Host Selection by the Walnut Twig Beetle, Pityophthorus

HOST SELECTION BY THE WALNUT TWIG BEETLE,

PITYOPHTHORUS JUGLANDIS,

IN CALIFORNIA

______

A Thesis

Presented

to the Faculty of

California State University, Chico

______

In Partial Fulfillment

of the Requirements for the Degree

Master of Science

Biological Sciences

______

Irene Lona

Spring 2019

HOST SELECTION BY THE WALNUT TWIG BEETLE,

PITYOPHTHORUS JUGLANDIS,

IN CALIFORNIA

A Thesis

Irene Lona

Spring 2019

APPROVED BY THE INTERIM DEAN OF GRADUATE STUDIES:

______Sharon Barrios, Ph.D.

APPROVED BY THE GRADUATE ADVISORY COMMITTEE:

______Donald G. Miller, Ph.D., Chair

______Colleen Hatfield, Ph.D.

______Richard Rosecrance, Ph.D.

______Steven J. Seybold, Ph.D.

TABLE OF CONTENTS

PAGE

List of Tables ...... iv

List of Figures ...... v

Abstract ...... vi

CHAPTER

I. Host selection behavior mediated by differential landing rates of the walnut twig beetle, Pityophthorus juglandis Blackman (Coleoptera: Scolytidae), on two western North American walnut species, Juglans californica and Juglans major ...... 1

Abstract ...... 2 Introduction ...... 4 Materials and methods ...... 7 Results...... 14 Discussion ...... 22 Acknowledgements ...... 27 References ...... 28

II. Effect of water stress on the susceptibility of commercial English walnut, Juglans regia, to the walnut twig beetle, Pityophthorus juglandis ...... 45

Abstract ...... 46 Introduction ...... 48 Materials and methods ...... 50 Results...... 58 Discussion ...... 63 Acknowledgements ...... 68 References ...... 68

iii

LIST OF TABLES

TABLE PAGE

1.1 Synthesis of measures of host susceptibility to walnut twig beetle, Pityophthorus juglandis, feeding and Geosmithia morbida inoculation from the literature ..... 33 1.2 Subcortical insect associates and predators of walnut twig beetle, Pityophthorus juglandis, trapped during flight monitoring at the NCGR ...... 34 1.3 Subcortical insect associates and predators of walnut twig beetle, Pityophthorus juglandis, trapped during flight monitoring at the ATRC...... 35 1.4 Landing rates of insects associated with Pityophthorus juglandis, Juglans spp., and other hardwood trees on acetate sticky sheets ...... 36 1.A Accession numbers of branches collected from Juglans californica and Juglans major trees ...... 43 2.1 Temperature, humidity, and baseline potential recorded during the dates when stem water potential (SWP) was measured on English walnut ...... 73 2.2 Number of main branches and twigs for each diameter class and number of sections with Pityophthorus juglandis entrance/emergence holes and Nathrius brevipennis or Gildoria sp. emergence holes ...... 74 2.3 Total number of insects trapped in two funnel traps each baited with an ethanol lure and two unbaited funnel traps (controls) ...... 75

LIST OF FIGURES

FIGURE PAGE

1.1 Branch sections of two Pityophthorus juglandis host plants used in the flight landing assays 1-4: Juglans major and Juglans californica ...... 37 1.2 Cardboard tube used as negative control ...... 37 1.3 Juglans branch section surrounded by a mesh screen bag and adhesive-coated acetate sheet ...... 38 1.4 Walnut twig beetle, Pityophthorus juglandis, flight at NCGR and ATRC...... 39 1.5 Weekly landing rates of male and female walnut twig beetle, Pityophthorus juglandis, female Xyleborinus saxeseni, female Hypothenemus eruditus, and male and female Nathrius brevipennis (Assays 1-4) ...... 41 1.6 Weekly landing rates of male and female walnut twig beetle, Pityophthorus juglandis, female Xyleborinus saxeseni, and female Hypothenemus eruditus (Assay 5) ...... 42 2.1 Juglans regia branch with lure of male Pityophthorus juglandis aggregation pheromone ...... 76 2.2 Juglans regia branch divided into sections of 27 cm ...... 77 2.3 Branch of Juglans regia with Pityophthorus juglandis entrance and emergence holes ...... 78 2.4 Twig from Juglans regia branch with Nathrius brevipennis and Gildoria sp. emergence holes ...... 78 2.5 Stem water potential of trees in the low water stress group, high water stress group, and control group ...... 79 2.6 Stem water potential of baited and unbaited branches ...... 80 2.7 Frequency distribution of Pityophthorus juglandis entrance or emergence holes by distance from the aggregation pheromone lure ...... 81 2.8 Number of Pityophthorus juglandis entrance or emergence holes on four aspects (top, left, right, and bottom) ...... 82 2.9 Distribution (by branch or twig diameter) of entrance or emergence holes from Pityophthorus juglandis, and emergence holes from Nathrius brevipennis and Gildoria sp...... 83 2.10 Number of Nathrius brevipennis emergence holes in relation to the distance from the lure of male Pityophthorus juglandis aggregation pheromone ...... 85 2.11 Number of Gildoria sp. emergence holes in relation to the distance from the lure of male Pityophthorus juglandis aggregation pheromone ...... 86 2.12 Flight response of female Xyleborinus saxeseni and male and female Nathrius brevipennis to two ethanol-baited and two unbaited traps ...... 87

ABSTRACT

HOST SELECTION BY THE WALNUT TWIG BEETLE,

PITYOPHTHORUS JUGLANDIS,

IN CALIFORNIA

Irene Lona

Master of Science in Biological Sciences

California State University, Chico

Spring 2019

Walnut (Juglans) and wingnut (Pterocarya) trees have been declining in response to Thousand Cankers Disease (TCD). The TCD pathogen, Geosmithia morbida, is vectored by a phloeophagous bark beetle, the walnut twig beetle (Pityophthorus juglandis

Blackman). To understand P. juglandis host selection, the following were investigated at two orchards in Northern California: 1) The susceptibility of two walnut species (Juglans californica and J. major) native to the western USA by comparing P. juglandis flight and landing responses to unbaited branch sections and 2) The relationship between English walnut (Juglans regia) tree health, represented by several levels of plant water stress, and

P. juglandis colonization success.

P. juglandis preferred to land on J. californica over J. major. With J. regia, there was no significant difference in P. juglandis activity between branches from trees in high water stress and low water stress groups (both baited with P. juglandis aggregation

pheromone). No P. juglandis colonization activity was observed in unbaited branches from the negative control group (unbaited). Activity by P. juglandis was found primarily on 4 cm diameter branches.

Other invasive bark and ambrosia beetles such as Hypothenemus eruditus and

Xyleborinus saxeseni; and an invasive longhorned beetle, Nathrius brevipennis preferred to land on J. major over J. californica. Collections of H. eruditus and an invasive ambrosia beetle, Xyleborus affinis, in Butte County reflect the northernmost records of these species in California. No significant relationships were found in J. regia between water stress and activity by N. brevipennis and a parasitoid Gildoria sp; however, activity by N. brevipennis and Gildoria sp. was found primarily on secondary 1 cm diameter branches (twigs).

In conjunction with this study, significant flight responses of N. brevipennis and

X. saxeseni to ethanol were recorded in a J. regia orchard. Overall, this study demonstrated that volatile cues from Juglans may influence host selection by P. juglandis and associated insects.

vii

Prepared for Entomologia Experimentalis et Applicata

CHAPTER I

Host selection behavior mediated by differential landing rates of the

walnut twig beetle, Pityophthorus juglandis Blackman (Coleoptera:

Scolytidae), on two western North American walnut species,

Juglans californica and Juglans major

Irene D. Lona1, Donald G. Miller III1, Colleen A. Hatfield1, Richard C. Rosecrance2, Lori J. Nelson3, Jackson P. Audley4, Yigen Chen5 & Steven J. Seybold3,4,*

1California State University Chico, Biological Sciences Department, 400 West First Street, Chico, CA 95929, USA 2California State University Chico, College of Agriculture, 400 West First Street, Chico, CA 95929, USA 3USDA Forest Service, Pacific Southwest Research Station, 1731 Research Park, Davis, CA 95616, USA 4Department of Entomology and Nematology, University of California, One Shields Avenue, Davis CA 95616, USA 5E. & J. Gallo Winery, 600 Yosemite Boulevard, Modesto, CA 95354, USA

*Correspondence: Steven J. Seybold, E-mail: [email protected]

Abstract

The walnut twig beetle, Pityophthorus juglandis Blackman (Coleoptera: Scolytidae), vectors a phytopathogenic fungus, Geosmithia morbida, which causes Thousand Cankers

Disease (TCD) in walnut (Juglans) and wingnut (Pterocarya) trees. We investigated an early point in disease inception in two walnut species (Juglans californica and J. major) native to riparian forests of the western USA by comparing P. juglandis flight and landing responses to small diameter branch sections. Twenty unbaited branch sections

(10 each of J. californica and J. major) were presented in a completely randomized design to populations of P. juglandis at the USDA Agricultural Research Service,

National Clonal Germplasm Repository (NCGR) Juglans collection located at Wolfskill

Experimental Orchards (Winters, California) and at the California State University,

Chico, Agricultural Teaching and Research Center (ATRC, Chico, California).

Concomitantly, weekly flight surveys over a three- to five-year period with aggregation pheromone-baited multiple funnel traps revealed that P. juglandis flight activity- abundance was higher at the NCGR than at the ATRC. For the landing rate assays, adhesive-coated acetate sheets were wrapped around the branch sections and exchanged weekly. Three assays were completed at the NCGR (Assays 1-3), whereas one assay was completed at the ATRC (Assay 4). Landing rates on these traps were compared between

J. californica and J. major. An additional assay (Assay 5) was completed at the NCGR to compare responses to branch sections of J. californica and to similarly-sized cardboard tubes (negative control). All five assays were completed over a two-year span during the three- to five-year weekly flight survey period. Pooled landing rates of male and female

P. juglandis (Assays 1-4) demonstrated a preference by both sexes for J. californica over

J. major. In Assay 5 there was no preference by males or females for J. californica over the negative control, perhaps due to the low flight activity-abundance of P. juglandis during the assay. Females of the invasive fruit-tree pinhole borer (an ambrosia beetle),

Xyleborinus saxeseni, and an invasive bark beetle, Hypothenemus eruditus (both

Scolytidae), showed relatively higher flight responses than P. juglandis during all assays, suggesting higher population densities of these two invasive species at the two orchards or a greater sensitivity to host volatiles. Xyleborinus saxeseni and H. eruditus preferred to land on J. major over J. californica. Similarly, an invasive longhorned beetle, Nathrius brevipennis (Cerambycidae), showed a significant preference for J. major. More male N. brevipennis were trapped than females at both study sites [sex ratio ranged from 10:1

(Assay 3) to 39:1 (Assay 4)], and flight occurred only in the spring and early summer months. Another ambrosia beetle trapped at the NCGR and ATRC, Xyleborus affinis, represented the first records of this species from western North America. In summary, flight responses recorded in some of our assays for P. juglandis and several other subcortical insects on Juglans indicate that host preference by these insects may be determined by long-range olfactory cues that do not involve pheromones.

Key words: ambrosia beetle, bark beetle, Cerambycidae, Coleoptera, Geosmithia morbida, host selection, Juglans

Introduction

The walnut twig beetle, Pityophthorus juglandis Blackman (Coleoptera: Scolytidae), vectors a phytopathogenic fungus, Geosmithia morbida (Kolařík et al., 2011), which causes Thousand Cankers Disease (TCD) in walnut (Juglans) and wingnut (Pterocarya) trees (Seybold et al., 2016). Walnut trees are used commercially for the production of nuts and high quality lumber (Leslie et al., 2010), and ornamentally in urban areas or along rural roads (Graves et al., 2009). Juglans also provides a food and cover source for various mammals and birds (Horton, 1949; Quinn, 1990; Burns & Honkala, 1990).

Through feeding, construction of brood galleries, or both activities, P. juglandis transfers the spores of the fungus to the phloem of host trees (Seybold et al., 2013a).

Affected trees have dark, wet stain spots on the bark surface and cankers in the phloem around P. juglandis entrance holes (Graves et al., 2009). As the numbers of beetle entrances accumulate, numerous cankers in the branches and stem can result in crown dieback and tree mortality (Tisserat et al., 2009; Seybold et al., 2013a, 2019).

Pityophthorus juglandis (Bright, 1981) and G. morbida (Kolařík et al., 2011;

Tisserat et al., 2011) were first collected and described, respectively, in the western USA, and since 2010 they have also spread to the eastern USA (Seybold et al., 2013a, 2016).

Surveys in the western USA suggest that P. juglandis may be the only subcortical insect to carry G. morbida (Kolařík et al., 2017). There have been reports from the eastern USA that other subcortical insects may vector G. morbida in Juglans, but the phoresy rates have been low; transmission has not been demonstrated experimentally; and the insects have tended to be species that target moribund trees (Juzwik et al., 2015). TCD has also spread to Italy where it occurs in five regions (Faccoli et al., 2016; Moricca et al., 2018).

Native walnut species in the western USA include southern California black walnut, Juglans californica, northern California black walnut, Juglans hindsii, and

Arizona black walnut, Juglans major. Native to the eastern USA is eastern black walnut,

Juglans nigra, which is grown widely for wood production (Burns & Honkala, 1990).

English walnut, Juglans regia, is native to central Asia and is used commercially worldwide for its nut production. Phylogeographic analyses suggest that J. major is the most likely native host for P. juglandis (Rugman-Jones et al., 2015), and G. morbida

(Zerillo et al., 2014). Conceivably, because of the long period of association among both pest organisms and J. major in the southwestern USA, there may have been some co- evolutionary events in this complex (Rugman-Jones et al., 2015) that have led to a relatively low susceptibility of J. major to the disease elements (Table 1). Although TCD in California was first recognized in 2008 (Flint et al., 2010), collection records of P. juglandis beginning in 1959 suggest that the disease may have been active in the state for much longer. The impact of the disease on J. californica planted in a USDA germplasm repository in northern California has been substantial, whereas the impact on J. regia in this repository and in California’s commercial orchards has been relatively modest (Table

1; Hishinuma, 2017; Seybold et al., 2019).

The impact of TCD has been evaluated by subdividing the disease into the susceptibility of host material to fungal infection and to the reproductive success of the beetle (Table 1). Juglans nigra, J. hindsii, and J. californica are highly susceptible to G. morbida (based on lesion length or size on inoculated branches and stems); J. major is less susceptible (Table 1; Utley et al., 2013). Similarly, reproduction of P. juglandis is more substantial in J. nigra, J. hindsii, and J. californica than in J. major (Table 1; Hefty

et al., 2016, 2018). Host selection by P. juglandis is a critical set of behavioral steps when spores of G. morbida are introduced to the phloem of walnut or wingnut trees, which results in TCD. However, little is known about the role that host selection by P. juglandis plays in TCD impact (Hishinuma, 2017).

Host plant selection by bark beetles is influenced by factors such as visual, olfactory, tactile, and gustatory cues (Raffa et al., 2016). The process may begin with undirected or directed flight and landing based on long-range olfactory signals (Graves et al., 2008). These signals may include behavioral chemicals from host and non-host plants, as well as from conspecific or heterospecific bark beetles (Graves et al., 2008).

Directed flight may be followed by shorter-range walking behavior guided by olfaction or feeding guided by gustatory stimulation and deterrence (Graves et al., 2008). After a host plant is selected, P. juglandis males produce an aggregation pheromone, which attracts both sexes (Seybold et al., 2015), resulting in increased initiation of brood galleries by new males.

In this study, we sought to determine if P. juglandis uses olfactory cues in its initial detection of the host. Volatile compounds released from living branches on trees of different species of Juglans may elicit varying landing rates by P. juglandis (Hishinuma,

2017). Hishinuma (2017) also studied flight responses by P. juglandis to cut branch sections from different species of Juglans by baiting the branch sections with aggregation pheromone lures, and then observing and comparing the P. juglandis landing rates. These studies showed a higher landing preference for J. californica than for J. major. However, landing responses on branch sections have not been investigated in the absence of pheromone lures, which might have influenced the observed preferences. We investigated

differential landing rates of P. juglandis and various associated insects on unbaited branch sections of two key walnut hosts of P. juglandis: J. californica and J. major. By following this approach, we sought to establish the rationale for pursuing a future comparative chemical analysis of bark volatiles from the two hosts in hopes of discovering chemical attractants or repellents.

Materials and methods

Study sites

Experiments were conducted in two different walnut orchards: 1) a multi-species planting at the United States Department of Agriculture, Agricultural Research Service, National

Clonal Germplasm Repository (NCGR) Juglans collection at Wolfskill Experimental

Orchards in Winters, California (Solano Co.) (38°30′00.4″N, 121°58′37.1″W) and 2) a commercial-style English walnut, J. regia, orchard at the California State University

Chico, Agricultural Teaching and Research Center (ATRC) in Chico, California (Butte

Co.) (39°41′41.3″N, 121°49′13.1″W). The NCGR site contained approximately 1,000

Juglans trees representing 15 species planted in 25 rows of 40 trees each, whereas the

ATRC site contained approximately 2,200 trees of one species planted in 44 rows of approximately 50 trees. Forty of the rows were planted with the 'Chandler' cultivar; four rows were planted with multiple cultivars, which included 12 named cultivars (e.g.,

'Durham,' 'Forde,' 'Gillet,' 'Hartley,' 'Ivanhoe,' 'Payne,' etc.), as well as numerous unnamed experimental crosses from the University of California Davis Walnut Breeding Program.

Rows were aligned from west to east at both sites.

Seasonal flight activity of walnut twig beetle

In order to determine P. juglandis flight activity within each orchard, four-unit funnel traps with wet cups (Lindgren, 1983; Seybold et al., 2013b) were maintained at each study site. At the NCGR, two traps were placed on one pole on the southern edge of the orchard, and two were placed on one pole near the center of the orchard. At the ATRC, four traps (one per pole) were placed on the eastern (two traps: one among 'Chandler' and one among multiple cultivars) and western (two traps: one among 'Chandler' and one among multiple cultivars) edges of the orchard and two (one per pole) were placed in the interior of the multiple cultivar block. These funnel traps were suspended on ~3 m long stainless steel conduit poles and baited with the male-produced aggregation pheromone of P. juglandis (Seybold et al., 2013b; Chen & Seybold, 2014). Traps were attached to the pole at 1.5 m and 2.75 m (NCGR) and at 2.75 m only (ATRC). The different arrangement of traps on the poles was a consequence of the earlier establishment of the

NCGR survey site as part of a broader orchard survey for P. juglandis in California.

Approximately 100 ml of ethanol-free marine antifreeze (Winter Safe Antifreeze, product

#31200SLVV3, Kinpak Inc., Montgomery, AL, USA) containing propylene glycol were added into the trap cups to immobilize any captured P. juglandis. Once per week, the contents of the cups were collected by pouring antifreeze laden with insects through conical paper paint strainers with nylon mesh (Basecoat/Clearcoat Strainers, Fine 190

Micron 1000CT, product # Gerson-10601, Toolrage.com, Dillsburg, PA, USA) and placing the filtered contents in plastic bags (Seybold et al., 2013b). Collected samples were frozen and then sorted in the laboratory to determine the total number of male and female P. juglandis caught each week, as well as associated insects such as the fruit-tree

pinhole borer, Xyleborinus saxeseni (Ratzeburg) and the minute bark beetle,

Hypothenemus eruditus Westwood (both Scolytidae); and Nathrius brevipennis (Mulsant)

(Cerambycidae).

Collection and handling of host material for landing rate assays

Methods for the landing rate assays generally followed the procedures described by

Hishinuma (2017) with some modifications as described below. Five flight landing assays were conducted during the Spring and Fall of 2015 and 2016 (Assay 1: 5 May to 6

June 2015, NCGR; Assay 2: 29 August to 24 October 2015, NCGR; Assay 3: 10 April to

24 June 2016, NCGR; Assay 4: 13 April to 21 June 2016, ATRC; and Assay 5: 27

August to 29 October 2016, NCGR). In the first four assays, P. juglandis landing rate was measured on translucent adhesive-coated acetate sheets (see below for details) surrounding branch sections of J. californica and J. major (Figure 1). In the final assay

(Fall 2016), P. juglandis landing rate was measured on the same type of sheets surrounding branch sections of J. californica and a negative control, which was a cardboard mailing tube (1.5 in. x 18 in., Tubes, Mailing, product # 37005-OD, LePage’s

2000 Inc., Taylor, MI, USA) cut to a length of 42 cm (Figure 2).

Branches of J. californica and J. major ranging in diameter from 1.3 cm to 4.7 cm were cut from trees at the University of California, Davis, Hutchison Road Juglans

Collection, New Stuke Block, HQ Californica, calif 97-040, Davis, California (Yolo Co.)

(38°32′21.2″N, 121°47′45.1″W) and at the NCGR (Supplemental Table 1). Branches on the orchard trees were selected for cutting without regard to location in the tree crown, but based on their diameter and the absence or low density of P. juglandis entrance or

emergence holes. The high population density of P. juglandis and the high susceptibility of J. californica to colonization in these areas made it difficult to locate branches of J. californica that did not show evidence of these holes (Hishinuma, 2017). These branches were cut into approximately 42 cm sections (Figure 1) and placed in a walk-in freezer (‒

25°C) for at least 3 days to kill any living insects in the branches. All branches were retrieved from the freezer and placed overnight in a walk-in cold room (4.4 °C) to thaw prior to their placement in the field.

Preparation of host material and placement in the field

For each assay, thawed branch sections of J. californica and J. major were examined again in the laboratory for any pre-existing entrance holes or old galleries of P. juglandis.

Branch sections showing a high density of P. juglandis entrance or emergence holes were discarded. Ten branch sections each of J. californica and J. major with zero or a low density of these holes were selected for the study. Depending on the P. juglandis activity on the available branches, the low densities ranged from 0 to 50 holes, and the holes were marked with a red paint pen. However, if we were unable to find sufficient branch sections with low densities for a given assay, a small number of branch sections that exceeded 50 P. juglandis holes were used (14 out of 50 sections of J. californica in five assays, and 0 out of 40 sections of J. major in four assays). However, these were checked carefully by removing a small amount of bark around the hole and galleries with a razor blade to ensure that no beetles were present.

All branch sections of both species were sealed at each end with wax (Gulf brand, product #203-060-005, Roswell, GA, USA) to reduce desiccation. Eyehole screws (#6

Zinc-Plated Screw Eyes, CrownBolt, Aliso Viejo, CA, USA) were attached to each end of the branch sections, which were then placed into mesh screen bags (100 x 100 mesh,

0.01 mm [T-316 STAINLESS STEEL100X100-0-0045-T-316-PW] Screen Technology

Group Inc., Santa Fe Springs, CA, USA), and the bags were sealed with hot glue. The bags were used to prevent incidental colonization by responding beetles.

For each assay, twenty branch sections (ten for each treatment) were placed in the field. Branch sections were assigned randomly to one of twenty trapping stations

(completely randomized design) located primarily between NCGR accessions of northern

California black walnut, J. hindsii, in the northeastern corner of the NCGR orchard and between 'Chandler' J. regia, in the center of the ATRC orchard. Each screened branch section was suspended horizontally with wires from a stainless steel conduit pole so that it hung ~1.5 m above the orchard floor (Figure 3). Cork stoppers were glued to the outside of each mesh bag with wood glue. Adhesive-coated, translucent acetate sheets

(27.94 cm x 43.18 cm, Apollo Transparency Film, product# VPP100C, ACCO Brands

Inc., Lincolnshire, IL, USA) were placed around each branch section and held in place with push pins attached to the cork stoppers (Figure 3). The use of the cork stoppers minimized any further damage to the bark of the branch sections. The trap adhesive used on the acetate sheets was Stikem Special (Formulation Regular, Seabright Laboratories,

Emeryville, CA, USA).

Every seven days, branch sections were re-randomized among the trapping stations. Prior to re-randomization, all insects of interest were removed from the adhesive-coated acetate sheets in the field and, during periods of peak flight, the acetate sheets were removed from the field for laboratory assessment of any remaining insects of

interest and replaced with fresh adhesive-coated acetate sheets. We conducted the first assay for 4 seven-day cycles, and the others for 8 to 11 seven-day cycles. Branch sections were removed at the end of the assays and placed into cold storage so that they could be evaluated in the laboratory.

Handling of insects and acetate sheets in the laboratory

Adhesive-coated acetate sheets were examined under a dissecting microscope (40X magnification), and insects of interest were excised or lifted from acetate sheets and transferred to vials of xylene to dissolve the trap adhesive. After soaking for at least 20 minutes, insects were separated from the pieces of acetate sheet and dried on paper towels or paper wipes and washed twice in 100% ethanol. Insects were then transferred into vials of 100% ethanol for storage. Eventually, insects were sorted into various taxonomic groups (to species where possible, but at least to genus or family). We focused on insects in the subcortical community associated with Juglans (Seybold et al., 2016).

After collection at the end of each assay, branch sections were dissected to confirm that no incidental P. juglandis had breached the wire mesh screening to attack and bore in, since the naturally-produced male aggregation pheromone released after attack would attract more P. juglandis and likely lead to an overestimate of our assessment of intrinsic landing rate.

Data handling and statistical analyses

Data from assays comparing P. juglandis landing rates on J. californica and J. major were grouped (Assays 1-4). For those species and instances where trap catch counts were

relatively high (>20 individuals per 10 acetate sheets), data were analyzed separately by insect species (i.e., P. juglandis, X. saxeseni, H. eruditus, and N. brevipennis) and sex (P. juglandis and N. brevipennis) by using a generalized linear mixed model [function glmer.nb() from package lme4 in R (R Core Team, 2016)] with location (i.e., NCGR or

ATRC) and walnut species as fixed factors. Assay number and sampling period were treated as random factors. The distributions of insects landing on the adhesive-coated acetate sheets were modeled as a negative binomial. The link function between the mean of the catches and the linear predictor was the default natural logarithm. Since there was no significant interaction between location and walnut species for either insect species or sex, these interactions were not included in the model.

For Assay 5, data were analyzed separately by insect species (i.e., P. juglandis, X. saxeseni, and H. eruditus) and sex (P. juglandis) by using a generalized linear mixed model (as described above) with walnut species as a fixed factor. Sampling period was treated as a random factor. The distributions of the insects landing on the adhesive-coated acetate sheets during Assay 5 were also modeled as a negative binomial. The link function between the mean of the catches and the linear predictor was the default natural logarithm.

Voucher specimens and nomenclature

All specimens of Coleoptera of interest were identified by SJS with consultation with Dr.

Donald E. Bright, Jr., Colorado State University, Fort Collins, Colorado, for bark and ambrosia beetles. Voucher specimens of adults were accessioned into the Entomology

Department at the California Academy of Sciences, San Francisco, California and the

Entomology Collection at California State University, Chico. For this study, we have elected to use the original nomenclature for bark and ambrosia beetles (Coleoptera:

Scolytidae) based on the arguments presented in Bright (2014, 2019). In essence, morphological and fossil evidence of adult scolytids supports the family-level treatment, whereas similarity in scolytid and curculionid larval morphology supports a subfamily placement. Because this issue is not entirely resolved, we prefer to take the more conservative approach of using the original nomenclature. A large number of green lacewings, Chrysopidae, and brown lacewings, Hemerobiidae (both Neuroptera), were collected in the funnel traps during the surveys at the NCGR and the ATRC. The specimens from the ATRC were identified by Dr. Catherine A. Tauber, University of

California, Davis (UCD). Voucher specimens of these adults were deposited in the UCD

Bohart Insect Museum. Similarly, snakeflies (Raphidioptera: Raphidiidae) were collected in the funnel traps and branch section traps at both the NCGR and ATRC. These were identified by Dr. Shaun Winterton, California Department of Food and Agriculture,

Sacramento, CA. Voucher specimens of these adults were deposited in the California

State Collection of Arthropods.

Results

Pityophthorus juglandis seasonal flight monitoring

Weekly trap catches revealed that P. juglandis flight activity-abundance was greater at the NCGR than at the ATRC (Figure 4). The seasonal flight data were used to guide the timing of the landing rate assays, which were initiated in Spring 2015 (Assay 1); Fall

2015 (Assay 2); Spring 2016 (Assays 3 and 4); and Fall 2016 (Assay 5). All assays were

conducted during periods of flight when the highest landing rates were anticipated, although Assays 2, 4, and 5 tended to miss these windows (Figure 4; Seybold et al., 2012;

Chen & Seybold, 2014). Flight peaks at the NCGR with a mean catch greater than 50 P. juglandis per trap per day were observed during Spring and Summer 2015, and during

Summer and early Fall 2018, whereas flight peaks in 2014, 2016, and 2017 were below

50 P. juglandis. At the ATRC, flight peaks with a mean catch greater than 2 P. juglandis per trap per day were observed during Fall 2015, Spring and Fall 2016, Spring 2017, and

Fall 2018. At both locations, primarily during flight peaks, twice as many females were trapped than males.

Pityophthorus juglandis landing rate on adhesive-coated acetate sheets

A total of 1153 P. juglandis (504 males and 649 females) were trapped during Assays 1-4

(Assay 1: 142 males and 152 females; Assay 2: 46 males and 54 females; Assay 3: 310 males and 432 females; and Assay 4: 6 males and 11 females). Pooled P. juglandis landing rates from Assays 1-3 (NCGR), and Assay 4 (ATRC) suggested host preferences by both male and female P. juglandis for J. californica over J. major (NCGR: nassay1-

3=230 branch sections and ATRC: nassay4=100 branch sections; male P. juglandis:

2 2 X1 =3.95, P=0.046; female P. juglandis: X1 =6.06, P=0.013) (Figure 5). Comparing locations, there was greater male and female P. juglandis flight activity-abundance at

NCGR (NCGR: nassay1-3=230 branch sections and ATRC: nassay4=100 branch sections;

2 2 male P. juglandis: X1 =52.91, P<0.001; female P. juglandis: X1 =81.67, P<0.001).

A total of 82 P. juglandis (21 males and 61 females) were trapped during Assay 5

(NCGR), and P. juglandis males and females showed no preference for J. californica

over the control (cardboard tube) (nassay5=90 branch sections; male P. juglandis:

2 2 X1 =0.048, P=0.826; female P. juglandis: X1 =0.35, P=0.550) (Figure 6). Dissection of branches at the end of each trial revealed no evidence of incidental attacks by P. juglandis under the fine mesh metal screening during the course of each assay.

Associated insects and seasonal flight monitoring

At the NCGR, seasonal survey trap catches for P. juglandis collected from January 2014 through February 2019 also yielded other herbivorous insects (primarily Coleoptera) or predaceous insects (primarily Neuropteroidea) thought to be associated with Juglans or other hardwood trees growing nearby (Table 2). Female X. saxeseni were trapped from

May to June 2014 (4), March to June 2015 (8), March to September 2016 (22), February to August 2017 (10), and March to June 2018 (9). Two cerambycid beetles were trapped during this survey: Phymatodes juglandis Leng (one specimen, 25 March to 2 April

2018) and Xylotrechus nauticus (Mannerheim), the latter of which was detected March to

July 2015 (11), April 2016 (1), May to July 2017 (2), and May to August 2018 (7).

Hypothenemus eruditus was trapped during 15-22 September 2014 (1), July to August

2015 (2), April to September 2016 (7), May to October 2017 (24), and February to

November 2018 (21). The western oak bark beetle, Pseudopityophthorus pubipennis

LeConte, was trapped in small numbers, April to May 2015 (2), April to June 2016 (2),

July 2017 (2), and August 2018 (1), as was the shothole borer, Scolytus rugulosus

(Müller) (both Scolytidae), which appeared in survey traps during the time periods of

March to November 2014 (4), April to August 2015 (4), April to June 2017 (3), and April to July 2018 (8). None of the latter were observed in 2016. A scolytid species that was

unexpectedly present at the NCGR (and later at the ATRC) was the ambrosia beetle

Xyleborus affinis Eichhoff. This species is native to eastern North America and Central and South America, but one specimen was trapped at the NCGR in a survey trap in

March 2015. Woodborers, such as the lead cable borer, Scobicia declivis LeConte

(Bostrichidae) and Polycesta californica LeConte (Buprestidae) were active in November

2014 (1), July 2015 (1), March to May 2016 (2), April to August 2017 (2), and April to

September 2018 (5); and May 2014 (3), May to June in 2015 and 2016 (9 total), and May

2018 (3), respectively. None of the latter were observed in 2017. This buprestid is common in California and Oregon and recorded hosts include: Alnus, Acer, Arbutus,

Quercus, Populus, Cercocarpus, and Salix (Furniss & Carolin 1977; MacRae, 2006).

Predaceous lacewings, Chrysopidae and Hemerobiidae (both Neuroptera), were trapped at NCGR throughout each year from 2014-2019 (179 Chrysopidae and 7

Hemerobiidae); however, they were not identified to species. Predaceous snakeflies,

Agulla sp. (Raphidioptera: Raphidiidae), were also trapped at the NCGR from 2014-2018

(March to June) with a total of 341 (67 males, 93 females, 181 not identified to sex).

From this catch, 6 males and 6 females (caught in April 2015) were identified as Agulla modesta Carpenter. The remaining were not identified to species.

At the ATRC, seasonal survey trap catches included female X. saxeseni from

February to June 2016 and March to June 2017 (at low abundance, <10 per year) (Table

3). They also included male and female N. brevipennis from April to May 2016, May to

June 2017, and May to July 2018 (at intermediate abundance, <50 per year), in accordance with its spring and early summer flight period (Linsley, 1963). Another longhorned beetle, Stenocorus nubifer (LeConte) (Cerambycidae), was trapped only at

the ATRC in seasonal survey traps in April and May during 2016, 2017, and 2018, corresponding with the flight peak (April to July) described by Linsley and Chemsak

(1972). This insect was not trapped during the NCGR seasonal survey or on the adhesive- coated acetate sheets from the landing rate study at either site. Its larval host plant in or around the ATRC is not known. In seasonal survey trap catches at the ATRC, one specimen of P. californica was trapped in late May 2016, whereas one female specimen of H. eruditus was trapped in September 2018. One male S. rugulosus was also trapped in this same weekly flight survey trap in April 2018. Thus, in most instances the flight activity-abundance of these subcortical insect associates of P. juglandis was relatively low at the ATRC.

Three species of Chrysopidae and one species of Hemerobiidae, Chrysopa quadripunctata Burmeister; Chrysoperla comanche (Banks); Chrysoperla carnea

(Stephens) species group (one or more of several possible species including C. adamsi, C. johnsoni, and C. plorabunda; and Hemerobius pacificus Banks) were trapped in the

2015-2017 ATRC seasonal survey traps. A total of 33 specimens of the green lacewing

C. comanche were trapped (maximum flight activity in the fall), whereas 37 specimens from the C. carnea species group and 12 specimens of H. pacificus were also trapped

(maximum flight activity in the spring for these taxa). Only one specimen of C. quadripunctata (a female) was trapped, and it was caught in the fall. For C. comanche, C. carnea group, and H. pacificus, more females were trapped than males. This difference was more pronounced with C. comanche (7 males and 26 females) and the C. carnea group (13 males and 24 females), than with H. pacificus (5 males and 7 females). From

2018-2019, a total of 70 Chrysopidae and one Hemerobiidae (2018) were trapped but were not identified to species. These insects may function as predators of P. juglandis.

A total of 38 Agulla sp. (Raphidioptera: Raphidiidae) were also trapped at the ATRC from 2016-2018 (no specimens were trapped during 2015). Five males and 16 females were trapped from April to June in 2016, whereas 2 males and 8 females were trapped from May to June in 2017 and 2 males and 5 females were trapped from May to June in

2018.

Associated insect landing rate on adhesive-coated acetate sheets

In addition to P. juglandis, trap catches on sticky traps in the various assays yielded other insects thought to be associated with P. juglandis on Juglans or perhaps directly with other hardwood trees growing nearby (Table 4). These included, Xyleborinus saxeseni,

Cyclorhipidion bodoanum (Reitter), Hypothenemus eruditus, Pseudopityophthorus pubipennis, Scolytus rugulosus (all Scolytidae); Nathrius brevipennis (Cerambycidae);

Polycesta californica (Buprestidae); Scobicia declivis (Bostrichidae); and Petalium californica Fall, Priobium punctatum LeConte, and Ptilinus sp. Müller (all Anobiidae)

(Seybold et al., 2016). Two probable parasitoids of P. juglandis collected in the assays were Plastonoxus westwoodi Kieffer (Hymenoptera: Bethylidae), and Neocalasoter pityophthori (Ashmead) (Hymenoptera: Pteromalidae) (Seybold et al., 2016). A third parasitoid was trapped in relatively large numbers in our assays [Gildoria sp. Hedqvist

(Hymenoptera: Braconidae: Doryctinae)] is likely associated with N. brevipennis.

Predaceous insects trapped in the assays included two species of Coleoptera:

Laemophloeidae, Narthecius simulator Casey and Parandrita cephalotes (LeConte)

(Seybold et al., 2016).

A total of 1606 female X. saxeseni were trapped at both the NCGR and ATRC during Assays 1-4 (Assay 1: 301; Assay 2: 33; Assay 3: 985; and Assay 4: 287 female X. saxeseni). Males of this species are flightless. Pooled results from Assays 1-3 (NCGR) and Assay 4 (ATRC) showed that female X. saxeseni preferred to land on acetate sheets surrounding J. major over acetate sheets surrounding J. californica (NCGR: nassay1-3=230

2 branch sections and ATRC: nassay4=100 branch sections; X1 =25.78, P<0.001) (Figure 5).

In these assays, the flight activity-abundance of female X. saxeseni at the NCGR was greater than at the ATRC (NCGR: nassay1-3=230 branch sections and ATRC: nassay4=100

2 branch sections; X1 =5.86, P=0.015). A total of 57 female X. saxeseni were trapped during Assay 5 (NCGR). In this assay, female X. saxeseni preferred landing on acetate sheets surrounding J. californica branch sections versus acetate sheets surrounding the

2 cardboard tubes (negative control) (nassay5=90 branch sections; X1 =5.86, P=0.015)

(Figure 6). Flight of X. saxeseni was primarily in the spring and early summer.

A total of 789 female H. eruditus were trapped at the NCGR (Assay 1: 6; Assay

2: 226; and Assay 3: 557 female H. eruditus), whereas none were trapped at the ATRC

(Assay 4). Similar to X. saxeseni, H. eruditus males are flightless. Pooled results from

Assays 1-3 (NCGR) showed that female H. eruditus preferred to land on acetate sheets surrounding J. major over acetate sheets surrounding J. californica (nassay1-3=230 branch

2 sections; X1 =38.18, P<0.001) (Figure 5). A total of 177 female H. eruditus were trapped during Assay 5 (NCGR). In this assay, female H. eruditus preferred landing on acetate sheets surrounding J. californica branch sections versus acetate sheets surrounding the

2 cardboard tubes (negative control) (nassay5=90 branch sections; X1 =19.08, P<0.001)

(Figure 6).

A total of 4653 N. brevipennis (4457 males and 196 females) were trapped during

Assays 1-4 (Assay 1: 99 males and 7 females; Assay 2: 0 males and 0 females; Assay 3:

1068 males and 104 females; and Assay 4: 3290 males and 85 females). Overall, N. brevipennis sex ratio ranged from 10:1 (Assay 3) to 39:1 (Assay 4) times more males than females (Figure 5). Pooled results from Assays 1-3 (NCGR) and Assay 4 (ATRC) suggested that male N. brevipennis preferred to land on acetate sheets surrounding J. major branch sections, whereas females showed no significant host preference (NCGR:

2 nassay1-3=230 branch sections and ATRC: nassay4=100 branch sections; males: X1 =7.25,

2 P=0.007; females: X1 =2.86, P=0.09) (Figure 5). There was no difference in flight activity-abundance of male and female N. brevipennis during the landing assays at the

NCGR and ATRC (NCGR: nassay1-3=230 branch sections and ATRC: nassay4=100 branch

2 2 sections; males: X1 =0.002, P=0.958; females: X1 =1.92, P=0.165). No comparison between N. brevipennis landing rates on J. californica versus control were carried out for

Assay 5 at the NCGR, since this species apparently flies during the spring and early summer and not the fall (our data and Linsley, 1963).

Thirteen other subcortical insects associated putatively with Juglans, other hardwood trees, or P. juglandis (Seybold et al., 2016) were trapped during the assays

(Table 4). Bark and ambrosia beetles trapped at the NCGR included C. bodoanum, P. pubipennis, and S. rugulosus. Cyclorhipidion bodoanum and P. pubipennis were not trapped during the fall at the NCGR. Scolytus rugulosus were trapped in all assays at the

NCGR (Assay 1: 9 specimens; Assay 2: 6 specimens; Assay 3: 41 specimens; and Assay

5: 2 specimens). One specimen of X. affinis was trapped in 2016 at the ATRC landing on a J. major branch section during Assay 4 (Table 4). A second specimen was caught in

2018 in an unrelated study conducted at the ATRC. Scobicia declivis and P. californica, were present in both locations and all assays except for Assay 5. Priobium punctatum was not caught during Assays 2 and 5 at the NCGR, suggesting that it does not fly in the fall. Ptilinus sp. was not caught during Assay 2 at the NCGR. Of potential P. juglandis parasitoids, N. pityophthori was trapped in all assays, but P. westwoodi was not caught during Assay 4 at the ATRC and Assay 5 at the NCGR. The N. brevipennis parasitoid,

Gildoria sp. was present in both locations in Assays 3-5. However, its presence during

Assays 1 and 2 is unknown because we were not aware of this parasitoid when the data were collected. The P. juglandis predator, Narthecius simulator was trapped in all assays, but a related predator, Parandrita cephalotes, was not trapped during Assays 2 and 5 at the NCGR or during Assay 4 at ATRC.

Discussion

Seasonal flight monitoring trap catches from the NCGR (2014 to 2019) and the ATRC

(2015 to 2019) showed that P. juglandis flight peaks in 1) late Spring/early Summer and

2) Fall (Figure 4), similar to flight patterns described in Seybold et al. (2012).

Pityophthorus juglandis trap catches at both sites can be used to compare and predict P. juglandis abundance in future studies based on population modeling (Chen & Seybold,

2014).

Assays of P. juglandis landing rate showed that branch sections of J. californica elicited a higher landing rate by P. juglandis than branch sections of J. major. This is

consistent with the work of Hishinuma (2017), who found that P. juglandis discriminated among various Juglans spp. by landing on live branches on trees and on branch sections of certain species baited with the aggregation pheromone. Specifically, P. juglandis showed a higher preference for landing on pheromone-baited J. californica than J. major.

The attraction of P. juglandis to unbaited J. californica branch sections in our work indicates that P. juglandis host preference may be influenced by long-range olfactory cues emitted by the host. Balanced sex ratios in the responses in our assays suggest the absence of volunteer attacks by male P. juglandis under the mesh metal screening, and indicate that the landing rate was not influenced by naturally produced aggregation pheromone, which tends to attract more females than males (Chen & Seybold, 2014).

Surprisingly, P. juglandis showed no preference for landing on J. californica over

J. major branch sections in Assay 4 at the ATRC. However, the low flight rate (17 beetles total) of P. juglandis at the ATRC (compared to the NCGR) may have precluded any discrimination amongst the treatments. This emphasizes the need for a sufficiently high P. juglandis flight activity-abundance in an orchard or other habitat to conduct studies of intrinsic landing rate (i.e., in the absence of aggregation pheromone). Even under conditions of high population density, landing rates in response to host volatiles alone are low, which may demand that researchers conduct multiple assays or use greater replication to accumulate enough data for statistical analysis.

Several factors may explain the low population density of P. juglandis that we noted at the ATRC relative to the density at the NCGR (Figure 4 and Assay 4). The walnut orchard at the ATRC, as well as nearby orchards, is a commercial-style planting of J. regia. In contrast, the orchard at the NCGR is comprised of at least 15 species of

Juglans, and the assays were positioned under the crowns of trees of northern California black walnut, J. hindsii. Adjacent to the NCGR is the Putah Creek riparian area, which includes a large number of J. hindsii dying from TCD. Hishinuma (2017) reported that in several assays J. regia was less attractive to P. juglandis than black walnuts such as J. californica, J. hindsii, and J. nigra. The presence of these latter host species at the NCGR and the greater rate of P. juglandis reproduction in these hosts (Hefty et al., 2018) may explain the higher number of P. juglandis caught in our assays at the NCGR. Since the branch sections were not cut from one accession each for J. californica and J. major, other possible factors influencing P. juglandis landing rates may be the differing branch sources with respect to potential variability in genetically-based, intraspecific volatile profiles. Also, wounding or stress cues (e.g., volatiles) of the individual cut branches due to handling, which includes cutting, freezing, and thawing prior to their placement in the field may have influenced the landing rate. During the assays, volatile composition emitted from the cut branches may have changed during the course of each assay, mainly for Assays 3 and 4, which lasted 10-11 weeks. Trap catches of P. juglandis, X. saxeseni, and N. brevipennis were low during the beginning and end of each assay but were largest during the middle of the assays.

It was also surprising that in Assay 5 P. juglandis showed no preference between

J. californica and the control (i.e., cardboard mailing tube). Pityophthorus juglandis seasonal flight monitoring at both study sites indicated a reduction in flight activity at the time of that assay, which likely resulted in low rates and minimal data collection during this landing assay. This assay would have shown whether P. juglandis uses olfactory cues to distinguish a host from a non-host cardboard tube (control). To fully test the

hypothesis that volatiles from J. californica branch sections are attractive, however, this assay should be conducted again during spring, when landing rates are expected to be higher or through the addition of more replicates for treatment and control.

This study also showed that branch sections of J. major elicited higher landing rates by female X. saxeseni and H. eruditus than branch sections of J. californica (Assays

1-4 and Assays 1-3, respectively). However, both invasive species preferred J. californica over the cardboard tube (control) in Assay 5. Although X. saxeseni was trapped in small numbers during Assay 5, which was conducted in the fall, seasonal survey trapping at both the NCGR and ATRC, suggested that this beetle was not as active in flight during the fall. This may have resulted from the difference in lures for the two trapping efforts. Results from all assays suggest that this insect may also use long-range olfactory signals to find hosts. This invasive ambrosia beetle bores into the xylem of hardwoods and conifers and is very responsive to ethanol, which may have emanated from the branch sections over time (Seybold et al., 2016).

Hypothenemus eruditus was trapped almost exclusively at the NCGR; however, one specimen was trapped in September 2018 at the ATRC during seasonal flight monitoring. There are no previous records of this beetle from Butte County, and this is the northernmost record of this invasive species in California (Seybold et al., 2016). Early in its invasion of California, it was found primarily along the southern Pacific Coast

(Bright & Stark, 1973) and then later in the Central Valley (Seybold et al., 2016). It has a broad host range (e.g., Bright, 2014). From a separate study at the ATRC, we trapped two other specimens of H. eruditus during summer 2018, confirming their presence at this site.

Branch sections of J. major also elicited higher landing rates by male N. brevipennis than branch sections of J. californica. A lack of a significant response for one host over the other by females may be attributable to the relatively low number of females caught in the study. Assays 1, 3, and 4 showed a significant difference in male and female N. brevipennis catches. This invasive beetle is native to Europe. In California it has been detected in Juglans, Quercus, and Ficus (Linsley, 1963). Not much is known biologically of this species, so it is difficult to explain the large difference in male and female catches in the landing assays. Seasonal trap catches at the ATRC showed a more even distribution of both sexes (Table 3), but this may have been a consequence of random capture in traps baited with the P. juglandis pheromone. Male N. brevipennis may precede females phenologically to the mating and oviposition site. Assay 5 was inconclusive for preference by N. brevipennis for J. californica over cardboard tubes

(control) because N. brevipennis does not fly during the fall. Gildoria sp., the potential parasitoid of N. brevipennis, is likely an undescribed species (R. Zuparko and Yves

Braet, personal correspondence). Flight of Gildoria was largely contemporaneous with the flight of N. brevipennis. However, a small number of Gildoria were trapped during

Assay 5 at the NCGR (i.e., fall of 2016). Rearing records (SJS unpublished data) suggest a major emergence of Gildoria in March through June from small diameter twigs of J. regia. For future studies, assays targeting N. brevipennis or Gildoria should be conducted during their flight peaks, which we have determined to be April to June, and earlier than the California flight peak (June to August) described by Linsley (1963).

Subcortical insect species from eight families were trapped in the two northern

California walnut orchards. Fewer Scolytidae were present at the ATRC (Table 4).

Surprisingly, one species, X. affinis, thought native to eastern North America, as well as

Central and South America (Rabaglia et al., 2006; Bright, 2014), was trapped in 2015 in a seasonal survey trap at the NCGR and in 2016 while attempting to land on J. major at the

ATRC. This species has not been reported previously from western North America. The presence of X. affinis at the ATRC was confirmed by the collection of a second specimen during an unrelated study conducted in the fall of 2018. A number of other herbivorous insects, parasitoids, and predaceous insects were present in both locations and most assays (Table 4).

Our research may serve as a guide for designing and scheduling future studies at

NCGR and ATRC, as well as similar sites throughout California. To further study the susceptibility of J. californica and J. major to P. juglandis, the next step should be to allow P. juglandis to land on cut branches of the two hosts and investigate their reproductive success rate. Since our study found indications of long-range olfactory cues, examining the volatiles including the amount of ethanol emitted from the two host plants may provide further insight on the host selection by P. juglandis and associated beetles such as X. saxeseni, H. eruditus, and N. brevipennis.

Acknowledgements

We thank Megan A. Siefker, and Mariah Quintanilla, (both UC-Davis Department of

Entomology and Nematology) for their excellent assistance in the lab and field. We also thank Stacy M. Hishinuma (formerly UC-Davis Department of Entomology and

Nematology) for some of the photographs and for reviewing initial drafts of this manuscript. We are grateful to John Preece and the USDA ARS NCGR for providing

access to the study site and permission to collect host material at Wolfskill Experimental

Orchards in Winters, California, and Charles A. Leslie, UC Davis, Department of Plant

Sciences, for permission to collect additional host material for the study. We are also very grateful to Jeff Boles and the ATRC for access to the second study site in Chico. We appreciated the assistance of Donald E. Bright (Colorado State University), Catherine A.

Tauber (University of California Davis), and Shaun Winterton (California State

Collection of Arthropods) with the identification of various insect taxa. Funding for this project was provided through a Joint Venture Agreement (16-JV-11272139-086) between the USDA Forest PSW Research Station and California State University Chico Research

Foundation.

References

Bright DE (1981) Taxonomic monograph of the genus Pityophthorus Eichhoff in North and Central America (Coleoptera: Scolytidae). Memoirs of the Entomological Society of Canada 113: 1–378.

Bright DE (2014) A catalog of Scolytidae and Platypodidae (Coleoptera), Supplement 3 (2000–2010), with notes on subfamily and tribal reclassifications. Insecta Mundi 356: 1– 336.

Bright DE (2019) A taxonomic monograph of the bark and ambrosia beetles of the West Indies (Coleoptera: Curculionoidea: Scolytidae and Platypodidae (Coleoptera). Sudies on West Indian Scolytidae 7. Occasional Papers of the Florida State Collection of Arthropods, Gainesville, Florida, 500 pp.

Bright DE & Stark RW (1973) The bark and ambrosia beetles of California, Coleoptera: Scolytidae and Platypodidae. Bulletin of the California Insect Survey 16: 1–169.

Burns RM & Honkala BH (1990) Black walnut, Juglans nigra L. Silvics of North America. Vol. 2: Hardwoods. USDA Forest Service Agriculture Handbook 654 pp. https://www.srs.fs.usda.gov/pubs/misc/ag_654/volume_2/silvics_v2.pdf (accessed November 17, 2018).

Chen Y & Seybold SJ (2014) Crepuscular flight activity of an invasive insect governed by interacting abiotic factors. PLoS ONE 9(8): e105945. doi:10.1371/journal.pone.0105945.

Faccoli M, Simonato M & Rassati D (2016) Life history and geographical distribution of the walnut twig beetle, Pityophthorus juglandis (Coleoptera: Scolytinae), in southern Europe. Journal of Applied Entomology 140: 697–705. doi: 10.1111/jen.12299.

Flint ML, Graves AD & Seybold SJ (2010) Thousand cankers disease of walnuts spreads in California. CAPCA Advisor Magazine 8(3): 36–39. http://entomology.ucdavis.edu/files/201360.pdf (accessed November 17, 2018).

Furniss RL & Carolin VM (1977) Western forest insects, Vol. 1339. US Department of Agriculture, Forest Service. 654 pp.

Graves AD, Holsten EH, Ascerno ME, Zogas KP, Hard JS et al. (2008) Protection of spruce from colonization by the bark beetle, Ips perturbatus, in Alaska. Forest Ecology and Management 256: 1825–1839.

Graves AD, Coleman TW, Flint ML & Seybold SJ (2009) Walnut twig beetle and thousand cankers disease: Field identification guide, UC-IPM Website Publication, 2 pp. Nov. 21, 2009. http://www.ipm.ucdavis.edu/PDF/MISC/thousand_cankers_field_guide.pdf (accessed November 17, 2018).

Hefty AR, Coggeshall MV, Aukema BH, Venette RC & Seybold SJ (2016) Reproduction of walnut twig beetle in black walnut and butternut. HortTechnology 26: 727–734. doi: 10.21273/HORTTECH03427-16.

Hefty AR, Aukema BH, Venette RC, Coggeshall MV, McKenna JR et al. (2018) Reproduction and potential range expansion of walnut twig beetle across the Juglandaceae. Biological Invasions 20: 2141–2155. https://doi.org/10.1007/s10530-018-1692-5.

Hishinuma SM (2017) Interactions among the walnut twig beetle, Pityophthorus juglandis, the pathogenic fungus, Geosmithia morbida, and host species in thousand cankers disease in California. Dissertation, University of California, Davis.

Horton JS (1949) Trees and shrubs for erosion control of southern California mountains. Berkeley, CA: U.S. Department of Agriculture, Forest Service, California [Pacific Southwest] Forest and Range Experiment Station; California Department of Natural Resources, Division of Forestry, 72 pp.

Juzwik J, Banik MT, Reed SE, English JT & Ginzel MD (2015) Geosmithia morbida found on weevil species Stenomimus pallidus in Indiana. Online. Plant Health Progress 16: 7–10. doi:10.1094/PHP-RS-14-0030. 29

Kolařík M, Freeland E, Utley C & Tisserat N (2011) Geosmithia morbida sp. nov., a new phytopathogenic species living in symbiosis with the walnut twig beetle (Pityophthorus jugandis) on Juglans in the USA. Mycologia 103: 325–332. doi:10.3852/10–124.

Kolařík M, Hulcr J, Tisserat N, De Beer W, Kostovčík M, et al. (2017) Geosmithia associated with bark beetles and woodborers in the western USA: Taxonomic diversity and vector specificity. Mycologia 109: 185–199. http://dx.doi.org/10.1080/00275514.2017.1303861.

Leslie CA, Seybold SJ, Graves AD, Cranshaw W & Tisserat N (2010) Potential impacts of thousand cankers disease on commercial walnut production and walnut germplasm conservation. 6th International Walnut Symposium, 25–27 February 2009, Melbourne, Australia, Acta Horticulturae 861: 431–434.

Lindgren BS (1983) A multiple funnel trap for scolytid beetles (Coleoptera). The Canadian Entomologist 115: 299−302.

Linsley EG (1963) The Cerambycidae of North America. Part IV. Taxonomy and Classification of the Subfamily Cerambycinae, Tribes Elaphidionini through Rhinotragini. Vol. 21. University of California Publications in Entomology, University of California Press, Berkeley, CA, USA. 165 pp.

Linsley EG & Chemsak JA (1972) Cerambycidae of North America. Part VI. No. 1. Taxonomy and Classification of the Subfamily Lepturinae. Vol. 69. University of California Publications in Entomology, University of California Press, Berkeley, CA, USA. 138 pp.

MacRae TC (2006) Distributional and biological notes on North American Buprestidae (Coleoptera), with comments on variation in Anthaxia (Haplanthaxia) cyanella Gory and A.(H.) viridifrons Gory. The Pan-Pacific Entomologist 82: 166–199.

Moricca S, Bracalini M, Benigno A, Ginnetti B, Pelleri F & Panzavolta T (2018) Thousand cankers disease caused by Geosmithia morbida and its insect vector Pityophthorus juglandis first reported on Juglans nigra in Tuscany, Central Italy. Plant Disease https://apsjournals.apsnet.org/doi/pdf/10.1094/PDIS-07-18-1256-PDN

Quinn, RD (1990) The status of walnut forests and woodlands (Juglans californica) in southern California. Endangered Plant Communities of Southern California (ed. by A.A. Schoenherr) Proceedings, 15th Annual Symposium; 1989 October 28; Fullerton, California. Special Publication No. 3. Southern California Botanists, Claremont, CA, USA. pp. 42−54.

R Core Team (2016) R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria. https://www.R-project.org.

Rabaglia RJ, Dole SA & Cognato AI (2006) Review of American Xyleborina (Coleoptera: Curculionidae: Scolytinae) occurring north of Mexico, with an illustrated key. Annals of Entomological Society of America 99: 1034–1056.

Raffa KF, Anderson MN & Schlyter F (2016) Host selection by bark beetles: playing the odds in a high-stakes game. Advances in Insect Physiology Vol. 50. (ed. by C. Tittiger & G.J. Blomquist) Academic Press, Oxford, UK. pp. 1–74.

Rugman-Jones PF, Seybold SJ, Graves AD & Stouthamer R (2015) Phylogeography of the walnut twig beetle, Pityophthorus juglandis, the vector of Thousand Cankers Disease in North American walnut trees. PloS One 10(2), p.e0118264.

Seybold SJ, Klingeman WE III, Hishinuma SM, Coleman TW & Graves AD (2019) Status and impact of walnut twig beetle in urban forest, orchard, and native forest ecosystems. Journal of Forestry 117: 152−163. DOI: 10.1093/jofore/fvy081.

Seybold SJ, King JA, Harris DR, Nelson LJ, Hamud SM et al. (2012) Diurnal flight response of the walnut twig beetle, Pityophthorus juglandis Blackman (Coleoptera: Scolytidae), to pheromone-baited traps in two northern California walnut habitats. The Pan-Pacific Entomologist 88: 231−247.

Seybold SJ, Haugen D, O’Brien J & Graves AD (2013a) Thousand cankers disease. USDA Forest Service, Northeastern Area State and Private Forestry Pest Alert, NA-PR- 02-10, originally published May, 2010, reprinted Aug. 2010, Oct. 2011, and Feb. 2013, 2 pp., https://www.fs.usda.gov/naspf/publications/pest-alert-thousand-cankers-disease- pest-alert-revised-february-2013-na-pr-02-10 (accessed November 17, 2018).

Seybold SJ, Dallara PL, Hishinuma SM & Flint ML (2013b) Detecting and identifying the walnut twig beetle: Monitoring guidelines for the invasive vector of thousand cankers disease of walnut. University of California Agriculture and Natural Resources, Statewide Integrated Pest Management Program, Oakland, California, 13 pp., March 13, 2013, http://ipm.ucanr.edu/PDF/PESTNOTES/WTB_trapping.pdf (accessed November 17, 2018).

Seybold SJ, Dallara PL, Nelson LJ, Graves AD, Hishinuma SM et al. (2015) Methods of monitoring and controlling the walnut twig beetle, Pityophthorus juglandis. US Patent 9, 137, 990 B2, 12 pp. + 7 Figs., 22 Sep 2015.

Seybold SJ, Penrose RL & Graves AD (2016) Invasive bark and ambrosia beetles in California Mediterranean forest ecosystems. Insects and Diseases of Mediterranean Forest Systems (ed. by TD Paine & F Lieutier) Springer International Publishing, AG, Cham, Switzerland, pp. 583−662.

Tisserat N, Cranshaw W, Leatherman D, Utley C & Alexander K (2009) Black walnut mortality in Colorado caused by the walnut twig beetle and thousand cankers disease. Phytopathology Supplement 99:S128. 31

Tisserat N, Cranshaw W, Putnam M, Pscheidt J, Leslie CA et al. (2011) Thousand cankers disease is widespread on black walnut, Juglans nigra, in the western United States. Online. Plant Health Progress (June 30, 2011) doi:10.1094/PHP-2011-0630-01- BR.

Utley C, Nguyen T, Roubtsova T, Coggeshall M, Ford TC et al. (2013) Susceptibility of walnut and hickory species to Geosmithia morbida. Plant Disease 97: 601–607. http://dx.doi.org/10.1094/PDIS-07-12-0636-RE, “First Look” posted December 5, 2012.

Zerillo MM, Caballero JI, Woeste K, Graves AD, Hartel C et al. (2014) Population structure of Geosmithia morbida, the causal agent of thousand cankers disease of walnut trees in the United States. PLoS ONE 9(11): e112847. doi:10.1371/journal.pone.0112847

Table 1.1 Synthesis of measures of host susceptibility to walnut twig beetle, Pityophthorus juglandis, feeding and Geosmithia morbida inoculation from the literature. Pityophthorus juglandis Geosmithia morbida Host Hefty et al. (2016)1 Hefty et al. (2018)1 Hishinuma (2017)2 Hishinuma (2017)3 Hishinuma (2017)3 Hishinuma (2017)3 Utley et al. (2013)4

Juglans ailantifolia − low − − − intermediate low Juglans californica − highest highest − highest highest high Juglans cathayensis − high − − − − − Juglans cinerea no difference low − − − lowest intermediate Juglans hindsii − high intermediate no difference low high high Juglans major − low lowest − low intermediate lowest Juglans mandshurica − low − − − − low Juglans microcarpa − lowest − − lowest intermediate high Juglans nigra no difference intermediate moderate − lowest − highest Juglans regia − low low no difference − low intermediate Pterocarya sp. − lowest − − − lowest − Carya illinoensis − unsuccessful − − − − lowest Carya ovata − unsuccessful − − − − lowest

1Number of P. juglandis progeny (i.e., measure of reproductive success). 2Tree mortality. 3P. juglandis landing rate. 4Number and development of G. morbida cankers in laboratory/greenhouse assays.

Table 1.2 Subcortical insect associates and predators of walnut twig beetle, Pityophthorus juglandis, trapped during flight monitoring at the NCGR1 with maximum periods of annual flight (2014 to 2019) 2014 Flight Range 2015 Flight Range 2016 Flight Range 2017 Flight Range 2018 Flight Range 2019 Flight Range Coleoptera Bostrichidae Scobicia declivis LeConte 1 Nov 1 Jul 2 May 2 Apr-Aug 5 Apr-Sep 0 − Buprestidae Polycesta californica LeConte 3 May 3 May-Jun 6 May-Jun 0 − 3 May 0 − Cerambycidae Nathrius brevipennis (Mulsant) male 0 − 0 − 0 − 0 − 0 − 0 − Nathrius brevipennis (Mulsant) female 0 − 0 − 0 − 0 − 0 − 0 − Stenocorus nubifer (LeConte) 0 − 0 − 0 − 0 − 0 − 0 − Xylotrechus nauticus (Mannerheim) 0 − 11 Mar-Jul 1 Apr 2 May-Jul 7 May-Aug 0 − Scolytidae Hypothenemus eruditus Westwood 1 Sep 2 Jul-Aug 7 Apr-Sep 24 May-Oct 21 all year 0 − Pseudopityophthorus pubipennis LeConte 0 − 0 − 2 Apr-Jun 0 − 0 − 0 − Scolytus rugulosus (Müller) 4 Mar-Nov 4 Apr-Aug 0 − 3 Apr-Jun 8 Apr-Jul 0 − Xyleborinus saxeseni (Ratzeburg) 4 May-June 8 Mar-Jun 22 Mar-Sep 10 Feb-Aug 9 Mar-Jun 0 − Xyleborus affinis Eichhoff 0 − 1 Mar 0 − 0 − 0 − 0 − Neuroptera Chrysopidae Chrysopa carnea (Stephens) species group − − − − − − − − − − − − Chrysopa comanche (Banks) − − − − − − − − − − − − Chrysopa quadripunctata Burmeister − − − − − − − − − − − − Chrysopa sp. 31 all year 25 all year 36 all year 17 all year 68 all year 2 all year Hemerobiidae Hemerobius pacificus Banks − − − − − − − − − − − − Hemerobius sp. 2 May 2 Mar 1 Aug 1 Oct 1 May 0 − Raphidioptera Raphidiidae2 Agulla modesta male − − 6 April − − − − − − − − Agulla modesta female − − 6 April − − − − − − − − Agulla sp. male 7 Mar-Jun 26 Mar-Jun 12 Apr-May 16 Mar-May 0 − 0 − Agulla sp. female 11 Mar-Jun 33 Mar-Jun 17 Apr-May 24 Mar-May 2 Mar-Apr 0 − 1United States Department of Agriculture, Agricultural Research Service, National Clonal Germplasm Repository (NCGR) Juglans collection at Wolfskill Experimental Orchards in Winters, California (Solano Co.) (38°30′00.4″N, 121°58′37.1″W). 2Specimens for 2014 (2 total) and 2015 (179 total) missing and therefore not listed/identified.

Table 1.3 Subcortical insect associates and predators of walnut twig beetle, Pityophthorus juglandis, trapped during flight monitoring at the ATRC1 with maximum periods of annual flight (2015 to 2019) 2015 Flight Range 2016 Flight Range 2017 Flight Range 2018 Flight Range 2019 Flight Range Coleoptera Bostrichidae Scobicia declivis LeConte 0 − 0 − 0 − 0 − 0 − Buprestidae Polycesta californica LeConte 0 − 1 May-Jun 0 − 0 − 0 − Cerambycidae Nathrius brevipennis (Mulsant) male 0 − 20 April-Jun 15 May-Jun 30 May-Jul 0 − Nathrius brevipennis (Mulsant) female 0 − 3 May-Jun 22 May-Jun 24 May-Jul 0 − Stenocorus nubifer (LeConte) 0 − 19 Apr-May 74 May 5 Apr-May 0 − Xylotrechus nauticus (Mannerheim) 0 − 0 − 3 May 1 Apr 0 − Scolytidae Hypothenemus eruditus Westwood 0 − 0 − 0 − 1 Sep 0 − Pseudopityophthorus pubipennis LeConte 0 − 0 − 0 − 0 − 0 − Scolytus rugulosus (Müller) 0 − 0 − 0 − 1 Apr 0 − Xyleborinus saxeseni (Ratzeburg) 0 − 8 Feb-Jun 7 Mar-Jun 0 − 0 − Neuroptera Chrysopidae Chrysopa carnea (Stephens) species group 0 − 17 all year 20 all year − − − − Chrysopa comanche (Banks) 6 Sep-Dec 22 all year 5 Mar-Jun − − − − Chrysopa quadripunctata Burmeister 1 Sep 0 − 0 − − − − − Chrysopa sp. 0 − 0 − 0 − 37 all year 12 all year Hemerobiidae Hemerobius pacificus Banks 0 − 6 May-Oct 6 Feb-Jul − − − − Hemerobius sp. 0 − 0 − 0 − 8 all year 6 all year Raphidioptera Raphidiidae Agulla modesta male − − − − − − − − − − Agulla modesta female − − − − − − − − − − Agulla sp. male 0 − 5 Apr-Jun 2 May-Jun 2 May-Jun 0 − Agulla sp. female 0 − 16 Apr-Jun 8 May-Jun 5 May-Jun 0 − 1California State University Chico, Agricultural Teaching and Research Center (ATRC) in Chico, California (Butte Co.) (39°41′41.3″N, 121°49′13.1″W).

Table 1.4 Landing rates (mean no./branch/week) of insects associated with Pityophthorus juglandis, Juglans spp., and other hardwood trees on acetate sticky sheets surrounding branch sections of Juglans californica, Juglans major, or a cardboard tube (control) for Assays 1-51, Solano and Butte Co., California Assay 1 Assay 2 Assay 3 Assay 4 Assay 5 Associated insects J. californica J. major J. californica J. major J. californica J. major J. californica J. major J. californica Control Coleoptera Anobiidae Petalium californica Fall 0.08 0.13 0.01 0 1.22 0.78 0.02 0 0 0 Priobium punctatum LeConte 0.63 0.48 0 0 1.16 1.33 0.17 0.07 0 0 Ptilinus sp. 0.05 0.15 0 0 0.02 0.01 0.01 0.01 0 0.01 Bostrichidae Scobicia declivis LeConte 0.08 0.15 0.20 0.29 0.25 0.32 0.02 0 0 0 Cerambycidae Nathrius brevipennis (Mulsant) male 1.95 0.50 0 0 3.73 5.98 10.56 22.34 0 0 Nathrius brevipennis (Mulsant) female 0.08 0.1 0 0 0.40 0.55 0.28 0.57 0 0 Laemophloeidae Narthecius simulator Casey 0.10 0.13 0.01 0.01 0.15 0.05 0.04 0.04 0.01 0.03 Parandrita cephalotes (LeConte) 0.33 0.58 0 0 0.11 0.15 0 0 0 0 Scolytidae Cyclorhipidion bodoanum (Reitter) 0.05 0.10 0 0 0.04 0.03 0 0 0 0 Hypothenemus eruditus Westwood 0.05 0.10 0.74 2.09 1.45 3.61 0 0 1.40 0.57 Pseudopityophthorus pubipennis 0.03 0.08 0 0 0.02 0.01 0 0 0 0 LeConte Scolytus rugulosus (Müller) 0.08 0.15 0.05 0.03 0.23 0.15 0 0 0.01 0.01 Xyleborus affinis Eichhoff 0 0 0 0 0 0 0 0.01 0 0 Xyleborinus saxeseni (Ratzeburg) 1.75 5.78 0.1 0.31 3.58 5.37 0.96 1.91 0.47 0.17 Hymenoptera Bethylidae Plastonoxus westwoodi Kieffer 0.45 0.40 0.19 0.15 0 0.01 0 0 0 0 Braconidae Gildoria sp. − − − − 0.45 0.34 0.89 0.76 0.01 0.03 Pteromalidae Neocalosoter pityophthori (Ashmead) 0.35 0.30 0.51 0.60 0.20 0.40 0.01 0.02 0.44 0.32 1Assays were conducted during the following date ranges: Assay 1 (5 May to 6 June 2015, N=40); Assay 2 (29 August to 24 October 2015, N=80); Assay 3 (10 April to 24 June 2016, N=110); Assay 4 (13 April to 21 June 2016, N=100); and Assay 5 (27 August to 29 October 2016, N=90).

Figure 1.1 Branch sections of two Pityophthorus juglandis host plants used in the flight landing assays 1-4: a) Juglans major (3.8 cm mean diameter, 42.1 cm mean length, N =

40) and b) Juglans californica (3.7 cm mean diameter, 42.2 cm mean length, N = 40).

Photograph by S.M. Hishinuma (UC Davis Dept. of Entomology and Nematology).

Figure 1.2 Cardboard tube (4 cm diameter, 42 cm length) used as a negative control to compare Pityophthorus juglandis landing rate on a host, Juglans californica (3.8 cm mean diameter, 40.7 cm mean length, N = 10) vs. a non-host without volatiles.

Photograph by I.D. Lona.

Figure 1.3 Juglans branch section surrounded by a mesh screen bag hung from a 1.5 m tall stainless steel conduit pole in the field (left). An adhesive-coated acetate sheet was placed around the bagged branch section (right) to trap adult Pityophthorus juglandis and associated insects as they attempted to land. Photograph by I.D. Lona.

Figure 1.4 Walnut twig beetle (WTB), Pityophthorus juglandis, flight at the National Clonal Germplasm Repository (NCGR) at

Wolfskill Experimental Orchards in Winters, California (2014 to 2019) and at the California State University Chico Agricultural

Teaching and Research Center (ATRC) in Chico, California (2015 to 2019). Note the different Y-axis scales for the graphs by location. Five flight landing assays were conducted during the following date ranges: Assay 1 (5 May to 6 June 2015, NCGR); Assay

2 (29 August to 24 October 2015, NCGR); Assay 3 (10 April to 24 June 2016, NCGR); Assay 4 (13 April to 21 June 2016, ATRC); and Assay 5 (27 August to 29 October 2016, NCGR). Arrow points to the date range (4 September to 11 September 2018) when a single female Hypothenemus eruditus was trapped at the ATRC. Daily maximum temperatures and precipitation (Winters A., CIMIS weather station #139 and Chico C., NCDC weather station #1715, CSU Chico University Farm) are: Assay 1 (19.4° C to 32.2° C, 1 mm), Assay 2 (20.6° C to 40° C, 0.3 mm to 1.8 mm), Assay 3 (17.8° C to 37.2° C, 0.3 mm to 5.6 mm), Assay 4 (16.7° C to 38.3° C,

0.5 mm to 16.5 mm), Assay 5 (17.2° C to 36.1° C, 0.3 mm to 9.6 mm).

Figure 1.5 Weekly landing rates (mean ± SE) of male and female walnut twig beetle

(WTB) Pityophthorus juglandis (N=330), female Xyleborinus saxeseni (N=330), female

Hypothenemus eruditus (N=230), and male and female Nathrius brevipennis (N=330) on adhesive-coated acetate sheets surrounding branch sections of Juglans californica and

Juglans major (N=330). Landing rates from four assays were combined (Assay 1: 5 May to 6 June 2015; Assay 2: 29 August to 24 October 2015; Assay 3: 10 April to 24 June

2016; and Assay 4: 13 April to 21 June 2016). Assays 1 through 3 were conducted at the

National Clonal Germplasm Repository (NCGR) Juglans collection at Wolfskill

Experimental Orchards in Winters, California (Solano Co.). Assay 4 was conducted at the

California State University Chico Agricultural Teaching and Research Center (ATRC) in

Chico, California (Butte Co.). Weekly landing rates of female Hypothenemus eruditus were only recorded at Wolfskill (Assays 1-3, N=230). Different letters and superscripted letters above pairs of histogram bars indicate significantly different mean responses compared by walnut species (P<0.05).

Figure 1.6 Weekly landing rates (mean ± SE) of male and female walnut twig beetle

(WTB), Pityophthorus juglandis, female Xyleborinus saxeseni, and female

Hypothenemus eruditus on adhesive-coated acetate sheets surrounding branch sections of

Juglans californica and cardboard tubes (negative control) (N=90). Landing rate Assay 5 was conducted at the National Clonal Germplasm Repository (NCGR) Juglans collection at Wolfskill Experimental Orchards in Winters, California (Solano Co.) from 27 August to 29 October 2016. Different letters and superscripted letters above pairs of histogram bars indicate significantly different mean responses compared between J. californica and the negative control (P<0.05).

Supplemental Table 1.A Accession numbers of branches collected from Juglans californica and Juglans major trees at locations NCGR1 and UC Davis2 for Assays (1-5) of Pityophthorus juglandis landing rates

Juglans californica Assays Accession Number Location Date Collected Days in Cold Storage 1 New Stuke Block UC Davis 2 May 2015 6 2 New Stuke Block UC Davis 22 Aug 2015 6 3 and 4 New Stuke Block UC Davis 4 Apr 2016 5 5 New Stuke Block UC Davis 26 Aug 2016 3 3 and 4 DJUG0014.0007A NCGR 2 Apr 2016 7 5 DJUG0014.0007A NCGR 2 Apr 2016 140 3 and 4 DJUG0017.0005A NCGR 2 Apr 2016 7 5 DJUG0017.0005A NCGR 2 Apr 2016 140 3 and 4 DJUG0018.0004A NCGR 2 Apr 2016 7 5 DJUG0018.0004A NCGR 2 Apr 2016 140 3 and 4 DJUG0020.0015A NCGR 2 Apr 2016 7 5 DJUG0020.0015A NCGR 2 Apr 2016 140 3 and 4 DJUG0021.0018A NCGR 2 Apr 2016 7 3 and 4 DJUG0024.0012A NCGR 2 Apr 2016 7 5 DJUG0024.0012A NCGR 2 Apr 2016 140 3 and 4 DJUG0023.0018A NCGR 2 Apr 2016 7 5 DJUG0023.0018A NCGR 2 Apr 2016 140 3 and 4 DJUG0059.0015A NCGR 2 Apr 2016 7 5 DJUG0059.0015A NCGR 2 Apr 2016 140 3 and 4 DJUG0028.0007A NCGR 2 Apr 2016 7 5 DJUG0028.0007A NCGR 2 Apr 2016 140 5 DJUG0022.0017A NCGR 26 Aug 2016 3

Juglans major Assays Accession Number Location Date Collected Days in Cold Storage 1 DJUG0069.0014A NCGR 2 May 2015 6 2 DJUG0069.0014A NCGR 22 Aug 2015 6 3 and 4 DJUG0069.0014A NCGR 2 Apr 2016 7 3 and 4 DJUG0082.0008A NCGR 2 Apr 2016 7 3 and 4 DJUG0083.0001A NCGR 2 Apr 2016 7 3 and 4 DJUG0084.0004A NCGR 2 Apr 2016 7 3 and 4 DJUG0079.0011A NCGR 2 Apr 2016 7 3 and 4 DJUG0078.0007A NCGR 2 Apr 2016 7 3 and 4 DJUG0076.0015A NCGR 2 Apr 2016 7 3 and 4 DJUG0075.0008A NCGR 2 Apr 2016 7 3 and 4 DJUG0070.0008A NCGR 2 Apr 2016 7 3 and 4 DJUG0068.0016A NCGR 2 Apr 2016 7 1United States Department of Agriculture, Agricultural Research Service, National Clonal Germplasm Repository (NCGR) Juglans collection at Wolfskill Experimental Orchards in 43

Winters, California (Solano Co.) (38°30′00.4″N, 121°58′37.1″W). 2University of California, Davis, Hutchison Rd. Juglans Collection, New Stuke Block, HQ Californica, calif 97-040, Davis (Yolo Co.), California (38°32′21.2″N, 121°47′45.1″W).

Prepared for Agricultural and Forest Entomology

CHAPTER II

Effect of water stress on the susceptibility of commercial English walnut, Juglans regia, to the walnut twig beetle, Pityophthorus juglandis

Irene D. Lona1, Donald G. Miller III1, Colleen Hatfield1, Richard Rosecrance2, Bruce D. Lampinen3, Yigen Chen4, and Steven J. Seybold5*

1Department of Biological Sciences, California State University, Chico CA 95929, USA,

2College of Agriculture, California State University, Chico CA 95929, USA,

3Department of Plant Sciences, University of California, One Shields Avenue, Davis CA 95616, USA,

4E. & J. Gallo Winery, 600 Yosemite Boulevard, Modesto, CA 95354, USA, and

5Pacific Southwest Research Station, USDA Forest Service, 1731 Research Park Drive, Davis, CA 95618, USA

*Correspondence: Steven J. Seybold, E-mail: [email protected] 45

Abstract

1 Drought increases susceptibility of some plants to attack by herbivorous insects, and

this is described formally in the plant-stress hypothesis. We investigated the

relationship between English walnut, Juglans regia, tree health, represented by

several levels of plant water stress, and walnut twig beetle, Pityophthorus juglandis

Blackman (Coleoptera: Scolytidae), infestation and reproductive success. We

predicted that walnut trees under water stress would have more P. juglandis attacks

and more developing brood, reflected in a greater number of entrance and emergence

holes through the outer bark of branches. We also applied this hypothesis to the

number of emergence holes from an invasive longhorned beetle, Nathrius brevipennis

(Mulsant) (Coleoptera: Cerambycidae), and a parasitoid likely associated with N.

brevipennis, Gildoria sp. Hedqvist (Hymenoptera: Braconidae: Doryctinae).

2 We found that P. juglandis aggregation pheromone-baited branches from trees in a

high water stress group had greater P. juglandis activity than unbaited branches from

trees in a control group (with generally intermediate water stress), which had zero

entrance or emergence holes. However, baited branches from trees in a low water

stress group had greater P. juglandis activity than baited branches from trees in the

high water stress group. The latter effect may have been a consequence of data from

one outlier branch in the low water stress group. Excluding the outlier, there was no

significant difference in P. juglandis activity between the two groups. There were

also no significant relationships between levels of N. brevipennis or Gildoria sp.

activity and water stress.

3 The majority of P. juglandis entrance and emergence holes on branches were near the

aggregation pheromone lures that were used to bait branches and on the aspect where

the lures were located (bottom of the branch). Holes associated with P. juglandis

occurred primarily on branches (4.0 cm mean diameter, 2.0 to 7.7 cm), whereas

emergence holes associated with N. brevipennis and Gildoria sp. occurred primarily

on secondary branches (twigs) (1.0 cm mean diameter, 0.3 to 4.5 cm).

4 Distances of N. brevipennis and Gildoria sp. emergence holes from the P. juglandis

aggregation pheromone lure were not correlated and did not suggest a kairomonal

effect of the P. juglandis pheromone for N. brevipennis.

5 In conjunction with this study, we surveyed an adjacent J. regia orchard for beetle

flight response to ethanol, a common plant stress signal. Although P. juglandis did

not respond to ethanol, two invasive woodborers, male and female N. brevipennis and

female Xyleborinus saxeseni, responded significantly to this signal. Two invasive

scolytids, Hypothenemus eruditus and Xyleborus affinis, were also caught in funnel

traps in this survey, establishing the northernmost records of these species in

California.

Keywords drought • Gildoria sp. • host colonization • Nathrius brevipennis • Scolytidae

• thousand cankers disease • walnut twig beetle

Introduction

Many species of walnut trees, Juglans, serve as host plants supporting the mating and development (i.e., reproductive hosts) of the walnut twig beetle, Pityophthorus juglandis

Blackman (Coleoptera: Scolytidae) (Seybold et al., 2016; Hefty et al., 2018). This beetle is native to the western USA and was first collected in 1959 in California (Bright, 1981).

Invasive populations were discovered in the eastern USA in 2010 (Seybold et al., 2013a,

2016) and in five regions of Italy by 2018 (Faccoli et al., 2016; Moricca et al., 2018).

Pityophthorus juglandis is a phloem feeder. Adult male P. juglandis bore through the outer bark and produce an aggregation pheromone that attracts both males and females, eliciting a mass attack on larger branches or stems of Juglans spp. (Seybold et al., 2016).

Pityophthorus juglandis vectors a phytopathogenic fungus, Geosmithia morbida (Kolařík et al., 2011), and the activity of the beetle and fungus results in thousand cankers disease

(TCD). Spores of G. morbida are transferred by adult P. juglandis through feeding and brood gallery construction in the phloem of the host plant (Seybold et al., 2013a). Trees with TCD have dark, wet stain spots on the bark surface and cankers in the phloem around P. juglandis entrance holes (Graves et al., 2009). Accumulation of P. juglandis galleries causes numerous cankers in the branches and stem, which can result in crown dieback and tree mortality (Tisserat et al., 2009; Seybold et al., 2013a, 2019).

The plant-stress hypothesis is a general ecological principle proposed by White

(1969) in relation to psyllid outbreaks, which may arise in response to stress in the host plant. Stress factors for trees are diverse and range from defoliation, root disease, and fire injury, to drought (Raffa et al., 2016). Water stress in trees can be a result of insufficient or too much water and is revealed by symptoms such as leaf wilt, and sunburn. In

California orchards, varieties of English walnut, Juglans regia L., are particularly sensitive to irrigation and water stress (Fulton & Buchner, 2006).

Drought might be a factor that increases susceptibility of Juglans spp. to P. juglandis attacks. The probability of bark beetle infestation and disease transmission likely depends on the health of individual trees. For example, Kelsey & Joseph (2001) reported more frequent attacks by a Douglas-fir engraver, Scolytus unispinosus, on water- stressed Douglas-fir branches but no attacks on control branches. Kelsey et al. (2014) also showed that Aleppo pine trees attacked by a shoot-feeding bark beetle, Tomicus destruens, had higher water stress and higher ethanol content in the phloem than unattacked trees. However, Huberty & Denno (2004) analyzed various studies on insect responses to water-stressed host plants and found that phloem-feeding insects responded positively to plants with intermediate stress and performed poorly on plants under continuous stress. Studies with flowering dogwood in nurseries showed higher ethanol content and greater attacks by an ambrosia beetle, the black stem borer, Xylosandrus germanus, in trees experiencing floods when compared to non-flooded trees (Ranger et al., 2010, 2013). This result was confirmed under field conditions on flood-intolerant dogwood and flood-tolerant species such as silver maple and swamp white oak (Ranger et al., 2015).

The susceptibility of water-stressed Juglans to P. juglandis has not been examined, so we investigated the relationship between this stress factor and P. juglandis infestation rates. We predicted that English walnut trees under water stress would have a greater number of P. juglandis attacks and more developing brood. We also applied this hypothesis to an invasive longhorned beetle, Nathrius brevipennis (Mulsant)

(Cerambycidae), and a parasitoid likely associated with N. brevipennis, Gildoria sp.

Hedqvist (Hymenoptera: Braconidae: Doryctinae). Both species co-occur with P. juglandis in J. regia in California. In conjunction with this study, we also surveyed beetle flight response to ethanol, a common plant stress signal. More broadly, we conducted this study to improve our understanding of how drought stress may affect colonization success by bark and woodboring beetles.

Materials and methods

Study site

This study was conducted in an English walnut, J. regia L., orchard at the California

State University Chico, Agricultural Teaching and Research Center (ATRC) in Chico,

California (Butte Co.) (39°41′33.9″N, 121°49′17.5″W). The orchard block chosen for the study comprised four rows, each with 53 trees consisting of multiple J. regia cultivars.

This block was chosen primarily for its irregular irrigation system, which provided variation in stem water potential amongst the trees. The survey of beetle flight response to ethanol was conducted in the center of a three-row block in an adjacent area of the orchard consisting only of J. regia 'Chandler' cultivars (39°41′41.96″N, 121°49′14.92″W).

Measuring stem water potential of Juglans regia

Stem water potential (SWP) is a plant-based measure of plant water stress (McCutchan &

Shackel, 1992). The measured SWP, which is recorded in units of negative megapascals

(1 MPa = 10 bar), equals the pressure only sufficient to bring water to the leaf stem cut surface. Juglans regia under low stress have a SWP value of approximately −0.4 MPa

(Fulton & Buchner, 2006; Fulton et al., 2014). Trees presenting lower SWP (more negative values) experience higher stress. The SWP of J. regia at the study site was measured on five occasions from 2015 to 2018: 15 Sep 2015; 15 July 2016; 6 Sep 2017;

15 Sep 2017; 27 Jul 2018; and 31 Jul 2018. We attempted to measure the SWP at least 7 days after irrigation to ensure that the trees were in a water-stressed state. Because the measurements recorded on 6 Sep 2017 were taken on a single day after the orchard was watered, we repeated the measurements on 15 Sep 2017. The SWP was measured on every third tree of the innermost three rows (N = 51 trees).

To measure midday SWP, one terminal leaflet from a compound leaf from each selected tree was covered with a reflective mylar foil bag for 1 hour, until the leaf water potential reached equilibrium with the SWP (McCutchan & Shackel, 1992; Fulton et al.,

2014). Leaflets were thus “bagged” at noon, and SWP was measured at midday from

13:00 to 15:00 when plant tension was highest (Fulton et al., 2014). The bagged leaflets were located on the northeast side of the trees, facing away from the sun, under the crown, and closest to the trunk to get a more accurate measurement of SWP. The leaflets were excised from the trees by using a razor blade and placed in a pressure chamber

[Plant Water Status Console (home built stand up model) from the laboratory of BDL], with the mylar foil bag still surrounding the leaflet (Goldhamer & Fereres, 2001; Fulton et al., 2014). Throughout the day on each measurement date, the temperature and humidity were also recorded (Table 1). The fully watered baseline was calculated as described in Fulton et al. (2014) by using temperature and humidity data from the nearby

CIMIS Weather Station (Durham, California Station #12).

Baiting trees with Pityophthorus juglandis aggregation pheromone

On 15 Sep 2015, ten J. regia with SWP measurements less negative than −0.87 MPa were selected and placed into a low water stress group, whereas ten J. regia with measurements more negative than −1.48 MPa were selected and placed into a high water stress group. These were empirically structured groups, established based on the data at the time. On 24 Aug 2016, one branch from each of these 20 trees was baited with the male-produced aggregation pheromone of P. juglandis (Seybold et al., 2013b; Chen &

Seybold, 2014) (Fig. 1). Lures were attached with a wire around the middle of each branch and left in place for about 2 years to elicit colonization by P. juglandis. The baited branches were 3 to 4.5 m from the orchard floor, on the lower to mid-crown and located generally on the west side of the walnut trees. Five more trees with an intermediate SWP

(−1.0 to −1.4 MPa) were placed in a control group and were not baited. The flight activity-abundance of P. juglandis at this location was monitored weekly through a funnel trap survey (Lona et al., 2019), and the population density here was inferred to be relatively low, given activity-abundances measured at other northern California locations

(Chen & Seybold, 2014; Lona et al., 2019).

Collecting branches in the field

Branches from baited and control trees (25 branches in total) were collected from June to

July 2018. Branches were cut at least 40 cm away from the lure, and the position where the lure hung was marked on the bark surface in the field with a Sharpie marking pen.

The five control branches were marked as though a lure had been placed (approximately at the middle of the branch). In the field, the leaves on the branches were trimmed;

however, side branches were left intact. Branches ranged from 150 to 300 cm in length.

In order to examine branches soon after they were removed from the field, five branches were collected at once and stored at 13 °C for a maximum of three days before dissection.

Branch dissection

To record and map insect activity from each branch in the laboratory, three 27-cm linear sections were assigned from the location of the lure to the proximal branch end (i.e., toward the tree stem), and three identical sections were assigned from the location of the lure to the distal end of the branch. The aspects of each branch section (i.e., right, left, top, and bottom) were also subdivided for analysis (Fig. 2 and 3). Right and left sides were assigned from the perspective of the proximal end extending toward the distal end.

Pityophthorus juglandis entrance and emergence holes were marked and counted for each section and aspect. The distance from each P. juglandis hole to the lure location was measured. At each hole, the bark was scraped off with a razor blade to measure the

P. juglandis gallery length and to record any P. juglandis adults and progeny (eggs and larvae). Presence or absence of cankers was also noted. Additionally, branches were examined for N. brevipennis and Gildoria sp. emergence holes, which were typically on the smallest diameter branches and twigs. Side branches and twigs were divided into sections of 20 cm and examined for P. juglandis, N. brevipennis, and Gildoria sp. holes

(Fig. 4). On the side branches and twigs, the distances of holes to the lure were also measured.

Flight monitoring

Trees under stress may produce and accumulate ethanol in their tissues, which may elicit a response from herbivorous insects such as bark and ambrosia beetles (Kirkland &

Kelsey, 2015). To investigate the abundance and diversity of ethanol-responding bark and ambrosia beetles at our study site, we added four 16-unit funnel traps with wet cups

(Lindgren, 1983) from 22 May - 11 Dec 2018. The traps were suspended on ~3 m long stainless steel conduit poles near trees with crown gall, a disease caused by the bacterial root pathogen, Agrobacterium tumefaciens. Crown gall is a concern with irrigation management (Fulton & Buchner, 2006) and may act as an additional stressor. We anticipated that these infected trees might provide an elevated source of ethanol- responsive beetles. Approximately 100 ml of ethanol-free marine antifreeze (Winter Safe

Antifreeze, product #31200SLVV3, Kinpak Inc., Montgomery, AL, USA) containing propylene glycol were added into the trap cups to immobilize any captured beetles.

Ethanol lure pouches (High Release Rate, Contech Enterprises, black plastic pouch,

121.5 g load, 280 mg/d at 20°C, product no. 200000362, Lot #14120, Ethanol Lure UHR) were placed on two of the funnel traps, whereas the other two traps were unbaited

(negative controls). The ethanol lures were placed initially on 31 May 2018 and replaced

31 Jul 2018. Every seven days, traps were randomized among four fixed location trap stations, and the contents of the trap cups were collected by pouring the antifreeze laden with insects through conical paper paint strainers with nylon mesh (Basecoat/Clearcoat

Strainers, Fine 190 Micron 1000CT, product Gerson-10601, Toolrage.com, Dillsburg,

PA, USA) and folding and placing the filters in plastic bags (Seybold et al., 2013b).

Collected samples were frozen and then sorted in the laboratory to determine the number of species of bark and ambrosia beetles and other woodboring insects.

Data handling and statistical analyses

Stem water potentials: Because trees under high and low levels of water stress were all baited and trees under an intermediate level of water stress served as unbaited controls,

SWPs were analyzed separately by water stress (three levels) and by bait status (2 levels: baited and unbaited).

To investigate if there was a correlation between the SWPs of the water stress groups and measurement date, SWP data were analyzed by a two-way ANOVA [function gls() in nlme package in R (R Core Team, 2016)]. Water stress groups (three levels: low, high, and control), and measurement dates (six levels: 15 Sep 2015; 15 Jul 2016; 6 Sep

2017; 15 Sep 2017; 27 Jul 2018; and 31 Jul 2018) were defined as two factors. Stem water potentials between different measurement dates from the same tree were correlated and allowed to vary [correlation function varSymm() in R (R Core Team, 2016)]. Water potential variances among the measurement dates varied greatly and were modeled by function varIdent() in R (R Core Team, 2016). Model assumptions were checked by various residual plots (i.e., standardized residuals against predicted values and each factor, quantile-quantile plot of standardized residuals). Because the interaction between water stress groups and measurement date was significant, effects of the water stress groups on SWP were further analyzed separately by measurement date. Multiple mean comparisons between water stress groups were adjusted by Tukey’s procedure. The

distance of study trees to the edge of orchard did not significantly affect water potential, and this factor was excluded from future models.

Stem water potentials in response to bait status and measurement date were analyzed similarly as described above. Because the interaction between bait status and measurement date was significant, the effect of bait status on SWP was analyzed separately by measurement date.

Insect response to water stress: The P. juglandis entrance or emergence holes in branches in the three water stress groups (low, high, and control) were analyzed by a generalized linear mixed model [function glmer.nb () in package lme4 in R (R Core

Team, 2016)], with water status (i.e., high and low) as a fixed factor and the diameter of branches as a covariate. Since no P. juglandis holes were found in control (unbaited) branches, these branches were excluded from the analyses. The distribution of the

“catches” (i.e., signs of colonization on the branch sections) was modeled as a negative binomial.

The numbers of N. brevipennis and Gildoria sp. emergence holes were analyzed by a generalized linear mixed model [function glmer.nb () in package lme4 in R (R Core

Team, 2016)] with water status (i.e., high and low) or bait status (i.e., baited and unbaited) as fixed factors, and secondary branches from each main branch (from each tree) as mixed factors. Diameter of side branches was a covariate.

Comparisons between branch diameters with P. juglandis, N. brevipennis, and

Gildoria sp. holes were analyzed by non-parametric Kruskal-Wallis test (function

“kruskal.test” in base R). A multiple mean comparison was adjusted by Bonferroni’s test

(α = 0.05/3).

The distribution of trap catches from the study of beetle flight response to ethanol was modeled as a negative binomial generalized linear mixed model [function glmer.nb() in package lme4 in R (R Core Team, 2016)]. Sampling period was treated as a random factor. Although traps were in place from May to December, only data within the flight period of each taxon were used in the analysis. Therefore, based on previous observations, N. brevipennis results after July and X. saxeseni results after October (zero catches in both cases) were excluded (Lona et al., 2019).

Voucher specimens and nomenclature

All specimens of Coleoptera of interest were identified by SJS with consultation with Dr.

Donald E. Bright, Jr., Colorado State University, Fort Collins, Colorado, for bark and ambrosia beetles. Voucher specimens of adults were accessioned into the Entomology

Department at the California Academy of Sciences, San Francisco, California and the

Entomology Collection at California State University, Chico. For this study, we have elected to use the original nomenclature for bark and ambrosia beetles (Coleoptera:

Results

Stem water potential measurements

Combining data from all measurement dates showed that the three water stress groups

2 had significantly different SWP (X1 = 243.46, P < 0.001). Walnut trees selected originally for the high water stress group had lower SWP values throughout the study, indicating that those trees were under higher water stress for the duration. In contrast, trees selected originally for the low water stress group had higher SWP values, indicating that low water stress was maintained for this group for the duration of the study. The control group with intermediate water stress maintained SWP values between the low and the high stress group. For most measurement dates, there were significant differences in

SWP among the three water stress groups; however, that difference declined over time

2 2 (Table 1, Fig. 5): 15 Sep 2015 (X1 = 294.56, P < 0.001); 15 Jul 2016 (X1 = 117.22, P <

2 2 0.001); 6 Sep 2017 (X1 = 2.63, P = 0.27); 15 Sep 2017 (X1 = 14.94, P < 0.001); 27 Jul

2 2 2018 (X1 = 7.04, P = 0.03); and 31 Jul 2018 (X1 = 1.78, P = 0.41). The SWP readings taken on 6 Sep 2017 were probably affected by the short period of time following the previous round of irrigation, which likely caused all trees to manifest comparatively low stress levels, resulting in no significant differences among experimental groups.

Measurements taken 9 days later indicated that the study trees were again experiencing significant stress. The trees that had not been watered for 14 days prior to the measurements on 31 Jul 2018 were all under high stress. Combining all water stress groups, SWP values decreased significantly over time, indicating greater water stress by

2 the end of the study (X1 = 336.71, P < 0.001). One tree within the low stress group was an outlier: Stem water potential values were lower (higher stress) on that tree than on

other trees in the low stress group for the measurement on 6 Sep 2017 (Fig. 5). For the subsequent measurements, the SWP of the outlier tree remained lower than the rest of the trees in the low stress group and even reached levels similar to the high stress group.

Overall, there was no significant difference in SWP values among trees with an aggregation pheromone-baited branch (low and high water stress groups) and trees with

2 an unbaited branch (controls) (X1 = 0.01, P = 0.92) (Fig. 6). There was also no significant difference in SWP measurements among trees in the baited and unbaited

2 groups within individual SWP measurement dates except for 15 Jul 2016 (X1 = 5.12, P =

2 0.02) and 15 Sep 2017 (X1 = 10.15, P = 0.001).

Pityophthorus juglandis colonization activity on branches

There was a marginally significant difference in number of P. juglandis entrance and

2 emergence holes among water stress groups (X1 = 3.59, P = 0.06). A total of 105 P. juglandis holes were recorded on branches in the high water stress group. A total of 261

P. juglandis holes were recorded on branches in the low water stress group; however, 194 of those holes were from a single branch. Therefore, without the outlier branch, only 67 holes were observed in the low water stress group. However, even with the omission of this outlier, there was no significant difference in the number of P. juglandis holes by

2 stress group (X1 = 0.01, P = 0.91). No P. juglandis entrance or emergence holes were recorded on control branches.

For both low and high water stress groups, the P. juglandis holes were located with a mean distance of 19 cm to the aggregation pheromone lures (including the outlier branch in the low stress group) (Fig. 7) and on the aspect of the branch where the lures

were located, i.e., the bottom of the branch (Fig. 8). Excluding the outlier group in the low water stress group, the mean distance of P. juglandis holes to the lures was 17 cm.

Combining data from all groups, there was no difference in the number of holes with respect to linear orientation around the lure (proximal versus distal ends of the branch).

Proximal mean distance was 21 cm and distal mean distance was 17 cm. Branch diameter

(all groups combined) did not have a significant effect on number of P. juglandis

2 entrance or emergence holes (X1 = 0.39, P = 0.53). However, P. juglandis holes were found primarily on branches with diameters ranging from 2 to 6.75 cm in the high water stress group (4.06 cm mean diameter), and 2.35 to 7.65 cm in the low water stress group

(4.3 cm mean diameter) (Table 2, Fig. 9). Only one adult P. juglandis (male) was found beneath the bark in a branch in the high water stress group. No larvae or eggs were found in galleries on any of the branches from the two groups. There were 105 galleries in branches from the high water stress group with a gallery length ranging from 0.1 to 1.2 cm (mean of 0.4 cm). In branches from the low water stress group (including the outlier branch) there were 261 galleries with a gallery length ranging from 0.1 to 1.1 cm (mean of 0.35 cm). Excluding the outlier, the low water stress group had a total of 67 galleries ranging from 0.1 to 1.1 cm in length (mean of 0.38cm). Because there were no entrance or emergence holes, we assumed that there were no P. juglandis galleries in branches from the control group.

Nathrius brevipennis and Gildoria sp. colonization activity on branches

Secondary branches from all water stress group levels (low, high, and control) had many

N. brevipennis and Gildoria sp. emergence holes. No holes were found on the main

branches, with the exception of the outlier branch from the low water stress group (15 holes from N. brevipennis and 2 holes from Gildoria sp.). There was no significant relationship between the number of N. brevipennis or Gildoria sp. holes and water stress

2 2 groups (N. brevipennis: X1 = 0.01, P = 0.99; Gildoria sp.: X1 = 2.71, P = 0.26) or branch

2 2 diameter (N. brevipennis: X1 = 1.97, P = 0.16; Gildoria sp.: X1 = 0.76, P = 0.38).

Nathrius brevipennis emergence holes were found in branches ranging from 0.4 cm to

2.55 cm in diameter in the high water stress group (0.84 cm mean diameter); 0.3 to 4.5 cm in the low water stress group (1.08 cm mean diameter); and 0.4 to 2.75 cm in the control group (0.83 cm mean diameter) (Table 2, Fig. 9). Gildoria sp. emergence holes were found in branches ranging from 0.4 cm to 2.4 cm in diameter in the high water stress group (0.88 cm mean diameter); 0.35 to 3.35 cm in the low water stress group (0.95 cm mean diameter); and 0.3 to 1.2 cm in the control group (0.81 cm mean diameter)

(Table 2, Fig. 9). Bait status of the branch sections did not have a significant effect on the

2 number of N. brevipennis or Gildoria sp. emergence holes (N. brevipennis: X1 = 0.002, P

2 = 0.96; Gildoria sp.: X1 = 2.68, P = 0.10). There was also no indication of response between N. brevipennis and Gildoria sp. holes and proximity to the P. juglandis aggregation pheromone lure (Fig. 10 and 11). However, holes of both insects tended to be further away from the aggregation pheromone lures than the P. juglandis holes. The mean distance of N. brevipennis emergence holes to the lure was 149 cm and the mean distance of Gildoria sp. holes to the lure was 142 cm. Based on the distance to the P. juglandis aggregation pheromone lure only, there was no correlation in the distance between N. brevipennis and Gildoria sp. emergence holes (Spearman correlation −0.14).

Branch diameter comparisons among the locations of P. juglandis, N. brevipennis,

2 and Gildoria sp. entrance and emergence holes showed a significant difference (X1 =

623.4, P < 0.001) with P. juglandis entrance and emergence holes located in branches with a mean diameter of 4.0 cm; N. brevipennis emergence holes located in twigs with a mean diameter of 0.96 cm; and Gildoria sp. emergence holes located in twigs with a mean diameter of 0.91 cm (Table 2, Fig. 9).

Flight monitoring for response to ethanol

Insects trapped in the ethanol-baited traps included N. brevipennis; the fruit-tree pinhole borer, Xyleborinus saxeseni (Ratzeburg); the minute bark beetle, Hypothenemus eruditus

Westwood (both Scolytidae); and the lead cable borer, Scobicia declivis LeConte

(Bostrichidae) (all Coleoptera) (Table 3). These insects responded positively to the ethanol lures. A total of 60 male and 146 female N. brevipennis were trapped from May to the end of July 2018. Eight N. brevipennis were collected in control traps and 198 were collected in the ethanol-baited traps. A total of 231 female X. saxeseni were trapped from

May to October 2018. A single female X. saxeseni was trapped in a control funnel trap.

Male X. saxeseni are flightless, so none were trapped. The ethanol-baited traps had more

2 2 male and female N. brevipennis (males: X1 = 9.43, P = 0.002; females: X1 = 17.76, P <

2 0.001) and female X. saxeseni (X1 = 26.11, P < 0.001) (Fig. 12).

Two female H. eruditus were trapped in August. Scobicia declivis (19 total) were trapped from May to October and only in the ethanol-baited traps. Between 26 Sep – 2

Oct 2018, one specimen of Xyleborus affinis Eichhoff (Scolytidae) was trapped in an unbaited (control) funnel trap. Another longhorned beetle, Xylotrechus nauticus

(Mannerheim) was trapped (6 total) from May to August, one of which was trapped in the control. One Gildoria sp. (July) and one Priobium punctatum LeConte (Anobiidae) (late

June) were trapped in funnel traps baited with ethanol lures.

Discussion

Variation in water stress on J. regia did not affect the activity patterns of P. juglandis.

Furthermore, a typical plant stress signal (ethanol) was unattractive to P. juglandis.

Ethanol was attractive to two other woodboring insects (X. saxeseni and N. brevipennis), the latter of which colonized branches on trees in the water stress study.

The SWP measurements from 2015 to 2018 of trees in low and high water stress groups were significantly different in the beginning of the study. However, by 2018, we recorded lower SWPs from trees in the low water stress group to the point that this cohort resembled trees in the high water stress group, even though the temperature remained similar across the measurement periods (~30 °C). Thus, in 2018, all study trees had low

SWP values, indicating that all study trees were under water stress. One factor influencing the level of water stress was that the irrigation schedule in 2018 changed from weekly to biweekly. The reduced difference in SWP values between the high and low water stress groups moderated our capacity to draw substantive conclusions about the effect of water stress on P. juglandis colonization and reproductive success.

Surprisingly, at first analysis there were marginally significantly more P. juglandis entrance and emergence holes on branches from trees in the low water stress group, compared to the high water stress group. However, this difference may have been due to the single outlier branch within the low water stress group. The outlier branch was

on a tree whose SWP measurements were lower than other trees in the low water stress group indicating that it may have been under greater water stress (Fig. 5). This outlier branch had many more entrance and emergence holes than any other branch in the study

(Fig. 3). When the data were re-analyzed without this outlier, the branches in the low water stress group had fewer P. juglandis holes than those in the high stress group; however, even with the omission of data from the outlier branch, the difference was not significant.

This study confirmed that the P. juglandis aggregation pheromone lure is useful for inducing P. juglandis attack on individual branches. Pityophthorus juglandis concentrated its attacks around the location of the aggregation pheromone lures that had been placed on tree branches of both low and high water stress groups. There were no P. juglandis attacks on any of the control (unbaited) branches, which suggests that the beetles responded primarily to the lure in our study and were not landing randomly at an appreciable rate in this orchard. Coleman et al. (2011) showed that colonization of

Quercus agrifolia by the goldspotted oak borer, Agrilus auroguttatus, may have been the cause of water stress. However, in this study, water stress between trees with baited branches (low and high water stress group) and control trees with unbaited branches

(intermediate water stress) was not significantly different except for two measurement dates (Fig. 6). This suggests that P. juglandis attacks did not result in increased water stress. Pityophthorus juglandis entrance and emergence holes occurred on branches with a mean diameter of 4 cm, and primarily on the bottom aspect of these branches (i.e., where the lure was positioned). However, P. juglandis did not seem to have a preferred location for colonization at a larger scale (i.e., distal or proximal ends of the branch) on

aggregation pheromone-baited branches. Although no P. juglandis eggs, larvae, or pupae were found, and we could not measure details of reproductive success, we did establish that there might be an effect of water stress on P. juglandis attacks from the numbers of entrance and emergence holes observed on the high water stress group and the outlier from the low water stress group. This study might profitably be repeated for a longer duration, on trees with more pronounced SWP differences, and with more frequent replacement of the aggregation pheromone lures. A follow-up study might also include more replication to reduce the confounding effect of individual branches.

Neither N. brevipennis nor Gildoria sp. showed a preference for colonizing branches on trees in different water stress groups. All branches were attacked by N. brevipennis and Gildoria sp. exclusively on the small side branches or twigs with an approximate mean diameter of 1 cm, and near the ends of the main branches. Nathrius brevipennis emergence hole distances to the male P. juglandis aggregation pheromone

(mean distance of 142 cm) indicate that the P. juglandis aggregation pheromone did not act as a kairomone for ovipositing female N. brevipennis. This is not surprising, given the

Nearctic origin of P. juglandis (as opposed to the Palearctic origin of N. brevipennis).

However, N. brevipennis and Gildoria sp. emergence holes were present in experimental trees from all water stress groups, which underscored the abundance of these taxa in this orchard.

Gildoria sp. holes were present near or on branches parallel to N. brevipennis holes; however, there was no significant correlation between the distance between N. brevipennis emergence holes and Gildoria sp. emergence holes. The mean distance to the

P. juglandis aggregation pheromone lure was 142 cm, which is similar to the mean

distance of N. brevipennis emergence holes to the lure. In this study, the distances between N. brevipennis and Gildoria sp. holes were compared by using distance from the

P. juglandis aggregation pheromone lures. To better understand if there is a correlation in distance between the emergence holes of these two species, the distance from each

Gildoria sp. hole should be measured to the nearest N. brevipennis hole.

Very little is known about this woodborer and its purported parasitoid in their native or naturalized ranges. Nathrius brevipennis is native to southern Europe and has spread to northern Europe and Africa, Great Britain, Argentina, and the USA (Linsley,

1963). Known host trees in California include Juglans, Quercus, and Ficus. The flight season is reported to occur from June to August (Linsey, 1963); however, data from studies conducted at the ATRC and at another site in Yolo Co. (2015 - 2016) recorded the flight season beginning as early as March and ending in August (Lona et al., 2019). Late- stage larvae were present in the xylem of J. regia twigs in the field in February and

March in northern California. Gildoria sp. specimens caught in these studies are likely an undescribed species (R. Zuparko and Yves Braet, personal correspondence). Preliminary data from rearing studies in our laboratory suggest a major emergence of Gildoria sp. from March through June out of small diameter J. regia twigs. Most species of Gildoria described in Achterberg (2003) are parasitoids of scolytids.

The flight survey with ethanol lures demonstrated a response from female X. saxeseni and male and female N. brevipennis. However, this study began in May, in the middle of the N. brevipennis flight period of March to August (Linsley, 1963, Lona et al.,

2019) and did not include the full flight season of this beetle. The survey also recorded more N. brevipennis females than males in the trap catches, which was the reverse of a

different flight bioassay conducted at the ATRC (Lona et al., 2019). In the latter, nearly

10 times more males were captured than females in response to Juglans branch sections as stimuli. To substantiate the response of N. brevipennis to ethanol, flight surveys should be repeated during the flight period in its entirety.

Hypothenemus eruditus was also trapped in the ethanol-baited funnel traps. This invasive beetle had not been recorded previously in Butte County, and this, along with other specimens collected from a different study at the ATRC, are the northernmost records of this species in California (Seybold et al., 2016, Lona et al., 2019).

Furthermore, one specimen of Xyleborus affinis was found in a funnel trap without the ethanol lure (control), confirming its presence in Butte Co. Lona et al. (2019) had recorded a previous specimen in the spring 2016 in a flight survey trap baited with the P. juglandis pheromone.

This exploratory study can be used as a basis for a more extensive investigation of the effect of water stress on the colonization success of P. juglandis and other insects, such as N. brevipennis, on J. regia. Although our analysis of P. juglandis activity on J. regia branches did not definitively establish an effect of tree water stress on colonization rates by the beetle, we did gain a deeper understanding of the relationship of this activity to the plant. We also learned how P. juglandis and N. brevipennis (and by association,

Gildoria sp.) partition Juglans branch tissue. Research on Juglans susceptibility under water stress to bark and woodboring beetles could be extended, with greater attention to timing and level of water stress.

Acknowledgements

We thank Megan A. Siefker, Jackson P. Audley (both UC-Davis Department of

Entomology and Nematology), and Hannes H. Bauser (Institute of Environmental

Physics, Heidelberg University) for their excellent assistance in the field. We also thank

Stacy M. Hishinuma (USDA Forest Service, Pacific Southwest Region) for reviewing the initial version of this manuscript. We are very grateful to Jeff Boles and the ATRC for access to the study site in Chico. Funding for this project was provided through a Joint

Venture Agreement (16-JV-11272139-086) between the USDA Forest PSW Research

Station and California State University Chico Research Foundation.

References

Achterberg, C. van. (2003) The West Palaearctic species of the genera Gildoria Hedqvist and Platyspathius Viereck, with keys to the species (Hymenoptera: Braconidae: Doryctinae). Zoologische Mededelingen, 77, 267−290.

Bright, D.E. (1981) Taxonomic monograph of the genus Pityophthorus Eichhoff in North and Central America (Coleoptera: Scolytidae). Memoirs of the Entomological Society of Canada, 113, 1–378.

Bright, D.E. (2014) A catalog of Scolytidae and Platypodidae (Coleoptera), Supplement 3 (2000–2010), with notes on subfamily and tribal reclassifications. Insecta Mundi, 356, 1– 336.

Bright, D.E. (2019) A taxonomic monograph of the bark and ambrosia beetles of the West Indies (Coleoptera: Curculionoidea: Scolytidae and Platypodidae (Coleoptera). Studies on West Indian Scolytidae 7. In Occasional Papers of the Florida State Collection of Arthropods, Gainesville, Florida, 500 pp.

Chen, Y. & Seybold, S.J. (2014) Crepuscular flight activity of an invasive insect governed by interacting abiotic factors. PLoS ONE, 9, e105945. doi:10.1371/journal.pone.0105945.

Coleman, T.W., Grulke, N.E., Daly, M., Godinez, C., Schilling, S.L., Riggan, P.J., Seybold, S.J. (2011) Coast live oak, Quercus agrifolia, susceptibility and response to goldspotted oak borer, Agrilus auroguttatus, injury in southern California. Forest Ecology and Management, 261, 1852–1865. doi:10.1016/j.foreco.2011.02.008.

Faccoli, M., Simonato, M. & Rassati, D. (2016) Life history and geographical distribution of the walnut twig beetle, Pityophthorus juglandis (Coleoptera: Scolytinae), in southern Europe. Journal of Applied Entomology, 140, 697–705. doi: 10.1111/jen.12299.

Fulton, A. & Buchner, R. (2006) The effect of water stress on walnut tree growth, productivity, and economics. UCCE Farm Advisors Draft Publication, University of California Cooperative Extension, Tehama County. Feb. 23, 2006. http://cetehama.ucanr.edu/files/23092.pdf (accessed 20 December 2018).

Fulton, A., Grant, J., Buchner, R. & Connel, J. (2014) Using the pressure chamber for irrigation management in walnut, almond, and prune. University of California Agricultural and Natural Resources. ANR Publication 8503. Available at: http://anrcatalog.ucanr.edu/pdf/8503.pdf (accessed 28 December 2018).

Goldhamer, D. & Fereres, E. (2001) Simplified tree water status measurements can aid almond irrigation. California Agriculture, 55, 32−37.

Graves, A.D., Coleman, T.W., Flint, M.L. & Seybold, S.J. (2009) Walnut twig beetle and thousand cankers disease: Field identification guide, UC-IPM Website Publication, 2 pp. Nov. 21, 2009. http://www.ipm.ucdavis.edu/PDF/MISC/thousand_cankers_field_guide.pdf (accessed 17 November, 2018).

Huberty, A.F. & Denno, R.F. (2004) Plant water stress and its consequences for herbivorous insects: a new synthesis. Ecology, 85, 1383–1398.

Kelsey, R.G., Gallego, D., Sanchez-Garcia, F.J. & Pajares, J.A. (2014) Ethanol accumulation during severe drought may signal tree vulnerability to detection and attack by bark beetles. Canadian Journal of Forest Research, 44, 554−561. doi: 10.1139/cjfr- 1013-0428.

Kelsey, R.G. & Joseph, G. (2001) Attraction of Scolytus unispinosus bark beetles to ethanol in water-stressed Douglas-fir branches. Forest Ecology and Management, 144, 229−238. doi:10.1016/S0378-1127(00)00387-X. 69

Kirkland, J. & Kelsey, R. (2015) What do cocktail parties and stressed trees have in common? Plenty of alcohol! Science Findings 173. Portland, OR: US Department of Agriculture, Forest Service, Pacific Northwest Research Station. 6 pp., 173.

Kolařík, M., Freeland E., Utley, C. & Tisserat, N. (2011) Geosmithia morbida sp. nov., a new phytopathogenic species living in symbiosis with the walnut twig beetle (Pityophthorus jugandis) on Juglans in the USA. Mycologia, 103, 325–332. doi: 10.3852/10-124.

Lindgren, B.S. (1983) A multiple funnel trap for scolytid beetles (Coleoptera). The Canadian Entomologist, 115, 299−302.

Linsley, E.G. (1963) The Cerambycidae of North America. Part IV. Taxonomy and Classification of the Subfamily Cerambycinae, Tribes Elaphidionini through Rhinotragini. Vol. 21. University of California Publications in Entomology, University of California Press, Berkeley, CA, USA, 165 pp.

Lona, I.D., Miller, D.G., Hatfield, C.A., Rosecrance, R.C., Nelson, L.J., Audley, K.P., Chen, Y. & Seybold, S.J. (2019) Host selection behavior mediated by differential landing rates of the walnut twig beetle, Pityophthorus juglandis Blackman (Coleoptera:Scolytidae), on two western North American walnut species, Juglans californica and Juglans major. Entomologia Experimentalis et Applicata (Submitted, April 10, 2019).

McCutchan, H. & Shackel, K.A. (1992) SWP as a sensitive indicator of water stress in prune trees (Prunus domestica L.cv French). Journal of the American Society of Horticultural Science, 117, 607−611.

Moricca, S., Bracalini, M., Benigno, A., Ginnetti, B., Pelleri, F. & Panzavolta, T. (2018) Thousand cankers disease caused by Geosmithia morbida and its insect vector Pityophthorus juglandis first reported on Juglans nigra in Tuscany, Central Italy. Plant Disease, 103, 369−369. https://apsjournals.apsnet.org/doi/pdf/10.1094/PDIS-07-18-1256- PDN.

R Core Team (2016) R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria. https://www.R-project.org.

Raffa, K.F., Anderson, M.N. & Schlyter, F. (2016) Host selection by bark beetles: playing the odds in a high-stakes game. In Advances in Insect Physiology, Vol. 50. Academic Press, pp. 1–74.

Ranger, C.M., Reding, M.E., Persad, A.B. & Hermst D.A. (2010) Ability of stress-related volatiles to attract and induce attacks by Xylosandrus germanus and other ambrosia beetles. Agricultural and Forest Entomology, 12, 177−185.

Ranger, C.M., Reding, M.E., Persad, A.B. & Oliver, J.B. (2013) Influence of flood-stress on ambrosia beetle host-selection and implications for their management in a changing climate. Agricultural and Forest Entomology, 15, 56−64.

Ranger, C.M., Schultz, P.B., Frank, S.D., Chong, J.H., & Reding, M.E. (2015) Non- native ambrosia beetles as opportunistic exploiters of living but weakened trees, PLos ONE, 10, e0131496. doi:10.1371/journal.pone.0131496

Seybold, S.J., Klingeman, W.E. III, Hishinuma, S.M., Coleman, T.W. & Graves, A.D. (2019) Status and impact of walnut twig beetle in urban forest, orchard, and native forest ecosystems. Journal of Forestry, 117, 152−163. doi: 10.1093/jofore/fvy081.

Seybold, S.J., Haugen, D., O’Brien, J. & Graves, A.D. (2013a) Thousand cankers disease. USDA Forest Service, Northeastern Area State and Private Forestry Pest Alert, NA-PR-02-10, originally published May, 2010, reprinted Aug. 2010, Oct. 2011, and Feb. 2013, pp. 2, https://www.fs.usda.gov/naspf/publications/pest-alert-thousand-cankers- disease-pest-alert-revised-february-2013-na-pr-02-10 (accessed 17 November, 2018).

Seybold, S.J., Dallara, P.L., Hishinuma, S.M. & Flint, M.L. (2013b) Detecting and identifying the walnut twig beetle: Monitoring guidelines for the invasive vector of thousand cankers disease of walnut. University of California Agriculture and Natural Resources, Statewide Integrated Pest Management Program, Oakland, California, 13 pp., March 13, 2013, http://ipm.ucanr.edu/PDF/PESTNOTES/WTB_trapping.pdf (accessed 17 November 2018).

Seybold, S.J., Penrose, R.L. & Graves, A.D. (2016) Invasive bark and ambrosia beetles in California Mediterranean forest ecosystems. In Insects and Diseases of Mediterranean Forest Systems (ed. by TD Paine & F Lieutier), Springer International Publishing, AG, Cham, Switzerland, pp. 583−662.

Tisserat, N., Cranshaw, W., Leatherman, D., Utley, C. & Alexander, K. (2009) Black walnut mortality in Colorado caused by the walnut twig beetle and thousand cankers disease. Phytopathology Supplement, 99, S128.

UC ANR (2018) Irrigation scheduling using stem water potential (SWP) measurements. http://informatics.plantsciences.ucdavis.edu/Brooke_Jacobs/index.php (accessed 31 July 2018).

White, T.C.R. (1969) An index to measure weather-induced stress of trees associated with outbreaks of psyllids in Australia. Ecology, 50, 905−909.

Table 2.1 Temperature, humidity, and baselinea potential recorded during the dates when stem water potential (SWP) was measured on English walnut, Juglans regiab, at the ATRCc

High Stress Group Low Stress Group Control Date Temperature °C Humidity % Baseline (MPa) mean SWP (MPa) mean SWP (MPa) mean SWP (MPa)

2 Sep 2015 34-36 23-34 −0.38 −1.60 −0.69 −1.19 15 Jul 2016 34-36 24-32 −0.52 −1.31 −0.70 −0.87 6 Sep 2017 31-33 44-49 −0.33 −0.45 −0.36 −0.36 15 Sep 2017 26-29 25-32 −0.35 −0.96 −0.57 −0.57 27 Jul 2018 33-36 28-31 −0.45 −1.25 −1.05 −1.14 31 Jul 2018 32-34 31-35 −0.43 −1.24 −1.11 −1.17 a UC ANR (2018). b Ten trees categorized as high stress, ten trees categorized as low stress (both baited with male Pityophthorus juglandis aggregation pheromone), and five control trees (unbaited) were measured between 13:00-15:00. c California State University Chico, Agricultural Teaching and Research Center (ATRC) in Chico, California (Butte Co.) (39°41′41.3″N, 121°49′13.1″W).

Table 2.2 Number of main branchesa and twigsb for each diameter classc and number of sections with Pityophthorus juglandis entrance/emergence holes and Nathrius brevipennis or Gildoria sp. emergence holes Branch/twig No. of No. of sections No. of sections with No. of sections with P. juglandis diameter (cm) class branch/twig with P. juglandis N. brevipennis/Gildoria sp. and N. brevipennis/Gildoria sp. 0.4 37 0 31 0 0.6 126 0 91 0 0.8 135 0 94 0 1 112 0 76 0 1.2 33 0 32 0 1.4 17 0 16 0 1.6 7 0 9 0 1.8 4 0 4 0 2 9 0 9 0 2.2 5 0 5 0 2.4 2 0 2 0 2.6 3 1 2 0 2.8 5 2 4 0 3 1 0 1 1 3.2 1 0 1 1 3.4 4 4 2 0 3.6 4 5 1 1 3.8 7 6 1 0 4 3 4 0 0 4.2 9 9 0 1 4.4 8 7 0 2 4.6 2 1 0 1 4.8 3 3 0 0 5 4 4 0 0 5.2 2 2 0 0 5.4 1 1 0 0 5.8 1 1 0 0 6 1 1 0 0 6.8 1 1 0 0 7 2 2 0 0 7.8 1 1 0 0 a Main branches (20 total) were divided into six sections. b Twigs or secondary small branches belong to sections of main branches. c Diameters were rounded and assigned to the nearest 0.2 cm class.

Table 2.3 Total number of insects trapped in two funnel traps each baited with an ethanol lure and two unbaited funnel traps (controls) in 2018 at the ATRCa

Ethanol Lure Control Date Range Coleoptera Anobiidae Priobium punctatum LeConte 1 0 26 Jun to 3 Jul Bostrichidae Scobicia declivis LeConte 19 0 29 May to 16 Oct Cerambycidae Nathrius brevipennis (Mulsant) male 59 1 29 May to 10 Jul Nathrius brevipennis (Mulsant) female 139 7 29 May to 31 Jul Xylotrechus nauticus (Mannerheim) 5 1 29 May to 7 Aug Scolytidae Hypothenemus eruditus Westwood female 2 0 14 Aug to 28 Aug Xyleborus affinis Eichhoff female 0 1 26 Sep to 2 Oct Xyleborinus saxeseni (Ratzeburg) female 231 1 22 May to 30 Oct Hymenoptera Braconidae Gildoria sp. Hedqvist 1 0 3 Jul to 10 Jul a California State University Chico Agricultural Teaching and Research Center in Chico, California (Butte Co.) (39°41′41.3″N, 121°49′13.1″W).

Figure 2.1 Juglans regia branch with lure of male Pityophthorus juglandis aggregation pheromone. Lures were placed on a branch of each of ten trees grouped as low water stress and on a branch of each of ten trees grouped as high water stress on 24 Aug 2016.

Lures were left for two years to permit P. juglandis to colonize branches. Photograph by

I.D. Lona.

Figure 2.2 Juglans regia branch (one of 25) divided into sections of 27 cm. Each section was divided into four aspects (right, left, bottom, and top). Pityophthorus juglandis entrance and emergence holes were recorded in each section (marked with red sharpie pen). Photograph by I.D. Lona.

Figure 2.3 Branch of Juglans regia with Pityophthorus juglandis entrance and emergence holes (marked with red sharpie pen). This branch was an outlier in the low water stress group. An ‘X’ was marked where the P. juglandis aggregation pheromone lure was placed. Photograph by I.D. Lona.

Figure 2.4 Twig from Juglans regia branch with a) Nathrius brevipennis and b) Gildoria sp. emergence holes. Photograph by I.D. Lona.

Figure 2.5 Stem water potential (mean ± SE) of trees in the low water stress group, each with a baited branch (N=10); trees in the high water stress group, each with a baited branch (N=10); and trees in the intermediate water stress group (control), each with no baited branch (N=5) from six different measuring dates from 2015 to 2018. Outlier reflects SWP measured from a tree that belongs to the low stress group. Different letters refer to a significant difference in water stress among trees in the three groups (P<0.05).

Figure 2.6 Stem water potential (mean ± SE) of pooled trees in the low water stress group and trees in the high water stress group, each with a baited branch (N=20); and trees in the intermediate water stress group (control), each with no baited branch (N=5).

Different letters refers to a significant difference in water stress between the baited groups (P<0.05).

(a) (b)

Figure 2.7 Frequency distribution of Pityophthorus juglandis entrance or emergence holes by distance from the aggregation pheromone lure on ten branches each of Juglans regia from trees under high water stress (a) or low water stress (b). Zero distance represents the lure placement on the branch. Negative distance represents the distal side

(closest to the branch terminus). Positive distance represents the proximal side (closest to tree stem). The low water stress data set (b) is pictured with (light bars) and without (dark bars) data from an outlier branch.

2,5 High Stress

Low Stress

2 SE) ±

1,5 entrance or emergence

0,5 holes/branch/section (mean Pityophthorus juglandis 0 Top Left Right Bottom

Figure 2.8 Number (mean ± SE) of Pityophthorus juglandis entrance or emergence holes on four aspects (top, left, right, and bottom) on ten branches of Juglans regia trees under high water stress (open bars) and ten branches of trees under low water stress (shaded bars), each baited with the male P. juglandis aggregation pheromone located on the bottom of each branch.

(a)

(b)

Figure 2.9 Distribution (by branch or twig diameter) of entrance or emergence holes from Pityophthorus juglandis (a) (N = 366), and emergence holes from Nathrius brevipennis (b) (N = 1182) and Gildoria sp. (b) (N = 482) in ten branches of Juglans regia from trees under high water stress and ten branches of J. regia from trees under low water stress (each branch from both groups baited with the male P. juglandis aggregation pheromone) and five control branches (unbaited). Water stress groups were pooled for each of the three species (there were no P. juglandis holes present in any of the control branches). Majority of N. brevipennis and Gildoria sp. emergence holes are from secondary branches and twigs of the larger (main) branches in which P. juglandis entrance and emergence holes were recorded.

(a) (b) (c)

Figure 2.10 Number of Nathrius brevipennis emergence holes in relation to the distance from the lure of male Pityophthorus juglandis aggregation pheromone in branches from ten Juglans regia trees under high water stress (a), ten J. regia trees under low water stress (b) and five J. regia trees under intermediate water stress (c). Branches from the latter five trees were used as controls and were not baited with the male P. juglandis aggregation pheromone. For these control branches, the distance from an assumed lure was measured as if a lure had been placed.

(a) (b) (c)

Figure 2.11 Number of Gildoria sp. emergence holes in relation to the distance from the lure of male Pityophthorus juglandis aggregation pheromone in branches from ten Juglans regia trees under high water stress (a), ten J. regia trees under low water stress

(b) and five J. regia trees under intermediate water stress (c). Branches from the latter five trees were used as controls and were not baited with the male P. juglandis aggregation pheromone. For these control branches, the distance from an assumed lure was measured as if a lure had been placed.

Figure 2.12 Flight response (mean ± SE) of female Xyleborinus saxeseni (N = 44) and male and female Nathrius brevipennis (N = 16) to two ethanol-baited (ethanol lure) and two unbaited (control) traps (placed from 22 May to 11 December 2018) at the California

State University Chico Agricultural Teaching and Research Center (ATCR) in Chico,

California (Butte Co.) (39°41′41.3″N, 121°49′13.1″W). Means were calculated from trap catches only during periods of flight activity for X. saxeseni (May 22 to October 30,

2018) and N. brevipennis (May 29 to July 31, 2018). Different letters and superscripts above pairs of histogram bars indicate significantly different mean responses compared between trap catches from ethanol-baited and control traps (P<0.05).