When Studying the Effects of Light on Herbivory, Should One Consider Temperature? The Case of hortaria F. (: Geometridae) Feeding on L. (Lauraceae) Author(s): Richard A. Niesenbaum and Emily C. Kluger Source: Environmental Entomology, 35(3):600-606. 2006. Published By: Entomological Society of America DOI: http://dx.doi.org/10.1603/0046-225X-35.3.600 URL: http://www.bioone.org/doi/full/10.1603/0046-225X-35.3.600

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 When Studying the Effects of Light on Herbivory, Should One Consider Temperature? The Case of F. (Lepidoptera: Geometridae) Feeding on Lindera benzoin L. (Lauraceae)

1 2 RICHARD A. NIESENBAUM AND EMILY C. KLUGER

Biology Department, Muhlenberg College, Allentown, PA 18104Ð5586

Environ. Entomol. 35(3): 600Ð606 (2006) ABSTRACT Light and temperature are ecological factors that are known to inßuence leaf quality and herbivory. Because temperature is likely to vary with light environment, the effects that each factor has on herbivore performance and seasonal herbivory rates are likely to be confounded. We studied how light and thermal environment inßuenced consumption and conversion of Lindera benzoin L. (Lauraceae) leaf material by Epimecis hortaria F. (Lepidoptera: Geometridae) in Þeld and laboratory settings. Ambient and leaf surface temperatures were higher in the sun than in the shade. Experimental enclosures increased feeding temperatures by Ͻ1ЊC. Larvae of E. hortaria feeding at warmer temperatures consumed more leaf material than their counterparts in cooler conditions in the laboratory. However, in the Þeld, herbivores tended to remove less leaf material from plants in warm, sunlit habitats where seasonal herbivory is lower. Conversion of leaf area eaten into biomass was higher in larvae feeding on L. benzoin leaves from sun habitats for all Þeld and laboratory treatments and was independent of temperature, suggesting that sun leaves are more nutritious for E. hortaria than shade leaves. While elevations in ambient temperature play a signiÞcant role in laboratory-based settings, the effects of ambient temperature on herbivory rates in the natural environment seem to be muted by differences in leaf quality and antiherbivore defenses in sun and shade populations of L. benzoin.

KEY WORDS herbivory, temperature, light, Epimecis hortaria; Lindera benzoin

Light environment is typically considered a major tially on rates of herbivory is that temperature and determinant of rates of herbivory. Numerous light intensity are expected to be tightly correlated studies have shown variation in herbivory between and can have contrasting inßuences on leaf consump- forest edge or gap and understory sites (Aide and tion by herbivores. Temperature can inßuence insect Zimmerman 1990, Niesenbaum 1992, Shure and Wil- herbivores directly. For example, elevations in ambi- son 1993, Sagers and Coley 1995, Fortin and Mauffette ent temperature result in increased rates of larval 2001), and along gradients of primary productivity consumption and growth and development (Sherman (Fraser and Grime 1998, Mazõ´a et al. 2004). This work and Watt 1973, Scriber and Lederhouse 1983, Knapp has almost exclusively been done within the context of and Casey 1986, Stamp and Bowers 1994a, Yang and the inßuence of light availability on leaf quality and Stamp 1996, Levesque et al. 2002). Studies on the defense as they affect rates of consumption by . effects of temperature, therefore, report that insects Most studies have found a strong relationship between feeding in warmer habitats consume more leaf tissue light environment and leaf quality, which in turn de- (Sherman and Watt 1973, Scriber and Lederhouse termined rates of herbivory (Mole and Waterman 1988, Dudt and Shure 1994, Nichols-Orians 1991, 1983, Knapp and Casey 1986, Stamp and Bowers 1994a, Louda and Rodman 1996, Jansen and Stamp 1997) Yang and Stamp 1996, Levesque et al. 2002), whereas such that light environment has been used as a proxy studies of light and herbivory generally report that for leaf quality when studying the inßuence of leaf rates of leaf damage are depressed in warm, sunlit quality on insect performance (Levesque et al. 2002). habitats (Coley 1983, Aide and Zimmerman 1990, A potential problem in determining the effects of Niesenbaum 1992, Shure and Wilson 1993, Dudt and light level on leaf quality and defense and consequen- Shure 1994, Sagers and Coley 1995, Crone and Jones 1999). The direct effects of temperature and light may not only be opposing, but are likely confounded in 1 Corresponding author, e-mail: [email protected]. 2 Current address: Department of Entomology, University of Illi- studies of herbivory rates in sun and shade plant pop- nois, Urbana-Champaign, Urbana, IL 61801. ulations.

0046-225X/06/0600Ð0606$04.00/0 ᭧ 2006 Entomological Society of America June 2006 NIESENBAUM AND KLUGER:LIGHT,TEMPERATURE, AND HERBIVORY 601

There are also indirect effects of temperature on have lower water content, and experience less her- insect feeding. Aizen and Patterson (1995) found that bivory than plants in the shade (Niesenbaum 1992). rates of herbivory decreased along an increasing tem- The primary herbivore in this system is the larvae of perature gradient because of variation in leaf phenol- Epimecis hortaria F. (Lepidoptera: Geometridae) or ogy in relation to the timing of insect feeding along tulip tree beauty. Larval abundances peak in late that gradient. Others have found that temperature can June and in August with two generations per season. inßuence rates of insect herbivory through its inßu- E. hortaria is oligophagous and has been observed in ence on leaf quality (Stamp and Bowers 1994a, Dury other habitats feeding on a limited number of species et al. 1998, Kuokkanen et al. 2001) and the release of including triloba (Annonaceae), organic volatile compounds from leaves (Pen˜ uelas and albidum (Lauraceae), and tulipifera Lluisa´ 2003). The latter may be of particular interest (Magnoliaceae); however, at our study sites, E. hor- given the recent focus on these compounds as poten- taria feeds almost exclusively on L. benzoin. We ob- tial cues for top-down regulation of insect herbivores served little or no herbivory on this host plant by other (Kessler and Baldwin 2001). Again, these indirect ef- insects. E. hortaria can be successfully reared in the fects of temperature are confounded with light and laboratory through multiple generations on leaves have the potential to counter or amplify them in the from L. benzoin. Þeld. Research was conducted at the Graver Arboretum Given the potential strong relationship between of Muhlenberg College in Northampton County, PA, light levels and herbivore thermal environment and which supports large numbers of L. benzoin in both the possibility for these variables to both inßuence sun and shade environments. Typical of the meso- insect feeding in either similar or opposing ways, stud- phytic deciduous forest of this region (Braun 1950), ies would beneÞt from considering them concur- the overstory is dominated by Acer rubrum, Fagus rently. However, very few studies that have consid- grandifolia, Liriodendron tulipifera, Juglans nigra, and ered the inßuence of light on insect herbivory have species of Quercus and Carya. The understory shrub also considered the inßuence of temperature. Con- layer is essentially a monoculture of L. benzoin. Sunny versely, the recent focus of temperature on plant qual- areas are those where the overstory is absent or quite ity and herbivory within the context of global warming thin, including the forest edge and large tree fall gaps, has essentially ignored interactions with light envi- and shade areas are under relatively intact overstory. ronment (Dury et al. 1998, Kuokkanen et al. 2001). Light and Temperature Measurement. Light inten- Here we examine the effects of light environment and sity, ambient temperature, and relative humidity were temperature on herbivory rates and herbivore perfor- monitored in three shade sites and three sun sites with mance. We did this by studying the relationship be- six Watchdog 450 series data loggers (Spectrum Tech- tween light and temperature and their inßuence on nologies, PlainÞeld, IL) in July 2003. The data loggers Epimecis hortaria F. (Lepidoptera: Geometridae) and temperature sensors were mounted next to plants feeding on Lindera benzoin L. (Lauraceae), a plant and shielded from rain, solar radiation, and other that experiences greater herbivory in the shade than sources of radiated and reßected heat that can distort in the sun (Niesenbaum 1992). SpeciÞcally we exam- measurement. Photosynthetically active radiation ined (1) the relationship between light and ambient (PAR) was measured using quantum light sensors. temperature in sun and shade populations of L. ben- Light, temperature, and relative humidity were logged zoin and in experimental enclosures and (2) how rear- every hour for 7 d. We assumed that light environment ing E. hortaria on leaves grown in sun or shade inßu- differences in temperature would only occur during ences rates of feeding, growth, and the efÞciency of daylight hours. We therefore compared the depen- conversion in the Þeld and at different temperatures dent variables of ambient temperature and PAR in sun in the laboratory. and shade using one-way analysis of variance (ANOVA) on the means for each data logger from daylight hours only and again for mid-day hours (1100Ð1300 hours). Leaf surface temperature, inci- Materials and Methods dent PAR, and transpiration rate at mid-day were Study Organisms and Study Site. Lindera benzoin L. measured using an LCi portable leaf chamber analysis (Lauraceae) or spicebush is a common native under- system (Dynamax, Houston, TX). Measurements story shrub in moist forests of eastern North America were made alternately on 42 arbitrarily selected leaves from southern Ontario to Florida. The plant is con- in each of two neighboring sun and shade habitats for sidered shade adapted, but occurs in a range of light a total of 84 measurements. The relationship between conditions from deep shade within the forest under- PAR and ambient temperature was examined using story to bright light on forest edges and in tree-fall simple linear regression on raw data. All data met the gaps. Leaves vary in physical attributes such as leaf assumptions of normality and homoscadasticity. thickness and toughness and have a diversity of sec- Field feeding trials are often conducted in cages or ondary carbon based chemicals such as terpenes. Leaf white mesh/organdy bags constructed of sown bridal damage by lepidopteran larvae is easily distinguished veil and placed over plants or parts of plants (Bonser and measured and has been shown to be greater in the and Reader 1995). Given the potential effects of mi- shade than in the sun (Niesenbaum 1992). Prior work croenvironment on feeding and the possibility that has shown that sun leaves are tougher, less palatable, temperatures may be greater within these enclosures, 602 ENVIRONMENTAL ENTOMOLOGY Vol. 35, no. 3

Table 1. Microclimate in sun and shade habitats

Shade Sun SigniÞcance ␮ 2 ϭ Ͻ PAR ( mol/m /s)Ñdaylight 14.39 (22.3) 214.08 (22.3) F1,4 158.11, P 0.001 o ϭ ϭ Ambient temp. ( C)Ñdaylight 21.16 (0.18) 22.1 (0.18) F1,4 16.53, P 0.06 ϭ ϭ Relative humidity (%)Ñdaylight 82.26 (0.90) 77.77 (0.90) F1,4 37.47, P 0.004 o ϭ ϭ Ambient temp. ( C)Ñmid-day 22.50 (0.30) 24.15 (0.29) F1,4 46.42, P 0.002 ϭ ϭ Relative humidity (%)Ñmid-day 77.54 (1.86) 70.21 (1.82) F1,4 42.76, P 0.002 ␮ 2 ϭ Ͻ Incident PAR ( mol/m /s) 48.01 (77.63) 658.30 (77.63) F1,82 30.93, P 0.001 o ϭ Ͻ Leaf surface temp. ( C) 23.76 (0.38) 28.34 (0.34) F1,82 49.76, P 0.001 2 ϭ Ͻ Leaf transpiration (mmol/m /s) 2.33 (0.19) 3.14 (0.19) F1,82 8.53, P 0.01

Means (SE) are reported. we simultaneously compared temperatures within and Field Feeding Trials. The Þeld feeding trial was outside of these enclosures using data loggers as de- conducted in July 2003. Three branches per plant were scribed above in shade and sun habitats. In each hab- arbitrarily selected for use on six plants in sun and itat, we measured temperature on the inside and out- shade habitats for a total of 18 branches per habitat, side of six enclosures. Temperature was logged every and enclosed in white organdy bags constructed of hour for 7 d, and means were analyzed for daylight sown bridal veil. Before bagging, each branch was hours as described above. checked for existing herbivores, and previous her- Laboratory Feeding Trials. To separate the effects bivory was marked with nontoxic ink. Thirty-six third- of temperature from other factors associated with sun instar larvae that were reared in the laboratory on and shade in the Þeld, we conducted laboratory feed- fresh L. benzoin leaves were randomly assigned to Þeld ing trials with 28 Þeld-collected larvae that were fed sun and shade treatments, starved for 12 h to clear gut leaves from the sun and shade at different tempera- contents, and weighed. They were immediately trans- tures. Third-instar larvae were starved for 12 h to clear ported to the Þeld where one larva was placed on an gut contents, weighed, and randomly assigned to four individual branch. After 48 h, larvae were removed treatments based on peak temperature (22 or 32ЊC) from the branches, starved for 12 h, and weighed. No and diet (sun or shade leaves) ina2by2factorial larvae initiated molting during this time period. design replicated seven times. Each larva was placed Leaves that incurred herbivore damage were re- in 1 of 28 9-cm petri dishes with moistened Þlter paper moved, and leaf area measurements were taken as and fed fresh leaves from sun or shade habitats. Larvae described above. Leaf area consumed (cm2), biomass were reared for6dintwoseparate Conviron CMP gained (mg), and conversion (mg/cm2) were ana- 4030 growth chambers (Conviron, Pembina, ND). Di- lyzed using one-way ANCOVA with habitat (sun/ urnal patterns of light (14/10) and relative humidity shade) as the main effect. Initial larval weight was used (65%) did not vary between chambers. Temperature as the covariate. The assumption of homogeneity of was controlled diurnally with the average in the high slopes to perform an ANCOVA was met in all in- temperature treatment 8ЊC higher than the colder stances. Prior work indicated that branches function temperature treatment and within the normal tem- essentially as independent units, making them appro- perature ranges seen in the Þeld. Fresh leaves from priate units of independent replication (Niesenbaum either plant in sun or shade habitats were replaced as 1996). needed, but no larvae required Ͼ3 leaves during the course of the experiment. After the 6-d period, larvae Results were starved for 12 h and reweighed. No larvae ini- tiated molt during this time period. Light and Temperature Measurements. During The amount of leaf area consumed was estimated daylight hours (0600Ð1900 hours), PAR was signiÞ- using a Delta-T (Dynamax, Houston, TX) leaf area cantly greater in sun plots than in shade plots (Table measurement system (Niesenbaum 1992). EfÞciency 1), and PAR and ambient temperature were linearly ϭ Ͻ 2 ϭ of conversion was calculated as the biomass gained related (F1,362 62.07; P 0.001, R 0.15). Accord- divided by leaf area consumed per individual larva. ingly, temperatures during daylight hours were sig- Leaf area consumed (cm2), proportion of total leaf niÞcantly higher in sun plots than in shade plots (Ta- area consumed, biomass gained (mg), and efÞciency ble 1). During daylight hours, relative humidity was of conversion (mg/cm2) were analyzed as 2 by 2 signiÞcantly greater in the shade than in the sun (Ta- factorial randomized design using a two-way analysis ble 1). The differences in the mean ambient temper- of covariance (ANCOVA) with temperature, habitat ature and relative humidity at sun and shade sites were source (sun/shade) of leaves, and their interaction as even greater at mid-day from1100 to 1300 hours (Table the main effects and initial larval weight as the co- 1). Leaf surface temperatures were linearly related to variate. Proportional data that did not meet the as- PAR measured concurrently at the leaf surface (F1,82 sumptions of normality or homoscadasticity were log- ϭ 7.446, P Ͻ 0.01), and incident PAR and leaf surface transformed. The assumption of homogeneity of temperatures were on average signiÞcantly higher at slopes to perform an ANCOVA was met in all in- sun sights (Table 1). Transpiration rates were signif- stances. icantly greater in the sun as well (Table 1). The tem- June 2006 NIESENBAUM AND KLUGER:LIGHT,TEMPERATURE, AND HERBIVORY 603

Fig. 1. Laboratory feeding trials. Sun and shade differences at 22 and 32ЊC in (A) mean leaf area consumed, (B) mean percent leaf area consumed, (C) mean larval biomass gained, and (D) mean conversion of leaf area into larval biomass. Error bars represent SE. Bars with the same letter are not signiÞcantly different from each other (P Ͼ 0.05). perature within experimental enclosures was signiÞ- Field Feeding Trials. In Þeld feeding trials, the ϭ Ͻ cantly warmer (F1,20 20.64, P 0.001), but only on amount of leaf area consumed was not signiÞcantly average by 0.33 Ϯ 0.05ЊC. different on plants at sun and shade sites (Table 3; Fig. Laboratory Feeding Trials. Both the absolute 2A). The amount of biomass gained by E. hortaria amount and the percentage of leaf area consumed was larvae also was not different between sun and shade signiÞcantly greater at warmer temperatures, but sites (Table 3; Fig. 2B). The conversion of leaf area these values were unaffected by whether the leaves came from the sun or shade microhabitats (Fig. 1A and B; Table 2). Rates of consumption of leaf material were Table 2. ANCOVAs for laboratory feeding trials nearly doubled at higher temperatures (Fig. 1A). Source Models df F ratio Feeding temperature, but not the source of habitat Leaf area Temperature 1 12.21b (sun/shade) of leaves, signiÞcantly inßuenced larval consumed Leaf source 1 2.45 biomass gain (Fig. 1C; Table 2). However, efÞciency Temp ϫ source 1 0.65 of conversion, i.e., larval weight gain per unit area of Covariate (initial larval weight) 1 11.18b b leaf consumed, exhibited the opposite effect. Tem- Percent leaf area Temperature 1 7.178 consumed Leaf source 1 3.095 perature did not inßuence the conversion of leaf area Temp ϫ source 1 2.875 consumed into larval biomass, but whether the leaves Covariate (initial larval weight) 1 104.711c came from the sun or shade did (Fig. 1D; Table 2). Biomass gained Temperature 1 5.139a Leaf source 1 0.068 Larvae fed leaves from sun habitats gained signiÞ- Temp ϫ source 1 0.541 cantly more weight per amount of leaf material con- Covariate (initial larval weight) 1 12.783b sumed than those fed on leaves from the shade inde- Conversion Temperature 1 3.242 pendent of feeding temperature. The interaction Leaf source 1 6.177a Temp ϫ source 1 3.904 between temperature and leaf source was never sig- Covariate (initial larval weight) 1 0.208 niÞcant, but the covariate, initial larval mass, was sig- niÞcant for most variables (Table 2). a P Ͻ 0.05; b P Ͻ 0.005; c P Ͻ 0.001; all else not signiÞcant, P Ͼ 0.05. 604 ENVIRONMENTAL ENTOMOLOGY Vol. 35, no. 3

Table 3. ANCOVAs for field feeding trials area consumed and biomass gained but not for efÞ- ciency of conversion (Table 3). Source Models df F ratio Leaf area consumed Habitat 1 1.991 Covariate (initial larval weight) 1 131.908b Discussion Biomass gained Habitat 1 0.05 Covariate (initial larval weight) 1 20.646b Feeding microclimate within a population of her- Conversion Habitat 1 3.965a bivores can vary greatly over small spatial scales. Here, Covariate (initial larval weight) 1 0.699 we show an order of magnitude difference in the mean incident PAR at plants within a few meters of each aP Ͻ 0.05; b P Ͻ 0.001; all else not signiÞcant, P Ͼ 0.05 other (Table 1). Higher PAR in sun habitats results in signiÞcantly greater ambient temperatures and leaf surface temperatures (Table 1). Although higher tem- into larval biomass was signiÞcantly greater on plants peratures result in increased transpiration, which in in the sun than in the shade (Table 3; Fig. 2C). The turn should cause cooling, at mid-day, the average leaf covariate, initial larval mass, was signiÞcant for leaf surface temperatures were still 5ЊC greater in sun habitats. These differences in the temperature of the microhabitats where herbivores feed are likely to in- ßuence aspects of herbivore consumption that are not typically considered in studies of light and herbivory. In the laboratory, larvae of E. hortaria consumed more leaf area, consumed a greater percentage of total leaf area, and gained more biomass at warmer tem- peratures. This result is consistent with previous stud- ies of the effects of temperature on herbivore perfor- mance in which lepidoptera larvae had higher consumption rates and developed faster in elevated temperatures (Sherman and Watt 1973, Scriber and Lederhouse 1983, Kingsolver and Woods 1997). Higher growth rates in lepidoptera reared in warmer temperatures have been attributed to increased con- sumption rates (Sherman and Watt 1973, Scriber and Lederhouse 1983, Kingsolver and Woods 1997, Levesque et al. 2002). Papilio glaucus reared at 30ЊC had relative consumption rates that were twice as high than when reared at 15ЊC, and larval growth rates were directly related to increases in food intake (Scriber and Lederhouse 1983). Malacosoma disstria has dis- played similar foraging behavior in response to in- creases in temperature (Levesque et al. 2002). In the Þeld, we consistently see greater rates of seasonal herbivory in cooler, shaded microhabitats than in warmer, sun habitats; as do others (Louda and Rodman 1983, Larson et al. 1986, Niesenbaum 1992, Lindroth et al. 1993, Dudt and Shure 1994, Sagers and Coley 1995). However, we saw no differences in rates of consumption in Þeld feeding trials in the sun and the shade. That larval feeding rates were not elevated in warmer, sunlit habitats in the Þeld and seasonal her- bivory is greater in the shade contradict our laboratory results and suggest that temperature may be less im- portant in determining feeding rates in the Þeld. A variety of ecological factors that inßuence insect per- formance and behavior in their natural environment, such as the risk of predation and more variable tem- peratures and microhabitats, are absent in laboratory experiments and might place an artiÞcial emphasis on the extent to which temperature can affect herbivory Fig. 2. Field feeding trials. Sun and shade differences in rates. In the Þeld, light-based variation in plant de- (A) mean leaf area consumed, (B) mean larval biomass fenses and nutritional quality might largely account gained, and (C) mean conversion of leaf area into larval for the absence of foraging patterns and differences in biomass. Error bars represent SE. Bars with the same letter insect performance associated with increases in tem- are not signiÞcantly different from each other (P Ͼ 0.05). perature. June 2006 NIESENBAUM AND KLUGER:LIGHT,TEMPERATURE, AND HERBIVORY 605

E. hortaria larvae consuming L. benzoin leaves from Acknowledgments sunlit environments had higher efÞciencies of con- This work was support in part by NSF-DEB Award 0128098 version than their counterparts feeding on leaves from and NSF-DBI Award 0442049 to R. Niesenbaum, Merck/ the shade, independent of temperature. Both in- AAAS, and Muhlenberg College. R. Kipp, B. Tavernia, and L. creases in the nutritional quality of plant material Mastro provided support in the Þeld, and M. Edwards, E. (Mattson 1980, Scriber and Slansky 1981, Lincoln and Iyengar, B. Casper, J. Thaler, S. Agosta, N. Muth, P. Ellsworth, Mooney 1984, Fortin and Mauffette 2001, Levesque et and two anonymous reviewers provided excellent comments al. 2002), and elevations in plant antiherbivore de- on earlier versions of this paper. fenses (Coley 1983, Dudt and Shure 1994, Sagers and Coley 1995, Crone and Jones 1999) can lead to de- pressed rates of herbivory and essentially counteract References Cited the effect that thermal environment has on E. hortaria Aide, M. T., and J. K. Zimmerman. 1990. Patterns of insect feeding in warm, sunlit habitats. Ecologists typically herbivory, growth, and survivorship of juveniles of a neo- measure herbivory as area removed because this es- tropical liana. Ecology 71: 1412Ð1421. timates loss of photosynthetic potential (Strauss and Aizen, M. A., and W. A. Patterson. 1995. Leaf phenology and Agrawal 1999); however, insect growth and conver- herbivory along a temperature gradient: a spatial test of sion is more typically assessed strictly on the basis of the phenological window hypothesis. J. Veg. Sci. 6: 543Ð 550. biomass (Gavloski and Lamb 2000). Leaf thickness Ayres, M. P., and M. J. Lombardero. 2000. Assessing the and water content are also likely to vary with light as consequences of climate change for forest herbivore and well (Field and Mooney 1983, Yun and Taylor 1986), pathogens. Sci. Tot. Environ. 262: 263Ð286. although these may vary even more with leaf age, Bonser, S. P., and R. J. Reader. 1995. Plant competition and nitrogen availability, and other factors. Future studies herbivory in relation to vegetation biomass. Ecology 76: relating leaf herbivory to insect growth and conver- 2176Ð2183. sion must consider ways to accurately relate or con- Braun, E. L. 1950. Deciduous forests of eastern North America. The Free Press, New York. vert the measure most relevant to plant physiology, Coley, P. D. 1983. Herbivory and defensive characteristics leaf area, to the measure most relevant to insect per- of tree species in a lowland tropical forest. Ecol. Monogr. formance, biomass gain, and should also take into 53: 209Ð233. account leaf water content and larval dry weights (van Crone, E. E., and C. G. Jones. 1999. The dynamics of carbon- Loon et al. 2005). nutrient balance: effects of cottonwood acclimation to Our Þeld experimental enclosures resulted in sig- short and long-term shade on beetle feeding preferences. J. Chem. Ecol. 25: 635Ð659. niÞcant, but small, increases in temperature. These Dudt, J. F., and D. J. Shure. 1994. The inßuence of light and differences could be exaggerated during periods of nutrients on foliar phenolics and insect herbivory. Ecol- high light and temperature. Given the common use of ogy 75: 86Ð98. enclosures in studies of insect herbivory, the lack of Dury, S. J., F. G. Good, C. M. Perrins; A. Buse, and T. Kaye. real alternatives for conducting Þeld feeding trials, and 1998. The effects of increasing CO2 and temperature on the clear potential for light and temperature to either oak leaf palatability and the implications for herbivorous directly or indirectly inßuence the outcome of these insects. Global Change Biol. 4: 55Ð61. Field, C., and H. A. Mooney. 1983. Leaf age and seasonal interactions, attention to enclosure effects should be effects on light, water, and nitrogen use efÞciency in a considered (Stamp and Bowers 1994b). California shrub. Oecologia (Berl.) 56: 348Ð355. A number of studies have considered various mi- Fortin, M., and Y. Mauffette. 2001. Forest edge effects on croclimatic effects on herbivory and insect perfor- the biological performance of the forest tent caterpillar mance. This includes studies of light environment in- (Lepidoptera: Lasiocampidae) in sugar maple stands. cluding UV light (Warren et al. 2002, Henriksson et al. Ecoscience 8: 164Ð172. Fraser, L. H., and J. P. Grime. 1998. Top-down control and 2003), temperature (Ayres and Lombardero 2000), its effect on the biomass and composition of three grasses and CO2, particularly within the context of global at high and low soil fertility in outdoor microcosms. Oe- climate change (Dury et al. 1998, Pen˜ uelas and Lluisa´ cologia (Berl.) 113: 239Ð246. 2003). However, light as a major determinant of her- Gavloski, J. E., and R. J. Lamb. 2000. SpeciÞc impacts of bivory is rarely, if ever, considered within the context herbivores: comparing diverse insect species on young of temperature. Because light inßuences feeding tem- plants. Environ. Entomol. 29: 1258Ð1267. perature, leaf defense, and leaf nutritional quality, Henriksson, J., E. Haukioja, V. Ossipov, S. Ossipova, S. Sil- lanpaa, L. Kapari, and K. Pihlaja. 2003. Effects of host future work on the effects of light on herbivory must shading on consumption and growth of the geometrid begin to tease apart these confounding factors. We Epirrita autumnata: interactive roles of water, primary suggest that the effects of light and temperature on and secondary compounds. Oikos 103: 3Ð16. herbivore feeding and growth rates can be partially Jansen, P. T., and N. E. Stamp. 1997. Effects of light avail- resolved by comparing laboratory data with herbivore ability on host plant chemistry and the consequences for performance in the Þeld and by incorporating more behavior and growth of an insect herbivore. Entomol. Exp. Appl. 82: 319Ð326. detailed analyses of ecological factors, such as preda- Kessler, A., and I. T. Baldwin. 2001. Defensive function of tion and plant chemistry, that inßuence phytophagous herbivore-induced plant volatile emissions in nature. Sci- insects in concert with their biophysical environment. ence 16: 2141Ð2144. 606 ENVIRONMENTAL ENTOMOLOGY Vol. 35, no. 3

Kingsolver, J. G., and H. A. Woods. 1997. Thermal sensitivity Niesenbaum, R. A. 1992. The effects of light environment of growth and feeding in Manduca sexta caterpillars. on herbivory and growth in the dioecious shrub Lindera Physiol. Zoo. 70: 631Ð638. benzoin (Lauraceae). Am. Midl. Nat. 128: 270Ð275. Knapp, R., and T. M. Casey. 1986. Thermal ecology, behav- Niesenbaum, R. A. 1996. Linking herbivory and pollination: ior and growth of gypsy and eastern tent caterpil- defoliation and selective fruit abortion in Lindera ben- lars. Ecology 67: 598Ð608. zoin. Ecology 77: 2324Ð2331. Kuokkanen, K., R. Julkunen-Tiitto, M. Keina¨nen, P. Niemela¨, Pen˜ uelas, J., and J. Lluisa´. 2003. BVOCs: plant defense and J. Tahvanainen. 2001. The effect of elevated CO2 against climate warming? Trends Plant Sci. 3: 105Ð109. and temperature on the secondary chemistry of Betula Sagers, C. L., and P. D. Coley. 1995. BeneÞts and costs of pendula seedlings. Trees Struct. Funct. 6: 378Ð384. defense in a neotropical shrub. Ecology 76: 1835Ð1843. Larson, S., A. Wiren, L. Lundgren, and T. Ericsson. 1986. Scriber, J. M., and R. C. Lederhouse. 1983. Temperature as Effects of light and nutrient stress on leaf phenolic chem- a factor in the development and feeding ecology of tiger istry in Salix dasclados and susceptibility to Galerucella swallowtail caterpillars, Papilio glaucus (Lepidoptera). lineola. Oikos 47: 205Ð210. Oikos. 40: 95Ð102. Levesque, K. R., M. Fortin, and Y. Mauffette. 2002. Tem- Scriber, J. M., and F. Slansky. 1981. The nutritional ecology perature and food quality effects on growth, consump- of immature insects. Annu. Rev. Entomol. 26: 183Ð211. tion, and post-ingestive utilization efÞciencies of the for- Sherman, P. W., and W. B. Watt. 1973. The thermal ecology est tent caterpillar Malacosoma disstria (Lepidoptera: of some Colias butterßy larvae. J. Comp. Physiol. 3: 25Ð40. Lasiocampidae). Bull. Entomol. Res. 92: 127Ð136. Shure, D. J., and L. A. Wilson. 1993. Patch-size effects on Lincoln, D. E., and H. A. Mooney. 1984. Herbivory on plant phenolics in successional openings of the Southern Diplacus aurantiacus shrubs in the sun and shade. Oeco- Appalachians. Ecology 74: 55Ð67. Stamp, N. E., and M. D. Bowers. 1994a. Effect of tempera- logia (Berl.) 64: 173Ð176. ture and leaf age on growth versus moulting time of a Lindroth, R. L., P. B. Reich, M. G. Tjoelker, J. C. Volin, and generalist caterpillar fed plantain (Plantago lanceolata). J. Olesyn. 1993. Light environment alters response to Ecol. Entomol. 19: 199Ð206. ozone stress in seedlings of Acer saccharum Marsh. and Stamp, N. E., and M. D. Bowers. 1994b. Effects of cages, hybrid LIII: consequences for performance of plant age and mechanical clipping on plantain chemistry. gypsy moth. New Phytol. 124: 647Ð651. Oecologia (Berl.) 99: 66Ð71. Louda, S. M., and J. E. Rodman. 1983. Concentration of Strauss, S. Y., and A. A. Agrawal. 1999. The ecology and glucosinolates in relation to habitat and insect herbivory evolution of plant tolerance to herbivory. Trends Ecol. for the native crucifer Cardamine conlifolia. Biochem. Evol. 14: 179Ð185. Syst. Ecol. 11: 199Ð207. van Loon, J.J.A., J. Casas, and S. Pincebourde. 2005. Nutri- Louda, S. M., and J. E. Rodman. 1996. Insect herbivory as a tional ecology of insectÐplant interactions: persistent major factor in the shade distribution of a native crucifer handicaps and the need for innovative approaches. Oikos (Cardamine cordifolia A. Gray, bittercress). J. Ecol. 84: 108: 194Ð201. 229Ð237. Warren, J. M., J. H. Bassman, and S. Eigenbrode. 2002. Leaf Mattson, W. J. 1980. Herbivory in relation to plant nitrogen. chemical changes induced in Populus trichocarpa by en- Annu. Rev. Ecol. Syst. 11: 119Ð161. hanced UV-B radiation and concomitant effects on her- Mazı´a, N., T. Kitzberger, and E. J. Chaneton. 2004. Inter- bivory by Chrysomela scripta (Coleoptera: Chrysomeli- annual changes in folivory and bird insectivory along a dae). Tree Phys. 22: 1137Ð1146. natural productivity gradient in northern Patagonian for- Yang, Y., and N. E. Stamp. 1996. Simultaneous effects of ests. Ecography 27: 29Ð32. temperature and multiple allelochemicals on the perfor- Mole, S., and Waterman, P. G. 1988. Light-induced varia- mance of a Solanaceae specialist caterpillar (Manduca tion in phenolic levels in foliage of rain-forests plants. II. sexta). Ecoscience 3: 81Ð92. Potential signiÞcance to herbivores. J. Chem. Ecol. 14: Yun, J., and E. Taylor. 1986. Adaptive implications for leaf 23Ð34. thickness for sun- and shade-grown Abutilon theophrasti. Nichols-Orians, C. M. 1991. The effects of light on foliar Ecology 67: 1314Ð1318. chemistry, growth and susceptibility of seedlings of a canopy tree to an attine ant. Oecologia (Berl.) 86: 552Ð Received for publication 12 March 2005; accepted 26 Janu- 560. ary 2006.