsp28. On the contrary, hollow body cavities humidities, it may be possible that low RH toleration observed in the dissected pupae may not be a result of may have evolved in our black soldier ßy colony as a desiccation, but instead, exhausted fat bodies because result of mass rearing in a controlled environment for of the postfeeding larvaeÕs delayed time to pupation. eight years in Tifton, GA. Evolution of low RH toler- Regardless, with 16 and 59% successful adult emer- ance is evident in other , Ferrero et al. gences in the 25 and 40% RH, respectively, RH did (2010) reared a low RH tolerant laboratory strain of signiÞcantly impact H. illucens development. over several generations and, therefore, it may One drawback to this study is the lack of observa- be possible to develop a black soldier ßy strain that can tions on adult fecundity. Although a proportion of thrive in low RH conditions. The black soldier ßyÕs adults successfully emerged in each RH treatment, RH natural tropical and subtropical habitats have likely effects on successful mating and oviposition are un- been crucial selective pressures through evolutionary known. RH effects on fecundity have been studied by August 2012 HOLMES ET AL.: RELATIVE HUMIDITY EFFECTS ON H. illucens 977 several authors in several systems (Cun- Clark, B. R., and S. H. Faeth. 1997. The consequences of nington 1985, Wilson 1982, Kitayama 1979, Coombs larval aggregation in the butterßy Chlosyne lacinia. J. Ecol. 1978, Richards 1947). For example, low RH environ- Entomol. 22: 408Ð415. ments delays oviposition in Acarus siro (L.) (Acarina: Clark, B. R., and S. H. Faeth. 1998. The evolution of egg ) (Cunnington 1985). Similarly, adult weight clustering in butterßies: a test of the egg desiccation and oviposition increased with higher RH in Crvpto- hypothesis. J. Evol. Ecol. 12: 543Ð552. lestes pusillus (Schonherr) (Coleoptera: Cucujidae) Coombs, C. W. 1978. The effect of temperature and relative ¨ humidity upon the development and fecundity of Derm- (Currie 1967). On the contrary, various species of estes lardarius L. (Coleoptera, Dermestidae). J. Stored spider mites have higher oviposition and longer lon- Prod. Res. 14: 111Ð119. gevity at lower RH (Kramer and Hain 1989, Boudreaux Cunnington, A. M. 1985. Factors affecting oviposition and 1957). In our study, the only indicator of adult Þtness fecundity in the grain Acarus siro L. (Acarina: Ac- we have is longevity, with adults in 70% RH living 2 to aridae), especially temperature and relative humidity. 3 d longer than adults subjected to lower RH envi- Exp. Appl. Acarol. 1: 327Ð344. ronments. Despite longevity trends, we have no evi- Currie, J. E. 1967. Some effects of temperature and humid- dence in this study that ovarian development and ity on the rates of development, mortality and oviposition successful mating followed by viable egg deposition, is of Crvptolestes pusillus (Schbherr) (Coleoptera, Cucuji- correlated to adult longevity. Additionally, the green- dae). J. Stored Prod. Res. 9: 97Ð108. house conditions with surrounding outbuildings can- Ferrero, M., C. Gigot, M.-S. Tixier, Y.M.V. Houten, and S. not be excluded as potentially impacting the results Kreiter. 2010. Egg hatching response to a range of air humidities for six species of predatory mites. Entomol. generated from this study. Exp. Appl. 135: 237Ð244. Reduced indoor ambient RH throughout Canadian Gullan, P. J., and P. Cranston. 2010. The insects: an outline winters is at a direct consequence to heating systems of entomology, 3rd ed. Blackwell Publishing, Malden, in greenhouses and homes. Before commencing this MA. study, unsuccessful rearing of the black soldier ßy was Han, R. D., M. Parajulee, Z. He, and F. Ge. 2008. Effects of attempted in a greenhouse without the addition of environmental humidity on the survival and development humidiÞers in Windsor, Ontario, during the winter of pine caterpillars, Dendrolimus tabulaeformis (Lepidop- season. The results of this study might explain prior tera: Lasiocampidae). Insect Sci. 15: 147Ð152. unsuccessful attempts of rearing the black soldier ßy Hogsette, J. A. 1992. New diets for production for produc- and suggests that in Canada and similar locations ex- tion of house-ßies and stable ßies (Diptera, Muscidae) in periencing cold weather, year-round waste manage- the laboratory. J. Econ. Entomol. 85: 2291Ð2294. ment using the black soldier ßy will require that steps Kitayama, K., R. E. Stinner, and R. L. Rabb. 1979. Effects of temperature, humidity, and soybean maturity on longev- be taken to increase RH in heated indoor facilities. ity and fecundity of the adult Mexican bean beetle, Epi- lachna varivestis. Environ. Entomol. 8: 458Ð464. Acknowledgments Kramer, D. A., and F. P. Hain. 1989. Effect of constant- humidity and variable: humidity and temperature re- We extend our warmest thanks to Tawni Crippen, Longyu gimes on the survival and development periods of Olig- Cheng, Andrew Fields, Michelle Sanford, Micah Flores, Jen- onychidae ununguis (Acarina, Tetranychidae) and nifer Pechal, Adrianne Brundage, Rachel Mohr, Eric Den- Neoseiulus fallacis (Acarina, ). J. Environ. emark, Trinh Nguyen, Jennifer Rosati, Luis Alvarez, Ashley Entomol. 18: 741Ð746. SummerÞeld, Craig Sheppard, Daniel Edelstein, Oliver Love, Kroeker, E. M., and V. K. Walker. 1991a. Developmental and Edwin Tam. We would also like to thank the Department expression and hormonal regulation of a desiccation of Entomology at Texas A&M University for hosting this stress protein in Tenebrio molitor. Insect Biochem. 21: research project. Lastly, funding for this research was made 631Ð640. possible by Natural Science and Engineering Research Kroeker, E. M., and V. K. Walker. 1991b. Dsp28: a desicca- Council, Essex-Windsor Solid Waste Authority and Gesing tion stress protein in Tenebrio molitor hemolymph. Arch. Consultants Ltd. Insect Biochem. 17: 169Ð182. Pappas, M. L., G. D. Broufas, and D. S. Koveos. 2008. Effect of relative humidity on development, survival and repro- References Cited duction of the predatory lacewing Dichochrysa prasina Booram, Jr. C. V., G. M. Newton, O. M. Hale, and R. W. (Neuroptera : Chrysopidae). Biol. Control 46: 234Ð241. Barker. 1977. Manure residue as a substrate for protein Quinn, G. P., and M. J. Keough. 2002. Experimental design production via Hermetia illucens larvae. Proceeding in the and data analysis for biologists. Cambridge University Cornell Agricultural Waste Management Conference, Press, New York, NY. Cornell University, Ithaca, N.Y. 599Ð604. Richards, O. W. 1947. Observations on grain-weevils, Ca- Booth, D. C., and C. Sheppard. 1984. Oviposition of the landra (Col.. Curcullonidae). I. General biology and ovi- black soldier ßy, Hermetia illucens (Diptera, Stratiomyi- position. Proc. Zool. Soc. Lond. 17: 1Ð43. dae): eggs, masses, timing and site characteristics. J. En- Schausberger, P. 1998. The inßuence of relative humidity viron. Entomol. 13: 421Ð423. on egg hatch in Euseius finlandicus, Typhlodromus pyri Boudreaux, B. H. 1957. The effect of relative humidity on and (, Phytoseiidae). egg-laying, hatching and survival in various spider mites. J. Appl. Entomol. 122: 497Ð500. J. Insect Physiol. 2: 65Ð72. Sheppard, D. C., J. K. Tomberlin, J. A. Joyce, B. C. Kiser, and Bruins, B. G., W. Scharloo, and G.E.W. Tho¨rig. 1991. The S. M. Sumner. 2002. Rearing methods for the black sol- harmful effect of light on Drosophila is diet dependent. dier ßy (Diptera: Stratiomyidae). J. Med. Entomol. 39: Insect Biochem. 21: 535Ð539. 695Ð698. 978 ENVIRONMENTAL ENTOMOLOGY Vol. 41, no. 4

Sheppard, D. C., G. L. Newton, S. A. Thompson, and S. Tomberlin, J. K., P. H. Adler, and H. M. Myers. 2009. De- Savage. 1994. A value-added manure management sys- velopment of the black soldier ßy (Diptera: Stratiomyi- tem using the black soldier ßy. Bioresource Technol. 50: dae) in relation to temperature. J. Environ. Entomol. 38: 275Ð279. 930Ð934. Stamp, N. E. 1980. Egg deposition patterns in butterßies: Wigglesworth, V. B. 1984. Insect physiology. Cambridge why do some species cluster their eggs rather than deposit University Press, New York, NY. them singly. Am. Nat. 115: 367Ð380. Wilson, K. G., R. E. Stinner, and R. L. Rabb. 1982. Effects of Smith, L. 1993. Effect of humidity on life history charac- temperature, relative humidity, and host plant on larval teristics of Anisopteromalus calandrae (Hymenoptera: survival of the Mexican bean beetle, Epilachan varivestis Pteromalidae) parasitizing maize weevil (Coleoptera: Mulsant. Environ. Entomol. 11: 121Ð126. Curculionidae) larvae in shelled corn. J. Environ. Ento- Willmer, P. G. 1982. Microclimate and the environmental mol. 22: 618Ð624. physiology of insects. Adv. Insect Physiol. 16: 1Ð57. Tomberlin, J. K., and D. C. Sheppard. 2002. Factors inßu- Woods, H. A., R. T. Bonnecaze, and B. Zrubek. 2005. Oxy- encing mating and oviposition of black soldier ßies (Dip- gen and water ßux across eggshells of Manduca sexta. J. tera: Stratiomyidae) in a colony. J. Entomol. Sci. 37: 345Ð Exp. Biol. 208: 1297Ð1308. 352. Zrubek, B., and H. A. Woods. 2006. Insect eggs exert rapid Tomberlin, J. K., D. C. Sheppard, and J. A. Joyce. 2002. control over an oxygen-water tradeoff. Proc. R. Soc. Biol. Selected life-history traits of black soldier ßies (Diptera Sci. 273: 831Ð834. : Stratiomyidae) reared on three artiÞcial diets. Ann. Entomol. Soc. Am. 95: 379Ð386. Received 19 February 2012; accepted 25 May 2012.