Effects of Insect Opportunists on a Four-Level Trophobiotic System

Home , Acraspis, Antistrophus, Apidae, Chalcid wasp, Cuckoo wasp, Disholcaspis

Eastern Illinois University The Keep

Masters Theses Student Theses & Publications

Fall 2020

Effects of insect opportunists on a four-level trophobiotic system involving nectar-producing galls of the cynipid wasp Disholcaspis quercusmamma (Hymenoptera: Cynipidae)

Stephanie L. Smith Eastern Illinois University

Follow this and additional works at: https://thekeep.eiu.edu/theses

Part of the Entomology Commons, Evolution Commons, and the Other Ecology and Evolutionary Biology Commons

Recommended Citation Smith, Stephanie L., "Effects of insect opportunists on a four-level trophobiotic system involving nectar- producing galls of the cynipid wasp Disholcaspis quercusmamma (Hymenoptera: Cynipidae)" (2020). Masters Theses. 4851. https://thekeep.eiu.edu/theses/4851

This Dissertation/Thesis is brought to you for free and open access by the Student Theses & Publications at The Keep. It has been accepted for inclusion in Masters Theses by an authorized administrator of The Keep. For more information, please contact [email protected]. Effects of insect opportunists on a four-level trophobiotic system involving nectar-producing galls of the cynipid wasp Disholcaspis quercusmamma (Hymenoptera: Cynipidae)

Stephanie L. Smith

A Thesis Submitted for the Requirements for a Degree of

Master of Science

Department of Biological Sciences, Eastern Illinois University

Charleston, IL, USA

Thesis Advisor:

Dr. Zhiwei Liu

Committee Members:

Dr. Ann Fritz

ABSTRACT

The induction of plant galls is considered an adaptive life history trait found in many insect groups. The formation of galls provides several advantages to the gall maker, such as enhanced nutrition, favorable microclimate, and protection from natural enemies, including parasitoids, inquilines, and predators. Order Hymenoptera has many gall- making species, belonging to the gall wasp family Cynipidae. As an extended phenotype of the gall makers, some galls exhibit very sophisticated adaptive mechanisms involving multilevel species interactions. In particular, the oak galls of the Cynipid species

Disholcaspis quercusmamma, found in much of Illinois, produce a palatable, sugary nectar-like secretion, attracting other insects. It is hypothesized that visiting insects discourage potential parasitoids and inquilines questing for oviposition sites within the galls, possibly producing an enemy-free space. Several studies on the species interactions involved in this four-level trophobiotic association have examined the role of visiting ant species in the system. However, other insect species have also been observed to be attracted to the extrafloral nectar of the galls, and their roles in the gall wasp-natural enemy interaction system have not yet been studied. This study aimed to examine the effects of all visiting insects, including ants, on the success of the gall-makers as a result of potentially reduced parasitism. Exclusion experiments were carried out on three oak trees in study sites in central Illinois. Statistical analyses of field data found no effect of treatment on gall wasp success, parasitism, or inquilinism. However, additional tests showed a positive correlation between gall wasp success and inquilinism rate compared to gall cluster size. Gall diameter correlated positively with gall maker success and

i negatively with inquilinism rate. This suggests an underlying complexity to the system.

Additional research is necessary to better understand this uniquely adapted relationship.

ACKNOWLEDGEMENT

First and foremost, I would like to extend my sincerest gratitude to my advisor,

Dr. Zhiwei Liu. His great passion for gall wasps has inspired me and given me the perseverance to complete this thesis. Thank you for all the encouragement and kind words along the way. You have been the best advisor I could have ever hoped for!

I would also like to extend thanks to my thesis committee, Dr. Ann Fritz and Dr.

Scott Meiners. Dr. Fritz’s introductory entomology course was another source of inspiration for this project, and I will always deeply appreciate her advice, both entomological and personal. Dr. Meiners’ melodious mathematical musings further cemented my interest in all things statistical, and his help with the dataset for this project was incredibly valuable.

I am also indebted to Dr. Britto Nathan, Dr. Gary Bulla, and Marschelle McCoy, who all had a part in making my time at Eastern Illinois University an enjoyable journey.

Thank you all so much for indulging my interest of insects during my time at EIU, and for making me feel like a part of the family.

Additionally, I extend a deep appreciation to Erin Shulk, who helped me with the experimental setup. She jumped headfirst into assisting me with this experiment and applied treatments to the tallest branches when my fear of heights got the best of me.

Thank you so much, Erin!

I appreciate the cooperation of the staff at Fox Ridge State Park. Without their permission and support, this experiment would never have taken place. Thank you for maintaining such a beautiful nature preserve.

iii

The obligatory ‘last but not least;’ my loving family. I will always be thankful for your support and interest in my project. Thank you so much for listening to my thought processes and providing such a wonderful sounding board, even when my ramblings made absolutely no sense. To my parents, who went with me to my first visit at EIU and provided unwavering support as soon as I knew that it was the program for me. To my grandparents, who allowed me to conduct part of my study on their property, and whose house always feels like home. To my aunt and godparents, who are happy to provide encouragement, a listening ear, or just a well-timed joke, even from afar. To my nephew, whose interest in insects I am always happy to indulge – for the best way to learn is to teach someone else. And to my husband, who would follow me to the ends of the earth to help support my dream. I love you all beyond measure.

TABLE OF CONTENTS

ABSTRACT ...... i

ACKNOWLEDGEMENT ...... iii

LIST OF FIGURES ...... vi

LIST OF TABLES ...... viii

INTRODUCTION...... 1

MATERIALS AND METHODS ...... 10

Study sites ...... 10 Treatments ...... 10 Observations...... 11 Data collection ...... 11 Statistical analyses ...... 12 RESULTS ...... 13

Observations...... 13 Statistical analyses ...... 14 DISCUSSION ...... 16

REFERENCES ...... 20

LIST OF FIGURES

Figure 1. The nectar-producing galls of Disholcaspis quercusmamma……………...….26

Figure 2. An inquiline of Disholcaspis quercusmamma……………...……………...….27

Figure 3. Map of study sites in Central and East Central Illinois……………..………...28

Figure 4. Mesh bags made of ½” mesh garden netting.……………………..…………..29

Figure 5. Fruit bags secured around the galled portion of the branches……...…………30

Figure 6. A Disholcaspis quercusmamma gall cut open to reveal the central gall chamber in which the larva resides…………….………...…………………………………….…..31

Figure 7. A carpenter ant (Camponotus sp.) visiting the nectar-producing galls…….....32

Figure 8. A northern paper wasp (Polistes fuscatus) visiting the nectar-producing galls Courtesy Zhiwei Liu………………..……………………………………………..……..33

Figure 9. A cuckoo wasp (Chrysis sp.) and housefly (family Muscidae) visiting the nectar-producing galls. Courtesy Zhiwei Liu……………………………………………34

Figure 10. A great black wasp (Sphex pensylvanicus) visiting the nectar-producing galls. Courtesy Zhiwei Liu………………………………………………………..…………....35

Figure 11. An Asian lady beetle (Harmonia axyridis) and several houseflies (family Muscidae) visiting the nectar-producing galls. Courtesy Zhiwei Liu.……………...……36

Figure 12. A wheel bug (Arilus cristatus) visiting the nectar-producing galls……...……………………………………………………………………………….37

Figure 13. Red-spotted purple butterflies (Limenitis arthemis) visiting the nectar- producing galls. Courtesy Zhiwei Liu.………...…………………………………….…..38

Figure 14. Black sooty mold on Disholcaspis quercusmamma galls late in the season, after they had ceased production of nectar……………………...……...………………..39

Figure 15. A boxplot of the Kruskal-Wallis one-way ANOVA comparing the number of successful galls for each exclusion treatment (p=0.35). Mesh cage excluded larger

vi insects, but not ants; Tanglefoot excluded ants, but not larger insects; Fruit bag excluded both ants and larger insects………………………………..……..………………..……..40

Figure 16. A boxplot of the Kruskal-Wallis one-way ANOVA comparing the number of parasitized galls for each exclusion treatment (p=0.83). Mesh cage excluded larger insects, but not ants; Tanglefoot excluded ants, but not larger insects; Fruit bag excluded both ants and larger insects...………………………………...…..………………..……..41

Figure 17. A boxplot of the Kruskal-Wallis one-way ANOVA comparing the number of inquilined galls for each exclusion treatment (p=0.78). Mesh cage excluded larger insects, but not ants; Tanglefoot excluded ants, but not larger insects; Fruit bag excluded both ants and larger insects…………………...………………..…………………….....……..42

Figure 18. A scatterplot comparing the number of inquilines found within the galls and gall diameter (rτ=-0.35, 60 d.f., p<0.005)………………………………………..………43

Figure 19. A scatterplot comparing the number of inquilines found within the galls and the number of galls in a cluster (rτ=0.46, 60 d.f., p<0.005)………………………….…..44

Figure 20. A scatterplot comparing the number of successful D. quercusmamma found within the galls and gall diameter (rτ=0.19, 60 d.f., p=0.04)………………………...…..45

Figure 21. A scatterplot comparing the number of successful D. quercusmamma found within the galls and the number of galls in a cluster (rτ=0.56, 60 d.f., p<0.005)………..46

Figure 22. A scatterplot comparing the number of parasitoids found within the galls and gall diameter (rτ=-0.05, 60 d.f., p=0.66)……………………………………………...….47

Figure 23. A scatterplot comparing the number of parasitoids found within the galls and the number of galls in a cluster (rτ=0.16, 60 d.f., p=0.15)………………………...……..48

vii

LIST OF TABLES

Table 1. A summary of the insect visitors of the D. quercusmamma galls……………..49

Table 2. A summary of the contents of the D. quercusmamma galls (n=272) on treated branches and successfully sampled branches (n=62).………...…….………………...…50

viii

INTRODUCTION

The induction of plant galls, or abnormal plant tissue growth, is a unique adaptation exhibited by many insect species in several orders. Hymenoptera is the insect order with the highest diversity of gall makers, mostly belonging to the families

Cynipidae, Tenthredinidae (subfamily Nematinae), and several families in the superfamily Chalcidoidea (Dreger-Jauffret and Shorthouse 1992). Galls develop because of stimulation from the feeding of the gall making insects, which either hatch from eggs laid in plant tissue as is the case for wasps of the family Cynipidae, or on the host, as for gall flies in the gall midge family Cecidomyiidae (Diptera) (Dreger-Jauffret and

Shorthouse 1992).

The Cynipidae is widely known as the gall wasp family and encompasses over

1300 known species worldwide apart from Australia (Ronquist and Liljeblad 2001), with all members being gall makers or otherwise exclusively associated with galls (Ronquist et al. 2015). In the Cynipidae, wasp larvae develop inside the gall and feed on the nutritious plant tissue lining the interior side of the galls until they pupate and eventually emerge as adults. Gall wasps are also highly host specific. Each wasp species in the tribe Cynipini has a host preference that only spans several, or even only one, closely related specie(s) of oak (Quercus spp.) (Abrahamson et al. 1998; Liljeblad and Ronquist 1998; Ronquist and Liljeblad 2001; Ronquist et al. 2015). Many gall wasps also tend to stay on the same individual tree for generations (Stone et al. 2009).

The development of galls provides several advantages to the insect inducer, such as enhanced nutrition, favorable microclimate, and protection from natural predators

(Price et al. 1987; Hardy and Cook 2010). However, these advantages are not without

1 their drawbacks; the sessile galls are often the target of various species of natural enemies. Vertebrate predators such as birds can easily break open the galls to feed upon the larvae inside (Tscharntke 1992; Poff et al. 2002). These natural enemies also include other insects - parasitoids and inquilines. Parasitoids are species often belonging to other families in the order Hymenoptera, such as the Eurytomidae, a gall wasp family which includes the genus Sycophila, a common parasitoid of cynipid galls (McEwen et al.

2014). These parasitoid species are host-specific, seeking out certain species of insect hosts to attack. Parasitoid females lay their eggs on or inside the host during a developing life stage, and their offspring feed upon and eventually destroy the host offspring (Washburn 1984).

Inquilines, like parasitoids, are also known as nest parasites. The inquilines of insect galls feed on the tissue of their host galls. Inquilines associated with the cynipid gall wasps are often also members of Cynipidae and are considered to be gall wasps that have lost their ability to induce galls, for example, the genus Synergus (Ronquist et al.

2018). Adult female inquilines lay eggs into the host gall at an early stage of the gall development. The resulting larvae may or may not kill the host larva upon maturation, unlike a parasitoid invasion (Craig et al. 2007). Beetles (Coleoptera) are also common inquilines that infest the galls produced by Cynipid wasps (Craig et al. 2007; Aranda-

Rickert et al. 2017). The larvae of inquilines of cynipid galls feed upon the outer portion of the gall parenchyma, oftentimes leaving the host larvae within the central gall chamber undisturbed. Nevertheless, their presence is detrimental to the gall makers’ larvae, as the invaders feed upon resources essential for the gall wasps’ development (Ritchie 1984;

Ronquist 1994; Ronquist and Liljeblad 2001; Ronquist et al. 2015).

Parasitoid and inquiline attack can cause devastating mortality in developing gall wasps (Washburn and Cornell 1979; Washburn and Cornell 1981; Aebi et al. 2007).

Parasitoids are so effective at decimating gall wasp populations that researchers have begun to use them in biological pest control. One example is the Oriental Chestnut gall wasp (OCGW) (Dryocosmus kuriphilus), which is endemic to eastern China and most likely also neighboring areas of the other continental Eastern Asian countries. The species began to be considered an invasive pest species in its native land of continental

Asia and Japan in the mid-20th century, and has later been accidentally introduced to

Europe and the Americas, becoming a devastating pest of various chestnut tree species in these regions (Aebi et al. 2007; Matošević et al. 2017; Ferracini et al. 2019). D. kuriphlius reduces fruit yield in chestnut (Castanea) by up to 75% and can even eventually kill the tree (Aebi et al. 2007). Multiple natural enemy species have been found to be useful in pest management projects for D. kuriphilus, including Torymus sinensis, a highly effective parasitic wasp native to Asia (Matošević et al. 2017; Ferracini et al. 2019). In the mid-2000s, this parasitoid was released into some of the most-affected countries, such as Italy, Hungary, Croatia, and Slovenia, and found to be extremely effective at controlling D. kuriphlius populations. Four years after T. sinensis’ release in

Italy, the D. kuriphlius infestation decreased by up to 95% in some areas, and continues to decline (Ferracini et al. 2019). In Hungary, Croatia, and Slovenia, the release of T. sinensis into areas affected by D. kuriphlius resulted in a parasitism rate of up to 90.8%

(Matošević et al. 2017).

Selective evolutionary pressure upon the gall wasp to prevent parasitoid and inquiline colonization has resulted in numerous adaptations to galls, including spines,

3 sticky resins, and hair-like coatings (Stone and Schönrogge 2003). Thirty-one species from five genera of Cynipidae have evolved another adaptation: the stimulation of the host plant to produce extrafloral nectar or honeydew from their gall tissue (Seibert 1993;

Aranda-Rickert et al. 2017; Nicholls et al. 2017). This unique adaptation entices other arthropods, including ants and other insects, to feed upon and even defensively guard the galls. It is thought that this increased presence of various visiting insects would intimidate potential parasitoid or inquiline females in the area, resulting in the decreased success of their oviposition into the potential host galls (Washburn 1984; Seibert 1993; Aranda-

Rickert et al. 2017; Nicholls et al. 2017).

The use of a honeydew-like nectar is an example of trophobiosis, or food-reward based mutualism (Yoshihisa 1988). Trophobiotic interactions usually involve three different organisms, or trophic levels – a herbivore that produces the nutritional reward

(most commonly, nectar), another insect species that consumes the reward, and in return, guards the herbivore, and finally, a parasite or predator of the herbivore that the guardian insect is, directly or indirectly, protecting the herbivore against (Yoshihisa 1988; Inouye and Agrawal 2004; Nicholls et al. 2017). Oftentimes, ants are the guardian in this relationship (Yoshihisa 1988; Inouye and Agrawal 2004); however, other insects have been known to also consume the nectar (Gagliardi and Wagner 2016; Aranda-Rickert et al. 2017) and may also offer protection.

A classic three-level trophobiotic interaction involving ants is the mutualism between the ant species Pseudomyrmex ferrugineus and the swollen-thorn acacia, Acacia cornigera. This six-million-year-old relationship is one of the most studied ant/plant mutualisms (Janzen 1966; Ward and Branstetter 2017). In this system, the ants receive

4 shelter in the form of nesting sites within the acacia’s thorns, as well as nutrition in the form of extrafloral nectaries and/or highly palatable leaf tips. In return, the ants defend the acacia from other insect invaders and even competing vegetation (Janzen 1966; Ward and Branstetter 2017).

Although three-level trophobiotic interactions are the most common, novel four- level trophobiotic interactions have been found to exist in systems involving certain species of gall wasps. This relationship involves all levels mentioned above – the primary producer, ants or other insect attendants attracted to the produced nectar, and parasitoids or inquilines attempting to attack the gall. However, this interaction boasts the addition of a fourth member: the gall maker (Nicholls et al. 2017). The gall maker’s host is the primary producer – most commonly, an oak species that otherwise cannot produce nectar on its own, as it lacks extrafloral nectaries (Washburn 1984).

In addition to producing the nectar that facilitates these trophobiotic interactions, gall wasp species exhibit another adaptation thought to increase their success -

‘clustering’ of galls, also known as compound galling (Washburn 1984; Yoshihisa 1992;

Seibert 1993; Eckberg and Cranshaw 1994; Nicholls et al. 2017). While some species of

Cynipidae induce solitary galls, many gall wasps form aggregates of multiple galls next to each other in one highly localized small area on the host plant. These are formed by a single female gall wasp ovipositing multiple eggs at a time in the same area (Nicholls et al. 2017). In the gall wasp genus Disholcaspis, clusters of galls may total up to a dozen or more (Yoshihisa 1992). Previous studies have found positive correlations between clustered galls and gall wasp success (Yoshihisa 1992; Inouye and Agrawal 2004), and predict that large clusters of galls are more attractive to helpful insect visitors than

5 solitary galls (Yoshihisa 1992; Nicholls et al. 2017). It has also been suggested that these clusters are a mandatory prerequisite that somehow better enables nectar production

(Nicholls et al. 2017).

Besides gall clustering, the size of galls themselves is also subject to selective pressure. Larger galls may be more successful depending on the types of parasitoids present (Fernandes et al. 1999). This hypothesized mechanism suggests that oviposition by the parasitoid female will become more difficult as the gall diameter increases. Since the attacking female parasitoid’s ovipositor must reach the central chamber of the gall to successfully lay her egg, larger galls may provide better protection for the host larvae by making this attack more difficult (Weis et al. 1985). An additional advantage of large galls is that ants have been observed to tend them more often than their smaller counterparts, possibly because their increased surface area provides more nectar for feeding (Fernandes et al. 1999).

It is through many unique adaptations that the galls of Disholcaspis quercusmamma, the gall wasp species of interest in this study, may have gained the capability to promote an ‘enemy-free space,’ the basis of the ‘enemy hypothesis,’ which postulates that these adaptations are the product of an evolutionary arms race between gall wasps and their enemies (Stone and Schönrogge 2003). Though galls offer some protection for the gall makers’ larvae, the sessile nature of the larvae makes them susceptible to parasitoid attack. In theory, morphological modifications of the galls that reduce the success of these attacks and produce an increase in gall wasp survival should be favored evolutionarily (Stone and Schönrogge 2003).

The nectar-producing galls of Disholcaspis quercusmamma commonly occur on a variety of oak species during late summer in much of North America (Figure 1). The hosts of D. quercusmamma all belong to the groups of oaks known as ‘white oaks’ in the genus Quercus section Quercus, most commonly, swamp white oak (Q. bicolor), overcup oak (Q. lyrata), and bur oak (Q. macrocarpa) (McEwen et al. 2014; Nixon 2020). The gall wasps induce large, woody galls on the twigs and buds of the tree (McEwen et al.

2014). D. quercusmamma exhibits an alternation of generations where an asexual generation alternates with a sexually reproducing generation (Melika et al. 2013;

McEwen et al. 2014). During late fall and early winter, the adult asexual females of

Disholcaspis quercusmamma emerge from the nectar-secreting twig galls and oviposit into the tree’s petioles. This oviposition event gives rise to small bud galls that emerge within the petiole during the late spring and early summer. The sexual generation of both male and female D. quercusmamma eclose from these bud galls. The females of the sexual generation must mate in order to oviposit, unlike the parthenogenic females of the asexual generation (McEwen et al. 2014). After mating, the sexual females oviposit into the new twig growth of the tree during the summer, after which the round twig galls containing the growing asexual female larvae emerge in the late summer and early fall and begin to secrete nectar. The galls mature by late fall and early winter, the asexual females then emerging from the twig galls, ovipositing into the tree’s petioles, and beginning the cycle anew (McEwen et al. 2014).

The insect community that is associated with the galls of Disholcaspis quercusmamma is closely tied to both the rate of both parasitism and the successful emergence of the gall makers. The group that has the most impact upon parasitism rates

7 are the many parasitoid and inquiline species that target D. quercusmamma galls specifically. These species include the chalcid wasp genera Sycophila, Torymus,

Mesopolobus, and Pteromalus (Figure 2) (McEwen et al. 2014). This gall-associated community also includes the insects attracted to the nectar produced by the galls, which may offer protection from parasitism and lead to a higher rate of successful gall maker emergence.

The most well-studied of the gall attendants are ant species (Yoshihisa 1988;

Inouye and Agrawal 2004). Ants have been observed to aggressively defend the nectar, which plays a role in discouraging parasitoids and inquilines from attacking the sessile galls. However, many other insects have been observed reaping the reward of nectar from

D. quercusmamma galls. The types of insects attracted to the nectar-producing galls vary widely in their size and behavior and include representatives from many insect orders.

Though they may not be as aggressive in behavior as the ant attendants, the sheer size of some visiting insects (large paper wasps, butterflies, moths, and large hemipterans such as those in the order Reduviidae) cannot be dismissed as another possible deterrent for potential parasitoids scouting for available galls in which to oviposit.

The sites of this study, located within Central and East Central Illinois, contain rich communities of galling insects. These regions in Illinois are located along the Central

Corn Belt Plains, which is dominated by both agriculture and hardwood forests; common vegetation includes beech (Fagus), maple (Acer), poplar (Populus), sycamore (Planatus), elm (Ulmus), hickory (Carya), ash (Fraxinus), walnut (Juglans), and numerous species of oak (Quercus) . Though oak (Quercus) species in this area contain some of the most diverse galling communities, other plants in the region are also common galling sites.

Goldenrod (Solidago) and rosinweed (Silphium), commonly found in the prairie ecosystem of Illinois, also play host to galling insects, including cynipids. For example, goldenrod (Solidago) is a host of the tephritid fly Eurosta solidaginis (Abrahamson et al.

1989), while rosinweed hosts the gall wasp genus Antistrophus (Tooker and Hanks 2004), both of which produce galls within the stem of their host plant.

The present study is designed to test the hypothesis that effective enemy-free space is created because of the nectar of the asexual summer galls of Disholcaspis quercusmamma. Several studies involving such systems have yielded results supporting the effectiveness of gall nectar in generating enemy free space (Washburn 1984;

Yoshihisa 1992; Fernandes et al. 1999; Inouye and Agrawal 2004). However, these studies only focused on the ant-gall interaction in this system. Other gall-attending insects have been mentioned as observations in previous research, but their influence has not been explicitly included in experimental treatments (Gagliardi and Wagner 2016;

Aranda-Rickert et al. 2017). As such, current literature describing the added effect of other insect consumers, or those that feed upon the nectar produced by the galls, is lacking. Additionally, most studies focused on other species of Disholcaspis; a study of this type involving D. quercusmamma has yet to be conducted. This study seeks to address how characteristics of the nectar-secreting galls of D. quercusmamma, including gall size and clustering, might create an enemy-free space that attracts visiting insects, ultimately reducing parasitism of the galls and increasing success of the gall makers.

MATERIALS AND METHODS

Study sites

Study sites were chosen by observing the abundance of the previous year’s galls on mature oak trees in two areas of central Illinois – Charleston in Coles County and

Barclay in Sangamon County, near Springfield (Figure 3). The first study site contained a specimen of swamp white oak (Quercus bicolor) located in Fox Ridge State Park in

Charleston, Illinois (Site 1); from this point forward, the tree will be referred to as Tree

A. The second study site contained a single specimen of bur oak (Quercus macrocarpa) located on private property in Barclay, Illinois (Site 2), and will be referred to as Tree B.

An additional specimen of Q. bicolor, approximately twelve meters from Tree 1 at Site 1 in Fox Ridge State Park, was also treated, albeit minimally, as it began to develop a modest number of galls late in the season. This tree will be referred to as Tree C.

Treatments

Fresh galls of the asexual generation of Disholcaspis quercusmamma began to first appear on Tree B in Barclay in mid-July. The galls appeared small and green and had not yet begun to produce nectar. At this point, 50 terminal branches upon which galls had begun to develop were chosen according to length (each was approximately 1’) and randomly tagged with numbers 1-50. A random number generator mobile application, Random Number Generator (UX Apps), was used to assign treatments to each numbered branch. To exclude all non-flying insects such as ants, Tanglefoot Insect

Barrier® (Contech Enterprises Inc) was applied to section of branch wrapped with

Tangle-Guard Tree Wrap® (Contech Enterprises Inc) approximately 35 cm from the branch’s end. To exclude large insects such as large wasps, stick insects, and

Lepidoptera, bags made of 13 mm mesh garden netting were affixed around the galled section of branch and secured with a cable tie (Figure 4). To exclude all insects larger than 1 mm in length, fruit bags of very fine mesh were affixed around the galled section of branch and secured with a cable tie (Figure 5). The total treatments included 10 fruit bags, 20 mesh bags, 10 Tanglefoot® barriers, and 10 controls, to which no treatment was applied.

This process was repeated for Tree A in Charleston when its galls appeared in

August, and again on Tree C in September when its galls began to appear. Fifty total treatments of the same proportions as Tree B were applied to Tree A. However, Tree C did not have as many newly galled branches as Trees A and B, so only 12 total treatments were applied. These included 3 exclusion bags, 3 mesh bags, 3 Tanglefoot® barriers, and

3 controls.

Observations

Once the treatments were applied, Site 1 was monitored at least once per week and Site 2 was monitored at least once per month. The exclusionary efficacy of the treatments was monitored, as well as the types of insects visiting the galls that were actively producing nectar.

Data collection

The galls were considered to be mature when they ceased production of nectar and began to darken in color. Small holes also began to appear on some galls, indicating that the mature Disholcaspis wasps had eclosed and subsequently emerged from the gall.

Maturation and collection for Trees A and B occurred in mid-October. For the additional, minimally treated tree at Site 1 (Tree C), maturation and collection occurred in early

November. At this point, all galls on the treated branches were collected. Shortly after collection, the galls were placed into small plastic cups covered with cheesecloth to prevent emerging insects from escaping. Each cup corresponded to one branch. The cups were kept in dark, room temperature conditions for approximately six weeks and checked twice per week. Any wasp emergence or other insect activity was recorded, and specimens that had emerged were preserved in vials containing 100% ethanol.

By early December, no wasps had emerged for over two weeks, so the galls were dissected to observe their contents (Figure 6). The number of galls per branch was recorded as the cluster size. Each gall was measured and its diameter in millimeters was recorded. An average gall diameter was calculated for each branch. Any specimens found inside the galls were preserved in vials containing 100% ethanol.

Upon dissection of the galls when all Disholcaspis should have emerged, or be close to emergence, galls were classified into three categories: 1) Parasitized galls, which contained either larvae or pupae in the gall wasp chamber, or hairy molts; 2) Inquilined galls, which contained larvae in the peripheral tissue of the gall; and 3) Galls not attacked, where either parasitoids or inquilines were not detected and only the gall maker larva, pupa, or adult form was present.

Statistical analyses

A Kruskal-Wallis one-way analysis of variance was conducted to examine effects of the three treatments on the success of Disholcaspis quercusmamma. A Kendall rank correlation test was calculated to examine potential correlations between gall diameter and gall cluster size and the number of parasitoids, inquilines, and successful gall wasps.

RESULTS

Observations

The applied treatments were regularly monitored to assess their effectiveness. The

Tanglefoot barrier was observed to be effective in excluding non-winged insects like worker ants from the galls but allowed other winged insects free access. The mesh cages were observed effectively excluding larger winged insects such as wasps and

Lepidoptera. No insects were observed as being able to gain access to the branches treated with fruit bags; these effectively excluded all arthropods except, possibly, the smallest of parasitoids.

Many insect species were observed visiting the galls while the latter were actively producing nectar (Table 1). The visiting ants belonged to multiple ant species

(Hymenoptera: Formicidae), of which carpenter ants (Camponotus pennsylvanicus) were the most common and were very common at both sites of the experiment (Figure 7).

Other Hymenopterans observed included honeybees (genus Apis), sweat bees (family

Halictidae), and wasps (family Vespidae) (Figures 8 through 10). In addition, many species of other insect groups were found to visit the nectar-secreting galls just as frequently. Many of these visitors were beetles (Coleoptera), including weevils and the

Asian lady beetle (Harmonia axyridis) (Figure 11). Visitors also included several species of flies (Diptera) (Figure 11). Hemipterans were also common at both sites – many wheel bugs (Arilus cristatus) (Figure 12), leaf-footed bugs (family Coreidae), and brown marmorated stink bugs (Halyomorpha halys) were observed feeding on the galls. Even moths and butterflies (Lepidoptera) were attracted to galls (Figure 13).

The majority of the insects dissected from the galls were in their larval stage and identification to the species level was impossible without DNA sequencing. However, since gall wasp larvae do not exhibit setae, the larvae were able to be identified as either gall wasp (absence of setae) or non-gall wasp (presence of setae) (Table 2). Parasitoids were pooled collectively in the dataset; the possibility existed that some had emerged before the galls were collected, and accurate numbers of each parasitoid species that invaded the galls would have been impossible to estimate. Additionally, two mature black vine weevils (Otiorhynchus sulcatus) were dissected from one gall from each site; these specimens were classified as inquilines.

Statistical analyses

Although 112 branches were originally treated, only 62 of them could be included in the analysis. Multiple reasons existed for omitting branches, including infestations of black sooty mold (Figure 14), which degraded the specimens inside to the point that they were unable to be identified. It is also possible that these degraded samples were old galls from previous years that had been mistaken as new galls. Also, many branches ultimately did not grow galls to maturation even though small, immature galls were originally seen early in the season. This was most commonly the case with the tree located at the Barclay site (Tree B).

The average gall size was found to be 10.6 mm, with gall size ranging from 5.8 to

16.5 mm; the average number of galls in a cluster was ~4, with a range of 1 to 11. No significant difference in the ultimate success of Disholcaspis quercusmamma was evidenced among the treatments in a Kruskal Wallis one-way analysis of variance (Figure

15; H=3.3, 3 d.f., p=0.35). No significant difference in the number of parasitized galls

14 was evidenced among the treatments in a Kruskal Wallis one-way analysis of variance

(Figure 16: H=0.89, 3 d.f., p=0.83). No significant difference in the number of inquilined galls was evidenced among the treatments in a Kruskal Wallis one-way analysis of variance (Figure 17; H=1.07, 3 d.f., p=0.78). Inquiline invasion and gall diameter (Figure

18; rτ=-0.35, 60 d.f., p<0.005) were negatively correlated, while inquiline invasion and gall cluster size were positively correlated (Figure 19; rτ=0.46, 60 d.f., p<0.005), as evidenced by a Kendall rank correlation coefficient. Additionally, there were positive correlations between gall wasp success and gall diameter (Figure 20; rτ=0.19, 60 d.f., p=0.04) and gall cluster size (Figure 21; rτ=0.56, 60 d.f., p<0.005). Parasitoid invasion was not correlated with either gall diameter (Figure 22; rτ=-0.05, 60 d.f., p=0.66) or gall cluster size (Figure 23; rτ=0.16, 60 d.f., p=0.15).

DISCUSSION

The factors relating to the success of Disholcaspis quercusmamma examined in this experiment show similar correlations as previous studies, even though the effect of exclusion treatment on gall maker success (Figure 15), parasitism (Figure 16), and inquilinism (Figure 17) showed no significant difference. Larger clusters of galls positively contributed in a significant manner to the eventual success of the gall maker

(Figure 21), an effect also supported by Yoshihisa (1992) and Inouye and Agrawal

(2004). Larger gall diameter was also positively correlated with the success of D. quercusmamma (Figure 20), a similar finding to Fernandes et. al (1999). This effect was also examined by Inouye and Agrawal (2004), however, their result differed in that they found no significant correlations between gall size and gall maker success.

Inquiline invasion was found to negatively correlate with gall diameter (Figure

18). Previous studies have demonstrated this same effect in regard to both inquiline and parasitoid attack (Weis et al. 1985; Fernandes et al. 1999; Craig et al. 2007). This effect was studied by Eckberg and Cranshaw (1994), who found that the galls of D. quercusmamma that were attacked by parasitoids were significantly smaller than galls that contained gall maker larvae. Inquilines oviposit into the gall parenchyma rather than the central gall chamber, as is required by parasitoids. Therefore, the reasoning behind the preference for smaller galls is less clear for inquilines than it is for parasitoids. It is possible that larger galls may be more heavily guarded by insect opportunists, discouraging inquilines from oviposition.

On the contrary, inquiline invasion had a strong positive correlation with the number of galls in a cluster (Figure 19), while parasitoid invasion did not (Figure 23).

Many galls in one area, like the larger clusters of nearly a dozen galls observed during this experiment, may provide an ideal condition for inquiline attack – the female inquiline has a wider selection of galls available in one area in which to oviposit. Gall maker success was also positively correlated with gall cluster size (Figure 21) and may also be attributed to the fact that inquiline invasion, though common among the clustered galls, may not be as devastating to the gall maker as parasitoid attack. Inquilines, though detrimental to the growing larvae’s nutritional needs, do not often parasitize the larvae, as is the case for parasitoids.

Though large galls in clusters may garner more protection by the insects attracted to the extrafloral nectary, inquilines may have more of an opportunity to oviposit simply due to the time required to do so. Since inquilines only need to oviposit to the outer margins of the gall, rather than the central chamber like in parasitoids, their oviposition is likely to take a shorter amount of time. This leaves more opportunity for a successful attack than in the lengthier process required of parasitoids, which could be more prone to interruption by ant guards or other insect visitors looking to claim the nectar on the gall surface. This fact could explain why inquilined galls were much more numerous than parasitized galls in this study. This temporal factor could warrant further investigation in future studies on the enemy hypothesis.

There was no significant effect of gall diameter (Figure 22) or cluster size (Figure

23) on parasitoid invasion. It is likely that the analysis was not as powerful as possible due to the small sample size of galls attacked by parasitoids (Table 2). Inquilined galls occurred much more often within the galls successfully sampled, and as such, the statistical testing regarding the morphological effects of galls on inquiline attack may be

17 more rigorous than those regarding parasitoid attack (Table 2). Nearly half of the experimental treatments were excluded from the dataset due to either damage by black sooty mold deeming the contents unidentifiable, the previous season’s galls being mistaken for new gall growth, or galls that were present early in the season not fully developing. This almost certainly eliminated valuable data involving parasitoid attack upon the galls. As previously stated, existing studies examined the effect of gall clustering and individual gall size on parasitoid attack have found a negative correlation

(Weis et al. 1985; Fernandes et al. 1999; Craig et al. 2007); a larger sample size would be essential to reveal if the population studied in this experiment demonstrate similar effects.

As was expected, ants were observed to be the most common visitors of the galls, but many other species of insects were also present (Table 1). Rarely during observation did a gall being actively tended by other insects also have potential parasitoids or inquilines present. Usually, parasitoids and inquilines were only spotted on galls that showed no other insect activity at the time. Though the experimental design did not quantify these observations, the insect activity observed shows how the enemy hypothesis may function in regard to the nectar-producing galls of Disholcaspis quercusmamma.

Defensive insect behaviors were observed at both study sites and included voracious protection of the nectar secreting galls, directed toward both insect intruders and human observers. These behaviors were most common among the Hymenopteran visitors, including the eusocial wasps (Vespidae) and ants (Formicidae). Other insects, not just ants, may play a role in the creation of an enemy free space for the gall makers – however, future studies would require a larger sample size and more thorough observation of visiting insect behavior to fully reveal the depth of this relationship.

It is evident that evolutionary strategies among the gall wasps, such as increasing gall diameter and gall cluster size, may contribute to the prevention of potentially harmful invaders such as parasitoids and inquilines. This system has important implications from both an ecological and evolutionary standpoint. Deeper knowledge of any evolutionary arms race, such as the one studied here, can aid in a better understanding of evolutionary pressure and anti-predator adaptation. This experiment provided encouraging results that warrant further study into the four-level trophobiotic interaction among the gall- associated community of Disholcaspis quercusmamma.

REFERENCES

Abrahamson WG, Melika G, Scrafford R, Csóka G. 1998. Gall-inducing insects provide insights into plant systematic relationships. Am J Bot. 85(8):1159–1165. doi:10.2307/2446348.

Abrahamson WG, Sattler JF, Mccrea KD, Weis, Arthur E. 1989. Variation in selection pressure on the goldenrod gall fly and the competitive interactions of its natural enemies.

Oecologia. 79(1):15–22.

Aebi A, Schönrogge K, Melika G, Alma A, Bosio G, Quacchia A, Picciau L, Abe Y,

Moriya S, Yara K, et al. 2007. Parasitoid recruitment to the globally invasive chestnut gall was Dryocosmus kuriphilus. Gall Arthropods Their Assoc. 103–121. doi:10.1007/4-

431-32185-3_9.

Aranda-Rickert A, Rothen C, Diez P, González AM, Marazzi B. 2017. Sugary secretions of wasp galls: A want-to-be extrafloral nectar? Ann Bot. 120(5):765–774. doi:10.1093/aob/mcx075.

Craig TP, Itami JK, Horner JD. 2007. Geographic variation in the evolution and coevolution of a tritrophic interaction. Evolution. 61(5):1137–1152. doi:10.1111/j.1558-

5646.2007.00099.x.

Dreger-Jauffret F, Shorthouse JD. 1992. Diversity of gall-inducing insects and their galls.

Biol Insect-induced Gall. 8–33.

Eckberg TB, Cranshaw WS. 1994. Occurrence of the oak rough bulletgall wasp,

Disholcaspis quercusmamma (Walsh) (Hymenoptera: Cynipidae), as a street tree pest in

Colorado. J Kansas Entomol Soc. 67(3):290-293.

Fernandes GW, Fagundes M, Woodman RL, Price PW. 1999. Ant effects on three- trophic level interactions: plant, galls, and parasitoids. Ecol Entomol. 24:411–415.

Ferracini C, Ferrari E, Pontini M, Saladini MA, Alma A. 2019. Effectiveness of Torymus sinensis: a successful long-term control of the Asian chestnut gall wasp in Italy. J Pest Sci

(2004). 92(1):353–359. doi:10.1007/s10340-018-0989-6.

Gagliardi BL, Wagner DL. 2016. “Northern” oak hairstreak (Satyrium favonius ontario)

(Lepidoptera: Lycaenidae): Status survey in Massachusetts, false rarity, and its use of non-nectar sugar resources. Ann Entomol Soc Am. 109(4):503–512. doi:10.1093/aesa/saw012.

Hardy NB, Cook LG. 2010. Gall-induction in insects: Evolutionary dead-end or speciation driver? BMC Evol Biol. 10(1):19–21. doi:10.1186/1471-2148-10-257.

Inouye BD, Agrawal AA. 2004. Ant mutualists alter the composition and attack rate of the parasitoid community for the gall wasp Disholcaspis eldoradensis (Cynipidae). Ecol

Entomol. 29(6):692–696. doi:10.1111/j.0307-6946.2004.00652.x.

Janzen DH. 1966. Coevolution of mutualism between ants and acacias in Central

America. Evolution. 20(3):249–275.

Liljeblad J, Ronquist F. 1998. A phylogenetic analysis of higher-level gall wasp relationships (Hymenoptera: Cynipidae). Syst Entomol. 23:229–252.

Matošević D, Lacković N, Kos K, Kriston E, Melika G, Rot M, Pernek M. 2017. Success of classical biocontrol agent Torymus sinensis within its expanding range in Europe. J

Appl Entomol. 141(9):758–767. doi:10.1111/jen.12388.

McEwen C, Digweed S, Nicholls JA, Cranshaw W. 2014. Description and biology of the sexual generation of Disholcaspis quercusmamma (Walsh and Riley) (Hymenoptera:

Cynipidae), with notes on associated parasitoids. Proc Entomol Soc Washingt.

116(3):294–310. doi:10.4289/0013-8797.116.3.294.

Melika G, Buss EA, Nicholls JA, Bird JP, Stone GN. 2013. Life cycle of Disholcaspis quercusvirens (Hymenoptera: Cynipidae) with a description of the sexual generation.

Florida Entomol. 96(3):991–1001. doi:10.1653/024.096.0337.

Nicholls JA, Melika G, Stone GN. 2017. Sweet tetra-trophic interactions: Multiple evolution of nectar secretion, a defensive extended phenotype in cynipid gall wasps. Am

Nat. 189(1):67–77. doi:10.1086/689399.

Nixon KC. 2020. Fagaceae. eFlora, Flora North Am. [accessed 2020 Nov 6]. http://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=10338.

Omernick J, Gallant A. Ecoregions of the upper Midwest states. 1988. U.S.

Environmental Protection Agency.

Poff AC, Haynes KJ, Szymanski M, Back D, Williams MA, Cronin JT. 2002. Bird predation and the host-plant shift by the goldenrod stem galler. Can Entomol.

134(2):215–227. doi:10.4039/Ent134215-2.

Price PW, Fernandes GW, Waring GL. 1987. Adaptive nature of insect galls. Environ

Entomol. 16(1):15–24. doi:10.1093/ee/16.1.15.

Ritchie AJ. 1984. A review of the higher classifiction of the inquiline gall wasps

(Hymenoptera: Cynipidae) and a revision of of the Nearctic species of Periclistus

(Foerster) [dissertation]. Ottawa, Ontario. Carleton University.

Ronquist F. 1994. Evolution of parasitism among closely related species: phylogenetic relationships and the origin of inquilinism in gall wasps (Hymenoptera, Cynipidae).

Evolution. 48(2):241–266. doi:10.2307/2410091.

Ronquist F, Liljeblad J. 2001. Evolution of the gall wasp-host plant association.

Evolution. 55(12):2503–2522.

Ronquist F, Nieves-Aldrey JL, Buffington ML, Liu Z, Liljeblad J, Nylander JAA. 2015.

Phylogeny, evolution and classification of gall wasps: The plot thickens. PLoS One.

10(5):1–41. doi:10.1371/journal.pone.0123301.

Ronquist F, Nylander JAA, Vårdal H, Nieves-Aldrey JL. 2018. Life history of Parnips and the evolutionary origin of gall wasps. J Hymenopt Res. 65:91–110. doi:10.3897/jhr.65.24115.

Seibert TF. 1993. A nectar‐secreting gall wasp and ant mutualism: selection and counter‐ selection shaping gall wasp phenology, fecundity and persistence. Ecol Entomol. 18:247–

253. doi:10.1111/j.1365-2311.1993.tb01097.x.

Stone GN, Hernandez-Lopez A, Nicholls JA, Di Pierro E, Pujade-Villar J, Melika G,

Cook JM. 2009. Extreme host plant conservatism during at least 20 million years of host plant pursuit by oak gallwasps. Evolution. 63(4):854–869. doi:10.1111/j.1558-

5646.2008.00604.x.

Stone GN, Schönrogge K. 2003. The adaptive significance of insect gall morphology.

Trends Ecol Evol. 18(10):512–522. doi:10.1016/S0169-5347(03)00247-7.

Tooker JF, Hanks LM. 2004. Impact of prescribed burning on endophytic insect communities of prairie perennials (Asteraceae: Silphium spp.). Biodivers Conserv.

13(10):1875–1888. doi:10.1023/B:BIOC.0000035872.24201.30.

Tscharntke T. 1992. Cascade effects among four trophic levels: bird predation on galls affects density- dependent parasitism. Ecology. 73(5):1689–1698.

Ward PS, Branstetter MG. 2017. The acacia ants revisited: Convergent evolution and biogeographic context in an iconic ant/plant mutualism. Proc R Soc B Biol Sci.

284(1850). doi:10.1098/rspb.2016.2569.

Washburn JO. 1984. Mutualism between a cynipid gall wasp and ants. Ecology.

65(2):654–656.

Washburn JO, Cornell H V. 1979. Chalcid parasitoid attack on a gall wasp population

(Acraspis hirta (Hymenoptera: Cynipidae)) on Quercus prinus (Fagaceae). Can Entomol.

111(4):391–400. doi:10.4039/Ent111391-4.

Washburn JO, Cornell H V. 1981. Parasitoids, patches, and phenology: Their possible role in the local extinction of a cynipid gall wasp population. Ecology. 62(6):1597–1607.

Weis AE, Abrahamson WG, McCrea KD. 1985. Host gall size and oviposition success by the parasitoid Eurytoma gigantea. Ecol Entomol. 10:341–348. doi:10.1111/j.1365-

2311.1985.tb00730.x.

Yoshihisa A. 1988. Trophobiosis between the gall wasp, Andricus symbioticus, and the gall-attending ant, Lasius niger. Appl Ent Zool. 23(1):41–44.

Yoshihisa A. 1992. The advantage of attending ants and gall aggregation for the gall wasp Andricus symbioticus (Hymenoptera: Cynipidae). Oecologia. 89:166–167.

Figure 1. The nectar-producing galls of Disholcaspis quercusmamma.

Figure 2. An inquiline of Disholcaspis quercusmamma.

Figure 3. Map of study sites in Central and East Central Illinois.

Figure 4. Mesh bags made of ½” mesh garden netting.

Figure 5. Fruit bags secured around the galled portion of the branches.

Figure 6. A Disholcaspis quercusmamma gall cut open to reveal the central gall chamber in which the larva resides.

Figure 7. A carpenter ant (Camponotus sp.) visiting the nectar-producing galls.

Figure 8. A northern paper wasp (Polistes fuscatus) visiting the nectar-producing galls. Courtesy Zhiwei Liu.

Figure 9. A cuckoo wasp (Chrysis sp.) and housefly (family Muscidae) visiting the nectar-producing galls. Courtesy Zhiwei Liu.

Figure 10. A great black wasp (Sphex pensylvanicus) visiting the nectar-producing galls. Courtesy Zhiwei Liu.

Figure 11. An Asian lady beetle (Harmonia axyridis) and several houseflies (family Muscidae) visiting the nectar-producing galls. Courtesy Zhiwei Liu.

Figure 12. A wheel bug (Arilus cristatus) visiting the nectar-producing galls.

Figure 13. Red-spotted purple butterflies (Limenitis arthemis) visiting the nectar- producing galls. Courtesy Zhiwei Liu.

Figure 14. Black sooty mold on Disholcaspis quercusmamma galls late in the season, after they had ceased production of nectar.

Figure 15. A boxplot of the Kruskal-Wallis one-way ANOVA comparing the number of successful galls for each exclusion treatment (H=3.30, 3 d.f., p=0.35). Mesh cage excluded larger insects, but not ants; Tanglefoot excluded ants, but not larger insects; Fruit bag excluded both ants and larger insects.

Figure 16. A boxplot of the Kruskal-Wallis one-way ANOVA comparing the number of parasitized galls for each exclusion treatment (H=0.89, 3 d.f., p=0.83). Mesh cage excluded larger insects, but not ants; Tanglefoot excluded ants, but not larger insects; Fruit bag excluded both ants and larger insects.

Figure 17. A boxplot of the Kruskal-Wallis one-way ANOVA comparing the number of inquilined galls for each exclusion treatment (H=1.07, 3 d.f., p=0.78). Mesh cage excluded larger insects, but not ants; Tanglefoot excluded ants, but not larger insects; Fruit bag excluded both ants and larger insects.

Figure 18. A scatterplot comparing the number of inquilines found within the galls and gall diameter (rτ=-0.35, 60 d.f., p<0.005).

Figure 19. A scatterplot comparing the number of inquilines found within the galls and the number of galls in a cluster (rτ=0.46, 60 d.f., p<0.005).

Figure 20. A scatterplot comparing the number of successful D. quercusmamma found within the galls and gall diameter (rτ=0.19, 60 d.f., p=0.04).

Figure 21. A scatterplot comparing the number of successful D. quercusmamma found within the galls and the number of galls in a cluster (rτ=0.56, 60 d.f., p<0.005).

Figure 22. A scatterplot comparing the number of parasitoids found within the galls and gall diameter (rτ=-0.05, 60 d.f., p=0.66).

Figure 23. A scatterplot comparing the number of parasitoids found within the galls and the number of galls in a cluster (rτ=0.16, 60 d.f., p=0.15).

Order Families Species (if identified) Coleoptera Chrysomelidae, Diabrotica undecimpunctata, Coccinellidae, Harmonia axyridis Curculionidae Diptera Calliphoridae, Musca domestica Muscidae, Syrphidae, Tabanidae Hemiptera Acanaloniidae, Arilus cristatus, Chinavia Coreidae, Lygaeidae, hilaris, Halyomorpha halys, Membracidae, Oncopeltus fasciatus Pentatomidae, Reduviidae Hymenoptera Apidae, Chrysididae, Apis sp., Camponotus Formicidae, Halticidae, pennsylvanicus, Chrysis sp., Sphecidae, Vespidae Dolichovespula maculata, Lasius sp., Polistes dominula, Polistes fuscatus, Sphex pensylvanicus Lepidoptera Noctuidae, Limenitis arthemis, Polygonia Nymphalidae, Pieridae interrogationis Phasmatodea Diapheromeridae Diapheromera femorata

Table 1. A summary of the insect visitors of the Disholcaspis quercusmamma galls.

Fox Ridge (Site 1) Barclay (Site 2) Total Branches sampled 57 8 62 Galls collected 239 33 272 D. quercusmamma 107 18 125 Inquilines 62 7 69 Parasitoids 20 7 27

Table 2. A summary of the contents of the D. quercusmamma galls (n=272) on treated branches and successfully sampled branches (n=62).