POPULATION ECOLOGY AND MANAGEMENT OF THE PALE-WINGED GRAY , EPHYRARIA WLK. (: GEOMETRIDAE)

by

Lauren Lynn Pinault

BSc. Biology, University of Ottawa, 2005

A Thesis, Dissertation or Report Submitted in Partial Fulfilment of the Requirements for the Degree of

Masters of Science in Forestry

in the Graduate Academic Unit of the Faculty of Forestry and Environmental Management

Supervisors: Dan Quiring, PhD., Faculty of Forestry and Environmental Management Graham Thurston, PhD., Canadian Forest Service

Examining Board: Eldon Eveleigh, PhD, Natural Resources Canada, Chair Gilles Boiteau, PhD, Agriculture Canada

This thesis, dissertation or report is accepted by the Dean of Graduate Studies

THE UNIVERSITY OF NEW BRUNSWICK

September, 2007

© Lauren Pinault, 2007 Library and Archives Bibliotheque et 1*1 Canada Archives Canada Published Heritage Direction du Branch Patrimoine de I'edition

395 Wellington Street 395, rue Wellington OttawaONK1A0N4 OttawaONK1A0N4 Canada Canada

Your file Votre reference ISBN: 978-0-494-63747-0 Our file Notre reference ISBN: 978-0-494-63747-0

NOTICE: AVIS:

The author has granted a non­ L'auteur a accorde une licence non exclusive exclusive license allowing Library and permettant a la Bibliotheque et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par telecommunication ou par I'lnternet, preter, telecommunication or on the Internet, distribuer et vendre des theses partout dans le loan, distribute and sell theses monde, a des fins commerciales ou autres, sur worldwide, for commercial or non­ support microforme, papier, electronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats.

The author retains copyright L'auteur conserve la propriete du droit d'auteur ownership and moral rights in this et des droits moraux qui protege cette these. Ni thesis. Neither the thesis nor la these ni des extraits substantiels de celle-ci substantial extracts from it may be ne doivent etre imprimes ou autrement printed or otherwise reproduced reproduits sans son autorisation. without the author's permission.

In compliance with the Canadian Conformement a la loi canadienne sur la Privacy Act some supporting forms protection de la vie privee, quelques may have been removed from this formulaires secondaires ont ete enleves de thesis. cette these.

While these forms may be included Bien que ces formulaires aient inclus dans in the document page count, their la pagination, il n'y aura aucun contenu removal does not represent any loss manquant. of content from the thesis.

1+1 Canada ABSTRACT

The pale-winged gray moth, Wlk. (Lepidoptera:

Geometridae) is an indigenous defoliator of eastern hemlock, L.

Carr, with no prior record of outbreak. Severe defoliation of mature hemlock stands in southwestern Nova Scotia began in 2002. The life stages, phenology, and natural enemies oil. ephyraria were described. The efficacy of sticky tape traps to monitor larval populations was assessed and larval densities explained 77% of current-year shoot defoliation.

Defoliation of mature trees attributable to I. ephyraria was highest on current- year foliage and was related to previous defoliation. Survival of early and late instars was highest on current-year and older foliage (> 1-year-old), respectively, supporting the ontogenetic hypothesis that nutritional requirements change with instar age.

However, both early and late-instar larvae had higher survival when provided access to all foliage age classes, also supporting the hypothesis that a balanced diet improves larval performance.

n ACKNOWLEDGEMENTS

I am indebted to my advisory committee: Dan Quiring, Graham Thurston,

Chris Lucarotti and Blair Pardy, for their tremendous assistance with this project. I am very appreciative to Chris McCarthy and Harry Delong of Parks Canada, and Eric

Georgeson, Robert Guscott, and Mike LeBlanc of Nova Scotia Department of Natural

Resources. Jim Crooker, Royce Ford, Andrew Ross, Tim Atkins and Andrea Wegerer graciously provided access to private woodlots for study. I am grateful to the following people for field assistance: Christa Brittain, Megan Crowley, Antoine Davy,

John Dixon (Bowater Inc.), Leah Flaherty, Jennifer Guy, Ryan Jameson, Zara

McKenna-Fuentes, Ryan McPhee, Vanessa Robichaud, Amanda Savoie, Kate Van

Rooyen, and Manon Vincent, Nova Scotia Power, and especially Lucie Carrat and

Ryan Jameson. I would also like to thank Marie-Paule McNutt and Colleen Teerling for help in the laboratory, Jean-Francois Landry, Jose Leonardo Fernandez, and

Robert Anderson, who assisted me with identifications, and Leah Flaherty,

Heidi Fry, Roger Graves, Steve Heard, Rob Johns, Jonathan Leggo, and Kate Van

Rooyen for helpful feedback on manuscripts. John Kershaw and Myriam Barbeau were very helpful with statistics.

This project was funded by an NSERC IPS in conjunction with Forest

Protection Limited. Additional support was provided by Parks Canada, Canadian

Forest Service, the Nova Scotia Department of Natural Resources, and NSERC.

Finally, I would like to thank my family and Greg Rampersad for so much encouragement along the way. L.L.P.

iii TABLE OF CONTENTS

Page

ABSTRACT ii

ACKNOWLEDGEMENTS iii

TABLE OF CONTENTS iv

LIST OF TABLES vi

LIST OF FIGURES vii

CHAPTER 1 INTRODUCTION 1

References 8

CHAPTER 2 LIFE HISTORY OF IRIDOPSIS EPHYRARIA, (LEPIDOPTERA: GEOMETRIDAE), A DEFOLIATOR OF EASTERN HEMLOCK IN EASTERN CANADA 14

2.1 Introduction 14 2.2 Methods 15 2.3 Results 22 2.4 Discussion 26 Figures 30 References 36

CHAPTER 3 SAMPLING STRATEGIES AND DENSITY-DEFOLIATION RELATIONSHIPS FOR THE PALE-WINGED GRAY MOTH, A DEFOLIATOR OF MATURE EASTERN HEMLOCK 41

3.1 Introduction 41 3.2 Methods 43 3.3 Results 47 3.4 Discussion 49 Tables and Figures 51 References 54

CHAPTER 4 THE INTERACTION OF FOLIAGE AND LARVAL AGE INFLUENCES THE PREFERENCE AND PERFORMANCE OF A CATERPILLAR ON EASTERN HEMLOCK 57

4.1 Introduction 57

iv 4.2 Methods 58 4.3 Results 62 4.4 Discussion 64 Tables and Figures 67 References 74

CHAPTER 5 GENERAL CONCLUSIONS 77

References 80

CURRICULUM VITAE

v LIST OF TABLES

Page

CHAPTER THREE

Table 1. Details of the study sites and years, where F = Foam trap, S = Sticky tape, and B = Beating sheet sampling (indicated with [*]). Site coordinates, previous defoliation (mean (±SE) percent defoliation of one-year-old shoot, in May), and mean (± SE) hemlock diameter at breast height are presented 51

CHAPTER FOUR

Table 2. Summary of nested ANOVA assessing the influence of date (D), foliage age (F), tree (T) and site (S) on the location where larvae were observed feeding 67

Table 3. Summary of split-plot ANCOVA evaluating the influence of foliage age on defoliation by /. ephyraria on mature hemlock trees in two years 68

Table 4. Summary of split-plot ANCOVA evaluating the influence of foliage age on defoliation by /. ephyraria on small, understory hemlock trees in two years 69

vi LIST OF FIGURES

Page

CHAPTER TWO

Figure 1. Map of the distribution of hemlock defoliation by I. ephyraria detected during aerial surveys between 2002-2006 30

Figure 2. Photographs of the life stages of/, ephyraria: (a) an egg, a few weeks after oviposition; (b) two late-instar larvae; (c) a prepupa; (d) apupa; and (e) an adult moth 31

Figure 3. Relationship between the vertical height up the bole of four mature hemlock trees, expressed as the proportion of the total height, and the log-transformed density of I. ephyraria eggs from bark samples 32

Figure 4. Frequency distribution of head capsule widths of/ ephyraria larvae from field collections, insectary rearings and first-instar larvae that emerged from overwintered eggs in the lab 33

Figure 5. The temporal distribution of/ ephyraria instars during summer 2005 34

Figure 6. Mean (± SE) percent survival of early and late instar larvae of/ ephyraria on eastern hemlock and five potential alternative host species in South Brookfield in 2006 35

CHAPTER THREE

Figure 7. Relationship between mean density per site of first instar / ephyraria larvae sampled from sticky tape on mature hemlock trees for 2004 (A), 2005 (•) and 2006 (•), and the mean amount of seasonal defoliation per site on (a) current year foliage, (b) current-year and one-year-old foliage, and (c) all age classes (i.e., current-year to 6 year-old) foliage. All regressions are significant (P<0.01) 52

Figure 8. Relationship between mean density of/ ephyraria larvae collected in beating sheets below young hemlock trees and the mean percentage of current-year foliage that / ephyraria larvae consumed by the end of the summer in 2005 (A) and 2006 (•), at 15 sites in southern Nova Scotia. This regression is significant (PO.01) 53

vn CHAPTER FOUR

Figure 9. Mean (± SE) number of larvae per branch section feeding on different-aged foliage on small, understory hemlock trees at four sites, in 2004 70

Figure 10. Mean (± SE) defoliation of different ages of foliage in mature hemlock trees in (a) 2005 and (b) 2006 71

Figure 11. Mean (± SE) defoliation of different ages of foliage in small understory hemlock trees in (a) 2004 and (b) 2005 72

Figure 12. Mean (± SE) survival of early (black bars) and late-instar (white bars) larvae feeding on current-year (C), or old (1-6- year-old) foliage, or both current and old foliage (mixed), of eastern hemlock in South Brookfield, NS, in 2006 73

viii 1

CHAPTER ONE

INTRODUCTION

Forest insect pests cause damage and tree mortality to several million hectares of Canadian forests every year, and constitute one of the most significant problems for forest ecosystem management. For instance, an estimated 13.1 million ha were damaged by in 2004 (Anonymous 2006). Defoliating Lepidoptera are very significant and common forest pests, and caterpillars such as the forest tent caterpillar,

Malacosoma disstria Hub. (Lassiocampidae) and the spruce budworm, Choristoneura fumiferana Clem. (Torticidae) have dominated the scientific attention of forest entomologists (Anonymous 2006). Lepidoptera can adopt a variety of feeding habits, such as feeding on different ages of foliage and crown levels within a host (Carroll &

Quiring 1994; Suomela et al. 1995; Fortin & Mauffette 2002), feeding on alternative hosts (Barbosa & Greenblatt 1979; deBoer & Hanson 1984), feeding on specific parts of needles or leaves (Wallin & Raffa 1997), and consuming foliage during specific hours of the day or night (Fitzgerald et al. 1988). These adaptations allow larvae to improve nutritional intake, avoid toxins, maintain suitable body temperatures and humidity levels, and evade natural enemies.

Commonly, lepidopteran pests such as the pine looper moth, Bupalus piniaria

L. (Geometridae) recur in cycles every few years (Kendall et al. 2005). New forest pests are often exotic species (Simberloff 1986; Liebhold et al. 1995), and the impact of introduced insects can be more severe than that of indigenous insects (Pimentel

1986). For example, the gypsy moth, Lymantria dispar L. (Lepidoptera:

Lymantriidae), was introduced accidentally to North America and has since 2

extensively defoliated a variety of plant hosts, most commonly oaks (Quercus species)

(Liebhold et al. 1992). On the other hand, it is uncommon for a very rare and

indigenous species to become a significant outbreak insect in a few years.

Prior to 2002, the pale-winged gray moth, Iridopsis ephyraria Wlk.

(Lepidoptera: Geometridae) was an innocuous species with rare abundance. Its range

extends from Nova Scotia to Alberta, and south as far as Texas (Ferguson 1954;

Rindge 1966; McGuffin 1977). There are no refereed publications of/, ephyraria ecology, and it is only mentioned in a government handbook as a minor, occasional pest of cranberry (Landry et al. 2002). Iridopsis ephyraria adults are described in

Rindge (1966) and McGuffin (1977), however description of the juveniles is incomplete in these studies (Ferguson 1985; McGuffin 1977) and all information of its life history relies on details noted during the collection of specimens.

In 2002, very severe defoliation of eastern hemlock, Tsuga canadensis (L.)

Carr. (Pinaceae) began in localized patches of southwestern Nova Scotia, specifically in and around Kejimkujik National Park and National Historic Site, initially affecting

507 ha (R. Guscott, NSDNR, pers. comm.). Mature hemlocks were nearly completely defoliated at some sites, while adjacent mature hemlock sites were not affected. The outbreak area expanded to encompass 821 ha in 2003, and 1549 ha in 2004 (R.

Guscott, NSDNR, pers. comm.). By 2005, many hemlock stands across southwestern

Nova Scotia were affected (approximately 23855 ha), although insect densities at sites had declined substantially. In 2006, approximately 43343 ha were defoliated (R.

Guscott, NSDNR, pers. comm.), although the majority of sites were only lightly defoliated. Iridopsis ephyraria also consumed a variety of plants in the understory, 3 consistent with previous studies that reported collecting larvae on a wide variety of hosts (Ferguson 1954; Rindge 1966; McGuffin 1977).

Eastern hemlock is a slow-growing tree of the pine family that can live for as long as 800 years, and grows across the eastern United States and Canada (Godman

& Lancaster 2006). Although the majority of hemlock stands are not unique in plant species composition (Rogers 1980), they provide habitat for white-tailed deer

(Odocoileus virginianus Zimmer.) (Anderson & Loucks 1979; Rooney & Waller

2003), redback salamanders (Plethodon cinereus) (Brooks 2001) and many bird species (Morrison 2006).

The hemlock woolly adelgid, Adelges tsugae Ann. (Hemiptera: Adelgidae), also causes loss of foliage and mortality of eastern hemlock (McClure 2002), and it is likely that defoliated stands in the pale-winged gray moth outbreak would be similarly affected. Following adelgid damage, more light reaches the forest floor, leading to reduced soil moisture and a higher pH (Orwig & Foster 1998; Cobb et al.

2005). Increased light incidence allows other plant species to move into the understory (Orwig & Foster 1998; Cobb et al. 2005; Eschtruth et al. 2006). Frass deposition can also cause nitrogen to fall to the forest floor, leading to leaching from the system (Jenkins et al. 1999; Cobb et al. 2005). Since young hemlock trees are often browsed by deer (Anderson & Loucks 1979; Rooney & Waller 2003), it is difficult for young hemlocks to fill forest gaps, and there is a danger that extensive defoliation might lead to widespread changes in forest composition in the long term. 4

The purpose of this research is to describe the life history and ecology of the pale-winged gray moth, in order to construct an effective management plan to limit future damage caused by this forest pest. The study objectives were threefold: first, to describe /. ephyraria juveniles and life history (Chapter Two); second, to develop density-defoliation relationships that utilize new sampling methods to predict subsequent defoliation on mature trees (Chapter Three); and third, to study the feeding patterns within the crown of a mature hemlock, specifically, the feeding patterns on different ages of foliage (Chapter Four).

Since some details of the pale-winged gray moth life history were unknown, it was necessary to describe all juvenile life stages and to determine the number of larval instars to permit field identification (Chapter Two). In addition, alternative hosts were identified and larval performance determined. Natural enemies were also identified, permitting a more thorough understanding of the population dynamics of this system.

From preliminary observations, /. ephyraria is similar in many aspects of its life history to the hemlock looper, Lambdina fiscellaria fiscellaria Guenee

(Lepidoptera: Geometridae), a well-studied defoliator of balsam fir,

(L.) Mill., in Atlantic Canada. Hemlock looper outbreaks have occurred sporadically in localized patches (Carroll 1956) and may last for as long as 7 years (Raske et al.

1995). Eggs overwinter (Carroll 1956), and larvae pass through four instars, consuming a variety of food wastefully, including balsam fir, eastern hemlock, spruce, , and red (Carroll 1956; Rose & Lindquist 1977). Young larvae often consume younger balsam fir foliage, whereas older larvae can feed on a variety of 5 foliage ages (Rose & Lindquist 1977). Eggs, larvae and pupae are parasitized by a variety of Diptera (Tachinidae) and Hymenoptera (Ichneumonidae and Scelionidae)

(Carroll 1956). Fungal pathogens, such as Entomophaga aulicae (Entomophthorales:

Zygomycetes), might cause declines in looper populations (Raske et al. 1995).

In Chapter Three, pale-winged gray moth density-tree defoliation relationships were described for three different insect sampling techniques. Density- defoliation relationships have been determined for a number of forest insect pests

(Lysyk 1990; Parsons et al. 2005; Johns et al. 2006), allowing managers to accurately predict seasonal defoliation from insect density counts taken earlier in the spring or previous summer. Since affected mature hemlock trees were too large to use standard sampling techniques, such as pole pruners or ladders (e.g., Jennings et al. 1990), three other sampling techniques were tested in the field. First, eggs were sampled using a foam oviposition trap design similar to that used in Hebert et al. (2003), which has successfully been used to sample hemlock looper eggs. First-instar larvae were sampled using a band of sticky tape placed around the bole of the tree. Late-instar larvae on low-hanging branches were sampled using the beating sheet method, evaluated previously by Harris et al. (1972). Tree defoliation was estimated using binoculars to observe the middle crown, as in Parsons et al. (2003). The objective was to find at least one sampling method which would correlate strongly with defoliation, and allow land managers to use the linear portion of the relationship to decide if intervention is necessary.

Chapter Four focuses on the feeding patterns of the pale-winged gray moth on differently-aged foliage of eastern hemlock. Similar to many species of sawfly and 6 the ramie moth, Arete coerulea (Lepidoptera: Noctuidae), which feed on a mixture of older foliage ages (Ikeda et al. 1977; Jensen 1988; Moreau et al. 2003; Parsons et al.

2003; Ide 2006), pale-winged gray moth larvae appear to feed on a mixture of foliage ages and seem to prefer the current-year shoot.

Feeding on multiple age classes might allow larvae to consume foliage of higher nutritional quality while reducing intake of harmful allelochemicals (balanced diet hypothesis) (Cates 1980; Moreau et al. 2003). Alternatively or concurrently, feeding on multiple age classes might also reflect different nutritional requirements of early and late instars (ontogenetic hypothesis) (Bernays et al. 1991; Hochuli 2001).

To assess both hypotheses, the performance of larvae on current-year, old (> 1-year- old), and a mixed diet of foliage ages was determined for both early and late instars.

Chapter Two has published in the Journal of the Acadian Entomological

Society (Pinault et al. 2007). There are eight other co-authors who contributed the following: Eric Georgeson (NSDNR) developed the methodological technique for bark washing, Robert Guscott (NSDNR) provided geographical data on the spread of the outbreak and helped to locate study sites, Ryan Jameson provided useful feedback in the field on methodology and helped carry out much of the field research in 2006,

Mike LeBlanc (NSDNR) helped to locate study sites, and carried out tree felling and bark washing, Chris McCarthy (Parks Canada) and Graham Thurston (CFS) both provided preliminary data on the early stages of the outbreak and personal observations, Christopher Lucarotti (CFS) provided information regarding the fungal pathogen, and Dan Quiring (UNB) contributed significantly to all methodology, analysis, and paper content. All authors contributed to manuscript review as well. 7

Chapter Three was submitted for publication in the journal Forest Ecology and

Management and was co-authored by Dan Quiring (UNB), who contributed extensively to the methodology and review of the paper. Chapter Four will be submitted for publication to an ecological journal and will be co-authored by Dan

Quiring (UNB) and Graham Thurston (CFS), who contributed to the design and implementation of experiments, as well as to the review of the manuscript. 8

REFERENCES

Anderson, R.C., and Loucks, O.L. 1979. White-tail deer (Odocoileus virginianus)

influence on structure and composition of Tsuga canadensis forests. J. Appl.

Ecol. 16(3): 855-861.

Anonymous. 2006. The state of Canada's forests 2005-2006: Forest industry

competitiveness. Nat. Res. Can., Can. For. Serv., Headquarters, Ottawa. ON.

Barbosa, P., and Greenblatt, J. 1979. Suitability, digestibility and assimilation of

various host plants of the gypsy moth Lymantria dispar L. Oecologia, 43:

111-119.

Bernays, E.A., Jarzembowski, E.A., and Malcolm, S.B. 1991. Evolution of insect

morphology in relation to plants [and discussion]. Philos. Trans.: Biol. Sci.

333(1267): 257-264.

Brooks, R.T. 2001. Effects of the removal of overstory hemlock from hemlock-

dominated forests on eastern redback salamanders. For. Ecol. Manage. 149:

197-204.

Carroll, A.L., and Quiring, D.T. 1994. Intra tree variation in foliage development

influences the foraging strategy of a caterpillar. Ecology, 75(7): 1978-1990.

Carroll, W.J. 1956. History of the hemlock looper, Lambdina fiscellaria fiscellaria

(Guen.), (Lepidoptera: Geometridae) in Newfoundland, and notes on its

biology. Can. Entomol. 88: 587-599.

Cates, R.G. 1980. Feeding patterns of monophagous, oligophagous, and polyphagous

insect herbivores: the effect of resource abundance and plant chemistry.

Oecologia, 46:22-31. 9

Cobb, R.C., Orwig, D.A., and Currie, S. 2005. Decomposition of green foliage in

eastern hemlock forests of southern New England impacted by hemlock

woolly adelgid infestations. Can. J. For. Res. 36: 1331-1341. deBoer, G., and Hanson, F.E. 1984. Food plant selection and induction of feeding

preference among host and non-host plants in larvae of the tobacco hornworm

Manduca sexta. Entomol. Exp. Appl. 35: 177-193.

Eschtruth, A.K., Cleavitt, N.L., Battles, J.J., Evans, R.A., and Fahey, T.J. 2006.

Vegetation dynamics in declining eastern hemlock stands: 9 years of forest

response to hemlock woolly adelgid infestation. Can. J. For. Res. 36: 1435—

1540.

Ferguson, D.C. 1954. The Lepidoptera of Nova Scotia 1. Macrolepidoptera. Bull. N.S.

Inst. Sci. 123(3).

Ferguson, D.C. 1985. Insects of eastern forests, U.S. Dept. Agric. Misc. Publ. 1426.

Fitzgerald, T.D., Casey, T., and Joos, B. 1988. Daily foraging schedule of field

colonies of the eastern tent caterpillar. Oecologia, 76: 574-578.

Fortin, M., and Mauffette, Y. 2002. The suitability of leaves from different canopy

layers for a generalist herbivore (Lepidoptera: Lasiocampidae) foraging on

sugar maple. Can. J. For. Res. 32: 379-389.

Godman, R. M., and Lancaster, K. 1990. Tsuga canadensis (L.) Carr. eastern

hemlock. In Silvics of North America. (Agricultural Handbook). Edited by R.

M. Burns, B.H. Honkala. Dept. Agri. For. Serv.. Washington, DC.

Harris, J.W.E., Collis, D.G., and Magar, K.M. 1972. Evaluation of the tree-beating

method for sampling defoliating forest insects. Can. Entomol. 104: 723-729. 10

Hebert, C, Jobin, L., Auger, M., and Dupont, A. 2003. Oviposition traps to survey

eggs of Lambdina fiscellaria (Lepidoptera: Geometridae). J. Econ. Entomol.

96(3): 768-776.

Hochuli, D.F. 2001. Insect herbivory and ontogeny: How do growth and development

influence feeding behaviour, morphology and host use? Austral Ecol. 26:

563-570.

Ide, J.-Y. 2006. Inter- and intra-shoot distributions of the ramie moth caterpillar,

Arete coerulea (Lepidoptera: Noctuidae), in ramie shrubs. Appl. Entomol.

Zool. 41(1): 49-55.

Ikeda, T., Matsumura, F., and Benjamin, D.M. 1977. Mechanism of feeding

discrimination between mature and juvenile foliage by two species of pine

sawflies. J. Chem. Ecol. 3(6): 677-694.

Jenkins, J.C., Aber, J.D.,and Canham, CD. 1999. Hemlock woolly adelgid impacts

on community structure and N cycling rates in eastern hemlock forests. Can. J.

For. Res. 29: 630-645.

Jennings, D. T., Diamond, J.B., and Watt, B.A. 1990. Population densities of spiders

(Araneae) and spruce budworms (Lepidoptera, Tortricidae) on foliage of

balsam fir and red spruce in east-central Maine. J. Arachnol. 18: 181-193.

Jensen, T. S. 1988. Variability of Norway spruce (Picea abies L.) needles;

performance of spruce sawflies {Gilpinia hercyniae Htg.). Oecologia, 77:

313-320.

Johns, R., Ostaff, D., and Quiring, D. 2006. Relationships between yellowheaded 11

spruce sawfly, Pikonema alaskensis, density and defoliation on juvenile black

spruce. For. Ecol. Manage. 228: 51-60.

Kendall, B. E., Ellner, S.P., McCauley, E., Wood, S.N., Briggs, C.J., Murdoch, W.H.,

and Turchin, P. 2005. Population cycles in the pine looper moth: dynamical

tests of mechanistic hypotheses. Ecol. Monogr. 75(2): 259-276.

Landry, J.-F., Roy, M., and Turcotte, C. 2002. Cranberry pests of Quebec: an

identification guide. Agri. and Agri-Food Can., Quebec.

Liebhold, A.M., Halverson, J.A., and Elmes, G.A. 1992. Gypsy moth invasion in

North America: A quantitative analysis. J. Biogeogr. 19(5): 513-520.

Liebhold, A.M., Macdonald, W.L., Bergdahl, D., and Mastro, V.C. 1995. Invasion by

exotic forest pests: A threat to forest ecosystems. For. Sci. Monogr. 30.

Lysyk, T. J. 1990. Relationships between spruce budworm (Lepidoptera: Tortricidae)

egg mass density and resultant defoliation of balsam fir and white spruce. Can.

Entomol. 122: 253-262.

McClure, M.S. 2002. Pest Alert: Hemlock Woolly Adelgid. U.S. Dept. Agri., For.

Serv., Northeastern Area.

McGuffin, W. C. 1977. Guide to the Geometridae of Canada (Lepidoptera) 2.

Subfamily . 2. Mem. Entomol. Soc. Can. 101.

Moreau, G., Quiring, D.T., Eveleigh, E.S., and Bauce, E. 2003. Advantages of a

mixed diet: feeding on several foliar age classes increases the performance of

a specialist insect herbivore. Oecologia, 135: 391-399.

Morrison, K.W. 2006. Ecological impacts of defoliation on forest birds in Kejimkujik 12

National Park and National Historic Site. B.Sc. Biology thesis, Dalhousie

University, Halifax, N.S.

Orwig, D. A., and Foster, D.R. 1998. Forest response to the introduced hemlock

woolly adelgid in southern New England, USA. J. Torr. Bot. Soc. 125(1): 60-

73.

Parsons, K., Quiring, D., Piene, H., and Farrell, J. 2003. Temporal patterns of balsam

fir sawfiy defoliation and growth loss in young balsam fir. For. Ecol. Manage.

184: 33^16.

Pimentel, D. 1986. Biological invasions of plants and in agriculture and

forestry. In Ecology of Biological Invasions of North America and Hawaii,

Ecological Studies. Edited by H.A. Mooney, J.A. Drake, Springer-Verlag,

New York. pp. 149-162.

Pinault, L., Georgeson, E., Guscott, R., Jameson, R., LeBlanc, M., McCarthy, C.,

Lucarotti, C, Thurston, G., and Quiring, D. 2007. Life history of Iridopsis

ephyraria, (Lepidoptera: Geometridae), a defoliator of eastern hemlock in

eastern Canada. J. Acad. Entomol. Soc. 3: 28-34.

Raske, A.G., West, R.J., and Retnakaran, A. 1995. Hemlock looper, Lambdina

fiscellaria. In Forest Insect Pests in Canada. Nat. Res. Can., Can. For. Serv.,

Ottawa, pp. 141-147.

Rindge, F.H. 1966. A revision of the moth genus (Lepidoptera,

Geometridae). Bull. Am. Mus. Nat. Hist. 132(3): 179-182.

Rogers, R.S. 1980. Hemlock stands from Wisconsin to Nova Scotia: Transitions in

understory composition along a floristic gradient. Ecology, 61(1): 178-193. 13

Rooney, T.P., and Waller, D.M. 2003. Direct and indirect effects of white-tailed deer

in forest ecosystems. For. Ecol. Manage. 181: 165-176.

Rose, A. H., and Lindquist, O.H. 1977. Insects of Eastern Spruces, Fir and Hemlock.

Dept. Env., Can. For. Serv., For. Tech. Rep. 23.

Simberloff, D. 1986. Introduced insects: A biogeographic and systematic perspective.

In Ecology of Biological Invasions of North America and Hawaii, Ecological

Studies. Edited by: H.A. Mooney, J.A. Drake, Springer-Verlag, New York, pp

3-26.

Suomela, J., Ossipov, V., and Haukioja, E. 1995. Variation among and within

mountain birch trees in foliage phenols, carbohydrates, and amino acids, and

in growth ofEpirrita autumnata larvae. J. Chem. Ecol. 21(10): 1421-1446.

Wallin, K. F., and Raffa, K.F. 1997. Association of within-tree jack pine budworm

feeding patterns with canopy level and within-needle variation of water,

nutrient, and monoterpene concentrations. Can. J. For. Res. 28: 228-233. 14

CHAPTER TWO

LIFE HISTORY OF IRIDOPSIS EPHYRARIA, (LEPIDOPTERA:

GEOMETRIDAE), A DEFOLIATOR OF EASTERN HEMLOCK IN EASTERN

CANADA1

2.1 INTRODUCTION

The pale-winged gray moth, Iridopsis ephyraria Wlk. (Lepidoptera:

Geometridae), recently caused high levels of defoliation of eastern hemlock, Tsuga canadensis (L.) Carr. (Pinaceae), in southwestern Nova Scotia. In 2002, severe defoliation began in a few isolated hemlock stands, and has since spread across the region (Fig. 1). There is no record in the refereed literature of a previous pale-winged gray moth outbreak. In addition, aside from systematic articles (e.g. Rindge 1966;

McGuffin 1977), and a government report (Landry et al. 2002), there is very little biological knowledge about this species.

Iridopsis ephyraria, (formerly referred to as Anacamptodes ephyraria), is widely distributed from Alberta to Nova Scotia, and as far south as Texas (McGuffin

1977; Ferguson 1954; Rindge 1966). Larvae of/, ephyraria have been described both as green (Ferguson 1985), and alternatively, brown to black with a herringbone pattern along their dorsum (McGuffin 1977). Iridopsis ephyraria adults have been well described in Rindge (1966) and McGuffin (1977), but immature stages have not been formally studied in the field. Similarly, although the months when larvae, pupae

1 Pinault, L., Georgeson, E., Guscott, R., Jameson, R., LeBlanc, M., McCarthy, C, Lucarotti, C, Thurston, G. and Quiring, D. 2007. Life history of Iridopsis ephyraria, (Lepidoptera: Geometridae), a defoliator of eastern hemlock in eastern Canada. J. Acad. Entomol. Soc. 3:28-34. 15 and adults were present has been reported previously (Landry et al. 2002), seasonal occurrence of different life stages had not been described in detail, nor the number of larval instars determined.

Iridopsis ephyraria larvae have been collected on a wide variety of trees and shrubs, both coniferous and deciduous, including hemlock (Schaffner & Griswold

1934), wild cherry, birch, (Ferguson 1954), maple, , ash, chokecherry, oak, gooseberry, balsam fir (Prentice 1963), arrow-weed, apple (McGuffm 1977) and cranberry (Landry et al. 2002). It is unlikely that the caterpillars feed extensively on all of these species, but this does suggest that feeding may occur on a wide range of hosts.

This study is intended to fill the gaps in the current descriptive knowledge to permit accurate field diagnosis of all I. ephyraria life stages and to improve understanding of the ecology of this occasional pest. We describe all juvenile life stages and their seasonal phenology. Performance of larvae on a variety of potential hosts is quantified and mortality of larvae and pupae is documented. We also present a successful method to rear pale-winged gray moth eggs.

2.2 METHODS

Description of Hemlock Stands

This study was carried out in hemlock-dominated (>60% of the upper canopy) stands in southwestern Nova Scotia, Canada, in and around Kejimkujik National Park, on trees whose diameter at breast height ranged from 30.7 - 49.1 cm. Common canopy species, in addition to eastern hemlock, included red maple, Acer rubrum (L.) 16

(Aceraceae), sugar maple, Acer saccharum (L.) (Aceraceae), red spruce, Picea rubens

(Sarg.) (Pinaceae), yellow birch, Betula alleghaniensis (Britt.) (Betulaceae), and white pine, Pinus strobus (L.) (Pinaceae). Immature trees in the understory included balsam fir, Abies balsamea (L.) P. Mill. (Pinaceae), red oak, Quercus rubra (L.)

(Fagaceae), American beech, Fagus grandifolia (Ehrh.) (Fagaceae), yellow birch, striped maple, Acerpennsylvanicum (L.) (Aceraceae), trembling aspen, Populus tremuloides (Michx.) (Salicaceae), and red maple. Sweet fern Comptonia peregrina

(L.) Coult. (Myricaceae), bracken fern, Pterinium aquilinum (L.) Kuhn.

(Hypolepidaceae), and mosses were also present in the understory.

Eggs

Description of Eggs and Oviposition Sites

Dissection of gravid females and observations of mated females in the laboratory and in the field suggested that /. ephyraria lays green eggs in deep bark crevices on the main bole of hemlock trees. Based on this information, eggs were obtained from both foam oviposition traps (Hebert et al. 2003) and bark samples from a private woodlot in South Brookfield, Nova Scotia, (N 44°23'45.6", W 65°00'17.1") in August 2005. To sample bark, several 10 x 10 cm squares of hemlock bark were cut to the depth of the phloem (< 2.5 cm), 1 m above the ground.

Bark samples were washed for eggs using the following procedure: 11 mL of

Triton X-100 was added to 5.0 L of water at 25°C in a large bucket, and lightly agitated and stirred every 10 minutes with bark samples, for 1 h. The liquid portion was poured through a #18 mesh sieve placed atop a #60 mesh sieve to filter out 17

particles too large or too small to be eggs. Bark pieces were rinsed again with water

above the sieve, to remove any remaining eggs. Materials collected by the #60 sieve

were poured onto dampened filter paper placed in a 15 cm Buchner funnel. The

funnel was mounted on a 2 L vacuum flask to drain off excess water, using a vacuum

pump drawing at 1.5 - 2.0 L/min. The filter paper was then placed under a microscope

to count the number of eggs per cm of bark surface.

We also obtained eggs from foam oviposition traps, in a manner similar to that

used for hemlock looper, Lambdina fiscellaria fiscellariaGuen . (Lepidoptera:

Geometridae) (Hebert et al. 2003). Foam oviposition traps consisted of white

polyurethane foam cut into 30 x 17 x 2.5 cm blocks. One foam trap was stapled to the bole of each of 30 hemlocks, located 10 m apart along a linear transect, just prior to

pupation in late July. We removed foam traps from the tree bole following the end of

adult activity, at the end of August, and examined foam traps under a microscope for

eggs.

Egg Rearing Protocol

In 2005, foam oviposition traps containing eggs were held outdoors throughout the winter to allow us to study emerging first-instar larvae the following

spring. In each foam trap, we indicated the position of viable-looking (i.e., green and rounded) eggs with a permanent marker. Foam traps were dipped in tap water, and permitted to air dry at 20°C for two days. Once dry, foam traps were placed in a nylon mesh bag outdoors in an area sheltered from snow, in Fredericton, N.B. In early

May, eggs were brought into the lab and removed from the foam traps. One hundred 18 forty-three viable-looking eggs were placed in groups of five eggs, (except 2 groups with six eggs, 1 with four eggs and 1 with two eggs), on damp filter paper, in 29 small plastic cups (240 mL, Solo). Eggs were lightly misted daily with water (only enough to slightly dampen the filter paper) until hatch occurred.

Vertical Egg Distribution in Tree Bole

To determine where eggs were laid in the bole of hemlock trees, four mature hemlock trees, 18.6 ± 1.8 m in height, were felled in South Brookfield. Strips of bark ranging from 90 cm2 to 945 cm2 were taken every 2.4 m along the length of the bole of each tree, beginning 1.3 m from the ground, and were washed for eggs, as described previously. Since the diameter of the tree bole is expected to decline sharply near the tree apex, no egg samples were taken <1.5 m from the top of the tree.

Egg densities were log-transformed to normalize the data, and sample heights were represented as a proportion of the tree height.

Larvae

Description of Larval Instars

Larval colouration was described and instars differentiated by head capsule widths. To collect first instar larvae just after egg hatch, in 2005 we placed a band of adhesive tape (5.1 cm wide, PheroTech Inc., sticky on both sides) around the lower bole of mature hemlock trees, thereby ensnaring first instars from the lower bole of the tree as they climbed up towards foliage in the crown. Adhesive bands were placed around 30 haphazardly selected mature hemlock trees 1.5 m from the ground, near the 19 group campground site in Kejimkujik National Park (N 44°24'00.4"; W 65°13'58.5") prior to egg hatch. A week following egg hatch, we removed first-instar larval corpses from the adhesive bands with forceps. In 2006, we examined first instars that emerged from overwintered eggs laid in oviposition traps from the previous year.

One hundred second-instar larvae were obtained from the field by beating hemlock tree foliage with a wooden dowel and collecting larvae on a 1 m2 beating sheet (Lucarotti et al. 1998). Individual larvae were placed in 13 x 7 x 7 cm transparent plastic cups with a fine nylon mesh bottom that were placed upside down on a Styrofoam cup base. Cups were placed in a field insectary that provided shelter from rain, but also allowed airflow with the outdoors, due to mesh walls on all sides.

A small branch of freshly cut hemlock foliage in water was placed in each cup.

Hemlock foliage was replaced with fresh foliage every 4-7 days, as needed. Head capsule widths were measured weekly (± 1 day) from 12 June until pupation

(approximately 21 July). In addition, groups of 40 larvae were collected weekly from the field using beating sheets, beginning 12 June, from the same site. Head capsule widths of collected larvae were measured to verify that insectary larvae were similar to those in the field and to determine the number of instars.

Larval Mortality

Larval mortality was estimated in 2006 during the declining phase of the current outbreak. One lower-crown hemlock branch, approximately at eye level, was selected and tagged in each of 10 trees at 3 sites. All trees were mature but located at the edge of stands. The number of larvae per branch was counted on 23 June, when most larvae were second instars, and 6 July, just prior to pupation. First instar larvae were too small to accurately count on hemlock foliage. It was assumed that immigration and emigration from the branch were approximately equal. The percent mortality of insectary-reared larvae was compared to field mortality rates, to see if the absence of parasitism or predation after the second instar (in the field insectary) increased survival.

Natural Enemies

In 2004 and 2005, groups of 100+ larvae were haphazardly collected every fourteen days from high population density sites in Kejimkujik National Park using a beating sheet. Larvae were reared on freshly cut hemlock foliage in groups of 10 - 30 larvae, in large mesh cages (40 x 25 x 15 cm) within the field insectary and observed daily for parasitoid emergence and fungal infection, until the end of pupation.

In 2006, 30 groups of five second-instar larvae were collected from hemlock and isolated in nylon mesh sleeve cages on hemlock branches at the same site. These were monitored until late in the third instar for fungal infection or parasitoid emergence. Field-collected third and fourth instars were reared similarly until pupation. Any parasitoids that emerged were captured and identified to species.

Potential Host Range

In early June, 2005, we observed early instar I. ephyraria feeding on red maple, white pine, sweet fern, balsam fir and red oak. To determine the suitability of these potential alternative host plants, 5 second-instar larvae collected from hemlock 21 were placed on each of 10 branches of each host (one branch per tree) and enclosed within nylon mesh sleeve cages in the field. When the majority of larvae were in the late third instar (after 10 days), cages were opened, and surviving larvae counted and removed. Five new late third or early fourth-instar larvae, collected from hemlock foliage, were then placed on the study branches and enclosed in sleeve cages until pupation (6 days). Survival was determined for early instars (i.e., second until late third/fourth instar), and for late instars (i.e., third/fourth instar until pupation). The influence of host plant and caterpillar age on survival was evaluated using a two-way

ANOVA(Zarl984).

Pupae

Description of Pupae

Descriptions of pupation and pupae were based on visual observations of I. ephyraria individuals developing in the field insectary in 2005. As pupation began, caterpillars were observed continuously and qualitative changes occurring during metamorphosis recorded.

Mortality of Pupae in Soil and Timing of Adult Emergence

To determine the rate of pupal mortality in the soil, we used soil-sifting methods to estimate the number of prepupae entering the ground, and adjacent emergence traps to measure the number of adults emerging from an adjacent area of soil. In 2005, one low-hanging branch on each of 10 trees at five sites in Kejimkujik

National Park was selected and tagged. As soon as all larvae had dropped from 22 branches, a 30 cm diameter circle of soil underneath the branch was dug to a depth of

11 cm, and carefully sifted for pupae. To estimate the number of emerging adults, one circular 30 cm diameter emergence trap was placed adjacent to the soil-sifted area beneath each branch, and the number of emerging counted and removed daily.

2.3 RESULTS

Eggs

Description of Eggs and Oviposition Sites

Iridopsis ephyraria eggs were easily identifiable as bright green with a reddish, flattened end, with regular rows of small, rounded swellings along their length, and were approximately 0.5 mm in width and 0.8 mm long (Fig. 2a). In the bark samples, eggs were regularly observed wedged into deep crevices, or underneath flaking pieces of lichen. Except for two cases where eggs were laid in pairs, eggs were laid singly on all bark samples and foam traps (n = 200).

Egg Rearing

Following overwintering, 87.3% (N = 182 eggs) of eggs in foam traps retained their original oval shape and were not deflated or desiccated. Larvae emerged from 70.8 ± 0.1% (n = 29 cups) of the 143 successfully transferred viable- looking eggs. Using this rearing method, the total survival rate from egg lay to larval emergence was 61.8%. No parasitoids or fungi were detected, and all egg mortality appeared to be due to desiccation. 23

Vertical Egg Distribution in Tree Bole

Egg density increased linearly from the bottom to the top of tree boles (Fig. 3).

The majority of eggs were laid on the bole within the live crown (108.94 ± 26.99 eggs/1000 cm2) rather than on the lower tree bole (17.43 ± 5.04 eggs/1000 cm2).

Larvae

Description of Larval Instars

The distribution of head capsule widths suggests that /. ephyraria passes through five instars, each with a non-overlapping range of head capsule widths (Fig.

4). The ranges for 1st, 2nd, 3rd, 4th, and 5th instars were 0.20 - 0.29 mm, 0.37 - 0.54 mm, 0.61 - 0.97 mm, 1.01 - 1.40 mm, and 1.68 - 2.20 mm, respectively. Only one head capsule width (1.55 mm) was not included in these five size ranges (Fig. 4).

Larvae were easily diagnosed by a rust-coloured head capsule with two prominent dorsal lobes. Larval bodies varied in colour from dark olive to a brilliant emerald green (Fig. 2b). First and second instars were often paler than older instars.

The dorsal herringbone pattern reported by McGuffin (1977) was present in only a few larvae, and then only very faintly.

Larvae developed approximately one week faster in the insectary (Fig. 5a) than in the field (Fig. 5b). Most instars were only present for 7-10 days, except for the

4th instars, which occurred for more than 2 weeks (Fig. 5b). There was substantial overlap in the timing of larval instars for individuals that developed in the insectary and in the field on almost all sample dates. At the end of larval development, larvae ceased to feed and became shortened, fattened, bright in colour, and glabrous (Fig. 2c). Prepupae subsequently dropped to the soil, burying themselves just under the surface to a maximum depth of < 10 cm, as determined by the soil-sifting method described previously.

Larval Mortality

Larval mortality varied significantly between the three study sites in 2006

(F2)29= 5.63; P < 0.01), and ranged between 35.9 ± 7.6% and 68.9 ± 7.1%. Larval mortality in the insectary in 2005 was 53%, which falls within this range.

Natural Enemies

A fungus, likely Entomophaga aulicae (Entomophthorales: Zygomycetes), was observed on late instar larvae in the insectary and in the field during 2004, 2005 and 2006. Larvae killed by the fungus appeared gray and rigid.

No parasitoids emerged from larvae during any of the three years of study. In the field, we occasionally observed predation of larvae by large arachnids, ants

(Hymenoptera: Formicidae) and wasps (various Hymentopera) early in the season, and by red squirrels (Tamiasciurus hudsonicus (Erxleben)) and birds later in the season.

Potential Host Range

Survival of early instars (71.4 ± 5.0%) was significantly greater than that of late instars (15.3 ± 3.4%) on alternative hosts (Fi)87 = 109.4; P < 0.001). Survival varied significantly among plant species (F4,g6 = 5.45; P < 0.002). Early instar 25 survival was high on all the potential hosts examined except balsam fir, where survival averaged 38% (Fig. 6). In contrast, late instars had the highest survival on sweet fern and eastern hemlock (Fig. 6), resulting in an interaction between plant and instar period (F^ = 2.76; P = 0.033). As no larvae survived on white pine, this species was removed from subsequent analyses.

Pupae

Description of Pupae

The transformation from fifth instar to pupa took less than one hour. Initially, pupae retained a greenish hue (Fig. 2d) before fading to reddish-brown. Pupal length ranged between 8.1 - 12.8 mm, with an average of 10.8 ± 0.2 mm (n = 44). Based on field observations and emergence traps, the duration of pupation in the soil was approximately 1-2 weeks.

Mortality of Pupae in Soil and Timing of Adult Emergence

The average mortality of pupae was 94.2 ± 2.9 % and did not differ between sites (F4,39= 1.23; P = 0.31). No parasitoids emerged from pupae in 2004 or 2005 but adult Pimpla pedalis Cress. (Hymenoptera: Ichneumonidae) emerged singly from approximately 6.2% (n = 73) of field-collected pupae in 2006.

Adults emerged between 27 July and 16 August 2005, about 2 weeks after the beginning of pupation (Fig. 5a). Casual field observations suggested that peak moth abundance occurred between 7 and 13 August 2005. 26

2.4 DISCUSSION

Pale-winged gray moth eggs were laid singly in deep bark crevices on hemlock, and females appear to oviposit preferentially in the upper bole. These behaviours may minimize egg predation (Warrington & Whittaker 1985) or mortality from fungal pathogens (Hajek 2001). Larvae emerging from eggs in the upper bole might also benefit from closer proximity to sun leaves (Fortin & Mauffette 2002; Ide

2006).

Given the large number of eggs collected from both foam traps and bark samples, both sampling techniques could potentially provide good estimates of/ ephyraria egg densities. Foam traps have already been used to sample hemlock looper eggs successfully (Hebert et al. 2003).

Although the descriptions of larval instars in the present study differed from those in previous reports, where body colour was substantially darker and a dorsal herringbone pattern was more apparent (McGuffin 1977; Landry et al. 2002), it is possible that these differences reflect a local phenotype. Head capsules of first and second instars had not been measured previously. Mean larval head capsule measurements reported in this chapter are consistent with the ranges for 4th and 5th instars and the single measurement value for the 3 rd instar presented in McGuffin

(1977), although our fourth instar ranges are broader. The broader range of head capsule widths of fourth instars observed in the present study might be due to feeding on different hosts (Bernays 1986), or localized differences in morphology. Pupae were morphologically similar to those described by McGuffin (1977). 27

Larvae developed one week more quickly in the field insectary than outdoors, most likely due to warmer ambient temperatures (Stamp & Bowers 1990) and a plentiful diet (Tammaru et al. 2004).

The fungal pathogen was present at all sites and appeared to kill a large number of larvae. Several years prior to this outbreak, E. aulicae infected the white- marked tussock moth, Orgyia leucostigma leucostigma (JE Smith) (Lepidoptera:

Lymantriidae) in the same area of Nova Scotia (van Frankenhuyzen et al. 2002), and

L. fiscellaria fiscellaria in New Brunswick (Lucarotti et al. 1998). Widespread infections in white-marked tussock moth larvae in Nova Scotia may have established a reservoir of E. aulicae in regional soils (Hajek 2001).

In contrast to high levels of parasitism in other caterpillars (Auerbach 1991;

Hawkins et al. 1997; Barbosa et al. 2001; Tanhuanpaa et al. 2001), parasitoids were not detected in this system until 2006, when parasitism rates were still very low. Low parasitism rates have been observed for other insects feeding on Tsuga, suggesting that this host plant may influence parasitoid behaviour or performance (Lill et al.

2002). Larval survival in the field was similar to that in the field insectary, suggesting that predation rates were also relatively low during this study.

Larvae survived well on all hosts except balsam fir and white pine, supporting the hypothesis that /. ephyraria has a wide host range (Prentice 1963; McGuffin

1977). The generalist diet of/, ephyraria is consistent with that of other geometrid larvae (Carroll 1956; Cuming 1961). Feeding on alternative hosts might allow /. ephyraria larvae to maximize a wider variety of nutrients, minimize hemlock toxins, and ensure adequate nitrogen uptake (e.g. Barbosa & Greenblatt 1979; Stockhoff 28

1993). Eastern hemlock has more canopy nitrogen and a lower nitrogen mineralization rate than other conifers (Ollinger et al. 2002), making it more nutritious for caterpillars (White 1984), potentially leading to the observed lower survival rates on balsam fir and white pine. The majority of larvae observed in the field remained on hemlock foliage for the duration of their larval period, and it is not known if larvae switch hosts or whether individual larvae could specialize on certain hosts. It should be noted that larvae for this study were obtained from hemlock foliage and might have been pre-adapted to feed on hemlock through previous experience (Jermy et al. 1968; deBoer & Hanson 1984).

Pupal mortality in the soil was very high and cannot be explained by the low level of pupal parasitism. Pupal mortality might be due to predation by insects (Frank

1967) or small mammals (Port & Thompson 1980; Valenti et al. 1998; Tanhuanpaa et al. 1999), as previously reported for other moths pupating in the soil. Some pupae may have also died from previous fungal infection.

Although outbreaks of I. ephyraria have not been previously reported, it has similar habits to a frequently outbreaking geometrid, the hemlock looper. The hemlock looper also overwinters as an egg, is univoltine (Raske et al. 1995), eats current-year as well as old foliage, is capable of generalist feeding, has a patchy feeding distribution (Carroll 1956), and is a wasteful feeder, not consuming the entire needle (Carroll 1956; Raske et al. 1995). As a result, it might be possible to effectively use some of the ecological knowledge and management tools already developed for the hemlock looper, such as foam oviposition traps (Hebert et al. 2003), to manage this new pest. Despite these similarities, the hemlock looper differs from I. ephyraria in several respects, such as the hemlock loopers ability to occasionally lay eggs in clusters (Carroll 1956) and to defoliate more hosts at moderate densities

(Rose & Lindquist 1977). Consequently, a complete understanding of the life history of the pale-winged gray moth and of the factors influencing its dynamics is indispensable for the establishment of a successful pest management program. 30

••TBI ~j, « *•« r—;

Fig. 1. Map of the distribution of hemlock defoliation by /. ephyraria detected during aerial surveys between 2002-2006. The extent of defoliation in the first three years of the outbreak are shaded as follows: 2002 yellow; 2003 pink; and 2004 purple. Defoliated sites in 2005 are indicated by dark blue circles and the extent is outlined by a thick black line. Defoliated sites in 2006 are indicated by red circles and undefoliated sites are indicated by green circles. 31

w

y.^' mm

a) b)

iJCOjU 1.0 cm

Ik ,3

Fig. 2. Photographs of the life stages of /. ephyraria: (a) an egg, a few weeks after oviposition; (b) two late-instar larvae; (c) a prepupa; (d) a pepa; and (e) an adult moth. * - •

y = 1.82x + 0.44 R2 = 0,45

0.0 i.2 0.4 0.6 0.8 1.0

Proportion of the distance up tree bole

Fig. 3. Relationship between the vertical height up the bole of four mature hemlock trees, expressed as the proportion of the total height, and the log-transformed density of I. ephyraria eggs from bark samples. 33

70

0.25 60 A

ifi 50 CD "D 1 40 1.23 30 E 0.44 3 0.80 20 1.97

10 A

I IH IIIII •I i ••LIUJIJUII il • i U ill ft I liui • l;ll •••••, 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 2.20 Head capsule size (mm)

Fig. 4. Frequency distribution of head capsule widths of /. ephyraria larvae from field collections, insectary rearings and first-instar larvae that emerged from overwintered eggs in the lab. Means of the widths for each instar are indicated above the bars (SE < 0.01 for all means), n = 125, 96, 76, 104, 118 for 1st, 2nd, 3rd, 4th, 5th instar larvae, respectively. 34

a)

00 - ....

\ 3rd Instar \ — - 4th Instar V 5th Instar 80 -I \ Pupae CD CO

CO 60- \ x

CD AV \ / 'X\ / -"^ \ 40 - A y \ / \ , \ A \ / \ / 20 - / \ / \ v /\ X

0 - 1 A- ^V * f-^- 1 12 June 19 June 26 June 4 July 11 July 18 July Date b)

... i.

3rd Instar 50 - 4th Instar 5th Instar

CD 5 40- " \ co \ • •""'•*». / _J »+— \ / "^ ' O 30- v t_ \ / / CD \ \ / E \ / >'' § 20- \ •••.. / x Z X '-< '\ 10 - // \ > \\ / \ 7 \ .' X / 0- ,-••'/ y . x. 12June 19June 26June 4 July 11 July 18 Juty Date

Fig. 5. The temporal distribution of I. ephyraria instars during summer 2005. The spline function of SigmaPlot (2001, Version 7.0) was used to estimate densities between the six sample dates. Distributions of the number of larvae in each instar are shown for: a) insectary-reared larvae, (n = 108, 69, 50, 39, 77, 38 for 1st, 2nd, 3rd, 4th, 5th instar larvae and pupae, respectively); and b) field-collected larvae (n = 76, 27, 26, 67, 44 for 1st, 2nd, 3rd, 4th, 5th instar larvae, respectively). 35

Early Instar (2nd - 3/4th instar) Late Instar (3/4th - pupation)

100

ITS

CO

5

c 0) o 0) mc

c> A^ ^e* v^°

Plant Species

Fig. 6. Mean (± SE) percent survival of early and late instar larvae of I. ephyraria on eastern hemlock and five potential alternative host species in South Brookfield in 2006. 36

REFERENCES

Auerbach, M. 1991. Relative impact of interactions within and between trophic levels

during an insect outbreak. Ecology, 72: 1599-1608.

Barbosa, P., and Greenblatt, J. 1979. Suitability, digestibility and assimilation of

various host plants of the gypsy moth, Lymantria dispar L. Oecologia, 43:

111-119.

Barbosa, P., Segarra, A.E., Gross, P., Caldas, A., Ahlstrom, K., Carlson, R.W.,

Ferguson, D.C., Grissell, E., Hodges, R.W., Marsh, P.M., Poole, R.W.,

Schauff, M.E., Shaw, S.R., Whitfield, J.B., and Woodley, N.E. 2001.

Differential parasitism of macrolepidopteran herbivores on two deciduous tree

species. Ecology, 82(3): 698-704.

Bernays, E. A. 1986. Diet-induced head allometry among foliage-chewing insects and

its importance for graminivores. Science, 231: 495^197.

Carroll, W.J. 1956. History of the hemlock looper, Lambdina fiscellaria fiscellaria,

(Guen.) (Lepidoptera: Geometridae) in Newfoundland, and notes on its

biology. Can. Entomol. 88: 587-599.

Cuming, F.G. 1961. The distribution, life history, and economic importance of the

winter moth, Operophtera brumata (L.) (Lepidoptera, Geometridae) in Nova

Scotia. Can. Entomol. 93: 135-142. deBoer, G., and Hanson, F.E. 1984. Foodplant selection and induction of feeding

preferences among host and non-host plants in larvae of the tobacco

hornworm. Entomol. Exp. Appl. 35: 177-193.

Ferguson, D.C. 1954. The Lepidoptera of Nova Scotia 1. Macrolepidoptera. Bull. N.S. 37

Inst. Sci. 123(3).

Ferguson, D.C. 1985. Insects of eastern forests, U.S. Dept. Agric. Misc. Publ. 1426.

Fortin, M., and Mauffette, Y. 2002. The suitability of leaves from different canopy

layers for a generalist herbivore (Lepidoptera: Lasiocampidae) foraging on

sugar maple. Can. J. For. Res. 32: 379-389.

Frank, J. H. 1967. The insect predators of the pupal stage of the winter moth,

Operophtera brumata (L.) (Lepidoptera: Hydriomenidae). J. Anim. Ecol.

36(2): 375-389.

Hajek, A. E. 2001. Larval behaviour in Lymantria dispar increases risk of fungal

infection. Oecologia, 126: 285-291.

Hawkins, B. A., Cornell, H.V., and Hochberg, M.E. 1997. Predators, parasitoids, and

pathogens as mortality agents in phytophagous insect populations. Ecology,

78: 2145-2152.

Hebert, C, Jobin, L., Auger, M., and Dupont, A. 2003. Oviposition traps to survey

eggs of Lambdina fiscellaria (Lepidoptera: Geometridae). J. Econ. Entomol.

96(3): 768-776.

Ide, J.-Y. 2006. Inter- and intra-shoot distributions of the ramie moth caterpillar,

Arete coerulea (Lepidoptera: Noctuidae), in ramie shrubs. Appl. Entomol.

Zool. 41:49-55.

Jermy, T, Hanson, F.E., and Dethier, V.G. 1968. Induction of specific food preference

in Lepidopterous larvae. Entomol. Exp. Appl. 11: 211-230.

Landry, J.-F., Roy, M, and Turcotte, C. 2002. Cranberry pests of Quebec: an

identification guide. Agri. and Agri-Food Can., Quebec. 38

Lill, J. T., Marquis, R.J., and Ricklefs, R.E. 2002. Host plants influence parasitism of

forest caterpillars. Nature, 417: 170-173.

Lucarotti, C.J., Leclerc, T.L. and Clopton, R.E. 1998. Incidence of Leidyana canadensis

(Apicomplexa: Eugregarinida) in Lambdina fiscellaria fiscellaria larvae (Lepidoptera:

Geometridae). Can. Entomol. 130: 583-594.

McGuffin, W. C. 1977. Guide to the Geometridae of Canada (Lepidoptera) 2.

Subfamily Ennominae. 2. Mem. Entomol. Soc. Can. 101.

Ollinger, S. V., Smith, M.L., Martin, M.E., Hallett, R.A., Goodale, C.L., and Aber,

J.D. 2002. Regional variation in foliar chemistry and N cycling among forests

of diverse history and composition. Ecology, 83: 339-355.

Port, G. R. and Thompson, J.R. 1980. Outbreaks of insect herbivores on plants along

motorways in the United Kingdom. J. Appl. Ecol. 17(3): 649-656.

Prentice, R. M. 1963. Forest Lepidoptera of Canada. Vol. 3. Lasiocampidae,

Drepanidae, Thyatiridae, Geometridae. Bull. 1034., Dept. For. Can., Ottawa.

Raske, A.G., West, R.J., and Retnakaran, A. 1995. Hemlock looper, Lambdina

fiscellaria. In Forest insect pests in Canada. Edited by J.A. Armstrong and

W.G.H. Ives. Nat. Res. Coun. Can., Can. For. Serv., Ottawa.

Rindge, F. H. 1966. A revision of the moth genus Anacamptodes (Lepidoptera,

Geometridae). Bull. Am. Mus. Nat. Hist. 132(3): 179-182.

Rose, A.H., and Lindquist, O.H. 1977. Insects of eastern spruces, fir and hemlock.

Nat. Res. Coun. Can., Can. For. Serv., Ottawa.

Schaffher, J.V., and Griswold, C.L. 1934. Macrolepidoptera and their parasites reared from field collections in the northeastern part of the United States. U.S. Dept.

Agri., Misc. Publ. 188, Washington, D.C.

Stamp, N. E., and Bowers, M.D. 1990. Variation in food quality and temperature

constrain foraging of gregarious caterpillars. Ecology, 71: 1031-1039.

Stockhoff, B. A. 1993. Diet heterogeneity: implications for growth of a generalist

herbivore, the gypsy moth. Ecology, 74: 1939-1949.

Tammaru, T., Nylin, S., Ruohomaki, K., and Gotthard, K. 2004. Compensatory

responses in lepidopteran larvae: a test of growth rate maximization. Oikos,

107:352-362.

Tanhuanpaa, M., Ruohomaki, K., Kaitaniemi, P., and Klemola, T. 1999. Different

impact of pupal predation on populations of Epirrita autumnata (Lepidoptera;

Geometridae) within and outside the outbreak range. J. Anim. Ecol. 68: 562-

570.

Tanhuanpaa, M., Ruohomaki, K., and Uusipaikka, E. 2001. High larval predation rate

in non-outbreaking populations of a Geometrid moth. Ecology, 82: 281-289.

Valenti, M. A., Berryman, A.A., and Ferrell, G.T. 1998. Natural enemy effects on the

survival oiSynaxis cervinaria (Lepidoptera: Geometridae). Environ. Entomol.

27:305-311. van Frankenhuyzen, K., Ebling, P., Thurston, G., Lucarotti, C, Royama, T., Guscott,

R., Georgeson, E., and Silver, J. 2002. Incidence and impact of Entomophaga

aulicae (Zygomycetes: Entomophthorales) and a nucleopolyhedrovirus in an

outbreak of the whitemarked tussock moth (Lepidotera: Lymantriidae). Can.

Entomol. 134: 825-845. Warrington, S., and Whittaker, J.B. 1985. An experimental field study of different

levels of insect herbivory induced by Formica rufa predation on sycamore

{Acerpseudoplanatus) I. Lepidopteran larvae. J. Appl. Ecol. 22: 775-785.

White, T.C.R. 1984. The abundance of invertebrate herbivores in relation to

availability of nitrogen in stressed food plants. Oecologia, 63: 90-105.

Zar, J.H. 1984. Biostatistical Analysis, 2nd Edition. Prentice-Hall Inc. Toronto. 41

CHAPTER THREE

SAMPLING STRATEGIES AND DENSITY-DEFOLIATION RELATIONSHIPS

FOR THE PALE-WINGED GRAY MOTH, A DEFOLIATOR OF MATURE

EASTERN HEMLOCK2

3.1 INTRODUCTION

The pale-winged gray moth, Iridopsis ephyraria Wlk. (Lepidoptera:

Geometridae) is an indigenous defoliator found throughout eastern and central North

America (McGuffin 1977). It has never been reported as a major pest, although it occasionally causes minor damage on cranberry, Vaccinium macrocarpon L. (Landry et al. 2002). However, since 2002 it has caused high levels of defoliation on juvenile and mature eastern hemlock trees, Tsuga canadensis (L.) Carr., in southern Nova

Scotia, Canada. Up to 90% of the understory and 40% of mature hemlocks have died

(J. Kershaw, UNB, pers. comm.). This high level of mortality is largely due to the feeding habits of larvae, which can consume all age classes of foliage in a single year

(Chapter Two). Unacceptably high densities of/, ephyraria can be suppressed with

Bacillus thuringiensis (Thurston 2005), but we currently lack sampling techniques and egg (or larval) density - subsequent defoliation relationships. These would be needed to determine when the application of suppression tactics would be warranted.

Here, we describe studies carried out; i) to develop sampling methods that estimate egg and larval densities of/, ephyraria, and ii) to determine whether the

2 Pinault, L.L. and Quiring, D.T. 2007. Sampling strategies and density-defoliation relationships for the pale-winged gray moth, a defoliator of mature eastern hemlock. Submitted August 2007 to Forest Ecology and Management. 42 density estimates of any of the sampling methods could be used to predict subsequent defoliation by this caterpillar. Sampling I. ephyraria was challenging because the affected mature hemlock trees were too large to use pole pruners (i.e., mid-crown branches were usually 20 m above the ground) and most eggs were laid in deep crevices on the tree bole (Chapter Two). Although I. ephyraria eggs can be extracted from the bark (Chapter Two), the process is very time consuming and not practical for monitoring a large number of sites.

We evaluated three different sampling methods. White polyurethane foam oviposition traps, which have been used successfully to estimate hemlock looper egg densities on balsam fir (Hebert et al. 2003), were placed on host trees to estimate egg densities of I. ephyraria. Assuming that first-instar larvae emerging from eggs laid on the lower bole would climb upwards to the foliated crown immediately following egg hatch, we also used bands of "sticky tape" placed around the lower bole of host trees, to estimate first instar densities. Third and fourth-instar larval densities were estimated using a beating sheet method modified from Harris et al. (1972). Although both foam oviposition traps and sticky bands could potentially estimate insect densities on very large and on juvenile trees, the beating sheet method could only be evaluated on juvenile trees or on mature trees with low-hanging foliage, since it requires direct access to foliage.

The utility of each method to predict subsequent defoliation was evaluated for sites with a range of insect densities. Density - defoliation relationships have been established for a variety of economically important forest pests, including Zeiraphera canadensis Mut. & Free. (Lepidoptera: Tortricidae) on white spruce (Picea glauca (Moench) Voss) (Carroll & Quiring 1993), Pikonema alaskensis (Ron.)

(Hymenoptera: Tenthredinidae) on black spruce (Picea mariana (Mill.)) (Johns et al.

2006), Lymantria dispar L. (Lepidoptera: Lymantriidae) on a variety of hardwood and softwood hosts (Liebhold et al. 1993), and Choristoneura fumiferana Clem.

(Lysyk 1990) and Neodiprion abietis Har. on balsam fir {Abies balsamea (L.) Mill.)

(Parsons et al. 2005). We predicted that estimates of insect densities provided by all three methods would be strongly related to subsequent defoliation of mature hemlocks, and that the beating sheet insect density-defoliation relationship would be the strongest, since sampling occurs during peak feeding.

3.2 METHODS

Description of Study Insect

The pale-winged gray moth is a defoliator of a wide variety of hosts, including eastern hemlock (Forbes 1948), sugar maple {Acer saccharum (Marsh.)), and balsam fir (Prentice 1963). Adult pale-winged gray moths are well described as

Anacamptodes ephyraria in Forbes (1948), Rindge (1966) and McGuffin (1977).

Oval, bright green eggs are laid singly deep in bark crevices and hatch in late May or early June. Green larvae have characteristic rusty-coloured head capsules, pass through five instars, and feed on both the understory and mature hemlock (Chapter

Two). Pupae are present in the soil in mid-July for about 1-2 weeks. Adults emerge in late July or early August and are active for one to two weeks (Chapter Two). Study Sites

From 2004 through 2006, a total of 20 hemlock-dominated sites were sampled during a population decline in southern Nova Scotia, Canada. At each of the study sites and years (Table 1), 30 mature (height > 10 m) hemlock trees were selected and tagged approximately every 5 m along a linear transect, and tree diameter at breast height (dbh) and defoliation (methods described below) were measured (Table 1). At some sites, trees were selected along 2-3 parallel transects, due to a small stand size.

Up to three additional parallel sets of transects, for a total of 30 trees, were also set up at each site to measure first instar densities on sticky tape, mid-instar densities on beating sheets, and egg densities in foam traps (methods described below). The mean dbh of study hemlock trees was similar at all sites, but previous defoliation levels were variable (Table 1).

Site-level Defoliation

Percent defoliation of mature hemlock trees was visually estimated using 10% defoliation classes, as in Parsons et al. (2005). To improve accuracy, observers were trained using manually defoliated branches, and observers carried a reference photographic guide of each defoliation class.

Defoliation was estimated separately for each age class of foliage along the main axis of one haphazardly selected mid-crown branch in each tree. The difference between defoliation estimates taken prior to egg hatch (late May) and following pupation (late July) provided an estimate of net seasonal defoliation. The mean net defoliation estimates were determined for different combinations of foliar age class, 45 to produce three defoliation estimates for each site: mean current-year defoliation, mean defoliation of current year and one-year old shoots, and mean branch defoliation (i.e., mean of defoliation on current year to 6-year old foliage).

Sticky Tape for First-Instar Larvae

Prior to egg hatch in early June, strips of double-sided yellow adhesive tape

(PheroTech Inc.) were placed 1.5 m above the ground around the bole of 30 selected trees located along transects in each site (Table 1). Larvae that emerged from eggs that hatched below this height crawled up, presumably in search of foliage, and became trapped on the outer and inner surfaces of the sticky tape. Sticky tape was collected once egg hatch ceased. The number of head capsules was counted on both the inner and outer surfaces and divided by the inner and outer surface area of the sticky tape to obtain a first instar larval density for the tree.

The potential use of sticky tape on small, understory trees (<5 m) to sample first instar larvae was also evaluated in 2005. At five of the study sites, (Sites 2, 5, 7,

9, 10 in Table 1), 30 small hemlock trees were selected along a linear transect approximately 15 m from the transect of mature study trees used in the study described above. Sticky tapes were wrapped around the bole of each tree at a height of 1.5 m and mean larval densities determined as in the mature tree study described above. 46

Foam Oviposition Traps

To determine whether females preferred to lay eggs in foam traps rather than natural substrates, we offered females a choice of oviposition substrates in the laboratory. A recently emerged (i.e., <24h old) virgin female was placed in each of 39 mesh arenas with two recently emerged males in the lab at 25°C and 16L:8D, and offered water. A piece of mature hemlock bark, a hemlock branch with foliage, and a foam trap were placed in each arena. Following female death, eggs were counted on all three oviposition substrates.

The ability of foam traps to estimate egg density and predict subsequent defoliation was tested at 8 sites (Table 1). Prior to pupation in mid July 2005, white polyurethane foam (mean pore size = 0.53 mm) was cut into 2.5 x 17 x 14 cm blocks, as described by Hebert et al. (2003), and stapled to the tree bole 1.5m from the ground on 30 trees at each site. The white colour of the foam presumably mimics lichens commonly coating hemlock boles in this area, although the foam discoloured to amber after two weeks. Foam traps were placed facing whatever cardinal direction minimized trap visibility to hikers and visitors. Cardinal direction did not influence oviposition site selection during behavioural observations in the field (L. Pinault, pers. obs.). At the end of the adult flight period, foam traps were collected and examined under a dissecting microscope at 40x magnification to determine egg density.

Flattened or desiccated eggs, which accounted for <10% of eggs, were presumed dead and were not included in analyses relating egg density to subsequent defoliation. 47

Beating Sheet Sampling for Late-Instar Larvae

At each of the beating sheet sampling sites (Table 1), 30 (2005) or 40 (2006) trees with low-hanging foliage were selected along a single linear transect. Beating sheet sampling was conducted once in early July, when most larvae were in the third instar, on dry, sunny days. At each tree, a lm2 sheet of white cotton stretched between two crossed wooden dowels was placed beneath a branch large enough to almost cover the entire sheet, and the branch was struck firmly three times with another wooden dowel. All I. ephyraria larvae that fell or silked down onto the beating sheet were counted.

Statistical Analyses

Linear regression was used to evaluate the relationship between the density of

/. ephyraria in each stand, estimated using the three sampling methods, and each of the three estimates of defoliation obtained later in the same year. A two-way analysis of variance was used to evaluate the influence of tree size (i.e., understory or dominant) and site on the density of first instars in 2005 (Zar 1984). A Student's T- test was used to determine whether females in the lab study laid more eggs on foam than on bark.

3.3 RESULTS

Sticky Tape for First-Instar Larvae

First instar densities on sticky tape were strongly related to subsequent estimates of defoliation, regardless of whether defoliation was only measured on 48 current-year shoots (Fig. 7 a), current-year and one-year-old shoots (Fig. 7b), or on the entire branch (current year to 6-year old foliage) (Fig. 7c). The relationships between defoliation on multiple foliage age classes (Figs. 7b, c) and first instar density depend heavily on measurements taken in 2004. The three sites used in 2004 were the only sites where previous defoliation was not measured, were therefore not included in defoliation estimates. If 2004 sites are disregarded, the regressions are still significant for defoliation of current-year shoots (r2 = 0.55, P<0.01) and for current-year and one- year-old shoots (r2 = 0.09, P<0.05), but not for current-year to six-year-old shoots (r2

= 0.04, P = 0.16). Larval densities did not differ significantly between large and small trees within a site (FM= 2.05, P = 0.22).

Foam Oviposition Traps

In the laboratory-based oviposition substrate selection study, 94.88 ± 2.90% of eggs were laid on foam, rather than hemlock bark (t = 46.46, P<0.001). No eggs were laid on hemlock branches. Despite the preference for foam as an oviposition substrate, egg densities in foam traps were only weakly and non-significantly related to all three estimates of defoliation means, (i.e., current year shoots (r2 = 0.23, P =

0.13), current and 1-year old shoots (r2 = 0.20, P = 0.17), current year to 6-year-old shoots (r2 = 0.003, P = 0.88)).

Beating Sheet Sampling for Late-Instar Larvae

Larval densities obtained using beating sheets were significantly related to defoliation on current-year shoots (Fig. 8), but very weakly and non-significantly related to defoliation on both current-year and one year old shoots (r = 0.06, P =

0.29), or on the entire branch (r2 = 0.09, P = 0.20).

3.4 DISCUSSION

Estimates of first instar larval density, obtained using sticky tape traps, were very strongly related to all three site-level defoliation indices and therefore sticky tape provides a robust sampling method to predict future defoliation by /. ephyraria.

Because sticky tape provides estimates of first instar density, and aerial application of

B. thuringiensis to suppress /. ephyraria would usually be directed at mid-instar larvae, this sampling method requires pest managers to determine whether they wish to apply a suppression tactic within a relatively short period of time (2-3 weeks).

However, unlike the density of eggs in foam oviposition traps, the density of first instars on sticky tape can be determined in the field. Sticky tape can also be used to detect the timing of egg hatch in the field.

Larval densities on sticky tape placed around understory tree boles were similar to those on adjacent large trees. Since eggs are deposited in the deep bark crevices present on mature hemlocks (Chapter Two), any larvae found on sticky trap tapes on small trees had likely dispersed to those trees via wind, and as a result, are perhaps not directly comparable to the sticky tape larval densities on mature trees.

Egg densities on foam traps provided poor predictions of subsequent defoliation. The inability of egg density estimates to predict defoliation might be due to variations in mortality rates among sites. During the time lag between egg density and defoliation estimations, natural enemies such as predation might have affected newly emerged larvae more severely at some sites than at others. The poor relationship between egg densities and subsequent defoliation might also be attributable to female oviposition preference. Foam traps were the preferred oviposition substrate in the laboratory, and therefore likely overestimate true egg densities on bark. Despite this, foam traps might be a useful and less harmful alternative to bark sampling to determine the presence or absence of I. ephyraria in an area, or to qualitatively estimate general levels of insect abundance.

Three limitations of the beating sheet method may explain the weak larval density-defoliation relationship obtained. First, beating sheets only sample the bottom part of the lower crown of trees, which might not have been representative of feeding throughout the rest of the tree crown. Heterogeneous feeding within a tree can result in uneven defoliation patterns, (e.g., Carroll & Quiring 1994, Wallin & Raffa 1997,

Moreau et al. 2003, Yamasaki & Kikuzawa 2003). Second, branches in the lower crown were not accessible on most mature trees, and thus most samples were carried out on smaller understory trees. Finally, sampling using beating sheets could only be carried out on dry, sunny days (Harris et al. 1972), which greatly limited sampling frequency and duration. While beating sheet sampling was useful to detect the presence or absence of/, ephyraria larvae on understory trees within a limited area, it only explained a significant proportion of the variation in defoliation on current-year shoots.

In summary, foam oviposition traps and beating sheets are useful to provide qualitative estimates of/, ephyraria density. Sticky tape could subsequently be used to estimate site-specific defoliation more precisely. 51

Table 1. Details of the study sites and years, where F = Foam trap, S = Sticky tape, and B = Beating sheet sampling (indicated with [*]). Site coordinates, previous defoliation (mean (±SE) percent defoliation of one-year-old shoot, in May), and mean (± SE) hemlock diameter at breast height (DBH) are presented.

Site Latitude Longitude Year F S B Prev defol DBH (cm)

1 N 44°24'50.7" W65°14'26.8" 2004 * * - 35.3±0.98 2 N44°23'55.6" W65°13'07.6" 2004 * * - 31.1 ±0.78 2005 * * 51.3 + 6.02 2006 * * 45.3 + 3.44 3 N 44°23*56.9" W65°12'58.8" 2004 * * 36.6 ±0.91 2005 * * 67.3 + 6.36 2006 * 32.1+6.88 4 N 44°27'46.7" W65°16'44.5" 2005 * * * 13.7 + 5.17 17.5 ±0.38 2006 * * 14.0 + 5.42 5 N44°26'09.7" W65°12'54.8" 2005 * * * 38.7 + 7.25 30.7 ±1.31 2006 * * 34.7 ±3.80 6 N44°24'14.4" W65°15'10.1" 2006 * * 36.9 + 4.11 26.9 ±1.26 7 N 44°24'00.4" W65°13'58.5" 2005 * * * 69.7 + 5.39 32.8 ±1.19 2006 * * 31.3 + 3.60 8 N 44°23'58.0" W65°13'45.5" 2006 * 41.1 + 3.41 29.1+0.71 9 N 44°22'22.6" W65°12'24.1" 2005 * * * 72.3 + 6.17 24.8 ±0.69 2006 * * 52.3 + 7.65 10 N44°17'56.8" W65°14'14.7" 2005 * * 20.3 + 5.94 24.0 ± 0.97 2006 * 17.3 + 5.36 11 N44°28'55.0" W65°12'59.5" 2006 * * 5.7 + 3.06 29.3+1.19 12 N44°23'23.1" W64°59'10.9" 2006 * * 41.7 + 7.66 28.5+0.76 13 N44°23'45.6" W65°00'17.1" 2006 * * 41.7 + 6.95 37.9 ±1.83 14 N44°22'10.3" W 65°07'26.8" 2006 * 26.0 + 6.62 35.0 ±1.73 15 N44°23'06.1" W 65°07'54.5" 2006 * 30.0 ±6.41 31.2 + 2.15 16 N44°24'27.5" W 65°05'40.7" 2006 * * 49.0 ±7.77 38.9±2.17 17 N44°15'42.9" W64°50'02.1" 2006 * * 8.0 ±3.40 35.8+ 1.60 18 N44°20'10.6" W 64°49'38.6" 2006 * * 2.7+ 1.06 26.1 + 1.34 19 N 44°23'02.8" W 64°52'29.3" 2006 * * 37.3 ±6.69 40.1 ±2.16 20 N 44°27'46.5" W65°02'35.3" 2006 * * 37.3 + 6.83 34.2 + 1.40 52

too- AA , , Current-year * y (a)

so- • / so- • „/ ,/* * • • . */•/ 40-

m 20- y=log25;27x+91.14 A * 2 • r =0.77

0- 10

100- Current-year and 1-year-oid A* - * * d*- 20- y=tog24.04x+81.76 r2=0.85 0- 0.001 0.01 0.1 10

1O0H Current-year to 6-year-old (C)

0.001 Larval density (per cm2]

Fig. 7. Relationship between mean density per site of first instar /. ephyraria larvae (log scale) sampled from sticky tape on mature hemlock trees for 2004 (A), 2005 (•) and 2006 (•), and the mean amount of seasonal defoliation per site on (a) current year foliage, (b) current-year and one-year-old foliage, and (c) all age classes (i.e., current- year to 6 year-old) foliage. All regressions are significant (P<0.01). 53

100

80

60

• A 40

20 y=1.33x+26.97 r2=0.35

—i 10 15 20 25 30

Larval density (per cmz)

Fig. 8. Relationship between mean density of I. ephyraria larvae collected in beating sheets below young hemlock trees and the mean percentage of current-year foliage that /. ephyraria larvae consumed by the end of the summer in 2005 (A) and 2006 (•), at 15 sites in southern Nova Scotia. This regression is significant (PO.01). 54

REFERENCES

Carroll, A.L., and Quiring, D.T. 1993. Influence of feeding by Zeiraphera canadensis

(Lepidoptera: Tortricidae) on growth of white spruce: Larval density-damage

and damage-shoot production relationships. J. Appl. Ecol. 30: 629-639.

Carroll, A. L., and Quiring, D.T. 1994. Intratree variation in foliage development

influences the foraging strategy of a caterpillar. Ecology, 75(7): 1978-1990.

Forbes, W.T.M. 1948. Lepidoptera of New York and neighboring states: Part II:

Geometridae, Sphingidae, Notodontidae, Lymantriidae. Cornell Uni. Agri.

Exp. Station. Mem. 274.

Harris, J.W.E., Collis, D.G., and Magar, K.M. 1972. Evaluation of the tree-beating

method for sampling defoliating forest insects. Can. Entomol. 104: 723-729.

Hebert, C, Jobin, L., Auger, M., and Dupont, A. 2003. Oviposition traps to survey

eggs of Lambdina fiscellaria (Lepidoptera: Geometridae). For. Entomol. 96(3):

768-776.

Johns, R., Ostaff, D., and Quiring, D. 2006. Relationships between yellowheaded

spruce sawfly, Pikonema alaskensis, density and defoliation on juvenile black

spruce. For. Ecol. Manage. 228: 51-60.

Landry, J.-F., Roy, M., and Turcotte, C. 2002. Cranberry pests of Quebec: an

identification guide. Agri. and Agri-Food Can., Quebec.

Liebhold, A.M., Simons, E.E., Sior, A., and Unger, J.D. 1993. Forecasting defoliation

caused by the gypsy moth from field measurements. Env. Entomol. 22(1): 26-

32.

Lysyk, T.J. 1990. Relationships between spruce budworm (Lepidoptera: Tortricidae) 55

egg mass density and resultant defoliation of balsam fir and white spruce. Can.

Entomol. 122: 253-262.

McGuffin, W.C. 1977. Guide to the Geometridae of Canada (Lepidoptera) II.

Subfamily Ennominae. Mem. Entomol. Soc. Can. 101.

Moreau, G., Quiring, D.T., Eveleigh, E.S., and Bauce, E. 2003. Advantages of a

mixed diet: feeding on several foliar age classes increases the performance of

a specialist insect herbivore. Oecologia, 135: 391-399.

Parsons, K., Quiring, D., Piene, H., and Moreau, G. 2005. Relationship between

balsam fir sawfiy density and defoliation in balsam fir. For. Ecol. Manage.

205:325-331.

Prentice, R. M. 1963. Forest Lepidoptera of Canada. Vol. 3. Lasiocampidae,

Drepanidae, Thyatiridae, Geometridae. Bull. 1034., Dept. For. Can., Ottawa.

Rindge, F.H. 1966. A revision of the moth genus, Anacamptodes (Lepidoptera:

Geometridae). Bull. Am. Mus. Nat. Hist. 132(2): 179-238.

Thurston, G.S. 2005. Evaluation of the efficacy of Foray 48B for management of

pale-winged gray (Iridopsis ephyrarid) larvae on eastern hemlock (Tsuga

canadensis). Can. For. Serv. Atlantic For. Cent. Rep. to N.S. Dept. Nat. Res.

and Valent Biosciences Ltd.

Wallin, K.F., and Raffa, K.F. 1997. Association of within-tree jack pine budworm

feeding patterns with canopy level and within-needle variation of water,

nutrient, and monoterpene concentrations. Can. J. For. Res. 28: 228-233.

Yamasaki, M., and Kikuzawa, K. 2003. Temporal and spatial variations in leaf

herbivory within a canopy of Fagus crenata. Oecologia, 137: 226-232. 56

Zar, J.H. 1984. Biostatistical Analysis, 2n Edition. Prentice-Hall Inc. Toronto. 57

CHAPTER FOUR

THE INTERACTION OF FOLIAGE AND LARVAL AGE INFLUENCES THE

PREFERENCE AND PERFORMANCE OF A CATERPILLAR ON EASTERN

HEMLOCK

4.1 INTRODUCTION

The nutritional quality of foliage for herbivorous insects often changes rapidly with foliage age, and consequently most insects specialize on either young or old foliage (e.g. Root 1973; Reichle 1973; Rausher 1981; Damman 1987; Carisey &

Bauce 1996). Some insects, however, consume a variety of different-aged foliage. For example, sawfly (Hymenoptera) larvae (Ikeda et al. 1977; Jensen 1988; Moreau et al.

2003; Parsons et al. 2003), and the ramie moth, Arete coerulea Guin. (Lepidoptera:

Noctuidae) (Ide 2006), consume a mixture of older age-classes of foliage. Presumably, a mixed diet of foliage ages could allow larvae to obtain a better balance of nutrients and water while minimizing the uptake of harmful allelochemicals (Cates 1980;

Schroeder 1986; Moreau et al. 2003). Recently both a laboratory study with a caterpillar (Carroll 1999) and a manipulative field study with a sawfly (Moreau et al.

2003) determined that herbivore performance was increased by feeding on multiple age classes of conifer foliage (balanced diet hypothesis). Feeding on different foliage ages might also allow larvae to avoid natural enemies at certain times of the day

(Damman 1987; Hunter 2003). 58

Alternatively, feeding on a mixed diet of foliage ages might result from variations in feeding patterns of early and late instars (ontogenetic hypothesis). Early and late instars might have different nutritional requirements with respect to water, nutrient and toxin content in foliage (Cates 1980; Schroeder 1986), or early instars might be limited to young foliage due to physical feeding barriers of toughened, older foliage (Bernays et al. 1991; Hochuli 2001). Some Lepidoptera, such as the buck moth Hemileuca lucina Hy. Ed. (Saturniidae), and the hemlock looper Lambdina fiscellaria fiscellaria Guen. (Geometridae), feed on the young foliage of current-year shoots as early instars but consume a mixed diet of foliage ages as older instars

(Stamp & Bowers 1990; Raske et al. 1995).

In this study, both the balanced diet and ontogenetic hypotheses were tested using surveys and manipulative field experiments with the pale-winged gray moth,

Iridopsis ephyraria Wlk. (Lepidoptera: Geometridae), an indigenous defoliator of eastern hemlock, Tsuga canadensis (L.) Carr. (Pinaceae), that feeds on current-year and older foliage.

4.2 METHODS

Study insect

Iridopsis ephyraria larvae were well suited to study the balanced diet and ontogenetic hypotheses, since preliminary observations suggested that caterpillars consume all ages of foliage, with young instars feeding primarily on current-year foliage, and later instars eating a mixture of older foliage ages. The pale-winged gray moth was responsible for severely defoliating stands of eastern hemlock in southwestern Nova Scotia between 2001-2003, followed by a gradual population 59 decline. There is no prior record of an outbreak of I. ephyraria in the refereed literature, although it is widely distributed from Alberta to Nova Scotia, and south to

Texas (Ferguson 1954; McGuffin 1977). Larvae emerge in May or June, pass through five instars, and pupate in mid-July (Chapter Two). Adults are active in early August, when females lay eggs singly on the boles of mature hemlock. Larvae consume a wide variety of hosts, most notably eastern hemlock (Chapter Two).

Foraging behaviour of larvae in nature

The location of larval feeding on a branch was observed at four study sites in

Kejimkujik National Park and National Historic Site (KNP), in southwestern Nova

Scotia, Canada, in 2004. At each site, one upper and one lower crown branch was tagged on each of 10 small (< 3m tall) understory trees, before egg hatch. Current- year to 3-year-old foliage on study branches was carefully searched for feeding larvae.

Larval position on the branch was recorded on 18 June, 28 June and 15 July, when most instars where second, third and fifth instars, respectively.

To assess caterpillar preference for different-aged foliage in mature trees, we measured defoliation attributable to I. ephyraria at seven study sites in 2005, and nineteen sites in 2006. All sites were located within 30 km of KNP. At each site, 30 mature hemlocks, each separated by at least 7 m, were selected along 1-3 linear transects. Using binoculars, one observer estimated defoliation of current-year to 3- year-old foliage along the main axis of one haphazardly selected mid-crown branch.

Natural needle fall occurred on foliage 4 years old and older, and these foliage ages were therefore not included in analysis. Defoliation estimates were categorized into 10% defoliation classes as in Parsons et al. (2003). Seasonal defoliation was calculated by taking the difference between estimates from late July, following pupation, and early June, prior to egg hatch.

To assess caterpillar preference for different-aged foliage in immature trees,

30 small (<3m tall) understory hemlock trees, each separated by at least 3m, were selected along a linear transect at four sites (2004) or five sites (2005) of the sites described above. Defoliation along the main axis of a mid-crown branch was visually estimated as described above.

Performance of larvae on different-aged foliage

To assess the performance of larvae on different ages of foliage, we caged larvae on mid-crown branches of 25 small (<3m) hemlock trees at a site with little

(<10%) previous defoliation in KNP in 2006. On each study tree, each of three branches was enclosed in a nylon mesh sleeve cage, and randomly assigned to one of three treatments: current-year foliage, old foliage only (1-6-year-old foliage); or a mixed diet of foliage ages (entire branch, which included current-year to 6-year-old foliage). Five 2nd-instar larvae were collected from nearby hemlock foliage and placed in each sleeve cage. Larvae were forced to feed on the current-year and old foliage treatments by twist ties fastened around the cage, preventing access to old and current-year foliage, respectively. Larvae were monitored throughout development to ensure that not all available foliage was consumed. When the majority of larvae were in the 3rd or 4th instar (halfway through larval development), all larvae were removed and survival determined. Five new 3rd or 4l instar larvae were collected from nearby 61 trees and placed in each study cage. Cages were removed after adult eclosion had finished and survival until pupation was determined for each cage. Pupal survival was not recorded since /. ephyraria pupate in soil and pupae in the sleeve cages desiccated quickly. Adult gender and wing length was measured when adults with undamaged wings were available.

Statistical Analysis

The number of larvae observed on different foliage ages was compared using a four-way ANOVA, with site and tree (nested within site) as random factors and date and foliage age as fixed factors, as well as all possible interactions.

For both mature and small trees, defoliation patterns along a branch were assessed with a split-plot analysis of covariance, since foliage ages were grouped by branch nested within site, with foliage age, site, and the interaction of site and foliage age as testable factors. Previous defoliation was included as a covariate, since this is a good estimate of foliage initially available to be consumed. Defoliation was assessed repeatedly between years at some sites, so separate analyses were conducted for each year. Mature tree defoliation in 2005 was power-transformed (X011), and small tree defoliation in 2005 was log-transformed, to ensure homogeneity of variances.

The performance of larvae in sleeve cages was compared using a split-plot analysis of variance, since study branches were grouped within tree, with instar period (early or late instars), foliage age treatment and the interaction of the two as factors. Proportion surviving was power transformed (X 25) to meet the assumption 62 of homogeneity of variances. Variation in adult wing length was assessed using a two-way ANOVA with treatment and moth gender as factors.

4.3 RESULTS

Foraging behaviour of larvae in nature

Foliage age explained a significant proportion of the variation in the location where larvae were observed feeding (Table 2), with most caterpillars found on current-year shoots (Fig. 9). During the first two sample dates, early instars were only observed feeding on current-year foliage (Fig. 9a,b). On 15 July, older instars consumed a mixture of foliage ages (Fig. 9c), resulting in a significant interaction between foliage age and date (Table 2). Late instars had a strong preference for current-year foliage only at one site (Fig. 9c), leading to a significant foliage age x date x site interaction (Table 2). Site, tree, and the interactions of site x foliage age, site x date, and foliage age x tree all explained some of the variation in larval density

(Table 2), likely reflecting a high natural variability between trees and between sites.

Foliage age explained a significant proportion of the variation in defoliation on mature trees in 2006, but was marginally insignificant in 2005 (Table 3). Larvae consumed a greater proportion of current-year than older foliage (Fig. 10). Previous defoliation explained most of the variation in defoliation (Table 3), since high levels of previous defoliation resulted in less foliage available for consumption on all older foliage ages (Fig. 10). In both years, defoliation varied among sites and was influenced by the interaction between site and foliage age (Table 3), the latter 63 presumably resulting from variations in the defoliation pattern at some of the study sites.

Similar to mature trees, pale-winged gray moth larvae consumed a higher proportion of current-year than old foliage on small understory trees (Fig. 11).

However, due to high levels of previous defoliation that varied among sites, only the interaction between foliage age and site, and not the main effect of foliage age, was significant (Table 4). Site and previous defoliation also explained a significant proportion of the variation in defoliation (Table 4).

Performance of larvae on different-aged foliage

Foliage age explained a significant proportion of the variation in larval survival (F2,48 = 24.56; P < 0.001). Both early and late instars survived best on a mixed diet of foliage ages (Fig 12). Although the survival of early and late instars was similar (Fi;24 = 3.14; P = 0.09), there was a significant interaction between larval age and foliage age (F2,48 = 6.08; P < 0.005). When forced to feed on only young or old foliage, the survival of early instars was higher on young than old foliage, whereas late instar survival was higher on old foliage (Fig. 12).

Wing lengths of adults that had developed on old foliage or on the mixed diet were not influenced by foliage age (Fi^i = 0.37; P = 0.49), gender (Fi^i = 0.01; P =

0.92), or the interaction of the two (F131 = 0.77; P = 0.39). Adult wing lengths from the current-year foliage treatment cages were not analyzed, since we only obtained three adults with undamaged wings from this treatment. 64

4.4 DISCUSSION

This study supports both the balanced diet and the ontogenetic hypotheses.

The superior performance of both early and late instars on a mixed diet supports the balanced diet hypothesis. The ontogenetic hypothesis was also supported, since early instars performed better on young than old foliage, and late instars performed better on old than young foliage. Generally, the low larval survival was consistent with natural mortality levels (Chapter Two).

Superior performance on a mixed diet may be due to the uptake of an optimal amount of nutritional components, such as nitrogen, while limiting the uptake of growth-limiting toxins often present in specific foliage ages (Cates 1980; Schroeder

1986; Moreau et al. 2003). It might be useful for future studies to examine the different dietary requirements of early versus late-instar larvae that might lead to the consumption of a balanced diet.

Higher early instar survival on current-year rather than old foliage may be due to higher water and nitrogen concentrations in foliar tissues (Cates 1980; Schroeder

1986; Hatcher 1990). Foliage water content had the strongest effect on larval growth of another geometrid caterpillar (Haukioja et al. 2002). Leaf toughness may also have reduced the performance of early instars on old foliage. Early instars, which have smaller mouthparts, may have difficulty biting into old foliage (Hochuli 2001). When they do succeed in cutting through cuticles, more bites are necessary to consume tougher food (Bernays et al. 1991). 65

Late instars had better survival on older foliage, perhaps due to lower concentrations of phenolics (Hatcher 1990), or their ability to overcome the physical barrier of toughened needles (Hochuli 2001). Some larvae specialize on old foliage to reduce the effects of natural enemies (Damman 1987), but natural enemies were excluded by sleeve cages in the present study.

Field surveys indicated that /. ephyraria prefers to feed on all age classes of foliage. Young larvae were only observed to feed on current-year shoots, while older instars fed on a mixture of older foliage ages. It is possible that young larvae may have fed on older foliage at times other than the few minutes they were observed during our field surveys.

Defoliation differed significantly between ages of foliage only on mature trees in 2006 (Table 3), although these differences were very small and gradual between adjacent foliage ages. Defoliation levels also varied significantly between sites

(Tables 3, 4). This was expected, since sites were chosen to represent a range of infestation levels, and previous defoliation varied significantly between sites. To a lesser extent, it is possible that previous defoliation might have altered foliage quality in subsequent years (Karban & Myers 1989). Even without variations in previous defoliation, foliage quality, most notably nitrogen content, can vary substantially between eastern hemlock stands (Pontius et al. 2006).

The results of the performance study suggest that the best strategy for survival of early and late instars is to feed on a mixture of foliage ages, likely allowing larvae to balance the effects of the uptake of nutrition and toxins. However, this study also 66 suggests that the relative ratio of young to old foliage required shifts from young to older foliage as larvae age. 67

Table 2. Summary of nested ANOVA assessing the influence of date (D), foliage age (F), tree (T) and site (S) on the location where larvae were observed feeding.

Source of Variation df (num) df (denom) F P Site 3 36 9.94 < 0.001 Tree (Site) 36 480 1.96 < 0.001 Foliage age 3 9 17.11 < 0.001 Date 2 6 2.59 0.150 SxF 9 108 9.83 < 0.001 SxD 6 72 22.53 < 0.001 FxD 6 18 3.88 0.012 F x T (S) 108 480 1.96 < 0.001 D x T (S) 72 480 1.06 0.350 S xF xD 18 216 20.83 < 0.001 F x D x T (S) 216 480 1.09 0.220 Error 480 68

Table 3. Summary of split-plot ANCOVA evaluating the influence of foliage age on defoliation by /. ephyraria on mature hemlock trees in two years. 2005 data was power-transformed (X° n) prior to analysis.

Year 2005 2006 Source of variation df F P df F P

Between plots Site 6 12.22 O.001 18 5.06 O.001 Error [Tree(Site)] 203 551

Within plots Age of foliage 3 2.89 0.064 3 8.08 O.001 Site x Age of foliage 18 4.41 O.001 54 5.80 <0.001 Error [Age of foliage x 608 1652 Tree(Site)]

Previous defoliation 1 145.37 <0.001 1 463.96 O.001 69

Table 4. Summary of split-plot ANCOVA evaluating the influence of foliage age on defoliation by I. ephyraria on small, understory hemlock trees in two years. 2005 data was log-transformed prior to analysis.

Year 2004 2005 Source of variation df F P df F P

Between plots Site 3 4.11 0.013 4 4.98 <0.001 Error [Tree (Site)] 36 145

Within plots Age of foliage 4 0.99 0.450 3 0.34 0.800 Site x Age of foliage 12 6.93 O.001 12 5.23 <0.001 Error [Age of foliage x 143 434 Tree (Site)]

Previous defoliation 1 68.89 O.001 1 132.02 O.001 70

60 (a) 50 18 June 40 30 20 10

0 -&• -K?- o e m 30 JQ i_ 0) 25 £L 28 June 0) 20 > js 15 i_ 0) 10 E 3 5 C C (0 0 -Q- 0>

8 -. (C) 15 July

C 1 2 3 Foliage age class

Fig. 9. Mean (± SE) number of larvae per branch section feeding on different-aged foliage on small, understory hemlock trees at four sites in 2004. Symbols represent different sites. Note differences in y-axis scales. 71

100 -, (a)

80 X I I 60

40

c 20 o lis is 0 2 (1) 100 (b) C TO

X 60

40

20 4

C 1 2 3 Foliage age class

Fig. 10. Mean (± SE) defoliation of different-aged foliage in mature hemlock trees in (a) 2005 and (b) 2006. Previous defoliation (black bars) and final defoliation (white bars), were measured just before and after feeding by larvae, respectively. Seasonal defoliation is the difference between the two. Means are based on (a) 7 and (b) 19 sites. 72

(a) I J,

c

(b)

C 1 2 Foliage age class

Fig. 11. Mean (± SE) defoliation of different-aged foliage in small understory hemlock trees in (a) 2004 and (b) 2005. Previous defoliation (black bars) and final defoliation (white bars), were measured just before and after feeding by larvae, respectively. Seasonal defoliation is the difference between the two. Means are based on (a) 4 and (b) 5 sites. 73

X i X Old Mixed diet

Foliage age treatment

Fig. 12. Mean (± SE) survival of early (black bars) and late-instar (white bars) larvae feeding on current-year (C), or old (1-6-year-old) foliage, or both current and old foliage (mixed), of eastern hemlock in South Brookfield, NS, in 2006. 74

REFERENCES

Bernays, E. A., Jarzembowski, E.A., and Malcolm, S.B. 1991. Evolution of insect

morphology in relation to plants [and discussion]. Phil. Trans: Biol. Sci. 333:

257-264.

Carisey, N., and Bauce, E. 1996. Impact of balsam fir foliage age on sixth-instar

spruce budworm growth, development and food utilization. Can. J. For. Res.

27: 257-264.

Carroll, A. L. 1999. Phyiological adaptation to temporal variation in conifer foliage

by a caterpillar. Can. Entomol. 131(5): 659-669.

Cates, R. G. 1980. Feeding patterns of monophagous, oligophagous, and polyphagous

insect herbivores: the effect of resource abundance and plant chemistry.

Oecologia, 46:22-31.

Damman, H. 1987. Leaf quality and enemy avoidance by the larvae of a pyralid moth.

Ecology, 68: 88-97.

Ferguson, D. C. 1954. The Lepidoptera of Nova Scotia 1. Macrolepidoptera. Bull.

N.S. Inst. Sci. 123.

Hatcher, P. E. 1990. Seasonal and age-related variation in the needle quality of five

conifer species. Oecologia, 85: 200-212.

Haukioja, E., Ossipov, V., and Lempa, K. 2002. Interactive effects of leaf maturation

and phenolics on consumption and growth of a geometrid moth. Entomol. Exp.

Appl. 104: 125-136. 75

Hochuli, D. F. 2001. Insect herbivory and ontogeny: How do growth and

development influence feeding behaviour, morphology and host use? Austral

Ecol. 26: 563-570.

Hunter, M.D. 2003. Effects of plant quality on the population ecology of parasitoids.

Agri. For. Entomol. 5: 1-8.

Ide, J.-Y. 2006. Inter- and intra-shoot distributions of the ramie moth caterpillar,

Arete coerulea (Lepidoptera: Noctuidae), in ramie shrubs. Appl. Entomol.

Zool. 41: 49-55.

Ikeda, T., Matsumura, F., and Benjamin, D.M. 1977. Mechanism of feeding

discrimination between mature and juvenile foliage by two species of pine

sawflies. J. Chem. Ecol. 3: 677-694.

Jensen, T. S. 1988. Variability of Norway spruce (Picea abies L.) needles;

performance of spruce sawflies (Gilpinia hercyniae Htg.). Oecologia, 77:

313-320.

Karban, R., and Myers, J.H. 1989. Induced plant responses to herbivory. Ann. Rev.

Ecol. Syst. 20:331-348.

McGuffin, W. C. 1977. Guide to the Geometridae of Canada (Lepidoptera) 2.

Subfamily Ennominae. 2. Mem. Entomol. Soc. Can. 101.

Moreau, G., Quiring, D.T., Eveleigh, E.S., and Bauce, E. 2003. Advantages of a

mixed diet: feeding on several foliar age classes increases the performance of

a specialist insect herbivore. Oecologia, 135: 391-399.

Parsons, K., Quiring, D., Piene, H., and Farrell, J. 2003. Temporal patterns of balsam 76

fir sawfly defoliation and growth loss in young balsam fir. For. Ecol. Manage.

184: 33-46.

Pontius, J. A., Hallett, R.A., and Jenkins, J.C. 2006. Foliar chemistry linked to

infestation and susceptibility to hemlock woolly adelgid (Homoptera:

Adelgidae). Environ. Ecol. 35:112-120.

Raske, A. G., West, R.J., and Retnakaran, A. 1995. Hemlock looper, Lambdina

fiscellaria. Pages 141-147 in J. A. Armstrong, Ives, W.G.H., editor. Forest

Insect Pests in Canada. NRC Canada, CFS, Ottawa.

Rausher, M. D. 1981. Host plant selection by Battusphilenor butterflies: The roles of

predation, nutrition, and plant chemistry. Ecol. Monogr. 51(1): 1-20.

Reichle, D. E., Goldstein, R. A., Van Hook, R. I., and Dodson, G. J. 1973. Analysis

of insect consumption in a forest canopy. Ecology, 54(5): 1076-1084.

Root, R.B. 1973. Organization of a plant- association in simple and diverse

habitats: The fauna of collards (Brassica oleracea). Ecol. Monogr. 43(1): 95-

124.

Schroeder, L. A. 1986. Changes in tree leaf quality and growth performance of

Lepidoptera larvae. Ecology, 67: 1628-1636.

Stamp, N. E., and Bowers, M. D. 1990. Variation in food quality and temperature

constrain foraging of gregarious caterpillars. Ecology, 71: 1031-1039. CHAPTER FIVE

GENERAL CONCLUSIONS

The research findings outlined in this thesis should help managers monitor and control Iridopsis ephyraria effectively. For example, although most eggs are laid on the upper tree bole, inexpensive foam oviposition traps could be used initially in the previous season to determine the presence or absence of/, ephyraria in hemlock stands. If eggs are detected in a stand or nearby, managers could then use sticky tape traps placed around the boles of trees to sample emerging first-instar larvae in late

May or early June. Due to a strong density-defoliation relationship for larvae collected on sticky tape traps, the density of larvae could be used to accurately predict subsequent hemlock defoliation for that season, most accurately for current-year foliage. Since the current-year shoot is slightly (although not always significantly) more heavily defoliated than other ages of foliage, it provides the most conservative estimate of tree damage for the year. If insect populations are deemed to be sufficiently high, it would be possible to implement control measures, such as selective hemlock cutting or the aerial application of Bacillus thuringiensis var. kurstaki, which was registered for /. ephyraria in 2005 (Thurston 2005). Although there was a significant density-defoliation relationship between densities of late instars sampled with beating sheets and defoliation of the current-year shoot, this sampling method may not be useful for management. Beating sheet sampling occurs very late in the season, once damage is already under way, and therefore the sampling might occur too late for intervention. 78

Head capsule measurements indicated that /. ephyraria develop through five larval instars. The first and second instars had not been described previously, since other studies had relied on field-collected samples of larger insects (McGuffin 1977).

Additionally, since the active period of I. ephyraria life stages has now been outlined, managers can time sampling and intervention programs to the insects' phenology.

The identification of a fungal pathogen in the system is important, since it appears to cause widespread larval mortality and possibly regulates the population naturally. Pupal mortality in the soil was also extremely high, and pupal predation often regulates other moth populations (Valenti et al. 1998; Tanhuanpaa et al. 1999).

It is vital to consider both of these mortality factors prior to intervention, since a warm and humid season favoring fungal growth (McDonald & Nolan 1995), or an unusually high activity rate of soil predators could lead to a rapid population decline.

Iridopsis ephyraria had been previously collected on a variety of plants, although in the field, larvae appeared to only eat alternative hosts when larval densities were extremely high. Larvae can complete development on several conifers, hardwoods and understory plants, although it appears from the extensive defoliation observed that eastern hemlock is the preferred host.

Within mature hemlock, larvae defoliated all foliage ages along a branch.

Although current-year foliage appeared to be more heavily defoliated, much of this pattern was attributable to previous defoliation, which often severely limited the availability of older foliage ages. In the understory, for instance, previous defoliation was approximately 80-90% on older foliage ages. 79

All larvae survived best on a mixed diet of foliage ages, supporting the balanced diet hypothesis. However, early instars had the second-highest survival when fed current-year foliage, whereas later instars had the second-highest survival on older foliage ages, also supporting the ontogenetic hypothesis. Generally, the observed feeding patterns were adaptive. Early (2n -3rd) instars were only observed feeding current-year foliage, although it is possible that larvae do eat older foliage ages and were simply not observed to do so. Later (4th-5th) instars fed on a mixture of all foliage ages. The performance experiment did not include first instars for practical reasons. It therefore remains unknown whether first instars would perform best on a mixed diet or on the current-year shoot only.

Iridopsis ephyraria is a useful study species for the study of feeding patterns within the crown of mature trees. Although many studies have evaluated heterogeneous feeding patterns of larvae within smaller study trees (e.g. Carroll &

Quiring 1994; Alonso & Herrera 1996; Anstey et al. 2002), few have examined feeding within the crown of such a sizeable mature tree, where microclimatic conditions could vary extensively (Rowe & Potter 1996; Fortin & Mauffette 2002;

Yamasaki & Kikuzawa 2003; Oishi et al. 2006). Iridopsis ephyraria larvae appear to more heavily defoliate the lower and middle crown than the upper crown of trees, a situation that appears to be uncommon in large trees. The sun leaves in the upper crown of trees are considered a more desirable food source (Fortin & Mauffette 2002), and many insects concentrate in the upper crown of trees (Rowe & Potter 1996;

Fortin & Mauffette 2002). Future studies might elucidate if and why larvae appear to avoid feeding on the upper crown of mature hemlocks. REFERENCES

Alonso, C, and Herrera, CM. 1996. Variation in herbivory within and among plants

of Daphne laureola (Thymelaeaceae): correlation with plant size and

architecture. J. Ecol. 84: 495-502.

Anstey, L.J., Quiring, D.T., and Ostaff, D.P. 2002. Seasonal changes in intra-tree

distribution of immature balsam fir sawfly (Hymenoptera: Diprionidae). Can.

Entomol. 134: 529-538.

Carroll, A. L., and Quiring, D.T. 1994. Intratree variation in foliage development

influences the foraging strategy of a caterpillar. Ecology, 75(7): 1978-1990.

Fortin, M., and Mauffette, Y. 2002. The suitability of leaves from different canopy

layers for a generalist herbivore (Lepidoptera: Lasiocampidae) foraging on

sugar maple. Can. J. For. Res. 32: 379-389.

McDonald, D. M. and Nolan, R.A. 1995. Effects of relative humidity and temperature

on Entomophaga aulicae conidium discharge from infected eastern hemlock

looper larvae and subsequent conidium development. J. Invert. Pathol. 65: 83-

90.

McGuffin, W.C. 1977. Guide to the Geometridae of Canada (Lepidoptera) II.

Subfamily Ennominae. Mem. Entomol. Soc. Can. 101.

Oishi, M., Yokota, T., Teramoto, N., and Sato, H. 2006. Japanese oak silkmoth

feeding preference for and performance on upper-crown and lower-crown

leaves. Entomol. Sci. 9: 161-169.

Rowe, W. J., and Potter, D.A. 1996. Vertical stratification of feeding by Japanese 81

beetles within linden tree canopies: selective foraging or height per se?

Oecologia, 108: 459^166.

Tanhuanpaa, M., Ruohomaki, K., Kaitaniemi, P., and Klemola, T. 1999. Different

impact of pupal predation on populations of Epirrita autumnata (Lepidoptera;

Geometridae) within and outside the outbreak range. J. Anim. Ecol. 68: 562-

570.

Thurston, G.S. 2005. Evaluation of the efficacy of Foray 48B for management of

pale-winged gray (Iridopsis ephyraria) larvae on eastern hemlock (Tsuga

canadensis). Can. For. Serv. Atlantic For. Cent. Rep. To N.S. Dept. Nat. Res.

and Valent Biosciences Ltd.

Valenti, M.A., Berryman, A.A., and Ferrell, G.T. 1998. Natural enemy effects on the

survival of Synaxis cervinaria (Lepidoptera: Geometridae). Environ. Entomol.

27:305-311.

Yamasaki, M., and Kikuzawa, K. 2003. Temporal and spatial variations in leaf

herbivory within a canopy of Fagus crenata. Oecologia, 137: 226-232. CURRICULUM VITAE

Candidate's full name: Lauren Lynn Pinault

Universities attended: University of Ottawa, BSc. Honours (Biology) September 2001 - April 2005

University of New Brunswick, MSc. (Forestry) May 2005 - September 2007

Publications:

Pinault, L., Georgeson, E., Guscott, R., LeBlanc, M., Jameson, R., McCarthy, C, Lucarotti, C, Thurston, G., and Quiring, D. Life history of Iridopsis ephyraria, (Lepidoptera: Geometridae), a defoliator of eastern hemlock in eastern Canada. J. Acad. Entomol. Soc. 3: 28-34.

Pinault, L. 2005. A revision of Dioptrophorus Faust and three new genera {Neodioptrophorus, Buckingorum and Chiapaneca (Coleoptera: Curculionidae), collected from the region of Chiapas, Mexico. Honours thesis, Department of Biology, Faculty of Science, University of Ottawa.

Conference Presentations:

2007 Pinault, L. Sampling the pale-winged gray moth, a defoliator of eastern hemlock, in southern Nova Scotia. University of New Brunswick Graduate Students Association, Annual Student Research Conference, Fredericton.

2006 Pinault, L. Influence of intra-plant heterogeneity on feeding patterns of the pale- winged gray moth, Iridopsis ephyraria. Entomological Society of Canada, Annual General Meeting, Montreal.

Pinault, L. A sampling strategy for the pale-winged gray moth in Kejimkujik National Park. Acadian Entomological Society, Annual General Meeting, Kentville, N.S. (also presented to the Kejimkujik National Park Annual Research Conference, Maitland Bridge, N.S.)

Pinault, L. Nocturnal adult behaviour of the pale-winged gray moth. University of New Brunswick Graduate Students Association, Annual Student Research Conference, Fredericton.

2005 Pinault, L., Anderson, R., Houseman, J. A revision of Dioptrophorus Faust and three new genera Neodioptrophorus, Buckingorum and Chiapaneca (Coleoptera: Curculionidae), collected from the region of Chiapas, Mexico. Poster presentation to the Acadian Entomological Society, Annual General Meeting, Fredericton.