J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 l

Journal of the

Entomological Society of British Columbia

Volume 116 December 2019 ISSN#0071-0733

Directors of the Entomological Society of British Columbia 2018-2019............cc cece eee y

PERSPECTIVES

Weather and insects in a changing climate

ARTICLES

Collections of fleas (Siphonaptera) from Pacific marten, Martes caurina (Carnivora:

Mustelidae), reveal unique host—parasite relationships in the Haida Gwaii archipelago.....17

Dispersal of Bactericera cockerelli (Hemiptera: Triozidae) in relation to phenology of

matrimony. vine (Lyciunr- apps Solanaceme ys .cekin Geutiewe cen eae Mohit MeN a vs va 5

Assessments of Rhagoletis pomonella (Diptera: Tephritidae) infestation of temperate,

tropical, and subtropical fruit in the field and laboratory in Washington State, U.S. ......... 40

SCIENTIFIC NOTES

Promachus dimidiatus Curran (Diptera: Asilidae): a robber fly genus and species new to

Factrats Coli 02 Poe cera cc er wl eels Utes ee ERR oa ds ee eae ee 59

Toxonevra muliebris (Harris) (Diptera: Pallopteridae): a European fly new to North

A TROT IG BLitd ee a hc eos tt heres v decaaslnin dle Re ard ae Pa nee a rr 64

OBITUARIES

Pater Beltote2cu. icy Tl a at es AL es a ee 69

ST Fees Ae I Ee BB ci asan as cttach ate eaiie'y a ba Gane thas phce $4 Inside Back Cover

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 Z

DIRECTORS OF THE ENTOMOLOGICAL SOCIETY

OF BRITISH COLUMBIA FOR 2018-2019

President:

Lisa Poirier (president@entsocbc.ca)

University of Northern B.C., Prince George

Ist Vice President:

Tammy McMullan

Simon Fraser University, Burnaby

2nd Vice-President:

Wim van Herk

AAFC, Agassiz

Past-President:

Jenny Cory ©

Simon Fraser University, Burnaby

Treasurer:

Ward Strong (membership@entsocbce.ca)

BC Ministry <a Lands and Natural Resource Operations and Rural Development, Vernon

Secretary:

Tracy Hueppelsheuser (secretary@entsocbc.ca)

B.C. Ministry of Agriculture, Abbotsford

Directors:

Tamara Richardson, Grant McMillan

Graduate Student Representative:

Dan Peach

Regional Director of National Society: Editor, Boreus:

Brian Van Hezewijk Gabriella Zilahi-Balogh(boreus@entsocbce.ca)

Canadian Forest Service, Victoria Canadian Food Inspection Agency

Web Editor: Editor Emeritus:

Brian Muselle (webmaster@entsocbc.ca) Peter & Elspeth Belton

University of British Columbia, Okanagan Simon Fraser University

Campus

Society homepage: http://entsocbe.ca Journal homepage: http://journal.entsocbe.ca

Editorial Committee of the Journal of the Entomological Society of British Columbia:

Editorial Board: Marla Schwarzfeld, Bo Staffan

Lindgren, Katherine Bleiker, Lisa Poirier, Lee

Humble, Bob Lalonde, Lorraine Maclauchlan,

Robert McGregor, Steve Perlman, Joel

Gibson, Dezene Huber

Editor-in-Chief:

Katherine Bleiker (journal@entsocbc.ca)

Canadian Forest Service, Victoria

Copy Editor: Monique Keiran Technical Editor: Alicja Muir

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 3

PERSPECTIVES

Weather and insects in a changing climate

V. NEALIS!

INTRODUCTION

The humourist Mark Twain is quoted as saying, “Everyone complains about

the weather, but no one does anything about it.”? A century later, we realize we

have been doing something about it all along. But unlike the intent of the joke,

human effects on weather and climate will have dire consequences for the future

of many biological systems. The drivers of this change are socio-economic, but

the consequences — and our ore to anticipate and mitigate them — require

insight from ecology.

The effects of weather on ee vital biological processes are obvious only

in the extreme. Our physiologies maintain steady thermal states and we have

extended our habitable range to adverse environments with clothing, shelter, and

fire. However, the development of modern ecology in the first decades of the 20

century gave entomologists an appreciation of the pervasive role of weather on

insects and the ecosystems they inhabit. Insects soon served as models of the

ecological relationships between climate and life systems, because of the

observable effects of weather on the behavior, development, and survival of

insects (Uvarov 1931). The small size, diverse and abundant populations, and

economic importance of insects has made entomology a major contributor to

theories and methods that are relevant to understanding ecological systems.

This essay explores the ecological relationships between insect populations

and climate and weather, which are aptly distinguished by Mark Twain as

“climate is what we expect, weather is what we get.” The subject is too large to

review critically in this forum; instead, I illustrate an approach that I believe

provides useful, general insights to the possible ecological consequences of

climate change, based mostly on my experience in forest entomology.

Insect outbreaks can propel major ecological disturbance in forests by

affecting forest composition, productivity, and structure. Forest ecosystems, in

turn, play a critical role in the terrestrial carbon budget and so have the potential

to mitigate the effects of climate change, if healthy, or exacerbate them, if not.

Compared to long-lived trees, which modulate the relatively slow transitions that

are characteristic of forest ecosystems, insect populations change rapidly — 1

part because their dynamics are more immediately responsive to ambient weather

than are the trees they inhabit. Insects are also highly mobile and can move to

more favourable areas as climate-related shifts in conditions occur. Their fast-

moving outbreak dynamics are indicators of emerging and future changes in

forest disturbance patterns.

Some of these changes are already underway. Our focus on insects as

disturbances to forests has resulted in detailed scientific knowledge of their

bionomics and in compilations of essential survey records. Forest entomologists

'Corresponding Author: Natural Resources Canada-Pacific Forestry Centre, Victoria, BC V8R 2H9

2The origin of the remark is attributed to Twain’s friend, Charles Warner

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 4

make significant contributions to field methods in ecology and analytical tools

that advance our understanding of quantitative ecology, population dynamics,

and systems analysis. Weather and climate affect ecological relationships at each

of these levels of inquiry.

MODELS FOR INFERENCE

Models to analyze the relationships between insects and climate may be

empirical or process models. Empirical models focus on statistical relationships

between, for example, observed patterns of insect damage and meteorological

variables. The structure of these models is not necessarily causative but describes

correlated events. In bioclimatic studies, empirical models are sometimes called

envelope or niche models, following early definitions of the niche as locations

where environmental conditions permit an organism to live — that is, its habitat.

By comparison, process models characterize functional variation in key

ecological relationships. These models reflect the contemporary view of the

niche as the ensemble of traits that determine survival and reproductive success

of an organism in particular environments — that is, its fitness (Chuine 2010).

Habitat models serve as the starting point of most quantitative investigations

in population ecology to test hypotheses and estimate model parameters. These

models draw. their evidence primarily from historical surveys of insect

distributions or their impacts, and weather. Their value is a function of the extent,

continuity, and accuracy of the data used. Canada is fortunate, because records

exist of annual overview and point surveys of insects and diseases compiled by

the Forest Insect and Disease Survey of the Canadian Forest Service between

1936 and 1996. These data provide baseline descriptions of where species have

been found and outbreaks observed.

Invasion and conservation ecologists, often confronted with little data on the

life histories of introduced or rare species, can make good use of habitat models

for initial estimates of climate suitability and relative risk. For example, plant-

hardiness zones are habitat models based on observed co-occurrence of plant

assemblages (McKenney ef al. 2007). Overlay of plant-hardiness zones on

survey records of invasions of winter moth, Operopthera brumata L., and balsam

wooly adelgid, Ade/ges piceae Ratzeburg, indicate boundaries in the frequency

of occurrence of these two species that are congruent with different plant-

hardiness zones. These results suggest that historic climatic gradients have, to

date, stalled further spread of these species in Atlantic Canada (Nealis ef al.

2016; Quiring ef al. 2008). However, climate change will shift the zones’

boundaries and possibly the area susceptible to accelerated invasion.

Where survey data are sufficiently rich, habitat models can reveal key

processes that characterize population dynamics of insects, especially where

apparent thresholds or steep environmental gradients correlate with the relative

abundance of an insect. Historical survey data from British Columbia (BC) show

relatively infrequent outbreaks of the mountain pine beetle, Dendroctonus

ponderosae (Hopk.), outside the —40° C minimum winter isotherm, thereby

identifying vulnerability of the beetle’s overwintering life stages to low

temperatures. This simple map accounted for previous incursions of mountain

pine beetle into high-elevation forests during warmer winters (Logan and Powell

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 5

2001) and recent range expansion into new host forests made susceptible by

warmer conditions (Safranyik ef a/. 2010).

Such direct and consequential effects of weather on insect populations,

however, are less common than indirect effects, which ripple through trophic and

physical interactions alongside other weather-dependent factors important to an

insect’s life. This higher-order complexity can yield problematic results for

habitat models. For example, there are so many possible meteorological variables

and indices to test that significant but spurious associations become probable and

inferences misleading. A related problem is significant interactions that defy

biological interpretation. Unless there is prior knowledge of system structure,

there are few criteria to determine which variables to include in a model. These

are pertinent limitations in projecting future states under climate change based on

historical correlations.

Process models focus on functional ecological relationships to improve

inferences regarding fitness under different environmental conditions. Fitness is

an attribute of individuals in nature. Process models can take advantage of

modern computing power to calculate the individual fitness of thousands of

members of a population that may vary slightly in their intrinsic responses to

weather variables. This enables realistic simulation of per capita reproductive

success and prediction of inter-generation rates of change in population density

under variable conditions (Régniére ef a/. 2012a).

In practice, empirical and process approaches to modelling ecological

relationships between weather and insects are complementary. Their successful

application often depends on practical limits of available information and model

objectives. However, ecological questions increasingly require dynamic models

to accommodate changing conditions in the fundamental processes underlying

the survival and fitness of organisms. There are few factors more fundamental to

terrestrial ecological systems than weather and we know it is changing. It is no

longer sufficient to predict future behaviour of forest ecosystems simply by

analyzing historical patterns of outbreaks and disturbance. We must aim for the

deeper inferences that come from analysis of the ecological processes that

determine the distribution and abundance of organisms today so those same

models can project the future.

PHENOLOGY MODELS

Phenology models reproduce the observed seasonal occurrence of critical

events in an insect’s life history. They are central to the analysis of ecological

relationships between weather and insects. At their simplest, phenology models

describe the empirical relationships between particular weather variables and

seasonal events, such as temperature and the timing of an insect’s emergence

from hibernation in a particular habitat. The eco-physiological relationships

included in phenology models reflect intrinsic, evolved characters that are not

likely to change as fast as ecological conditions. When we scale the entire life

history of an insect and the ecological events that determine its fitness via a

phenology model, the model becomes a tool to probe the species’ specific

seasonal experience and the ecological processes that determine its fitness in any

location and time.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 6

Temperature is the weather variable with the most profound eco-physiological

effect on insects. There are also many practical reasons to use temperature as the

driving variable in phenology models. Temperature is the easiest and most

commonly measured meteorological variable. Historic normal and real-time

records are available for thousands of fixed points, and realistic interpolations

can be calculated using environmental gradients known to affect temperature —

for example, elevation — to construct climate surfaces. Temperature fluctuates

periodically within normal limits on diurnal, seasonal, and annual scales, in

patterns associated with explicit, fixed geo-references. Day-to-day temperatures

are autocorrelated, allowing realistic, short-term daily temperature records to be

simulated from climatic normals. Although other meteorological variables such

as precipitation may also be important to an organism’s fitness, fluctuations in air

temperature are fluctuations in thermal energy that drive variation in most other

micro-scale factors in an insect’s environment. Gradients in thermal energy also

account for regional-scale variation in atmospheric pressure and consequently

most large-scale weather patterns. Given these correlations between temperature

and weather conditions, adding meteorological covariates may confer marginal

additional benefits for the practical objective of scaling eco-physiological events

within the realistic limits of predicting future weather.

Statistical models of the relationship between the rates of insect development

and ambient temperature transform insect time to a temperature-dependent scale.

Originally, these models were derived from field observations of the duration of

life stages at ambient temperature. The best-performing phenology models today

use estimates of rates of growth and development under controlled, laboratory

conditions with sufficient sample sizes to capture individual variability. The

added advantage of laboratory estimates is they are independent of observed,

seasonal events in both the weather and the insect population: they measure the

process directly. This means empirical field data can be used to test and calibrate

working phenology models. Degree-day models are the simplest models to fit

such data and have wide applicability in entomology (Gilbert and Raworth

1996). However, the availability of more detailed experimental data and

computational tools to fit flexible, non-linear functions enables simulation of

variable, stage-specific development rates over the full range of seasonal

temperatures experienced by insects in nature (Régniére ef a/. 2012a).

Modelling phenology over an entire year rather than just over the insect’s

active season allows simulation of a sequence of consecutive ‘generations’,

which translates the cumulative effects of variable weather to a measure of .

climatic suitability based on fitness. For example, the ability to reproduce the

phenology of a complete generation of gypsy moth, Lymantria dispar L. —

including its oviposition, diapause and active development periods — made it

possible to evaluate the probability of its survival over consecutive generations —

and therefore its relative invasion risk — in present and future environments

(Régniére et al. 2009). There are also immediate, practical applications. The

gypsy moth phenology model supports a decision-support protocol enabling

more efficient deployment of traps for detection, evaluation of probable

persistence of founder populations, and optimal timing of pesticide applications

for eradication (Nealis 2009).

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 7

These applications illustrate how phenology models’ can be used to provide a

relevant scale to examine other ecological processes in their appropriate spatial

and temporal contexts and to go beyond simple correlations between weather and

observed fluctuations in insect populations.

A CASE STUDY: SPRUCE BUDWORMS

Spruce budworms (Choristoneura) are among the most well-known forest

insects in Canada. Many significant advances in forest entomology owe their

development to spruce budworm research, including life tables, simulation and

mathematical models in population dynamics, and systems analysis in resource

management.

The influence of weather always figures in research on spruce budworm

ecology, but how weather influences fitness is only partially known, and the

evidence is difficult to apply to practical questions. Isolated weather events such

as killing frosts or the migration of moths on storm fronts may contribute to

survival of insect generations and egg recruitment, but the effects are variable

and often compensatory over the life cycle. More to the point, notable

meteorological events are not especially predictable and often affect populations

only temporarily or locally. Entomologists need models that wed normal weather

to seasonal events for several generations of an insect’s life history if they want

to understand general ecological processes that determine change over areas

where species persist now and may or may not persist in the future.

Wellington et al. (1950) first noted spruce budworm outbreaks were preceded

by years of warm, dry weather in the month of June. This was a habitat-based

inference. It could identify the times and places affected by weather but not how

or why this occurs. June is the main feeding period for budworms, and

behavioural studies had shown the caterpillars feed more continuously and

develop faster in dry, warm conditions — conceivably conferring survival

benefits. In addition, warm, dry summers promote production of pollen cones in

mature conifers, an early-season resource for small budworms that also provides

a beneficial “greenhouse” effect in cool northern environments (Wellington

1950). Both weather-related factors hinted at the importance of seasonal timing,

but despite concerted research, their net effect on fitness remained unclear

(Greenbank 1956).

In terms of population analysis, the problem stemmed partly from difficulties

in sampling early-stage budworms (Fig. 1). Eggs, overwintering, and spring-

foraging budworms each occupy different parts of the sample universe and are

either cryptic or in motion, so between-stage rates of survival are difficult to

estimate. Consequently, the focus of quantitative analyses shifted to survival of

large, feeding larvae for which sufficient population density estimates were

available. The analyses showed that during the feeding and subsequent pupal

stages, natural enemies — not weather — had a greater effect on survival and, as a

result, spruce budworm population dynamics came to be viewed as a predator—

prey relationship, with weather a stochastic effect imposing variation but not

trend (Royama 1984). However, during the early stages of the budworm’s life

history, the insect neither feeds nor is greatly affected by natural enemies,

leaving weather the likely dominant driver of environmental variation in their

survival. By relegating the role of weather to ‘noise’ in the analysis of population

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 8

dynamics, spruce budworm studies of this time essentially dismissed survival of

small budworm larvae from analysis of generation survival.

ob Egg parasitoids

Neonate dispersal losses

- Overwinter survival

* Spring dispersal losses

Budworm per kg of follage

Natural enemies

" 7

of 4, < fe Cok

. |

e Pi ca

Life Stage

Figure 1. Within-generation changes in western spruce budworm per kilogram of

foliage with key factors affecting stage-specific survival. Means (SE) calculated

from annual census data from several outbreak populations in the southern interior of

British Columbia, 1997-2015.

Yet, for budworms, a great deal goes on in these early stages, and weather

conditions run the gamut of annual extremes. Eggs hatch in summer, and

neonates disperse immediately away from their egg mass in search of hibernation

sites. Larvae must then survive a long dormant/diapause period, first in warm

summer conditions, then months of minimum northern temperatures, then

variable spring conditions. The budworms do not feed during these early stages

and must endure this prolonged period solely on nutrition provided in the egg.

Perhaps because of dwindling energy reserves, they emerge in spring well in

advance of fresh buds and forage throughout the forest canopy for sustenance. It

is not surprising that variation in field estimates of survival, derived after these

various processes have taken their toll, has bewildered the search for patterns of

survival.

I began to unpack the black box on survival of early-stage budworm when I

observed how critical early-season pollen cones were to the survival of jack pine

budworms, C. pinus pinus Free. All conifer-feeding budworms emerge from their

overwintering hibernacula well before current-year buds are available to them.

Their spring dispersal through the forest canopy in search of the earliest buds can

result in significant losses to populations (Nealis 2016). Nonetheless, field

observations of western spruce budworm, C. occidentalis Free., estimated the

optimal ‘head-start’ for emerging budworms was more than two weeks before

50% of new buds were available (Thomson ef a/. 1984). Insight into this counter-

intuitive adaptive syndrome meant going back to the beginning of the insect’s

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 9

life cycle and measuring changes in fitness as life stages passed through seasonal

pressure points. As these processes were evaluated, the phenology model was

enriched iteratively until it reproduced measured population densities, then

observed historical outbreak patterns, and — finally — the likely future of forest

disturbance under climate change.

In our calendar time, the nine-month period from oviposition to spring

emergence constitutes three-quarters of budworms’ annual life, during which

considerable weather-related mortality occurs (Fig. 1). Eggs are laid in mid- to

late-summer and, unless exposed to significant frost conditions, the effects of

weather at this stage appear to be inconsequential. The probability of killing

frosts for budworms depends on where and when eggs are laid. This is

determined by the seasonal period of moth dispersal and oviposition, which are

governed by the temperature-dependent pace of feeding and maturation earlier in

the season — that is, the timing of oviposition is determined entirely by

temperature. To test this, we simulated seasonal development with local

temperatures in a phenology model. The results demonstrated that the calendar

period of adult activity can be predicted and compared with historical

meteorological data to calculate the likelihood of killing frosts during the adult

and egg stages at any location (Nealis and Régniére 2014). In budworms, this

direct effect of cool, seasonal weather on development time and subsequent

likelihood of exposure of adults and eggs to killing frosts sets explicit northern

and elevational limits on climatic suitability (Régniére et al. 2012b; Régnicre and

Nealis 2019a).

After eggs hatch, larvae disperse to hibernation sites and prepare for winter.

Once the larvae are sheltered in their hibernacula, cool weather favours their

survival, whereas warm weather during this period demands energy and drains

the fixed energy reserves available to dormant larvae. As during the egg stage,

the weather conditions encountered by these dormant larvae are related to

previous, local phenology. In warmer locations where oviposition occurs earlier

in the season, progeny often must endure a longer period of heat stress, with

negative consequences for survival. Already in this earliest period of budworm

development, weather exerts a “pinching” effect on the primary environmental

range of budworm. In cool northern and high-elevation forests, delayed

maturation and oviposition exposes moths and eggs to killing frost, whereas in

warmer latitudes and at lower elevations, early oviposition results in budworms

consuming energy reserves required to survive winter.

Winter weather appears relatively benign to budworms over much of their

range. The crucial events occur later, after physiological diapause is complete in

late winter but before the weather warms up enough for dormant budworms to

emerge. It is during this transition period that the cumulative effects of energy

consumption begun during the warm days of the previous season take their toll:

budworms with insufficient energy reserves to endure the final three months of

dormancy perish. Warm weather during this late-winter period exacerbates stress

and further increases mortality. It is not cold winters but warm temperatures

during transition seasons that determine local fitness of small budworm larvae

between hatching and emergence nine months later (Nealis and Régniére 2016).

The precocious spring emergence of budworms weeks before budburst

suggests an urgency to secure a feeding site. This hypothesis prompted Thomson

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 10

and Benton (2007) to develop a habitat-based model that proposed an observed

change in outbreak history of the western spruce budworm on Vancouver Island

during the 20‘ century was related to local sea-level warming. The increase in

temperature caused earlier emergence of budworms but apparently no change in

the timing of bud flush. The inference was the resulting greater asynchrony

increased mortality of foraging budworms and reduced the likelihood of

outbreaks. However, simulations using our fitness-based phenology model,

corroborated by direct field measures of budworm survival during the foraging

period, show budworm synchrony with the host during this stage is a very robust

process. Budworms can survive several weeks of foraging in the spring before

bud-flush. The presence of pollen cones helps bridge this gap, and old needles

also provide sustenance. Field studies showed that losses during this dispersal

period do influence dynamics in important ways, but the losses are associated

with previous defoliation and stand characteristics, not with weather-mediated

synchrony of emergence and bud-flush (Nealis and Régni¢re 2009).

Spring emergence marks the beginning of the feeding period, with budworms

and their food supply developing at their respective, temperature-dependent

rates. A model predicting relative fitness of budworms as a function of normal,

seasonal change in host-foliage quality examined how variation in this seasonal

interaction affected inter-generation rates of change. Parameters were derived

from field measures of emergence dates and contemporaneous suitability of

developing foliage in the spring (Nealis and Nault 2005). The phenology model

then tracked temperature-dependent rates of growth and development of feeding

budworms and ultimate deterioration of foliage quality at the end of the growing

season, also determined by independent field measurements. Together, these

defined a phenological window for feeding and maturation, and resulting fitness

(Nealis 2012). The result was a trade-off between time of spring emergence and

subsequent time available for feeding before foliage became unpalatable. Early-

emerging budworm gamble an early period of deprivation against the greater

likelihood of maturing while foliage is most accessible. Consequently, the cost of

early emergence may be offset by the benefits of improved survival and greater

fecundity resulting from large feeding stages that coincide with succulent,

rapidly growing foliage (Régniére and Nealis 2018). The phenology model

allowed us to simulate the process under different seasonal conditions to

understand the outcome for fitness across landscapes and through time.

The most recent fitness model includes all of these temperature-related effects

on budworm across the range of its host (Régniére and Nealis 2019a). Spatial

and temporal characteristics of seasonal temperatures appear to explain a great

deal about the geographic range and population dynamics of western spruce

budworm. At the most northerly latitudes and highest elevations, cool

temperatures slow maturation of feeding budworms, decreasing their fitness

through the effects of a rapid decline in host-plant quality at the end of the short

season and exposing surviving moths and their eggs to killing frosts. At southern

and lower elevations, warm weather exhausts energy reserves of dormant larvae,

also reducing fitness. Between these limits, the model predicts climate suitability

and identifies areas of optimal fitness. This is where increases in budworm

populations are most likely.

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 1

The temporal view was equally interesting. Since 1950, climatic conditions in

the BC Interior have improved the fitness of western spruce budworm steadily.

The greatest increases have occurred since 2000, and match observed outbreak

behaviour over the same period (Maclauchlan ef a/. 2018). It seems much of the

historic, coarse-scale outbreak behaviour of western spruce budworm

populations coincides with spatial and temporal weather effects on phenology

which, in turn, influences fitness in the insect’s annual acquisition of resources.

When likely climate change scenarios are applied, they predict up-slope

improvement in climate suitability for budworm survival will continue. Southern

portions of the spruce budworm range in the USA will become too warm for

dormant budworm, except at ever-higher elevations, with the insect eventually

running out of host forests within climatically suitable territory. But in the north

of the range, a greater area of climatic suitability will intersect susceptible host

forests, and budworm disturbances will continue in northern and high-elevation

forests and will also expand into forests where they have not occurred in historic

times. |

Just as an earlier, generalized phenology model of spruce budworm informed

refinement of a fitness model for western spruce budworm, the western spruce

budworm model provides inferences about spruce budworm.

One such inference concerns differences among host trees exploited by

budworms in their respective forest eco-regions. The best place for any budworm

to live is in a forest full of food where early- and late-season weather conditions

are moderate, and where larvae emerge early in the race to find flushing buds

without excessive risk. Getting there early is critical because the quality of host-

foliage deteriorates quickly at the end of the season. In places and times where

this phenological window favours feeding budworms, population increases are

more likely.

Now consider the different host trees. Eruptive outbreaks of spruce budworms

in the past century have been most damaging in fir-dominated forests — balsam

fir, Abies balsamea (L.) Mill., east of the continental divide and other Abies

species to the west. As a resource for budworms, the distinguishing characteristic

of true firs over spruces (Picea), Douglas-fir (Pseudotsuga), and especially pines

(Pinus) is that the old foliage of true firs remains relatively soft and palatable

even at the end of the growing season. This makes the phenological window for

true firs functionally wider than for other host trees, because budworms can

continue to feed on fir later in the season and even back-feed on old foliage,

causing even more damage. The result is that forests dominated by fir within

regions of favourable weather support greater fitness of spruce budworms. In

forestry terms, true firs are more vulnerable than other host conifers. This

process-based assessment is consistent with empirical survey records, hazard

ratings, and habitat models that associate the origin and intensity of spruce

budworm outbreaks with particular forest types — now it has a fitness-based

explanation.

To return to western spruce budworm in Canada, at present it is mostly

associated with Douglas-fir, and tree mortality is relatively limited — again

because of the budworm’s difficulty in exploiting older foliage (Dodds ef al

1996). With climate suitability increasing at higher elevations in Canada,

however, western spruce budworm may increasingly inhabit high-elevation

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 12

forests of subalpine fir (A. lasciocarpa (Hook. Nutt.). Disturbance patterns in

those forests could change to resemble those of spruce budworm in the east

(Nealis 2005).

A related question regarding disturbance patterns associated with climate

change arises in eastern Canada, where climate suitability for spruce budworm

also is improving northward (Régniere ef al. 2012b) towards regions where

spruce—fir forests are replaced by forests dominated by black spruce (P. mariana

(Mill.) BSP). Black spruce is less susceptible to spruce budworm, because its

buds typically flush later in the season than those of balsam fir and white spruce

(P. glauca (Moench) Voss). Pureswaran ef al. (2015) hypothesize that warmer

conditions in the future will advance bud flush in black spruce, thereby

increasing its susceptibility to spruce budworm. Although warmer weather may

advance the calendar date of bud flush in trees, an increase in susceptibility of

black spruce will occur only if warmer weather has no effect on the timing of

decline in foliage quality the end of the season. In other words, if spruce

budworm fitness is going to increase because of a climate-related change in

black spruce phenology, it should result from a wider phenological window for

foliage suitability, not simply from an advance in the season for existing, relative

phenologies. A more likely change in the risk of significant budworm disturbance

for northern black spruce forests is the incursion of more susceptible balsam fir

and white spruce, whether this occurs as a result of management, climate change,

or both. Resulting mixed-wood boreal forests would certainly increase the

likelihood of significant damage to associated black spruce simply because of

their proximity to highly susceptible balsam fir (Nealis and Régnicre 2004).

WEATHER AND TROPHIC RELATIONSHIPS

Our ecological inferences agree with Wellington’s 1950 observation, noted

above, that budworms flourish in forests of their preferred hosts during warm,

dry Junes. It strengthens our knowledge and predictions about the geographic

range of budworms, what constitutes a suitable climate where populations may

increase to outbreak levels, interpretation of changing disturbance patterns, and

the extent to which specific, key trophic relationships vary with weather. Despite

this, however, the model provides only a proximate explanation of the

mechanism. We know the phenological window is important, but the analogy

eventually fails us because we still cannot see through that window to the

ultimate cause. A more fundamental, fitness-based explanation is overdue.

Shortly after I published observations on the relationship between pollen

cones and fitness of jack pine budworms (Nealis 2016), I received a letter from

Prof. T.C.R. White in Australia, recently deceased. Tom “gently” (and

gentlemanly) disagreed with my interpretation. Where I saw evidence of a

delayed and reciprocal, density-dependent relationship between the host tree and

the jack pine budworm, he saw further evidence that herbivores, including

budworms, normally live in a ‘nutritional desert’. Their change in fortune,

marked by outbreaks, is the result of environmental stress that accelerates

senescence in the foliage of mature trees, releasing higher levels of soluble

amino acids and thereby increasing food quality. White (1993) argued the

presumption of a uniform source of adequate food for herbivores is untenable

and contrary to the evidence from natural history. His alternative hypothesis of

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 13

outbreaks focusses on the fate of small, feeding larvae and their dependence on

the availability of soluble amino acids. Availability peaks in perennial plants at

the beginning of the growing season, when reserves stored in the roots are

recruited to produce new foliage, and increases again at the end of the season,

when nutrients synthesized by the foliage are shuttled back to storage in

perennial structures. This is the reason why so many herbivores are either ‘flush’

or ‘senescence’ feeders. Stress, such as drought, increases plant quality for

herbivores by further mobilizing amino acids. According to White, the greatest

survival benefit of this change in nutritional quality accrues to small larvae as

any improvements in their normally dismal survival rates increases fitness

overall. In eruptive species such as spruce budworms, the positive effects on

population growth will be most apparent in the years of transition from low to

high densities when survival of early life stages increases (White 2018).

White’s hypothesis implies an optimal feeding time with spatial and temporal

variability in the adequacy of the resource, just as in our phenology model.

Results of specific experiments that comprise that model are consistent with this

idea. For example, repeated bioassays on both early- and late-stage budworm

larvae on foliage at different stages of phenological development always show

optima, indicating a seasonally dynamic condition in host-plant quality that is

stage specific. The sharp decline in budworm fitness at the end of the season

implies plant quality is a time-limited commodity. Regrettably, we have sparse

quantitative population data on early-season nutrition and survival, particularly

during the incipient stages of an outbreak. But what we infer from White’s

hypothesis is consistent with most of what we do know. Budworm outbreaks do

tend to occur in drier parts of the budworms’ ranges and are often preceded by

notably dry, warm summers, especially in western populations (Maclauchlan ef

al. 2018). Defoliation is most intense in contiguous stands of over-mature trees

that have copious, senescing foliage and a greater propensity to produce pollen

cones. The greater fitness and higher density of budworm populations in these

areas then actively export gravid moths, which homogenizes regional population

densities and results in extensive and prolonged outbreaks (Régniére and Nealis

2019b).

The nutritional adequacy of the resource supporting these patterns varies in

space and time. This can be generalized and incorporated with fitness-based

models as seasonal processes scaled by temperature. If environmental stress does

improve the nutritional quality of budworms’ host trees, then climate change will

have an additional effect on disturbance ecology beyond a geographic shift in

climate suitability described above. Trees in all current forests survived because

their genotypes were best suited to conditions prevailing during their growth

under historic conditions. They will be stressed wherever climate change results

in less favourable conditions. This could increase the area of susceptibility to

outbreaks from small, scattered refuges envisaged by White (2018) to many

areas where nutritional quality of trees has improved as a result of the stress of

climate change. The result will be more frequent, explosive, and severe budworm

outbreaks, as we now see in western Canada.

A premise of White’s alternative hypothesis of budworm outbreaks is that

changes in the survival of the small larvae associated with host quality yield

commensurate changes in the density of subsequent generations. However,

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 14

improved survival early in budworm life history alone will not necessarily lead

to greater generation survival, particularly where survival later in the life cycle is

associated with density —- whether dependent, compensatory, or circumstantial.

Fitness models examine survival throughout the life cycle and at as many levels

as possible to identify critical events affecting survival (Fig. 1). Nonetheless,

White’s alternative hypothesis is eminently testable. The nutritional ecology of

spruce budworms has been well studied, and nitrogen levels have been found to

be most significant for budworm performance (Mattson ef al. 1991). Efforts to

identify other foliage-related indicators of budworm fitness have been equivocal

but do provide methods that could be applied to estimating parameters that relate

foliar nitrogen to phenology and stage of the outbreak cycle. These parameters

would inform a new generation of process models that examine how seasonal

fluctuations in foliage nutritional quality in different places and at different

stages of an outbreak affect budworm fitness, as has been done for phenology at

landscape scales (Régni¢re and Nealis 2019a). If White’s alternative hypothesis

bears out, we will have a much more objective way to identify places, times, and

tree species where the risk of spruce budworm outbreaks is changing.

EPILOGUE

The aphorism that opened this essay seems quaintly naive today. However,

Mark Twain was well-informed about the science of his day. The 19th-century

scientific advances that he witnessed, and their application to industry and trade,

enabled western countries to enter the 20 century at the peak of their economic

power. At the time, some people believed even control of weather was only a

clever invention away. In hindsight, these same scientific advances accelerated

the climate crisis we face today. We are connected to Twain in history and its

consequences.

It is these connections that ecology must decipher. I have always been

interested in how weather and climate shape biological systems. I now realize, as

many of us do, that climate change and future weather events will push and bend

those systems in ways that are highly uncertain and will result in unwelcome

surprises. Climate change is probably the gravest global ecological threat that

humans have ever faced knowingly. Entomologists’ contributions to our

understanding of climate—insects—forest systems may seem as small as insects

themselves, but, in ecology, small things add up, and they never function in a

vacuum. The emergent result of those many small, ecological interactions is what —

we seek to observe and model when we study insect populations and phenology

to understand the future of ecosystem behaviour.

We may still be able to do little about the weather, but we can do more than

just talk about it.

ACKNOWLEDGEMENTS

My thanks to the Entomological Society of British Columbia for the

invitation to prepare this essay, and to Jacques Régniére and Tom White for the

collaborations that shaped it. I dedicate this essay to Prof. T.C.R. White and his

indomitable contributions to ecology.

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 15

REFERENCES

Chuine, I. 2010. Why does phenology drive species distribution? Philosophical

Transactions of the Royal Society B, 365: 3149-3160. doi:10.1098/rstb.

2010.0142.

Dodds, K.A., Clancy, K.M., Leyva, K.J., Greenberg, D., and Price, P.W. 1996.

Effects of Douglas-fir foliage age class on western spruce budworm oviposition

choice and larval performance. Great Basin Naturalist, 56: 135-141.

Gilbert, N. and Raworth, D.A. 1996. Insect and temperature — a general theory. The

Canadian Entomologist, 128: 1-13.

Greenbank, D.O. 1956. The role of climate and dispersal in the initiation of

outbreaks of the spruce budworm in New Brunswick I. the role of climate.

Canadian Journal of Zoology, 34: 453-476.

Logan, J.A. and Powell, J.A. 2001. Ghost forests, global warming, and the mountain

pine beetle (Coleoptera: Scolytidae). American Entomologist, 47: 160-173.

Maclauchlan, L.E., Daniels, L.D., Hodge, J.C., and Brookes, J.E. 2018.

Characterization of western spruce budworm outbreak regions in the British

Columbia Interior. Canadian Journal of Forest Research, 48: 783-802.

Mattson, W.J., Haack, R.A., Lawrence, R.K., and Slocum, S.S. 1991. Considering

the nutritional ecology of the spruce budworm in its management. Forest Ecology

and Management, 39: 183-210.

McKenney, D.W., Pedlar, J.H., Lawrence, K., Campbell, K., and Hutchinson, M.F.

2007. Beyond traditional hardiness zones: using climate envelopes to map plant

range limits. BioScience, 57: 929-937.

Nealis, V.G. 2005. Diapause and voltinism in western and 2-year cycle spruce

budworms (Lepidoptera: Tortricidae) and their hybrid progeny. The Canadian

Entomologist, 137: 584-597.

Nealis, V.G. 2009. Still invasive after all these years: keeping gypsy moth out of

British Columbia. The Forestry Chronicle, 85: 593-603.

Nealis, V.G. 2012. The phenological window for western spruce budworm: seasonal

decline in resource quality. Agricultural and Forest Entomology, 14: 340-347.

Nealis, V.G. 2016. Comparative ecology of conifer-feeding spruce budworms

(Lepidoptera: Tortricidae). The Canadian Entomologist, 148: S33—S57.

Nealis, V.G., DeMerchant, I., Langor, D., Noseworthy, M.K., Pohl, K., Shanks, E.,

Turnquist, R., and Waring, V. 2016. Historical occurrence of alien arthropods and

pathogens on trees in Canada. Canadian Journal of Forest Research, 46: 1-9.

Nealis, V.G. and Nault, J.R. 2005. Seasonal changes in foliar terpenes indicate

suitability of Douglas-fir buds for western spruce budworm. Journal of Chemical

Ecology, 31: 683-696.

Nealis, V.G. and Régniére, J. 2004. Insect—host relationships influencing disturbance

by the spruce budworm in a boreal mixedwood forest. Canadian Journal of Forest

Research, 34: 1870-1882.

Nealis, V.G. and Régniére, J. 2009. Risk of dispersal in western spruce budworm.

Agricultural and Forest Entomology, 11: 213~223.

Nealis, V.G. and Régniére, J. 2014. An individual-based phenology model for

western spruce budworm (Lepidoptera: Tortricidae). The Canadian Entomologist,

146: 305-320.

Nealis, V.G. and Régniére, J. 2016. Why western spruce budworms travel so far for

the winter. Ecological Entomology, 41: 633-641.

Pureswaran, D.S., De Grandpré, L., Paré, D., Taylor, A., Barrette, M., Morin, H.,

Régniére, J., and Kneeshaw, D.D. 2015. Climate-induced changes in host tree-

insect phenology may drive ecological state-shift in boreal forests. Ecology, 96:

1480-1491.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 16

Quiring, D., Ostaff, D., Hartling, L., Lavigne, D., Moore, K., and DeMerchant, I.

2008. T emperature and plant hardiness zone influence distribution of balsam

wooly adelgid damage in Atlantic Canada. The Forestry Chronicle, 84: 558-562.

Régniére, J. and Nealis, V.G. 2018. Two sides of a coin: host-plant synchrony fitness

trade-offs in the population dynamics of the western spruce budworm. Insect

Science, 25: 117-126.

Régniére, J. and Nealis, V.G. 2019a. Influence of temperature on historic and future

population fitness of the western spruce budworm, Choristoneura occidentalis.

International Journal of Pest Management, 65: 228-243.

Régniére, J. and Nealis, V.G. 2019b. Density dependence of egg recruitment and

moth dispersal in spruce budworms. Forests, 10(8): 706. doi.org/10.3390/

£10080706.

Régniére, J., Nealis, V., and Porter, K. 2009. Climate suitability and management of

the gypsy moth invasion into Canada. Biological Invasions, 11: 135-148.

Régniére, J., Powell, J., Bentz, B. and Nealis, V. 2012a. Effects of temperature on

development, survival and reproduction of insects: experimental design, data

analysis and modelling. Journal of Insect Physiology, 58: 634-647.

Régniére, J., St-Amant, R., and Duval, P. 2012b. Predicting insect distributions under

climate change from physiological responses: spruce budworm as an example.

Biological Invasions, 14: 1571-1586.

Royama, T. 1984. Population dynamics of the spruce budworm Choristoneura

fumiferana. Ecological Monographs, 54: 429-462.

Safranyik, L., Carroll, A.L., Régniére, J., Langor, D.W., Riel, W.G., Shore, T.L.,

Peter, B., Cooke, B.J., Nealis, V.G., and Taylor, S.W. 2010. Potential for range

expansion of mountain pine beetle into the boreal forest of North America. The

Canadian Entomologist, 142: 415-442.

Thomson, A.J. and Benton, R. 2007. A 90-year sea warming trend explains outbreak

patterns of western spruce budworm in Vancouver Island. The Forestry

Chronicle, 83: 867-869.

Thomson, A.J., Shepherd, R.F., Harris, J.W.E., and Silversides, R.H. 1984. Relating

weather to outbreaks of western spruce budworm, Choristoneura occidentalis

(Lepidoptera: Tortricidae), in British Columbia. The Canadian Entomologist, 116:

375-381.

Uvarov, B.P. 1931. Insects and Climate. Transactions of the Entomological Society

of London, 79: 1-232.

Wellington, W.G. 1950. Effects of radiation on the temperatures of insectan habitats.

Scientific Agriculture, 30: 209-234.

Wellington, W.G., Fettes, J.J., Turner, K.B., and Belyea, R.M. 1950. Physical and .

biological indicators of the development of outbreaks of the spruce budworm.

Canadian Journal of Research D, 28: 308-331.

White, T.C.R. 1993. The inadequate environment. Nitrogen and the abundance of

animals. Springer-Verlag, Berlin.

White, T.C.R. 2018. An alternative hypothesis explains outbreaks of conifer-feeding

budworms of the genus Choristoneura (Lepidoptera: Tortricidae) in Canada.

Journal of Applied Entomology, 142(8): 725—730. doi: 10.1111/jen.12523.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 3

Collections of fleas (Siphonaptera) from

Pacific marten, Martes caurina

(Carnivora: Mustelidae), reveal unique

host—parasite relationships in the

Haida Gwaii archipelago

C.M. BERGMAN! T.D.GALLOWAY?2, AND

P. SINKINS!

ABSTRACT

Fleas and their host~parasite relationships are understudied in many parts of

Canada, yet such relationships may contribute to our knowledge of

ecosystems in ways we have yet to understand. A collection of 57 fleas from

Pacific marten (Martes caurina (Merriam)) in Haida Gwaii, off the coast of

British Columbia, Canada, led to the collection of five taxa of fleas: the

European rat flea, Nosopsyllus fasciatus (Bosc), a squirrel flea, Ceratophyllus

(Amonopsyllus) ciliatus protinus (Jordan), a mustelid flea, Chaetopsylla

floridensis (1. Fox), Hystrichopsylla (Hystroceras) dippiei, likely ssp. spinata

Holland, a parasite of mustelids and mephitids, and a generalist bird flea,

Dasypsyllus gallinulae perpinnatus (Baker). All five species are first records

for Haida Gwaii, and C. floridensis is recorded from Canada for the first time.

Two new host—parasite relationships support a previous dietary study of

marten in Haida Gwaii. This provides further evidence that fleas infesting

predators may indicate prey composition within their home ranges.

INTRODUCTION

Of 154 species of fleas (Siphonaptera) reported in Canada (Galloway 2019),

eight species are known to infest mammals and birds living in the remote

archipelago of Haida Gwaui, off the coast of British Columbia, Canada (Holland

1985). Among the first western scientific collections are those of Opisodasys

keeni (Baker) from Keen’s Mouse (Peromyscus keeni (Rhoads)) made by

Reverend John Henry Keen in 1895 (Sealy 2018).

In addition to hosting their own characteristic species of fleas, carnivores are

commonly infested with fleas from moribund prey and their nests (Rust ef ail.

1971). This makes predators useful targets for surveillance of flea-borne

pathogens such as plague (Yersinia pestis (Lehmann & Neumann)) (Gage ef al.

1994; Salkeld and Stapp 2006; Brown ef a/. 2011), which is an effective

monitoring strategy in addition to sampling prey alone, because predators may be

exposed to many individual prey animals of multiple species over time. For

example, predators are known to be accidental hosts for at least 40 of 50 flea

species that carry plague (Gage ef al. 1994). Thus, while many species of fleas

| Gwaii Haanas National Park Reserve, National Marine Conservation Area Reserve and Haida Heritage

Site, Parks Canada Agency, Skidegate, BC, VOT 1S]

2 Corresponding author: Department of Entomology, University of Manitoba, Winnipeg, MB, R3T 2N2;

Terry.Galloway@umanitoba.ca

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 18

have evolved close associations with certain hosts, they may also be found

commonly on predators of those host species. Because these fleas would not

normally infest their predators as hosts, their presence on predators is incidental.

This can allow researchers to infer predator—prey interactions from the species of

fleas found on predators (Zielinski 1984).

At least 25 species of fleas have been found on martens (American marten,

Martes americana (Turton), and Pacific marten, (Martes caurina (Merriam)) in

North America, most of which were likely acquired from their prey. In addition

to the 19 species/subspecies reported by Holland (based on current

nomenclature, 21 species/subspecies in Holland (1985)), six others have been

identified more recently in the United States by Zielinski (1984) and Scharf

(2017). Numbers of fleas collected from individual marten are often small,

averaging about 3—4 fleas per host individual (DeVos 1957; Zielinski 1984). Of

all the species of fleas documented from marten, none are monoxenous parasites

of marten, though several have mustelids as their primary hosts (Holland 1985).

Two each of Monopsyllus vison (Baker) and Kuichenliupsylla atrox (Jordan)

were collected from American marten (DeVos 1957) in Ontario. Holland (1949)

also documented Nearctopsylla grahami Holland on a marten in Ontario.

Nearctopsylla hyrtaci (Rothschild) was found on marten in Montana (Senger

1966), and Chaetopsylla floridensis (1. Fox) was first documented on marten in

Alaska (Hopla 1965). In what is perhaps the largest study of marten

ectoparasites, seven species were found among 70 fleas collected from 13

captures and recaptures of /. americana on 20 occasions in California (Zielinski

1984). These included Ceratophyllus (Amonopsyllus) ciliatus Jordan, C.

floridensis, Aetheca wagneri (Baker), Megarthroglossus spp., Orchopeas nepos

(Rothschild), Eumolpianus eumolpi (Rothschild), and Oropsylla idahoensis

(Baker). Orchopeas caedens (Jordan) was found on marten in the Yukon (Haas

and Johnson 1981; Haas ef al. 1989), and both C. ciliatus and Hystrichopsylla

dippiei spinata (Holland) have been found on marten in southeastern Alaska

(Haas ef al. 1989, 2005). Chaetopsylla lotoris (Stewart), Nearctopsylla genalis

(Baker), and Orchopeas howardi (Baker) infested marten in Michigan (Scharf

2017).

Understanding host—flea relationships gives insight into predator diets, may —

complement traditional diet studies by enlightening our understanding of

predator diets when such studies are scant or incomplete, and may provide

important links in epidemiological studies. In Haida Gwati (known for a time as

the Queen Charlotte Islands), located off the north coast of British Columbia,

only one study of marten diet has been undertaken (Nagorsen ef a/. 1991), and

samples were limited to areas of industrial forest harvest where access was

provided by logging roads. Here, we document new records for species of fleas

for Haida Gwaii and for Canada, as well as two new host—parasite associations

for marten.

MATERIALS AND METHODS

A female Pacific marten, accidentally struck and killed by a vehicle on the

morning of 5 February 2018 on the outskirts of the Village of Queen Charlotte

(53.2487°, —132.0286°), provided a large collection of fleas. Prior to necropsy

performed the same day, the animal had been placed in a freezer for

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 19

approximately six hours. Upon removal from the freezer, the carcass was placed

on a white surface, and live fleas were collected by hand from the fur as they

appeared at the surface during warming. On 17 January 2019, a male Pacific

marten was found dead on the only paved road on northern Moresby Island, a

15-km stretch connecting the community of Sandspit with the Alliford Bay ferry

terminal (53.2240°, ~131.9440°). The cold carcass was sealed in a bag, and fleas

were collected that evening using the method described above.

Fleas were also collected from two black rats (Rattus rattus (Linnaeus)) on

the Swan Islands (52.3321°, ~131.3063°), 28 March 2018, and another from the

Village of Queen Charlotte (53.2554°, -132.0757°), 31 January 2019.

Fleas were first frozen, then preserved in 95% ethanol. Fleas were mounted in

Canada balsam using the method described by Richards (1964). Voucher

specimens were deposited in the Royal BC Museum (Victoria, British

Columbia), Haida Gwaii Museum at Kay Llnagaay (Skidegate, Haida Gwaii,

British Columbia) and the J.B. Wallis/R.E. Roughley Museum of Entomology

(Department of Entomology, University of Manitoba, Winnipeg, Manitoba).

RESULTS

Fifty-four fleas were collected from the female marten. Gross physical

examination of the marten indicated the animal was older, as evidenced by

significant tooth wear and poor body condition. Forty-one (37 females, 4 males)

of the fleas collected were northern rat fleas, Nosopsyllus fasciatus (Bosc), 11 (3

males; 8 females) were squirrel fleas, Ceratophyllus (Amonopsyllus) ciliatus

protinus (Jordan), one (female) was a mustelid flea, Chaetopsylla floridensis, and

one female was Hystrichopsylla (Hystroceras) dippiei, likely ssp. spinata

Holland (males are needed for positive identification (Holland 1985; Lewis and

Lewis 1994)), a parasite of mustelids and mephitids. Three fleas were collected

from the male marten. Two of these were squirrel fleas, C. ciliatus protinus (1

male, 1 female), and the third was a generalist bird flea, Dasypsyllus gallinulae

perpinnatus (Baker) (1 female). All fleas collected from rats were northern rat

fleas, N. fasciatus. Rats from the Swan Islands were infested with one male and

three females, and the one from Queen Charlotte with one male and seven

females.

DISCUSSION

Chaetopsylla floridensis is a new flea record for Canada. This flea is a known

parasite of mustelids in Alaska and has been recorded from islands in the nearby

southern end of the Alaskan panhandle (Alexander Archipelago). Although the

one we found is a first record for Canada, its late detection is probably a result of

a general lack of study. The species has an odd history: it was originally

described in 1939 from specimens gathered from leaf mold in Gainesville,

Florida (Ewing and Fox 1943). Specimens collected since then have come from

Alaska (Hopla 1965; Haas ef a/.1978) and Colorado (Eads et al.1979). Zielinski

(1984) found that 31% of all fleas collected from 13 American marten (M.

americana) in California were this species, although C. /floridensis was

previously unreported in that state. Chaetopsylla floridensis is of particular

biogeographic interest because of its close relationship with marten. The extreme

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 20

isolation of MZ caurina in the Haida Gwaii archipelago (Dawson ef al. 2017)

suggests future research on the genetic uniqueness of C. floridensis in Haida

Gwaii would be of interest. In contrast to marten populations on some islands in

the Alexander Archipelago (SE Alaska), where introduced M. americana have

likely interbred with native MZ caurina, no introductions of M. americana are

known to have occurred in Haida Gwaii.

On a smaller spatial scale, our collections from two marten increase the

known flea fauna of the Haida Gwaii archipelago by 63%. Two of the five

species found on the marten, C. floridensis and H. dippiei spinata, have

mustelids as usual hosts, whereas D. gallinulae, C. ciliatus protinus, and N.

fasciatus are likely secondary infestations from marten prey species. Their usual

hosts, birds, red squirrel and rats, respectively, have all been documented as prey

items for marten in Haida Gwaii (Nagorsen ef a/. 1991). Unlike squirrels and

birds, rats were rare in marten diet in this study. None of these five flea species

has been documented in Haida Gwaii previously and, although lack of sampling

may contribute to a historical paucity of data, at least some of these species may

be relatively recent arrivals to the Haida Gwati archipelago, given the

introduction mostly during the past century of several species of mammals

(Golumbia et al. 2002). Ceratophyllus ciliatus protinus likely arrived with the

red squirrel (Jamiasciurus hudsonicus (Erxleben)) in 1950 (Golumbia ef al.

2002; Sealy 2012), whereas the introduction of N. fasciatus with rats (both R.

rattus and R. norvegicus (Berkenhout)) probably occurred much earlier, perhaps

in the 18th century. It is thus possible that the introduction of several mammal

species to Haida Gwaii has increased flea loads on some native mammals and

has perhaps contributed to a decline in fitness as a result. In contrast, D.

gallinulae is a bird flea presumed to be native to the islands.

Ceratophyllus ciliatus protinus is a Pacific coast flea species that infests red

squirrels, but is often recorded from predators of squirrels, including marten

(Ziclinski 1984; Haas et al. 1989). Red squirrels are now common.on Graham

and Moresby Islands and also on numerous smaller islands in the archipelago.

Another species of squirrel flea, O. caedens, which is common on red squirrels

throughout most of their range (Holland 1985) and is also documented in other |

mainland marten populations (Haas and Johnson 1981; Haas et a/. 1989; TD

Galloway, unpublished data), was not found on our two marten specimens.

Neither are there documented records of O. caedens from Vancouver Island

(Holland 1985). Vancouver Island was the source for Haida Gwaii red squirrels,

and the absence of this flea species on Vancouver Island would explain its

absence in Haida Gwaii.

Because the prey volume in the Haida Gwaii marten diet comprises twice the

red squirrels compared to rats (Nagorsen eft a/. 1991), one would predict C.

ciliatus protinus to occur on marten in higher numbers than rat fleas. Although

this prediction holds true for the collection made from the male marten (67%

squirrel fleas; 0% rat fleas), the opposite pattern was observed for the female

marten (76% rat fleas; 20% squirrel fleas). Even though these patterns are

inconsistent, they can be explained by the prey composition in the martens’ home

ranges. Rats are not present at the location where the male marten was collected,

but they are abundant in the area where the female marten was collected (Gwaii

Haanas 2019). As an avenue of future investigation, we suggest that marten

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 21

living in sympatry with rats might prefer rats over squirrels as prey. The same

may be true of other North American mesocarnivores that also include red

squirrels in their diet. If so, this food preference might modify small mammal

community composition and also may help explain why rats (Rattus spp.) are

absent from certain areas of the North American landscape. Unlike red squirrels,

rats did not co-evolve with marten and may be ill-adapted to avoid marten

predation. This could mean they may be selected preferentially by marten in

areas where both prey species occur. Similarly, in the United Kingdom, the

newly re-introduced European pine marten (Martes martes (Linnaeus)) prefers

the non-native grey squirrel (Sciurus carolinensis Gmelin) to the native red

squirrel (S. vulgaris Linnaeus), and their differential predation is affecting

community structure to the extent that the introduced grey squirrel may

eventually be extirpated (Sheehy e¢ a/. 2018). |

Hystrichopsylla dippiei spinata occurs on Vancouver Island, in the British

Columbia Lower Mainland and at Williams Lake, British Columbia (Holland

1957, 1985), and has been recorded in Oregon (Lewis ef a/. 1988). There are also

records from southern Alaska (Haas ef a/. 2005). It is a parasite mainly of

mustelids (Haas eft al. 1978), with a number of records from marten (Haas ef al.

2005); Haas et al. (2005) suggested that marten are the true hosts, but additional

study is needed to confirm this relationship. We are fairly certain our specimen is

this subspecies, based on location and host, but male specimens are needed as

they have the diagnostic features for the subspecies. Because H. dippiei spinata

is a winter flea, and our sampling effort was limited, it is likely this species has

been present in Haida Gwaii for some time without being detected. Further

sampling will no doubt provide new information about the occurrence of this

flea.

There are no published records for two of the five flea species new for Haida

Gwaii, N. fasciatus or D. gallinulae, infesting marten in Canada (Holland 1985,

p. 480). These host~parasite relationships may be unique to the Haida Gwaii

archipelago, where birds comprise a much higher proportion of marten diet

compared to what has been observed in marten populations elsewhere (Nagorsen

et al. 1991), and where rats are present only on small, off-shore islands where

marten do not occur and in association with human settlements on the two largest

islands. Specimens of marten on which Nagorsen ef a/. (1991) based their diet

study were collected from traplines radiating from towns and accessed by

logging roads. While historic studies of rat distribution in Haida Gwaii show rats

present on many islands in the archipelago (Bertram and Nagorsen 1995;

Golumbia et a/. 2002), a more recent, finer-scale study of rat distribution

demonstrates that rats are functionally absent from habitats where marten occur

and human disturbance is low—that is, the majority of the land area of the

archipelago’s two largest islands, Graham and Moresby (Gwaii Haanas 2019).

Thus, rats occur on small Haida Gwaii islands where marten are absent, and on

islands where marten are present but only in near proximity to areas of high

human disturbance, including all towns in the archipelago (Gwaii Haanas 2019).

As a result, there is little overlap between populations of marten and rats in

Haida Gwaii, except at the margins of areas populated by humans. This would

explain the low occurrence of rats in Nagorsen’s marten diet study, and yet the

prevalence of rat fleas on our female marten which, given its location of death,

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 22

apparently inhabited a rat-infested home range due to its proximity to human

habitation. Less than a month after collection of this specimen, another marten

was observed along the same stretch of road carrying a rat in its mouth and, as

with the individual hit by a car, was travelling across the road from shore to

forest in the early dawn hours.

Nosopsyllus fasciatus, the northern rat flea, is an immigrant to North America

and is widely distributed with its hosts, including along coastal British Columbia

(Holland 1940, 1985) and Alaska, USA. Rat fleas predominated in our sample,

which may suggest rats are a large component of the diet of marten in this

location. Moreover, this was the only flea species found on the black rat

specimens collected as part of this study. Although not commonly documented as

prey items for marten, rats are included in the diet of marten on both on

Vancouver Island (Nagorsen ef a/.1989) and in Haida Gwaii (Nagorsen ef al.

1991).

The large number (54) of fleas collected from the one female marten appears

atypical, given the much smaller numbers collected from individual hosts in

other studies. The only other similarly documented large collection from an

individual mustelid was described by Holland (1985): Martha Fern Munroe and

her husband (Frank Banfield) collected 38 fleas from a mink, stating that “there

were about 200 other fleas” on the mink. Holland goes on to state that this was

remarkable, and that he “supplied most of the flea museums of the world with

this species from this collection (Megabothris atrox (now in the genus

Kuichenliupsylla))...” There is one record of 88 fleas infesting a female

American marten in Manitoba (TD Galloway, unpublished data), consisting

predominantly of the tree squirrel flea, O. caedens (n=84), plus Chaetopsylla

lotoris, Nearctopsylla hygini, and K. atrox. Kuichenliupsylla atrox is considered

to parasitize mustelids as its true hosts. It is possible that such large flea

collections from a single host are not actually unusual, but rather related to the

length of time after death until the host and its fleas are contained or frozen.

Future studies should consider this factor if it is desirable to maximize the

numbers of fleas collected from each individual host.

While ecologists often attempt to explain the presence of species, rarely is

work done to explain a species’ absence. Rats are troublesome vermin in cities

and agricultural landscapes across North America, yet they have not invaded

many extensive tracks of northern forest where mesocarnivores such as marten

are present. Why is this the case? On a continental scale, mesopredators such as

marten may be responsible for preventing the spread of rats into such habitats.

Our discovery of a new host—parasite relationship may indicate this valuable

ecosystem service that mesocarnivores provide—one that has not yet been

recognized or valued and may be threatened by overharvesting through

commercial trapping of mustelids.

ACKNOWLEDGEMENTS

We thank Gwaii Haanas for access to their archives and unpublished

information on rat distribution in Haida Gwaii. TDG thanks the Department of

Entomology and the Faculty of Agricultural and Food Sciences, University of

Manitoba, for their continued support. Ralph Eckerlin (Natural Sciences

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 23

Division, Northern Virginia Community College, Annandale, Virginia, USA)

provided supplemental information on fleas in the United States.

REFERENCES

Bertram, D.F. and Nagorsen, D.W. 1995. Introduced rats, Rattus spp., on the Queen

Charlotte Islands: implications for seabird conservation. Canadian Field-

Naturalist, 109: 6—10.

Brown, H.E., Levy, C.E., Enscore, R.E., Schriefer, M.E., DeLiberto, T.J., Gage, K.L.,

and Eisen, R.J. 2011. Annual seroprevalence of Yersinia pestis in coyotes as

predictors of interannual variation in reports of human plague cases in Arizona,

United States. Vector-Borne and Zoonotic Diseases, 11: 1439-1446.

Dawson, N. G., Colella, J.P., Small, M.P., Stone, K.D., Talbot, S.L., and Cook, J.A.

2017. Historical biogeography sets the foundation for contemporary conservation

of marten (genus Martes) in northwestern North America. Journal of

Mammalogy, 98: 715-730.

DeVos, A. 1957. Pregnancy and parasites of marten. Journal of Mammalogy, 38:

412.

Eads, R.B., Campos, E.G., and Barnes, A.M. 1979. New records for several flea

(Siphonaptera) species in the United States, with observations on species

parasitizing carnivores in the Rocky Mountains. Proceedings of the

Entomological Society of Washington, 81: 38-42.

Ewing, H.E. and Fox, I. 1943. The Fleas of North America: Classification,

Identification, and Geographic Distribution of These Injurious and Disease-

Spreading Insects. U. S. Department of Agriculture.

Gage, K.L., Montenieri, J.A., and Thomas, R.E. 1994. The role of predators in the

ecology, epidemiology and surveillance of plague in the United States.

Proceedings of the Sixteenth Vertebrate Pest Conference, February, 1994. Pp.

200-206.

Galloway, T.D. 2019. Siphonaptera of Canada. /n: Biota of Canada — A biodiversity

assessment. Part 1: The terrestrial arthropods. D.W. Langor and C.S. Sheffield

(Eds). ZooKeys, 819: 455-462. doi.org/10.3897/ zookeys.819.25458

Golumbia, T.E., Bland, L., Moore, K., and Bartier, P. 2002. History and current

status of introduced vertebrates on Haida Gwaii. Jn A.J. Gaston, T.E. Golumbia,

J.-L. Martin, and S.T. Sharpe (eds.), Proceedings from the Research Group on

Introduced Species 2002 Symposium: Lessons from the Islands, 2002. Queen

Charlotte, BC: Canadian Wildlife Service Special Publication. Pp. 8-31. http://

publications.gc.ca/collections/collection_2019/eccc/CW69-20-1-2007-eng. pdf.

Gwaii Haanas. 2019. Rats On Board (pamphlet). Gwaii Haanas archives, Gwaii

Haanas National Park Reserve and Haida Heritage Site, Skidegate, BC.

Haas, G.E. and Johnson, L. 1981. A Siphonaptera collection from small mammals by

the Canol Road, Yukon Territory. The Canadian Entomologist, 113: 567.

Haas, G.E., Kucera, J.R., Runck, A.M.Y.M., MacDonald, S.O., and Cook, J.A. 2005.

Mammal fleas (Siphonaptera: Ceratophyllidae) new for Alaska and _ the

southeastern mainland collected during seven years of a field survey of small

mammals. Journal of the Entomological Society of British Columbia, 102: 65—

76. http://journal.entsocbe.ca/index.php/journal/article/view/58.

Haas, G.E., Barrett, R.E., and Wilson, N. 1978. Siphonaptera from mammals in

Alaska. Canadian Journal of Zoology, 56: 333-338. https://doi.org/10.1139/

z78-045.

Haas, G.E., Wilson, N., Zarnke, R.L., and Johnson, L. 1989. Mammal fleas

(Siphonaptera) of Alaska and Yukon Territory. Canadian Journal of Zoology, 67:

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 24

394—405.

Holland, G.P. 1940. A survey of the rat fleas of the southern British Columbia Coast

with relation to plague studies. Journal of the Entomological Society of British

Columbia, 37: 1—5. https://www.cabdirect.org/cabdirect/abstract/19422900519.

Holland, G.P. 1949. The Siphonaptera of Canada. Dominion of Canada, Department

of Agriculture, Publication 817, Technical Bulletin 70.

Holland, G.P. 1957. Notes on the genus Hystrichopsylla Rothschild in the New

World, with descriptions of one new species and two new subspecies

(Siphonaptera: Hystrichipsyllidae) The Canadian Entomologist, 89: 309-324.

Holland, G.P. 1985. The fleas of Canada, Alaska and Greenland (Siphonaptera).

Memoirs of the Entomological Society of Canada, No. 130.

Hopla, C. 1965. Alaskan hematophagous insects, their feeding habitas and potential

as vectors of pathogenic organisms. I: The Siphonaptera of Alaska. Fort

Wainright Aeromedical Laboratory, AAL-TR-64-12, Volume I, xiii + 267 pp.

https://apps.dtic.mil/dtic/tr/fulltext/u2/469666.pdf.

Lewis, R.E. and Lewis, J.H. 1994. Siphonaptera of North America north of Mexico:

Hystrichipsyllidae sens. str. Journal of Medical Entomology, 31: 795-812.

Lewis, R.E., Lewis, J.H., and Maser, C. 1988. The fleas of the Pacific Northwest.

Oregon State University Press.

Nagorsen, D.W., Campbell, R.W., and Giannico, G.R. 1991. Winter food habits of

marten, (Martes americana), on the Queen Charlotte Islands. Canadian Field-

Naturalist, 105: 55—59.

Nagorsen, D.W., Morrison, K.F., and Forsberg, J.E. 1989. Winter diet of Vancouver

Island marten (Martes americana). Canadian Journal of Zoology, 67: 1394-1400.

https://doi.org/10.1139/z89-198.

Richards, W.R. 1964. A short method for making balsam mounts of aphids and scale

insects. The Canadian Entomologist, 96: 963-966.

Rust, J.H., Cavanaugh, D.C., O’Shita, R., and Marshall, J.D. 1971. The role of

domestic animals in the epidemiology of plague. I. Experimental infection of

dogs and cats. Journal of Infectious Diseases, 124: 522-526. https://doi.org/

10.1093/infdis/124.5.522.

Salkeld, D.J. and Stapp, P. 2006. Seroprevalence rates and transmission of plague

(Yersinia pestis) in mammalian carnivores. Vector-Borne and Zoonotic Diseases,

6: 231-239.

Scharf, W. 2017. Recent Siphonaptera host and distribution records from northern

Michigan. The Great Lakes Entomologist, 50: 31-34. http://scholar.valpo.edu/

tglehttp://scholar.valpo.edu/tgle/vol50/iss 1/6.

Sealy, S.G. 2012. Voucher specimens of red squirrels introduced to Haida Gwaii

(Queen Charlotte Islands), British Columbia. Wildlife Afield, 9: 59-65. http://

www.wildlifebc.org/pdfs/9_1 Red Sqirrel Sealy R.pdf.

Sealy, S.G. 2018. Insect taxa named for the Rev. John H. Keen, early naturalist on

the Queen Charlotte Islands and at Metlakatla, British Columbia. Journal of the

Entomological Society of British Columbia, 114: 15-21. https://

journal.entsocbe.ca/index.php/journal/article/view/972.

Senger, C.M. 1966. Notes of fleas (Siphonaptera) from Montana. Journal of the

Kansas Entomological Society, 39: 105~109. https://doi.org/10.2307/25083495.,

Sheehy, E., Sutherland, C., O’Reilly, C., and Lambin, X. 2018. The enemy of my

enemy is my friend: native pine marten recovery reverses the decline of the red

squirrel by suppressing grey squirrel populations. Proceedings of the Royal

Society B: Biological Sciences, 285 (20172603). https://doi.org/10.1098/rspb.

2017.2603.

Zielinski, W.J. 1984. Plague in pine martens and fleas associated with its occurrence.

Great Basin Naturalist, 44: 170-175.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 B ie

Dispersal of Bactericera cockerelli

(Hemiptera: Triozidae) in relation to

phenology of matrimony vine (Lycium

spp.; Solanaceae)

W, RO COOPER: D. ROTORTON2,

J. THINAKARANS, AND A. V. KARASEV4

ABSTRACT

Bactericera cockerelli (Sulc) (Hemiptera: Triozidae) is a key pest of potato

(Solanum tuberosum; Solanaceae) in western North America. Native species

of Lycium (Solanales: Solanaceae) in the southwestern U.S. have been known

since the early 1900s to support populations of B. cockerelli. These shrubs are

adapted to survive arid habitats by entering a summer dormancy characterized

by partial or complete defoliation. Summer leaf fall by native Lycium in the

southwestern U.S. triggers the dispersal of B. cockerelli to new seasonally

available hosts, including potato. Recently, B. cockerelli was found to occur

on non-native species of Lycium (L. barbarum and L. chinense), collectively

known as matrimony vine in the Pacific Northwest (Washington, Oregon, and

Idaho). Monitoring of matrimony vine in previous years suggested

qualitatively that these non-native shrubs also entered a summer dormancy

with effects on B. cockerelli populations. Our study had two principal

objectives: 1) document when and under what conditions matrimony vine

enters summer dormancy, and 2) determine whether summer leaf fall is

associated with dispersal of B. cockerelli from these plants. In this report, we

demonstrate that matrimony vine exhibits xerophytic phenological traits

similar to the Lycium species native to the southwestern United States, and we

provide evidence that psyllid dispersal from matrimony vine is associated

with the onset of the host plant’s summer dormancy. These results may be

beneficial for the development of predictive models to forecast B. cockerelli

pressure in potato based upon populations occurring on matrimony vine in

early spring.

INTRODUCTION

The potato psyllid, Bactericera cockerelli (Sulc) (Hemiptera: Triozidae) is a

key pest of solanaceous crops (Solanales) including potato (Solanum tuberosum)

and tomato (S. /ycoperiscum) in western North America (Munyaneza 2012). This

psyllid is a primary vector of “Candidatus Liberibacter solanacearum” (= “Ca. L.

psyllaurous”’), the pathogen associated with zebra chip disease of potato (Hansen

'Corresponding author: USDA-ARS- Yakima Agricultural Research Laboratory, 5230 Konnowac Pass

Rd. Wapato, WA, 98951; Rodney.Cooper@usda.gov

2USDA-ARS- Yakima Agricultural Research Laboratory, 5230 Konnowac Pass Rd. Wapato, WA 98951

3Karunya Institute of Technology and Sciences, School of Agriculture and Biosciences, Karunya Nagar,

Coimbatore India 641114

4University of Idaho, Department of Plant, Soil, and Entomological Sciences, 875 Perimeter Dr. MS

2339 Moscow, ID 83844

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 26

et al. 2008; Liefting et al. 2009; Munyaneza 2012). No direct methods currently

exist for controlling Liberibacter, so zebra chip disease is managed using

prophylactic calendar-based applications of insecticides to suppress populations

of the vector. Bactericera cockerelli develops on nearly all genera of Solanaceae

examined and on many species of Convolvulaceae (Solanales), including many

weeds that require management by growers (Crawford 1914; Essig 1917;

Knowlton and Thomas 1934; Pletsch 1947; Wallis 1955; Murphy e¢ al. 2013;

Thinakaran e¢ al. 2017; Kaur et al. 2018; Cooper et al. 2019a). A major challenge

in the management of B. cockerelli and zebra chip disease is the inability to

predict when and in what fields the psyllid is likely to first colonize. The

difficultly in making such predications is largely due to uncertainty of the

primary weed sources of B. cockerelli entering fields of potato.

In the southwestern United States (U.S.), Lycium spp. (Solanales: Solanaceae)

are important non-cultivated host plants for B. cockerelli (Romney 1939). This

genus includes about 80 species worldwide with an area of high species richness

occurring in the southwestern U.S. (Hitchcock 1932; Chiang-Cabrera 1981;

Levin and Miller 2005). Lycium species that are native to the southwest exhibit

two seasonal intervals during which new foliar growth occurs, separated by a

summer dormancy characterized by total or partial defoliation (Hanley and

Brady 1977; Ackerman et al. 1980). The summer dormancy allows Lycium

species to survive periods without precipitation by limiting periods of growth

and flowering to the spring and autumn when precipitation occurs. Large

populations of B. cockerelli occur particularly on Lycium during the intervals of

leaf flush in spring and autumn (Romney 1939) but are thought to disperse when

Lycium enters summer dormancy, and to colonize seasonally available annual

Solanaceae weeds (e.g., Solanum, Nicotiana, Datura, etc.) (Wallis 1955; Horton

et al. 2016). A portion of B. cockerelli dispersing from Lycium are thought to also

colonize commercial plantings of potato or tomato.

Although Lycium does not naturally occur in the Pacific Northwest (PNW =

Washington, Oregon, and Idaho), herbaria records (http://pnwherbaria.org)

indicate that two introduced Eurasian species — L. barbarum L. (=L. halimifolium

Miller) and L. chinense Miller — occur in this region. Floral morphology and

sequence analyses support the occurrence of both species in the PNW, but

evidence also suggests the presence of intermediate forms, possibly due to

hybridization (Horton et al. 2016). These closely related species are both referred

to collectively and without distinction as matrimony vine, wolfberry, or Goji

berry (hereafter referred to as matrimony vine) (Horton ef a/. 2016). Matrimony

vine has recently been identified as an important host plant of B. cockerelli in the

PNW, particularly during the spring and summer, when annual host plants of the

psyllid are unavailable (Horton e¢ al. 2016; Thinakaran et al. 2017; Cooper et al.

2019a).

Matrimony vine appears to enter a period of summer dormancy marked by

defoliation much like the native Lycium species of the southwestern U.S. (Horton

et al. 2016; Thinakaran ef a/. 2017), but the relationship between matrimony vine

phenology and B. cockerelli dispersal is still unclear. Anecdotal evidence

suggests that leaf fall of matrimony vine in the PNW does indeed trigger

dispersal of B. cockerelli to annual herbaceous hosts, including potato. This

evidence includes observations that populations of B. cockerelli are highest

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 Za

during the intervals of leaf flush and begin arriving in fields of potato at about

the same time they begin declining on matrimony vine (Horton ef al. 2016;

Thinakaran ef al. 2017). Further evidence is provided by results of gut content

analysis that indicated that at least some B. cockerelli present on matrimony vine

had arrived there from potato following harvest (Cooper eft al. 2019b). A long-

term goal of our research is to develop a model to predict B. cockerelli pressure

in potato fields during summer based upon populations of B. cockerelli occurring

on matrimony vine during spring (Cooper ef a/. 2019c). For this model to be

effective, there is a need to document when and under what conditions

matrimony vine enters summer dormancy and to determine whether summer

leaf-fall is associated with dispersal of B. cockerelli from these plants. For this

study, we monitored both wild and propagated stands of matrimony vine in

Washington State (WA) and used canonical correlation analysis to identify

associations among weather events, plant phenology, and presence or dispersal of

B. cockerelli.

MATERIALS AND METHODS

Seasonal phenology of naturalized stands of matrimony vine

Psyllid presence and plant phenology were monitored weekly in naturalized

stands of matrimony vine (two stands in 2017, three in 2018) located in central

WA (Table 1). Sampling took place from spring until autumn to include both

periods of leaf flush and the period of summer dormancy. These three stands

were chosen because they were known to harbor large populations of B.

cockerelli in previous years (Thinakaran ef a/. 2017).

Table 1. Locations and characteristics of matrimony vine stands used to examine the

relationship between plant phenology and dispersal of B. cockerelli. Sampling dates

are also provided.

Approx. stand

Location GPS coordinates iat Months sampled

Prosser, WA, Gap 46° 13'24.8"N 20 Apr. to 29 Sep.

Road 119° 47’ 24.5" W 20x5m 2017

2 May to 7 Nov. 2018

Richland, WA, Port 46° 20' 57.4" N 18 May to 29 Sep.

of Benton Blvd 119° GL 1O,8 Weir 2017

| iy 2 May to 7 Nov. 2018

Richland, WA, Horn 46° 21'3.2"”N |

Bactericera cockerelli populations were monitored by dislodging adults from

branches onto a 0.5-m? beat sheet by tapping the branches with a 0.2-m rubber

hose at five locations from each stand. Dispersal of adult B. cockerelli from the

matrimony vine stands was monitored by placing three yellow sticky cards (30

cm x 23 cm, Alpha Scents Inc., West Linn, OR) attached to metal stakes at a

height of 1.5 m about 3 m from the edge of each stand.

Variables associated with foliar growth and senescence included leaf drop,

leaf density, and chlorophyll content. Leaf drop was monitored by placing three

45 x 30 cm pans filled with water at the base of each stand and counting the

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 28

number of leaves dropped in each pan (Horton et al. 1993, 1994). Leaf density

was monitored by counting the number of leaves per centimetre of shoot length

on five 15—20 cm shoots from each stand. Chlorophyll content was estimated

from 25 leaves per stand using a chlorophyll meter (Opti-Sciences, Model

CCM-300) (Gitelson ef al. 1999).

Abiotic variables included soil moisture, air and soil temperature, and

precipitation. Soil moisture was estimated by collecting three 15-cm-deep soil

plugs from each stand and measuring the difference in weight between freshly

collected soil plugs and soil that had been air-dried in a detached greenhouse.

The samples were weighed weekly until the dry weight was unchanged. The

greenhouse was not cooled and achieved high temperatures with low humidity in

the eastern WA summer climate. Daily weather data (minimum, maximum, and

average temperatures, average soil temperature, and precipitation) were obtained

from the Washington State University AgWeatherNet (www.weather.wsu.edu)

“WSU Prosser” (5 km from the Prosser stand of matrimony vine) and “WSU Tri

Cities” (2-3 km from the two Richland stands) weather stations. The data for

each abiotic variable were averaged over the week preceding each weekly

sampling.

Correlations among variables were assessed with three independent canonical

correlation analyses using PROC CANCORR of SAS 9.4 (SAS Institute Inc.

2013). This method identifies and measures associations among two sets of

variables by determining orthogonal linear combinations of variables within each

dataset that best explain the variability within and between datasets. The number

of orthogonal linear combinations, called canonical variates, is equal to the

number of variables in the smaller of the two datasets. The canonical variates are

then interpreted in terms of the original variables by inspecting canonical

structures where larger values indicate greater correlations between variables,

and opposite signs indicate inverse relationships. Data were averaged by week at

each location, and each sampling week in 2017 and 2018 was ordinally

categorized with the reference ‘week 1’ being the first week of January. Both

years were included in each of the three analyses. The first analysis assessed

correlations between weather variables (mean air temperature, mean soil

temperature, soil moisture, and precipitation) with plant growth characteristics

(leaf drop, chlorophyll content, and leaf density). The second analysis assessed

correlations between plant growth characteristics and presence or dispersal of B.

cockerelli (numbers on plants and numbers on sticky traps). The third analysis

assessed correlations between weather variables and B. cockerelli movement

variables. In each analysis, the Wilks’ Lambda statistic was used to determine

overall significance, and correlations among variables were estimated by

canonical loadings.

Seasonal phenology of experimental stands of Lycium

Cuttings collected from a naturalized stand of Lycium near Selah, WA, in June

2014 were propagated in soil within a greenhouse and transplanted into

experimental plots (n=4) located at the USDA experimental farm near Moxee,

WA, in July 2014. Each plot consisted of four plants separated from one another

by ~0.5 m. Adjacent plots were separated by 3—5 m of fallow ground. Each plot

was enclosed in a 2x2x2-m organdy cage from 2014, when the plots were

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 29

established, until spring of 2017. By 2017, the plants had spread to fill the 2x2-m

space of each cage, with many new shoots from spreading rhizomes. A

pyrethrum fogger (Doktor Doom 70ml, Ultrasol Industries Ltd., Edmonton,

Alberta) was used to treat the plots in late March 2017 to eliminate resident

insects prior to the study. Pyrethrum was chosen because it is an effective contact

insecticide with a residual activity limited to only several hours. Each cage was

then infested with 500 B. cockerelli of the northwestern haplotype (Swisher ef al.

2012) obtained from a laboratory colony on 17 April 2017 and again on 26 April

2017, 2-3 weeks after the pyrethrum treatments. Cages were removed from

plants on 20 May 2017, when eggs were present on leaves and shoots. Weekly

sampling of B. cockerelli presence and dispersal and foliar characteristics began

on 5 June 2017. All four plots received weekly irrigation until late June.

Plant characteristics, soil moisture, and B. cockerelli populations were

estimated as described for the naturalized stands, except that four sticky cards

were placed at each cardinal direction around each plot. Weather variables were

obtained from the WSU AgWeatherNet station “Moxee East”, located about 10

km from the USDA experimental farm. The relationships between weather

variables, plant characteristics, and B. cockerelli dispersal were assessed using

canonical correlation analysis as described for naturalized stands.

Simulated drought using potted plants

Shoot cuttings collected from a stand of Lycium located near Richland, WA,

(Horn Rapids Rd) were propagated in 10-cm pots within a greenhouse

maintained at 24° C and with supplemental lighting to ensure a 16:8 (L:D) hour

photoperiod. All plants received weekly watering before the start of the

experiment until the plants had established roots, new shoots, and new foliage

(about 6 weeks).

Chlorophyll, leaf density, and soil moisture were measured for each plant

(n=12) for seven weeks. Plants were either watered weekly or were water-

stressed during weeks 2, 3, and 4 of the study. Plants were arranged in six blocks,

with each block consisting of one plant of each treatment. Plants received ~250

ml of water weekly immediately after plant and soil variables were recorded:

these included chlorophyll meter readings from five leaves per plant, and leaf

density with counts of the number of leaves on a 10-cm section of stem from

each plant. The stem sections were marked at the beginning of the study so that

the same section was assessed each week. Soil moisture was recorded using a

soil probe (Spectrum Technologist Inc., Aurora, IL, U.S.; model TDR300). Each

variable (chlorophyll, number of leaves per centimetre of shoot, and soil

moisture) was analyzed using PROC GLMMIX of SAS 9.4 (SAS Institute Inc.

2013). In each analysis, week, treatment (watered versus dry), and the week by

treatment interaction were included as fixed effects, and block and treatment by

block interaction were included as random effects. When the overall univariate

analysis indicated a significant main effect interaction, differences between

treatments were assessed using a Tukey adjustment for multiple comparisons

(ADJUST=TUKEY of the LSMEANS statement) while including the SLICE

option of the LSMEANS statement to limit comparisons within each week.

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 30

RESULTS AND DISCUSSION

As observed in previous years (Horton et a/. 2016; Thinakaran et al. 2017),

Lycium introduced to the PNW underwent leaf flushes in spring and autumn

separated by a period of senescence (leaf yellowing) and leaf fall during summer

(Fig. 1). The degree of defoliation varied among stands, from complete

defoliation at the Richland Port of Benton stand to partial defoliation at the

Richland Horn Rapids Road and Prosser Gap Road stands. Populations of B.

cockerelli were considerably low on matrimony vine and in potato fields

throughout the Columbia Basin in 2017 and 2018, relative to previous years

(Thinakaran et al. 2017; Cooper et al. 2019c). Populations were low not only in

wild stands of matrimony vine but also in experimental plots that were infested

with colony-reared B. cockerelli in the spring of 2017. Factors responsible for the

low populations 2017 and 2019 remain unknown.

Figure 1. Appearance of matrimony vine during the spring leaf flush in mid-May

(A), leaf senescence in mid-July (B), and the autumn leaf flush in late-September

(C). Location=Prosser, Gap Road.

Similar trends in weather, plant phenology, and psyllid numbers were

observed on the naturalized stands in 2017 (Fig. 2) and 2018 (Fig. 3). Although

the 2017 and 2018 datasets were analyzed together, they are presented separately

in Figs. 2 and 3 to visually show correlations between weather events and plant

phenology. The canonical correlation analysis of plant variables with weather

variables from naturalized stands indicated that the first canonical variate was

significant, but the remaining variates were not (Table 2A). Since only the first

variate was significant, only the first pair of variables need to be identified from

the canonical structures. The weather dataset was associated primarily with air

and soil temperatures (Table 2A; 0.98 and 0.86, respectively) and was negatively

associated with precipitation (Table 2A; —0.37). The plant dataset was negatively

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 3]

associated with leaf density (Table 2A; —0.90), and positively associated with

leaf fall (Table 2A; 0.35). From these results, we can conclude that leaf growth

was associated with cooler temperatures and precipitation, and that leaf fall is

associated with warmer temperatures and decreasing precipitation. Plants began

to visibly decline on 22 June 2017 and 6 June 2018, when maximum air

temperatures the week prior to sampling were >25° C and soil moisture was <5%

(Figs. 2 and 3). Late summer/early autumn precipitation elicited marked

increases in leaf density and chlorophyll content in plants (2017, Fig. 2; 2018,

Fig. 3). These late-season precipitation events followed 6—7 weeks without

recorded precipitation. Autumn leaf growth occurred from mature stems without

growth of new shoots, and leaf densities typically surpassed those observed

during the spring interval of leaf flush, when new leaves grew from elongating

shoots (2017, Fig. 2B; 2018, Fig. 3B).

A oe Air Temperature (min/max shaded)

++» Soil Temperature- -- - Soil Moisture

_ Weekly precipitation

om)

VWweekly Precipitation (:

Mean Daily Air/Sail Temperature (C}

Mean Daily Soil Moisture (%)}

* > See *

"gi na es Semen pees . e

2, BY >A fate Ue AL aa

C. Oo Xe Xe Re

Ra RE,

@ Leavesitray

4 Chlorophyll

aA |

piel. ai 1000

Mean No. Leaves per Shoot or Tray

Mean Chlorophyll Content (mg/m)

An cy ean 8, > Fa en FE Fan 2 S Fd : J, Fy 2, Vy A. 2 oO <9,

RQ Mg 7K RO Ny BRE Ng 28 QR Sy 7x78 eK Ay Fe SZ SR

Oe Oa a, & yy yy ,& & &

uv 1 Yon Yon My yy? yy % yy" Ge %4%G%% &&Y

g E ‘ "Beat Sheet Captures iS we

} a-« Trap catch : Be

Yio 4 K wt 8-8 oe 8 OH i 4

a 3 ‘ gots 10 a0

SG ° 7 1 5 s

cw ‘'f a Of

6 aR 6 ee ame on il a

= “oa 2 *Q “> tgs & 78 eS # & mW as “9 a “2 > % *, Wy ” az 2, 7

8 oe eis hoo Ge ee Ce YE Ue URS,

Figure 2. Relationships among weather variables (A), pig vine phenology

(B), and occurrence and dispersal of Bactericera cockerelli (C) on matrimony vine

stands located in Washington State in 2017.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 32

Canonical correlation analysis also indicated significant associations between

psyllid movement and plant growth characteristics for both canonical variates

(Table 2B), but no associations between psyllid movement and weather (Table

2C). The first canonical variate suggested an inverse relationship between psyllid

trap catch and leaf density (2017, Fig. 2; 2018, Fig. 3). The second canonical

variable suggested a correlation between leaf drop and detection of psyllids on

the plants (Table 2B), likely because adult populations peaked on plants at about

the same time that leaf fall began, and psyllids were largely absent from plants

after rates of leaf fall declined (2017, Fig. 2C; 2018, Fig. 3C). Overall, these

results are consistent with the hypothesis that summer leaf drop triggers dispersal

of B. cockerelli from stands of introduced matrimony vine in the PNW (Horton ef

al. 2016).

<2 60 ee so 2.5

» 3 meee Ait Temperature (min/max shaded) ‘i

en «+ «+ Soil Temperature» «> +» Soil Moisture 5

o 5 40 Veekly precipitation =

eo s

e. 2 Se :

p= % a

D a 20 | 8

>a =

& = 10} § s

@ a

£ = vs

2 GH er sige Tg S RR KE

| wie ot, RS Wg 'Q oy :

> B | @ Leaves/tray a

yen 1, bedi oe © Leaves/shoot E

. 150 } a’ A ‘the pe AcChloraphyl |. >

be ‘ ak | i900 £

H : a” ny eating rare Canis roy Or catatar =

@ 100 | é S

wi are é€ 4

a +. % ake

6 7 + e-* J500

“— 5

5 50b 4 "hie Po, a 2

rm ‘ 28>g~ 079. ‘ ~ Oe. o- O \ g .

S 970° iene m4 $" sis ong. : o-¢” 2

= a oO” i. 08 Ba 8 ~O. “eo ‘ asin

+ “07

2,8, Za e5 RF 2% Res G eh 2 Wp Be 2p 2Q Or K-77 Trp hy

My 4p DR A Sr) yk Rg eS a RAR? ‘O67 5% Ae,

Be tat te O60 COS Us BRO Go, Ga

i @

Ga § ' a

fe iG [___]Beat Sheet Captures

= 2 4] : 4 8

> 8) e-8 Trap Catch a

fe 3 i-8-8-8—~8—-e- eB 3 a0

. = 3 . er er ee ee ee ee ae s oa

e & ; ) | |

= ip < &. ve J See OB 2 4s a. ee B, & va 5 %, 7 7» wa 2

tee 4792 ROAST “alae BAe 2923 Wra%o3 4 My,

246% 69 GOGY COTO GY Ue BE EY GU eae

Figure 3. Relationships among weather vatieles (A), matrimony vine phenology

(B), and occurrence and dispersal of Bactericera cockerelli (C) on matrimony vine

stands located in Washington State in 2018.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 a3

Table 2. Canonical correlation analyses of multivariate datasets obtained from

naturalized stands of matrimony vine in Washington State.

Proportion First dataset Second dataset

Likelihood ratio of model canonical canonical

variance structures4 structures4

Canonical Canonical

comparisons correlation

A. Plant! versus weather variables?

~ Fs3,8: af=12, 291:

Can. variate] 0.51 P<0.001 0.80 Air li 0. 98 Beat fall: 0. 35

Soil temp: Chlorophyll:

0.86 0.16

Precipitation: Leaves/shoot:

0.37 ~-0.90

Soil moisture:

F=1.6; d=6, 222 pen

¢ i + =1.6; =O, ) Sot) eke | AAR! oe Wc

Can. variate2 0.26 P=0.14 Oa} cs

Can. variate3 0.12 *% eee relied 0.03 ve vo

B. Psyllid3 versus plant variables

Can. variate] 0.28 PT 8222; cs) Feat fall: 0.18. tap catch:

P=0.006 0.96

Chlorophyll: —_ Beat trays:

0.77 O53

Leaves/shoot:

~0.64

Can, variate?” Oat GO Tae dent tall 00. ee

Chlorophyll: — Beat trays:

0.42 0.95

Leaves/shoot:

C. Psyllid versus weather variables

' ; = ir temp: rap catch:

Can. variate | 0.21 P=0.5] 0.79 0.23 0.58

Soil temp: Beat trays:

0.01 —0.80

Precipitation:

0.46

Soil moisture:

0.43

F=0,51; df=3, 117;

P=0.67

'Plant variables included leaf drop (no. leaves per tray), chlorophyll content (mg/m2), and leaf density

(no. leaves per cm of shoot)

2Weather variables included avg. daily air temperature, avg. daily soil temperature, soil moisture, and

weekly precipitation.

3Psyllid variables included no. of psyllids collected from matrimony vine and no. of psyllids captured on

traps.

4_arger values indicate greater correlations between variables, and opposite signs indicate inverse

relationships.

Can. variate 2 0.11 eR aang aL AOR gy

In addition to monitoring naturalized stands of matrimony vine, we used

experimental plots to correlate weather patterns, matrimony vine phenology, and

psyllid dispersal. The use of experimental plots allowed us to replicate plots at a

single location and to corroborate observations at wild stands. Analysis of plant

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 34

and weather variables suggested that chlorophyll content was positively

correlated with soil moisture and negatively correlated with soil temperature

(Table 3A; Fig. 4). Analysis of psyllid and plant variables from the experimental

plots did not indicate significant patterns (Table 3B), but the upper bound

statistic, Roy’s Greatest Root, warranted assessment of canonical structures

(F=2.3; df=3, 48; P=0.09) which suggested a positive relationship between the

presence of psyllids on plants and leaf density (Table 3B; Fig. 4). Finally,

analysis of the psyllid and weather variables suggested that the capture of

psyllids on traps was negatively correlated with precipitation and soil moisture

and positively correlated with air and soil temperatures (Table 3C; Fig. 4).

Overall, results from the experimental plots were consistent with those from the

naturalized stands by suggesting that senescence and leaf fall in matrimony vine

are triggered by hot and dry conditions (average and maximum air temperature

was 22° C and 31° C respectively, and soil moisture below 5%) and is associated

with dispersal of B. cockerelli from these plants (Fig. 4).

Table 3. Canonical correlation analyses of multivariate datasets obtained from

experimental stands established at the USDA experimental farm near Moxee, WA.

Proportion First dataset | Second dataset

of model canonical canonical

_variance _structures* __structures*

Canonical Canonical Likelihood

comparisons correlation ratio

A. Plant! versus weather? variables

| =e SP) iy i

Can. variate 1 0.72 119: P<0,001 0.64 Air temp: —-0.31 Leaf fall: -0.07

; Chlorophyll:

Soil temp: —0.62 we y

Precipitation: | Leaves/shoot:

0.47 0.34

Soil moisture:

0.89

F=4,4; df=6,

Can. variate 2 0.57 92: P<0.001 0.29 Air temp: 0.75 ~~ Leaf fall: 0.66

oe Chlorophyll:

peirtemp: 0.65. 0.95

Precipitation: | Leaves/shoot:

0.86 0.47

Soil moisture:

—0.26

ea - weet F=2.5; df=2, ze

Can. variate 3 0.31 47: P=0,09 Ce ge Fe

B. Plant versus psyllid’ variables

: F=1,5; df=6, Trap catch:

Can. variate | 0.36 94: P=0.18 0.75 Leaf fall: -0.78 “0.28

neat aactad Beat trays: 0.93

Leaves/shoot:

0.70

ee Fel 2, ge, wd

Can. variate 2 0.22 48: P=0.31 0.25 mn

C. Psyllid versus weather variables

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 So

F=3.2; df=8, ;

Can. variate | 0.59 100: mate 0.96 Air temp: —0.76 Trap catch: 0.95

Beat trays:

Soil temp: —0.81 “0.18

Precipitation:

0.92

Soil moisture:

0.91

: F=0.4; df=3,

Can. variate 2 0.15 51: P-0.77 0.04 ie Ra:

1Plant variables included leaf drop (no. leaves per tray), chlorophyll content (mg/m2), and leaf density

(no. leaves per cm of shoot)

2Weather variables included avg. daily air temperature, avg. daily soil temperature, soil moisture, and

weekly precipitation.

3Psyllid variables included no. of psyllids collected from matrimony vine and no. of psyllids captured on

traps.

4Larger values indicate greater correlations between variables, and opposite signs indicate inverse

relationships.

———— Avg. Air Temperature (min/max shaded)

~ + «+ Avg. Soil Temperature: «« « Avg. Soil Moisture

| Veekly precipitation

Mean Daily Soll Moisture (%)

Weekly Precipitation (crm)

Mean Daily Air/Soil Termperature (C)

@ Leavesstray

iia O Leaves/shoot

4 Chlorophyll

60 te i yg 1000

Mean No. Leaves per Shoot or Tray

o ¥ »

Mean Chioraphyll Content (maim?)

mH OH S&S &

Beat Sheet Captures

®~s Trap Catch

Mean No. Psyllids per

Cumulative

Trap Catch

5 Beat Sheet Samples

Figure 4. Relationships among weather variables (A), matrimony vine phenology

(B), and occurrence and dispersal of Bactericera cockerelli (C) on experimental

matrimony vine stands located at the USDA experimental farm in Moxee, WA in

2017.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 36

Soil temperature, air temperature, and soil moisture are related (precipitation

declines during the hottest part of summer), making it difficult to discern which

weather variables are most responsible for eliciting changes in seasonal

phenology of matrimony vine. In addition, the difference between wet and dry

soil weights served only as an indirect estimation for soil moisture under stands

of matrimony vine. We therefore performed a controlled greenhouse experiment

to assess the effects of soil moisture on matrimony vine phenology while keeping

air temperature relatively constant. Analysis of soil moisture indicated a

significant week by treatment (watered versus dry) interaction (F=9.6; df=6, 34;

P<0.001). As expected, soil moisture content dropped in the dry treatment

between weeks 2 and 5 when plants were not watered (Fig. 5A). Analysis of leaf

density also indicated a week by treatment interaction (F=2.5; df=6, 31; P=0.04;

Fig. 5B). Statistical differences were not observed between treatments during the

dry weeks, but leaf density increased sharply during week 5 on plants that were

previously deprived of water (Fig. 5B). There was also a week by treatment

interaction for chlorophyll content (F=5.8; df=6, 42; P<0.001). Statistically

significant differences in chlorophyll content were observed in weeks 4 through

7 (Fig. 5C). The lack of water was associated with a decrease in chlorophyll

content, but chlorophyll increased sharply when watering resumed on week 5

(Fig. 5C). In fact, chlorophyll content was significantly higher during weeks 5

through 7 in previously water-stressed plants compared with plants that received

weekly watering (Fig. 5C). These results demonstrate that matrimony vine

rapidly recovers from water deprivation with a hyper-response that leads to

higher leaf density and chlorophyll content when compared with unstressed

plants. Based on this greenhouse study, we conclude that soil moisture likely has

a larger role in eliciting changes in matrimony vine phenology — and therefore

dispersal of B. cockerelli — than air or soil temperature.

Results of our study demonstrate that the introduced matrimony vine in the

PNW is ecologically similar to native Lycium species that occur in the

southwestern United States. Like native species of Lycium (Hanley and Brady

1977; Ackerman ef al. 1980), the introduced matrimony vine undergoes a

summer dormancy triggered by low soil moisture and rapidly recovers after even

modest rainfall in autumn. This trait presumably allows these plants to survive in

the arid regions of inland PNW. Native Lycium species are an important seasonal

host plant for B. cockerelli in the southwest, where psyllid populations peak on

these plants just preceding the plants summer dormancy (Wallis 1955). This was

also the trend we observed on matrimony vine in WA, where psyllid populations

peaked prior to or during summer leaf fall and were undetectable during summer

dormancy. Results of our study confirm that matrimony vine serves as a host for

B. cockerelli during the spring and autumn when annual host plants are not

available (Thinakaran et al. 2017) and demonstrate that summer dormancy of

matrimony vine triggers the dispersal of B. cockerelli to new host plants, which

likely includes potato. By documenting the relationship among abiotic factors,

matrimony vine phenology, and B. cockerelli dispersal, this report should aid in

developing prediction tools with which to forecast B. cockerelli pressure based

upon psyllid populations occurring on matrimony vine in early spring (Cooper ef

al. 2019c).

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 37

@ Watered

0.8 a V Unwatered

4 a

7 ~~ ie al

Soil Moisture

Wk 1? Wik 2! 'Wk 3' 'Wk 4! 'Wk 5! 'Wk 6! ‘Wk 7’

A Leaves per Shoot

Wk 1 'Wk 2! 'Wk 3° 'Wk 4! 'Wk 5' 'WkK 6! 'Wk 7!

A Chorgphyll Content (mg/m?)

- 1500 | :

Wik 1! 'Wk 2' Wk 3 Wi 4! 'Wk 5! 'Wk 6° ‘Wk 7'

Figure 5. Soil moisture (A), change in leaf density (B), and change in chlorophyll

content (C) in a simulated drought study performed in a greenhouse. The dotted lines

in B and C indicate the starting variable; values below the dotted lines indicate

decreases in leaf density or chlorophyll content while and values above the dotted

lines indicate increases in these measurements. The shaded areas denote the standard

errors of the mean.

ACKNOWLEDGEMENTS

Pauline Anderson, Heather Headrick, Millie Heidt, Sara Shellenberger, and

Jerome Lael provided technical assistance. Funding was provided by the

Washington State Department of Agricultural Specialty Crop Block Grant project

#K1761, Northwest Potato Research Consortium, Washington State Commission

for Pesticide Registration, and from the USDA-NIFA-SCRI Project

#2015-51181-24292.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 38

DISCLAIMER

Mention of trade names or commercial products in this article is solely for the

purpose of providing specific information and does not imply recommendation

or endorsement by the United States Department of Agriculture. USDA is an

equal opportunity provider and employer.

REFERENCES

Ackerman, T.L., Romney, E.M., Wallace, A., and Kinnear, J.E. 1980. Phenology of

desert shrubs in southern Nye County, Nevada. Great Basin Naturalist Memoirs,

4: 4-23.

Chiang-Cabrera, Fernando. 1981. A taxonomic study of the North America species of

Lycium (Solanaceae). Ph.D (thesis), University of Texas. Austin, TX.

Cooper, W.R., Horton, D.R., Miliczky, E., Wohleb, C.H., and Waters, T.D. 2019a.

The weed ‘link in zebra chip epidemiology: Suitability of non-crop Solanaceae

and Convolvulaceae to potato psyllid and “Candidatus Liberibacter

solanacearum’. American Journal of Potato Research, 96: 262—271.

Cooper, W R., Horton, D.R., Wildung, M.R., Jensen, A.S., Thinakaran, J., Rendon,

D., Nottingham, L.B., Beers, E.H., Wohleb, C.H., Hall, D.G., and Stelinski, L.L.

2019b. Host and non-host “whistle-stops” for psyllids: Molecular gut content

analysis by high-throughput sequencings reveals landscape-level movements of

Psylloidea (Hemiptera). Environmental Entomology, 48: 554-566.

Cooper, W.R., Horton, D.R., Miliczky, E., Wohleb, C.H., and Waters, T.D. 2019c.

The weed link in zebra chip epidemiology. Potato Progress, 19(4): 1-7.

Crawford, D.L. 1914. A monograph of the jumping plant-lice or Psyllidae of the new

world. US Natural History Museum Bulletin 85, 18 pp.

Essig, E.O. 1917. The tomato and laurel psyllids. Journal of Economic Entomology,

10: 433-444.

Gitelson, A.A., Buschmann, C., and Lichtenthaler, H.K. 1999. The chlorophyll

fluorescence ration F735/F700 as an accurate measure of chlorophyll content in

plants. Remote Sensing of Environment, 69: 296-302.

Hanley, T.A. and Brady, W.W. 1977. Seasonal fluctuations in nutrient content of feral

burro forages, lower Colorado River Valley, Arizona. Journal of Range

Management, 30: 370-373.

Hansen, A.K., Trumble, J.T., Stouthamer R., Paine, T.D. 2008. New Huanglongbing

(HLB) Candidatus species, "Ca. Liberibacter psyllarous" found to infect tomato

and potato is vectored by the psyllid Bactericera cockerelli. Applied

Environmental Microbiology, 73: 7531-7535.

Hitchcock, C.L. 1932. A monographic study of the genus Lycium of the western

hemisphere. Annals of the Missouri Botanical Garden, 19: 179-374.

Horton, D.R., Burts, E.C., Unruh, T.R., Krysan, J.L., Coop, L.B., and Croft, B.A.

1993. Intraorchard changes in distribution of winterform pear psylla (Homoptera;

Psyllidae) associated with leaf fall in pear. Annals of the Entomological Society

of America, 86: 599608.

Horton, D.R., Burts, E.C., Unruh, T.R., Krysan, J.L., Coop, L.B., and Croft, B.A.

1994. Phenology of fall dispersal by winterform pear psylla (Homoptera:

Psyllidae) in relation to leaf fall and weather. Canadian Entomologist, 126:

111-120.

Horton, D.R., Thinakaran, T., Cooper, W.R., Munyaneza, J.E., Wohleb, C.H., Waters,

T.D., Snyder, W.E., Fu, Z., Crowder, D.W., and Jensen, A.S. 2016. Matrimony

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 39

vine and potato psyllid in the Pacific Northwest: A worrisome marriage? Potato

Progress, XVI: 14. 12 pp.

Kaur, N., Cooper, W.R., Duringer, J.M., Badillo-Vargas, I.E., Esparza-Diaz, G.,

Rashed, A., and Horton, D.R. 2018. Survival and development of potato psyllid

(Hemiptera: Triozidae) on Convolvulaceae: Effects of a plant—fungus symbiosis

(Periglandula). PLOS One, 13: e0201506.

Knowlton, G.F. and Thomas W.L. 1934. Host plants of the potato psyllid. Journal of

Economic Entomology, 27: 547.

Levin, R.A. and Miller, J.S. 2005. Relationships within tribe Lycieae (Solanaceae):

Paraphyly of Lycium and multiple origins of gender dimorphism. American

Journal of Botany, 92: 2044-2053.

Liefting, L.W., Weir, B.S., Pennycook, S.R., and Clover, G.R.G. 2009. ‘Candidatus

Liberibacter solanacearum’, associated with plants in the family Solanaceae.

International Journal of Systematic and Evolutionary Microbiology, 59: 2274—

2276.

Munyaneza, J.E. 2012. Zebra chip disease of potato: Biology, epidemiology, and

management. American Journal of Potato Research, 89: 329-350.

Murphy, A.F., Rondon, S.I., and Jensen, A.S. 2013. First report of potato psyllids,

Bactericera cockerelli, overwintering in the Pacific northwest. American Journal

of Potato Research, 90: 294—296.

Pletsch, D.J. 1947. The potato psyllid Paratrioza cockerelli (Sulc), its biology and

control. Montana Agricultural Experiment Station Bulletin, 446: 95.

Romney, V.E. 1939. Breeding areas of the tomato psyllid, Paratrioza cockerelli

(Sulc). Proceedings of the Utah Academy of Science, 12: 233-239.

SAS Institute Inc. 2013. SAS release 9.3 ed. SAS Institute, Cary, NC.

Swisher, K.D., Munyaneza, J.E., and Crosslin, J.M. 2012. High resolution melting

analysis of the cytochrome oxidase I gene identifies three haplotypes of the

potato psyllid in the United States. Environmental Entomology, 41: 1019-1028.

Thinakaran, J., Horton, D.R., Cooper, W.R., Jensen, A.S., Wohleb, C.H., Dahan, J.,

Mustafa, T., Karasev, A.V., and Munyaneza, J.E. 2017. Association of potato

psyllid (Bactericera cockerelli; Hemiptera: Triozidae) with Lycium spp.

(Solanaceae) in potato growing regions of Washington, Idaho, and Oregon.

American Journal of Potato Research, 94: 490-499.

Wallis, R.L. 1955. Ecological studies on the potato psyllid as a pest of potatoes.

USDA Technical Bulletin No. 1107.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 40

Assessments of Rhagoletis pomonelila

(Diptera: Tephritidae) infestation of

temperate, tropical, and subtropical fruit

in the field and laboratory in Washington

State, U.S.

W.L YEE AND R. 3B GCOUGHN OU RFR?

ABSTRACT

To understand the likelihood of any risk of apple maggot, Rhagoletis

pomonella (Walsh) (Diptera: Tephritidae), to domestic and foreign fruit export

markets, knowledge of its host plant use is needed. Here, assessments of R.

pomonella infestation of temperate, tropical, and subtropical fruit were made

in the field and laboratory in Washington State, U.S. In field surveys in 2010—

2017 in central Washington, 6.7% of Crataegus douglasii and 6.1% of feral

Malus domestica trees (both temperate plants) in fly-managed (insecticide-

treated) sites were infested by larvae. In unmanaged sites, 54.1% of C.

douglasii and 16.3% of feral 4. domestica tree samples were infested. In field

surveys of 36 types of temperate fruit in 2015-2018 in southwestern

Washington, new host records for R. pomonella were one species and three

hybrids of Crataegus, as well as Prunus domestica subsp. syriaca — all of

which produced adult flies. In addition, Prunus avium was a new host record

for Washington State, producing one adult fly. Prunus armeniaca x Prunus

salicina and Vitis vinifera exposed to flies in the laboratory produced adult

flies. Of 37 types of tropical and subtropical fruit hung in fly-infested M.

domestica trees in southwestern Washington, only Mangifera indica produced

puparia. Out of nine tropical and subtropical fruit types in laboratory tests,

Musa acuminata x balbisiana produced puparia but no adult flies. Results

provide a basis for further research and hypotheses concerning host use by R.

pomonella and its potential impact on protecting both U.S. and tropical and

subtropical fruit markets.

INTRODUCTION

The apple maggot, Rhagoletis pomonella (Walsh) (Diptera: Tephritidae), is a

quarantine pest of cultivated apple (Malus domestica) (Rosaceae) in western

North America whose ancestral hosts are hawthorns, Crataegus spp. (Rosaceae)

(Bush 1966). Native to eastern North America and Mexico, R. pomonella in

western North America was first detected in VM. domestica in 1979 in Portland,

Oregon, in the Pacific Northwest (PNW) of the U.S. (AliNiazee and Penrose

1981). It is now found throughout the PNW west of the Cascade Mountain

(Corresponding author: United States Department of Agriculture-Agricultural Research Service,

Temperate Tree Fruit & Vegetable Research Unit, 5230 Konnowac Pass Road, Wapato, WA, 98951;

wee. yee@usda.gov

2Washington State University Clark County Extension, 1919 NE 78" Street, Vancouver, WA, 9866

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 4]

Range, a relatively humid region with little commercial tree fruit production.

However, arid or semi-arid regions in the PNW east of the range in central

Washington State (Washington), Oregon, and Idaho, as well as in British

Columbia, Canada (Canadian Food Inspection Agency 2016), where commercial

apples are grown are mostly free of the fly. Preventing fly movement from

infested to pest-free areas across the PNW and preventing increases in fly

numbers within quarantine areas are high priorities for U.S. state departments of

agriculture. To date, there have been no reports of commercially grown apples

from the PNW infested by R. pomonella larvae (Washington State Department of

Agriculture 2018).

Washington is the PNW’s biggest apple producer. It exports about 30% of its

crop, which is valued at ~US$2.26 billion a year (NASS 2017), to overseas

markets (Anonymous 2018). About 20 of 60 export markets have requirements

or restrictions for apple import due to R. pomonella. These markets include

China, Japan, South Korea, India, South Africa, Indonesia, Australia, Brazil,

Chile, and Mexico. Some of these countries include regions with humid,

subtropical climates. Although R. pomonella is a temperate species, ecological

niche models indicate there are marginal to favorable habitats for the fly in such

climates (between 15°N and 30°N latitude; e.g., southern China, northern Laos,

Vietnam, and the Philippines) (Kumar ef a/. 2016). Furthermore, low

temperatures are not a requirement for adult emergence (AliNiazee 1988). Thus,

R. pomonella might become established if it were introduced into a subtropical

country.

To understand the likelihood of any risk of R. pomonella to domestic and

foreign fruit export markets, knowledge of its host-plant use is needed.

Rhagoletis pomonella is known to develop in at least 60 plant taxa (Yee and

Norrbom 2017), therefore movement of these taxa needs to be restricted.

However, more knowledge of its host-plant use could help to further reduce the

perceived risk. Areas in which greater knowledge is needed include (1)

frequencies of infestations of Crataegus douglasii (black hawthorn) and MM.

domestica trees, (2) additional host plants infested by the fly, and (3) fly

infestation of tropical and subtropical fruit.

With respect to (1), C. douglasii and feral M. domestica occur spottily around

commercial apple orchards in central Washington, where <10% and <1%,

respectively, of trees were found to be infested by R. pomonella in 2004—2006

(Yee 2008). This suggests frequencies of infestations of both species are low, but

that C. douglasii — which is native to the region — is more frequently infested and

thus a greater source of flies. However, reassessments of infestation frequencies

of the two plants over time may show that the frequencies change and thus can

affect the trees’ importance in fly control. In addition, Washington State

Department of Agriculture (WSDA) and county pest boards detect R. pomonella

in C. douglasii and feral M. domestica trees at sites near apple orchards using

traps and then treat fly-positive trees with insecticides, but these entities do not

control flies in C. douglasii and feral M. domestica trees at sites farther from

commercial orchards. Whether differences in frequencies and patterns of larval

infestations of C. douglasii versus M. domestica trees in fly-managed

(insecticide-treated) and unmanaged sites occur has yet to be determined.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 42

With respect to (2), there may be host plants of R. pomonella that are not yet

identified that could increase the risk of the fly spreading should their fruit be

moved from infested to uninfested areas within the PNW. Crataegus (hawthorn)

species would be likely candidates, as 30 of the 60 plant taxa that are hosts for R.

pomonella belong to this genus (Yee and Norrbom 2017).

With respect to (3), R. pomonella attacks Mangifera indica (mango)

(Anacardiaceae) and Carica papaya (papaya) (Caricaceae) hung in M. domestica

trees in the field and in the laboratory and are suitable hosts that produced adult

R. pomonella (Yee and Goughnour 2017). However, suitability of other tropical

and subtropical fruit has not yet been determined. No commercial tropical and

subtropical fruit belong to the Rosaceae, and because they differ from Crataegus

and Malus species in many respects, they may not be attractive to R. pomonella.

If that is the case, there may be no (zero) or minimal threat of R. pomonella

attacking most tropical or subtropical fruit in subtropical environments.

Here, our objective was to assess R. pomonella infestation of temperate,

tropical, and subtropical fruit through surveys and tests in the field and

laboratory in Washington. Specific goals were to determine (1) R. pomonella

infestation frequencies of C. douglasii versus feral M. domestica in central

Washington, both in fly-managed and unmanaged sites; (2) whether there are

unrecorded host plants of the fly in southwestern Washington; and (3) whether

various tropical and subtropical fruit are suitable as developmental hosts for R.

pomonella.

MATERIALS AND METHODS

Field surveys. In July to September 2010—2017 in central Washington, field

surveys were conducted of infestation by R. pomonella of C. douglasii versus M.

domestica. Fruit from C. douglasii and feral M. domestica were collected at 10

sympatric sites (Appendix 1). Within each site, trees of the two species were

~10—5,000 m apart, with numbers of each varying widely across sites. Sites were

in sagebrush, bunchgrass, or ponderosa pine ecosystems (Lyons and Merilees

1995). The three fly-managed sites — in arid sagebrush and bunchgrass habitats —

had an active fly detection and control program using insecticides run by WSDA

and county pest boards. The seven unmanaged sites — mostly in less arid

ponderosa pine habitat — had no history of fly control or had no control for up to

20 years before surveys. Each site was sampled for 1 to 3 years from 2010-2015.

Exceptions were Klickitat, which was sampled only in 2010 and 2012, and Nile,

where an additional C. douglasii sample took place in 2017. Both tree species

occurred along creeks, along roadsides beside ditches, in meadows, along trails

in wooded areas away from creeks, and in pastures. Fruit from both species were

collected when ripe: C. douglasii from mid-July to late August, and M.

domestica from mid-August to early October. About 800 C. douglasii fruit were

picked per tree, depending on fruit load. About 50 M. domestica fruit were

collected from beneath each tree about 1 week after they had dropped.

In July to November 2015-2018 in southwestern Washington, field surveys

were conducted of 36 types of temperate fruit — mostly non-native species —

including hybrids, subspecies, and varieties (see Table 1 for a list of temperate

fruit surveyed or tested for infestation by R. pomonella). The focus was on fruit

of unrecorded hosts, but fruit of known hosts were also collected for comparison.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 43

Fruit types collected included most of the accessible fruit present at the 10 sites

in the coast forest ecosystem (Appendix 1). Collections were made in parks,

along roadsides, in urban areas, and in demonstration tree plantings. All sites

were unmanaged, except for the Vancouver Orchard. Malus domestica fruit were

collected from the ground beneath trees, whereas other ripe fruit were collected

off trees or bushes, except the fruit of Prunus persica, which were collected off

the ground at the Vancouver Orchard. To gain additional information, the colour,

diameters, and weights of at least 20 individual ripe fruit of newly identified

Crataegus hosts were recorded. Fruit were collected for measurements in

November 2018 from the same trees that had produced fruit positive for R.

pomonella larvae in previous years.

Table 1. Temperate fruit sampled or tested for Rhagoletis pomonella infestation in

field surveys or in the laboratory in southwestern Washington State, U.S.

Common name Scientific name Family

Apple? Malus domestica Borkhausen Rosaceae

Dolgo Crabapple Malus x 'Dolgo' Rosaceae

Fruiting Crabapple Malus sp., unknown cultivar Rosaceae

Harvest Gold Flowering Malus x ‘Harvest Gold' Rosaceae

Crabapple

Black Hawthorn Crataegus douglasii Lindley Rosaceae

Red Sun Chinese Crataegus pinnatifida Bunge Rosaceae

Hawthorn

Autumn Glory Hawthorn Crataegus laevigata (Poiret) de Candolle x Rosaceae

Crataegus mexicana Mocino & Sessé ex de

Candolle

Lavalle Hawthorn Crataegus x lavalleei Hérincq ex Lavallée Rosaceae

(hybrid of C. mexicana x probably C.

calpodendron (Ehrhart) Medikus

Toba Hawthorn Crataegus x mordensis Boom (hybrid of Rosaceae

Crataegus laevigata (Poiret) de Candolle x

Crataegus succulenta Schrader ex Link)

Cockspur Hawthorn Crataegus crus-galli Linnaeus Rosaceae

Washington Hawthorn Crataegus phaenopyrum Borkhausen Rosaceae

Sweet Cherry Prunus avium (Linnaeus) Linnaeus Rosaceae

Tart Cherry Prunus cerasus Linnaeus Rosaceae

Mirabelle Plum Prunus domestica Linnaeus subsp. syriaca Rosaceae

Italian Plum Prunus domestica Linnaeus Rosaceae

French Petite Plum Prunus domestica Linnaeus Rosaceae

Friar Black Plum Prunus domestica Linnaeus Rosaceae

Cherry Plum Prunus cerasifera Ehrhart Rosaceae

Japanese Plum Prunus salicina Lindley Rosaceae

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 44

Choke Cherry Prunus virginiana Linnaeus Rosaceae

Peach (White, Mexican, Prunus persica (Linnaeus) Batsch Rosaceae

California)?

Dapple Dandy Pluot? ~30% apricot (Prunus armeniaca Linnaeus); Rosaceae

~70% plum (Prunus salicina)

European Quince Cydonia oblonga Miller Rosaceae

Pineapple Quince Cydonia oblonga Miller Rosaceae

Kosui Asian Pear Pyrus pyrifolia (Burman) Nakai Rosaceae

20th Century Asian Pear Pyrus pyrifolia (Burman) Nakai Rosaceae

Bartlett Pear Pyrus communis Linnaeus Rosaceae

Winter Pear Pyrus communis Linnaeus Rosaceae.

Bosc Pear Pyrus communis Linnaeus Rosaceae

Cotoneaster. | Cotoneaster sp. Rosaceae

Western Mountain Ash Sorbus scopulina Greene Rosaceae

Red Currant Ribes rubrum Linnaeus Grossulariaceae

White Currant Ribes rubrum Linnaeus (albino of red currant) Grossulariaceae

Goumi Berry Elaeagnus multiflora Thunberg Elaeagnaceae

Twinberry Honeysuckle Lonicera involucrata (Richardson) Banks ex _Caprifoliaceae

Sprengel

Highbush Blueberry Vaccinium corymbosum Linnaeus Ericaceae

Red Globe Grape? Vitis vinifera Linnaeus Vitaceae

Jiro Fuyu Persimmon Diospyros kaki Linnaeus the Younger Ebenaceae

4used in laboratory tests. The only plants native to Washington are C. douglasii, P. virginiana, S.

scopulina, R. rubrum, and L. involucrata; the rest originated from eastern North America, Europe, or

Asia.

For both central and southwestern Washington surveys, precise fruit counts

were made after collections. Fruit from individual trees were held in separate

tubs at ~15—27 °C outdoors for 2—3 months for larval emergence. Tubs were

checked for puparia every 1-3 days. Puparia were counted and identified using

puparial traits (Yee and Goughnour 2016). In addition, for the southwestern

Washington surveys, puparia from any fruit not previously recorded as a host

were placed in cups with moist soil, held at 3—4 °C for ~4. months, and then at

21-23 °C for adult fly emergence. Adult flies were identified as R. pomonella

using morphological characters (Bush 1966). Voucher specimens of reared adult

flies from select hosts are held at the USDA-ARS Temperate Tree Fruit &

Vegetable Research Unit in Wapato, Washington.

Infestation of tropical and subtropical fruit in the field, southwestern

Washington. Field tests of infestation by R. pomonella of tropical and

subtropical fruit were conducted in 2015, 2016, 2017, and 2018, in southwestern

Washington at T.G., Devine, and Woodland sites (three of the same sites that

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 45

were included for temperate fruit surveys; Appendix 1). These sites had high R.

pomonella populations; use of these sites was intended to increase the chances

fruit would be attacked.

A total of 37 types of tropical and subtropical fruit were hung 3—4 m above

ground, according to the methods described in Yee and Goughnour (2017), in M.

domestica trees infested by R. pomonella (see Table 2 for a list of tested tropical

or subtropical fruit types, including cultivars of the same species, if known;

Psidium guajava [guava apple and giant guava] were tested in the laboratory but

not the field). Malus domestica (Gala variety) fruit were hung in the same trees

as tropical and subtropical fruit to serve as positive controls. At each site, there

were 14—30 M. domestica trees (3-9 m tall and 3—8 m wide), each with 20—60

test fruit at any one time. Fruit were hung >1 m apart.

Fruit were tested in July and August, when adult flies were most abundant.

Over the four years, 5—189 total fruit per type (1,708 total fruit) were exposed to

flies. As M. indica, C. papaya, and Citrus spp. are particularly important

commercial tropical or subtropical fruit, emphasis was placed on testing these

species. After 2~3 week exposures, fruit were removed, placed in tubs, and

monitored for larval emergence.

Table 2. Tropical and subtropical fruit tested for Rhagoletis pomonella infestation in

the field or laboratory in southwestern Washington State, U.S.

Common name Scientific name Family

Yellow Mango (Ataulfo) Mangifera indica Linnaeus Anacardiaceae

Red Mango# Mangifera indica Linnaeus Anacardiaceae

Carambola> Averrhoa carambola Linnaeus Oxalidaceae

Blue Java Banana Musa acuminata Colla x Musaceae

balbisiana (ABB Group) ‘Blue

Java'

Red Banana Musa acuminata Colla (AAA Musaceae

Group)

Cavendish Banana Musa acuminata Colla Cavendish Musaceae

subgroup of the AAA Group

Pineapple Ananas comosus (Linnaeus) Bromeliaceae

Merrill

Pink Pineapple Ananas comosus (Linnaeus) Bromeliaceae

Merrill

Passion Fruit Passiflora edulis Sims Passifloraceae

Cherimoya> Annona cherimola Miller Annonaceae

Hawaiian Papaya» Carica papaya Linnaeus Caricaceae

Mamey Sapote Pouteria sapota (Jacquin) Harold Sapotaceae

Emery Moore & Stearn

Pineapple Guava Acca sellowiana (Otto Berg) Myrtaceae

Burret

Mexican Guava Psidium guajava Linnaeus Myrtaceae

Pink Guava

Psidium guajava Linnaeus

Myrtaceae

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019

Guava Apple®

Thai Guava

Giant Guava>

Horned Melon

Mangosteen?

Hass Avocado

Red Dragon Fruit

Yellow Dragon Fruit

Orange

Blood Orarige>

Navel Orange

Valencia Orange

Daisy Mandarin

Mandarin Orange>

Tangerine

Murcott Orange

Ortanique Tangerine

Clementine Orange

Grapefruit

Lemon

Lime

Key Lime

Fuzzy Kiwifruit

Smooth Skin Kiwifruit

Psidium guajava Linnaeus

Cucumis metuliferus Ernst Meyer

Garcinia mangostana Linnaeus

Persea americana Miller

Hylocereus costaricensis (Frederic

Albert Constantin Weber) Britton

& Rose

Hylocereus megalanthus

(Karl Schumann ex Vaupel) Ralf

Bauer

Citrus < sinensis

Citrus x sinensis pummelo x

mandarin orange

Citrus reticulata Blanco

Citrus tangerina Tanaka (hybrids)

Mandarin x sweet orange hybrid

Citrus reticulata x C. sinensis

hybrid

Citrus x clementina Mandarin

orange x sweet orange

Citrus < paradisi Macfadyen

Citrus limon (Linnaeus) Osbeck

Hybrid of Citrus spp.

Citrus aurantifolia Swingle

Actinidia deliciosa (Auguste

Chevalier) Chou-Fen Liang &

Allan Ross Ferguson

Actinidia chinensis (Golden)

Myrtaceae

Cucurbitaceae

Clusiaceae

Lauraceae

Cactaceae

Rutaceae

Rutaceae ~

Rutaceae

Actinidiaceae

‘Includes Tommy Atkins, Kent, and Palmer varieties.

bUsed in laboratory tests.

A caveat to results from tropical and subtropical fruit tests in the field and in

the laboratory tests of temperate, tropical, and subtropical fruit described in the

next section of this paper is that the fruit used were obtained from markets rather

than from the field. Thus, there is a possibility insecticides in fruit killed the

larvae and, therefore, no infestation was detected. However, in laboratory

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 47

studies, non-organic M. indica and M. domestica exposed to R. pomonella adults

consistently produce larvae (Yee and Goughnour 2017; W.L.Y., unpublished).

Thus, levels of any insecticides in market fruit must have been at sufficiently low

levels as to be harmless to at least some larvae. Also, no insecticide-based

method is accepted for disinfesting M. domestica of R. pomonella larvae for

export; only cold treatment is accepted as a method for killing larvae (Canadian

Food Inspection Agency 2017). Thus, it is unlikely any residues would have

killed larvae. Finally, there is a possibility that adult flies were prevented from

Ovipositing into market fruit due to insecticide residues on the fruit surfaces.

However, the fact that larvae were produced from market fruit (see results) and

that no research has shown insecticide residues on market fruit are effective

oviposition deterrents reduce this possibility. Nevertheless, the use of market

fruit needs to be kept in mind when interpreting results.

Laboratory tests. Laboratory tests of infestation of temperate fruit by R.

pomonella were conducted in March to May 2017. Fruit were obtained from

markets in Vancouver, Washington. Test flies were 14 d post-emergence in age,

and were reared from larvae collected from naturally infested M. domestica in

2016 in southwestern Washington. Six types of temperate fruit, including M.

domestica (Gala variety) as a positive control and three types of peaches (Table

1; a superscripts), were exposed to two to five females and three to five males

per 1.9-litre (16.2 cm inner diameter x 10.5 cm inner height) paper container.

One individual fruit or two fruits (Vitis vinifera) were exposed to flies at any one

time inside a container with dry food (80% sucrose:20% yeast extract, wt:wt) on

a paper strip and water at 22-25 °C, 40-50% relative humidity, and 16:8 L:D.

Two or three successive one- or two-fruit exposures took place, each for 2 weeks.

Fruit were removed after a 2-week period and held for larval emergence. Tests

were replicated 10 or 15 times for each fruit type.

Laboratory tests of infestation of tropical and subtropical fruit were similarly

conducted from December 2016 to June 2017. All fruit were obtained from

markets in Vancouver or Yakima, Washington. Nine types of fruit including apple

(Table 2; b superscripts) were exposed to flies following methods described for

temperate fruit in tests above. There were three to 17 replicates per fruit type.

Statistics. Frequencies of C. douglasii and feral M. domestica trees infested

with R. pomonella in central Washington surveys within fly-managed and

unmanaged sites were compared using a test of two proportions (Zar 1999).

Within fly-managed and unmanaged site categories, the total number of infested

C. douglasii or M. domestica trees were divided by the total number of C.

douglasii or M. domestica trees sampled. The frequencies for the two tree species

within fly-managed and unmanaged tree categories were compared. Within tree

species, the frequencies of trees infested in fly-managed versus unmanaged sites

were similarly calculated. In addition, to summarize data and provide a

descriptive measure of variability (rather than raw data from each site per year),

mean frequencies of infestation + SE were calculated. Frequencies of infested

trees across years within a site were averaged, and means of the frequencies

across sites — with each site serving as an observation — were generated. For

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 48

laboratory tests, tests of two or more proportions or Fisher’s exact test were

conducted to compare percentages of larvae-positive fruit types.

RESULTS

Infestation of C. douglasii and M. domestica, central Washington field

surveys. In central Washington, both C. douglasii and M. domestica were

infested in some sites in all survey years. In the three fly-managed (insecticide-

treated) sites, 6.7% of C. douglasii and 6.1% of M. domestica were infested (y? =

0.01; P = 0.90). In fly-managed sites, the mean frequency of C. douglasii trees

that were infested was numerically greater than the mean frequency of M.

domestica (Table 3) trees infested, due to one site (Ellensburg) where 40.0% of

trees were infested. In the seven unmanaged sites, 54.1% of C. douglasii versus

16.3% of M. domestica samples were infested (v2 = 63.77; P < 0.0001) (because

fruit from some of the same survey trees were collected in more than one year,

numbers of fruit samples were slightly greater than those of trees). In unmanaged

sites, the mean frequency of infested C. douglasii trees was numerically greater

than the mean frequency of infested M. domestica trees (Table 3). The frequency

of C. douglasii samples infested in fly-managed sites was lower than in

unmanaged sites (vy? = 37.71; P < 0.0001), but for 4. domestica, the difference

was not significant (P = 0.07).

Table 3. Surveys of infestation by Rhagoletis pomonella in Crataegus douglasii

(black hawthorn) and feral Malus domestica (apple) trees at three fly-managed

(insecticide-treated) and seven fly-unmanaged sites in central Washington State,

U.S., 2013-2017.

Nos. Mean frequencies

Se ee Nos. trees = Nos. Nos. i eae eins aa

Host plant Years sampled sampled fruit puparia oar of - ene

Crataegus 2013, 2014 60 51,815 7 0.000096 13.3 13.3

douglasii

Malus 2013, 2014 a3 3,405 6 0.001762 5.0+2.9

Fly- snaged sit s(n=7 dis ae

Crataegus 2010, 2012, 139a 224,249 5,515 0.024593 36.7 + 13.6

douglasii 2013, 2015,

2017

Malus 2010, 2012, 1238 13,668 366 0.026778 ID Si&S.9:

domestica ___ 2013, 2015

aSome trees were sampled multiple times across years. Frequencies of infested trees across years within a

site were averaged, and means of these frequencies across sites (each site as an observation) were

generated,

Infestation of temperate fruit, southwestern Washington field surveys. In

southwestern Washington, Malus x ‘Dolgo’ (Dolgo crabapple), Malus sp.

(unknown cultivar; fruiting crabapple), and C. douglasii — all known hosts —

produced R. pomonella puparia (Table 4).

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 49

Table 4. Infestation of temperate fruit in field surveys by Rhagoletis pomonella in

southwestern Washington State, U.S., 2015-2018.

ray ; Fruit collection Nos. Nos. Nos. Puparia/

Fruit Sites date plants fruit puparia fruit

July-Sep 2015;

Malus domestica —‘Heying, July 2017. "4° 39) a .s816

nixed VarCHe®) |: Nondignd. “SSNS: 1, 100 S0Re” arsae ects

Malus sp. H.O. 18 Aug, Nov2015 4 640 104 0.1625

(Dolgo, others)

Crataegus TG, Jul 2015; Aug 2017s 1 853 71 0.0832

douglasii Devine Boosie: Megha 1,716 0.4112

Crataegus HQ. Sep, Nov 2015 1 644 7 0.0109

pinnatifida

Crataegus H.O. Sep 2015 l 1,155 32 0.0277

laevigata x

Crataegus

mexicana :

C. mexicana x C. HO. Sep, Nov 2015 l 645 69 0.1070

calpodendron

Crataegus x H. O. Sep 2015 ] 1,031 8 0.0078

mordensis

Crataegus crus- H.O., T.G: Sep, Oct 2015 3 Oe 5 0.0029

galli

Crataegus Leverage Park Aug 2015; Sep 3 1,005 0 0

phaenopyrum Salmon Creek 2015 0 0

Prunus TG., AO;, July, Aug 2015; 5 1,063 153 0.1439

domestica Devine Aug, Sep 2017

Prunus Pye sn July 2015 ] 356 0 0

cerasifera

Prunus salicina Cherry Grove Aug, Sep 2015; 3 285 0.0175

| Sep 2017

Woodland 2018 I 150 0 0

(town) June 2018 2 498 0 )

Devine June 2018

Prunus cerasus — Cherry Grove July 2017 2-9 4,180 0 0

Devine June 2018 1 209 0 0

Prunus T.G, Sep 2017 2 743 0 0

virginiana

Prunus persica Vancouver July, Aug 2017 20 672 0 0

Orchard

Cydonia Tks. Sep 2015 5 84 0 0

oblonga‘

Pyrus pyrifolias T.G. Sep, Aug 2015 4 107 0 0

Pyrus communis® TG, Sep 2015 5 103 0 0

Cotoneaster sp. H.O Sep 2015 10 2 O67 0 0

Sorbus scopulina Devine Aug 2017 1 785 7 0.0089

Ribes rubrum H.O. July 2017 2 1,841 0 0

Elaeagnus T.G. Sep 2017 5 660 0 0

multiflora

Lonicera Devine July 2017 8 463 0 0

involucrata

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 50

Vaccinium H. Farm, July 2017 12 10,992 0 0

corymbosum Upper Field

Diospyros kaki H.O. Nov 2015; Sep | 113 0 0

2016; Nov 2018

aSite names, coordinates, and elevations are in the appendix 1.

bComprising subsp. syriaca, and French Petite and Friar Black varieties of P. domestica.

¢European and pineapple quince.

dKosui and 20 Century.

eBartlett, Winter, and Bosc.

fRed and albino.

In addition, four Crataegus that had never been reported previously as hosts

were identified. These were C. pinnatifida (Red Sun Chinese hawthorn), C.

laevigata x C. mexicana (Autumn Glory hawthorn), C. Javalleei (Lavalle

hawthorn), and C. x mordensis (Toba hawthorn), the latter three being hybrids.

Adult R. pomonella were reared from all four respective hosts: two males and

one female; six females; one male and three females; and one female. Larvae-

infested fruit were collected on the following dates: C. pinnatifida — 9 and 28

September and 2 November 2015; C. laevigata x C. mexicana — 2 and 28

September 2015; C. lavalleei — 28 September and 2 November 2015; C. x

mordensis — 2 and 28 September 2015. Mean diameter (mm) and weight (g) + se,

respectively, of fresh ripe fruit (n = 20) were: C. pinnatifida — 32.3 + 0.5 and

12.06 + 0.43; C. laevigata x C. mexicana — 19.7 + 0.5 and 3.92 + 0.28; and C.

lavalleei — 16.4 + 0.4 and 2.81 + 0.16. Measurements of C. x mordensis fruit

were not made, but they were ~10 mm in diameter. Crataegus crus-galli

(cockspur hawthorn) — a known host _ was also infested, but C. phaenopyrum

(Washington hawthorn) — not a known host — was not.

Prunus domestica was positive for R. pomonella puparia (Table 4). Of

particular note, P. domestica subsp. syriaca was found to be infested for the first

time on record, and adult flies were reared from it, although fly numbers were

not recorded; infested fruit of this subspecies were collected 15 July 2015 and 19

and 30 July 2017. Infestations by R. pomonella were also detected in Italian,

French Petite, and Friar Black varieties of P. domestica, as well as in P. salicina

(Japanese plum) and S. scopulina (western mountain ash) (Table 4). In addition,

a sample of Royal Anne P. avium (sweet cherry) collected 2 July 2017 was

infested, with one adult female R. pomonella reared from it — the first record of

P. avium being a host of the fly in the field in Washington. Black Republican and

Bing varieties of P. avium in the same grove as Royal Anne P. avium did not

produce puparia, nor did P. cerasus (tart cherry) (Table 4).

Infestation of tropical and subtropical fruit, southwestern Washington.

Results from tropical and subtropical fruit tests at the three sites were combined,

and some fruit varieties are pooled for presentation (Table 5). Malus domestica

fruit that were hung in trees produced R. pomonella puparia, but of the tropical

and subtropical fruit, only M. indica (red and yellow mangoes) produced puparia.

In 2018, all tropical and subtropical fruit from the three sites tested negative —

even at Woodland, where 17,287 R. pomonella puparia collected from 5,008 M.

domestica fruit on the ground indicated high fly pressure at this site.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019

Table 5. Infestation of Malus domestica and tropical and subtropical fruit by

Rhagoletis pomonella when hung in M. domestica trees at three sites with known

populations of the insect in southwestern Washington State, U.S., 20154, 2016, 2017,

and 2018.

Fruit

OS.

Nos. fruit

tested puparia positive (%)

Malus domestica (as control)

Mangifera indica (Yellow, ataulfo)

Mangifera indica (Red)4

Averrhoa carambola

Musa acuminata x balbisiana

Musa acuminata (Red)

Musa acuminata (Cavendish)

Ananas comosus

Ananas comosus (Pink Pineapple)

Passiflora edulis

Annona cherimola

Carica papaya

Pouteria sapota

Acca sellowiana

Psidium guajava (Mexican, Pink, and Thai Guavas)

Cucumis metuliferus

Garcinia mangostana

Persea americana

Hylocereus costaricensis

Hylocereus megalanthus

Citrus * sinensis (Orange)

Citrus < sinensis (Blood Orange)

Citrus < sinensis (Navel Orange)

Citrus x sinensis (Valencia Orange)

Citrus reticulata (Daisy Mandarin)

Citrus reticulata (Mandarin Orange)

Citrus tangerina

Mandarin x sweet orange hybrid (Murcott Orange)

137

116

29 (36.2a)

7 (4.5b)

ett.70)

— Ge £2.34 “So ee Se eS oe eS OM OS ee ee ee el DOSS CS. ee

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 $2

Citrus reticulata x C. sinensis 28 0 0

Citrus x clementina 10 0 0

Citrus * paradisi 28 0 0

Citrus limon 103 0 0

Hybrid of Citrus spp. (Lime) 114 0 0

Citrus aurantifolia 23 0 0

Actinidia deliciosa 30 0 0

Actinidia chinensis 10 0 0

ain 2015, 16 M. indica were hung. For positive fruit only: y2 = 70.46; df= 2; P < 0.0001; percentages.

with same letters are not significantly different (P > 0.05).

Laboratory tests. In the laboratory tests using temperate fruit (Table 6), M.

domestica , P. persica (white peach) and P. armeniaca x P. salicina (Dapple

Dandy pluot) and V. vinifera (red globe grape) produced R. pomonella puparia.

Adult flies were reared from P. armeniaca x P. salicina, but their precise

numbers were not recorded. The percentages of M. domestica and P. armeniaca

x P. salicina that produced puparia did not differ, but both were significantly

greater than the percentage of P. persica that produced puparia (Table 6). One

adult female R. pomonella was reared from the five puparia from V. vinifera.

Table 6. Infestation of temperate fruit by Rhagoletis pomonella inside 1.9-litre

containers in laboratory exposures in Vancouver, Washington State, U.S., 2017.

OS. OS. Jos. fruit ,

Malus domestica 10 48 30 “11°36.7a)

Prunus persica (White) 10 a 30 2 (6.7b)

Prunus persica (Mexican) 10 0 20 0

Prunus persica (California) 10 0 10 0

Prunus armeniaca x P. salicina 10 42 Me 13 (86.7a)

Vitis vinifera 15 5 85 —

«Two to five females and three to five males per replicate container.

bEach with two fruit; number of fruit positive not recorded.

‘For positive fruit only: y?2 = 12.11; df= 2; P = 0.002; percentages with same letters are not significantly

different (P > 0.05).

In laboratory tests using tropical and subtropical fruit (Table 7), M. domestica

produced R. pomonella puparia. Of nine tropical and subtropical fruit, only M.

acuminata x balbisiana (Blue Java banana) produced puparia. However, none of

the 11 puparia from M. acuminata x balbisiana produced adult flies.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 53

Table 7. Infestation of tropical and subtropical fruit by Rhagoletis pomonella inside

1.9-litre containers in laboratory exposures in Vancouver, Washington State, U.S.,

2016-2017.

Bruié ce BP ees, ON ke cee ar CO ems <r NRE

| replicates’ _ puparia tested positive (%)>

Malus domestica 10 48 30 11 (36.7a) |

Averrhoa carambola 17 0 a0 0

Musa acuminata x balbisiana 9 1] 17 3 (17.6a)

Psidium guajava 15 0 40 0

Carica papaya 13 0 ya 0

Annona cherimola : 0 6 0

Citrus reticulata Re: 0 10 0

Citrus x sinensis 5 0 iS 0

Garcinia mangostana 5 0 5 0

aTwo to five females and three to five males per replicate container.

bFor positive fruit only: y2 = 1.88; df= 1; P = 0.17; percentages with same letters are not significantly

different (P > 0.05).

DISCUSSION

Results show that infestation of C. douglasii and feral M. domestica trees by

R. pomonella in central Washington is not a rare occurrence, despite the region’s

dry habitat. In this study, 6.7% of C. douglasii and 6.1% of M. domestica trees

were infested even in fly-managed (insecticide-treated) sites. Results suggest

that, while frequencies of C. douglasii and M. domestica that are infested do not

differ in fly-managed sites, C. douglasii is infested at a higher frequency in

unmanaged sites, possibly due to several reasons. In fly-managed sites, fly

populations were low due to control efforts. Also, the dry sagebrush ecosystem

and climate is suboptimal for fly survival (Wakie ef al. 2019), and C. douglasii

and feral M. domestica trees are relatively rare and are often spaced far apart.

The combination of disruption of fly populations caused by insecticide sprays,

low fly numbers, and widely spaced trees may result in random infestations. In

unmanaged sites, flies were not controlled; the ponderosa pine ecosystem with its

milder climate is more suitable for fly survival (Wakie ef a/. 2019), resulting in

higher fly populations, and more C. douglasii and feral M. domestica trees were

spaced more closely together than in the sagebrush ecosystem. Any inherent

preference by R. pomonella for C. douglasii, as suggested by frequencies of

infested trees, would be more detectable under these conditions.

If the frequency of infested C. douglasii and feral M. domestica is a function

of how many R. pomonella can survive in a region, then population increases in

central Washington could result in more infested trees. Rhagoletis pomonella

captures on WSDA survey traps from 2006 to 2017 increased 3—4 times. In 2006

and 2007, 0.0042 and 0.0029 flies/trap (4,260 and 4,482 traps), respectively,

were caught. In 2014, 0.0122 flies/trap (4,673 traps) were caught (Yee ef al.

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 54

2012; Klaus 2014). In 2017, 0.0384 flies/trap (5,420 traps) were caught (Klaus

2017). In previous C. douglasii and feral M. domestica surveys conducted in

July, August, and September 2004-2006 at four or five sites (fly-managed and

unmanaged, combined) in central Washington, means of 7.7% of C. douglasii

trees and 0.2% of M. domestica trees were infested (based on mean frequency

per site) (Yee 2008). The higher frequencies of infestation in the current study

may be due in part to increased R. pomonella populations in central Washington

in the years since the 2004—2006 surveys.

In southwestern Washington, most Crataegus and Prunus species appear to be

suitable ‘natural’ hosts for R. pomonella. Specifically, C. pinnatifida, C

laevigata x C. mexicana, C. lavalleei, C. x mordensis, and P. domestica subsp.

syriaca are newly recorded ‘natural’ hosts of R. pomonella. In addition, P. avium

and P. salicina, previously recorded as hosts (Yee and Norrbom 2017), were also

infested. In the current study, these natural hosts are distinguished from

“unconfirmed” hosts in which fruit were not from the field but were infested in

the laboratory. According to the International Standards for Phytosanitary

Measures 37 (FAO 2016), a natural host is “a plant species or cultivar that has

been scientifically found to be infested by the target fruit fly species under

natural conditions and able to sustain its development to viable adults.”

The four newly recorded Crataegus hosts have red or orange fruit that ripen

in September to October, similar to Crataegus monogyna Jacquin (English

hawthorn), a species frequently attacked in late summer and fall (Tracewski et al.

1987). Other Crataegus species with these physical and phenology traits thus

may also be attacked. In addition, the four newly recorded hosts all have in

common relatively large fruit (compared with C. douglasii: 10 mm diameter;

0.65 g), an additional trait that may make them attractive to R. pomonella.

Seven Prunus species in the subgenus Prunus, called “plums”, have been

recorded as natural hosts for R. pomonella across North America (Yee and

Norrbom 2017). However, varieties or subspecies of infested plums are rarely

mentioned (Yee and Goughnour 2008). For example, unidentified varieties of

plum and prunes were listed as R. pomonella hosts in Oregon in the mid-1980s

(AliNiazee and Brunner 1986). Here, Italian, French Petite, and Friar Black

varieties of P. domestica, as well as P. domestica subsp. syriaca, are identified as

potentially highly susceptible plums in addition to P. salicina — suggesting most

forms of plums are susceptible to attack in Washington.

Prunus avium (sweet cherry), in the subgenus Cerasus, is reported as a

natural host of R. pomonella in the PNW for the first time (in 2017), despite

many years when cherry collections were negative for the fly. Specifically, R.

pomonella was detected from P. avium collections made in July to August over

10 years (2008-2017) neither in Roslyn at sites documented to have R.

pomonella (W.L.Y., unpublished) nor at other western Washington sites in earlier

studies (Yee and Goughnour 2008). This is unlike the situation in Utah, where P.

avium, as well as P. mahaleb L. (mahaleb cherry) and P. cerasus (tart cherry),

were reported hosts for R. pomonella in 1985-1986 surveys (Allred and

Jorgensen 1993). In addition, P. cerasus was reported as a host for the fly in

Wisconsin (Shervis et al. 1970). This, combined with the fact that plums in the

PNW were recorded as hosts in the mid-1980s (AliNiazee and Brunner 1986),

suggests that cherries are less likely than plums to be attacked by R. pomonella

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 S53

in the PNW. The positive P. avium tree in the Cherry Grove site in Vancouver in

2017 was ~5 m from positive P. salicina and ~50 m from infested M. domestica,

possible sources of the infestation.

In addition to natural Prunus hosts, P. armeniaca (30%) x P. salicina (70%)

(both subgenus Prunus), known by the common name Dapple Dandy pluot

(Anonymous 2019), is a newly recorded host for R. pomonella in the laboratory.

Both of its parental fruit are natural hosts for R. pomonella (Lienk 1970; Yee and

Goughnour 2008). Based on its infestation rate relative to that of MM. domestica,

Dapple Dandy pluot appears highly suitable for larvae development. Whether

other P. armeniaca x P. salicina hybrids with different per cent parentages are

similarly suitable for R. pomonella will require further testing.

Another newly recorded host for R. pomonella in the laboratory is V. vinifera

(Red Globe grape). While V. vinifera is a natural host for Mediterranean fruit fly,

Ceratitis capitata (Wiedemann) (e.g., Roditakis et al. 2008), there is no record of

it being a natural host for R. pomonella. Thus, the Red Globe variety of V.

vinifera was probably attacked because other fruit were absent. In addition, its

large size (25~27 mm in diameter) and smooth surface apparently made it

acceptable for ovipositing flies.

More field research than is presented in the current work is needed to

determine whether tropical or subtropical fruit could be hosts for R. pomonella in

nature. Future studies include exposing more fruit to flies in the field, as well as

using fruit documented to be untreated with insecticides. However, of the 37

tropical and subtropical fruit exposed to R. pomonella in the field, only M. indica

produced puparia, raising the possibility that most of the tested fruit are

unsuitable for the fly. The presence of natural apple fruit in trees could have

deterred R. pomonella’s use of the fruit hung in trees, but because fly populations

in test trees were high, at least some flies probably encountered the tropical and

subtropical fruit. If so, flies did not oviposit in most of these fruit, eggs did not

hatch, or larvae could not complete development in them.

In the laboratory tests of tropical and subtropical fruit, M acuminata x

balbisiana (Blue Java banana) was noted as a newly recorded host and the only

such fruit infested. Unlike in . domestica trees, where 55 M. acuminata x

balbisiana were hung, the lack of alternative fruit in the laboratory may have

forced the flies to oviposit in it. The skin of M. acuminata x balbisiana is ~1 mm

thick, versus ~2—3 mm for M. acuminata (Cavendish banana): this perhaps

allowed easier ovipositor penetration by flies, especially when the fruit were ripe

(as for Bactrocera invadens Drew, Tsura & White; Cugala ef a/. 2013).

In conclusion, findings indicate R. pomonella consistently infests C. douglasii

and feral M. domestica in central Washington, with C. douglasii being more

frequently infested in sites where the fly is not managed. Most Crataegus and

most types of Prunus in the subgenus Prunus in southwestern Washington may

be suitable hosts for R. pomonella. Additional assessments are needed to

determine the suitability of tropical and subtropical fruit. Despite the need for

more work, the current results provide a basis for further research and

hypotheses concerning host use by R. pomonella and its potential impact on both

U.S. and tropical and subtropical fruit markets.

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 56

ACKNOWLEDGMENTS

We thank Dana Jones and Peter Chapman (USDA-ARS) for field assistance,

Doug Stienbarger and Justin O’Dea (Washington State University Clark County

Extension) for providing space and logistical support to conduct the work,

Nicanor Liquido (USDA-APHIS-PPQ-S&T-CPHST, PERAL, Honolulu, HI,

U.S.), Bradley Sinclair (Canadian National Collection of Insects and Ottawa

Plant Laboratory, Ottawa, Ontario, Canada), and two anonymous referees for

reviews that improved the manuscript, and the U.S. Department of Agriculture —

Foreign Agricultural Service for partial funding.

REFERENCES

AliNiazee, M.T. 1988. Diapause modalities in some Rhagoletis species. In Ecology

and Management of Economically Important Fruit Flies. Edited by M. T.

AliNiazee, Agricultural Experiment Station Oregon State University, Corvallis,

OR. Pp 13-25.

AliNiazee, M.T. and Brunner, J.F. 1986. Apple maggot in the western United States:

a review of its establishment and current approaches to management. Journal of

the Entomological Society of British Columbia 83: 49-53.

AliNiazee, M.T. and Penrose, R.L. 1981. Apple maggot in Oregon: a possible threat

to the Northwest apple industry. Bulletin of the Entomological Society of

America, 27: 245-246.

Allred, D.B. and Jorgensen, C.D. 1993. Hosts, adult emergence, and distribution of

the apple maggot (Diptera: Tephritidae) in Utah. The Pan-Pacific Entomologist,

69: 236-246.

Anonymous. 2019. Dapple Dandy pluot® tree (semi-dwarf). GrowOrganic.com

[accessed 11 May 2019].

Anonymous. 2018. Washington apple crop forecast slightly lower amid tougher

export scenario. https://www.freshfruitportal.com/news/2018/08/08/washington-

apple-crop-forecast-slightly-lower-amid-tougher-export-scenario/ [accessed 3

December 2018].

Bush, G.L. 1966. The taxonomy, cytology, and evolution of the genus Rhagoletis in

North America (Diptera, Tephritidae). Bulletin of the Museum of Comparative

Zoology, 134: 431-562.

Canadian Food Inspection Agency. 2017. D-00-07 Phytosanitary requirements to

prevent the introduction and spread of apple maggot, Rhagoletis pomonella

(Walsh). https://www.inspection.gc.ca/plants/plant-pests-invasive-species/

directives/date/d-00-07/eng/1323819375916/1323819868990 [accessed 29

August 2019].

Canadian Food Inspection Agency. 2016. Apple maggot (Rhagoletis pomonella).

Pest alert. British Columbia Ministry of Agriculture. https://www?2.gov.bc.ca/

assets/gov/farming-natural-resources-and-industry/agriculture-and-seafood/

animal-and-crops/plant-health/phu-apple-maggot-pestalert.pdf [accessed 31 May

2019].

Cugala, D., Ekesi, S., Ambasse, D., Adamu, R.S., and Mohamed, S.A. 2013.

Assessment of ripening stages of Cavendish dwarf bananas as host or non-host to

Bactrocera invadens. Journal of Applied Entomology, 138: 449-457.

FAO. 2016. International standards for phytosanitary measures 37 (ISPM 37).

Determination of host status of fruit to fruit flies (Tephritidae). Secretariat of the

International Plant Protection Convention, Rome, Italy.

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 57

Klaus, M.W. 2017. 2017 apple maggot survey summary report. Washington State

Department of Agriculture, Plant Protection Division, Yakima, WA. 22 pp.

Klaus, M.W. 2014. 2014 apple maggot survey summary report. Washington State

Department of Agriculture, Plant Protection Division, Yakima, WA. 20 pp.

Kumar, S., Yee, W.L., and Neven, L.G. 2016. Mapping global potential risk of

establishment of Rhagoletis pomonella (Diptera: Tephritidae) using MaxEnt and

CLIMEX niche models. Journal of Economic Entomology, 109: 2043-2053.

Lienk, S.E. 1970. Apple maggot infesting apricot. Journal of Economic Entomology,

63: 1684.

Lyons, C.P. and Merilees, B. 1995. Trees, shrubs and flowers to know in Washington

and British Columbia. Lone Pine Publishing, Redmond, Washington.

NASS. 2017. 2017 State agriculture overview. https://www.nass.usda.gov/

Quick _Stats/Ag Overview/stateOverview.php?state=WASHINGTON [accessed

3 December 2018].

Roditakis, E., Tsagkarakou, A., and Roditakis, N.E. 2008. Extensive damage on

white variety table grapes by the Mediterranean fruit fly Ceratitis capitata

(Wiedemann) in Greece. OEPP/EPPO Bulletin, 38: 216-219.

Shervis, L.J., Boush, M.G., and Koval, C.F. 1970. Infestation of sour cherries by the

apple maggot: confirmation of a previously uncertain host status. Journal of

Economic Entomology, 63: 294-295.

Tracewski, K.T., Brunner, J.F., Hoyt, S.C., and Dewey, S.R. 1987. Occurrence of

Rhagoletis pomonella (Walsh) in hawthorns, Crataegus, of the Pacific Northwest.

Melanderia, 45: 19—25.

Wakie, T.T., Yee, W.L., Neven, L.G., and Kumar, S. 2019. Modeling the abundance

of two Rhagoletis fly (Diptera: Tephritidae) pests in Washington State, U.S.A.

PLOS ONE. doi.org/10.1371/journal.pone.0217071.

Washington State Department of Agriculture. 2018. Apple maggot.

https://agr.wa.gov/plantsinsects/insectpests/applemaggot/default.aspx [accessed 9

December 2018].

Yee, W.L. 2008. Host plant use by apple maggot, western cherry fruit fly, and other

Rhagoletis species (Diptera: Tephritidae) in central Washington state. The Pan-

Pacific Entomologist, 84: 163—178.

Yee, W.L. and Goughnour, R.B. 2017. Development in mango (Mangifera indica)

and other tropical and temperate fruit by Rhagoletis pomonella and R. indifferens

(Diptera: Tephritidae) in the laboratory. The Florida Entomologist, 100: 157-161.

Yee, W.L. and Goughnour, R.B. 2016. Peach is an occasional host for Rhagoletis

pomonella (Walsh, 1867) (Diptera: Tephritidae) larvae in western Washington

state, U.S.A. The Pan-Pacific Entomologist, 92: 189-199.

Yee, W.L. and Goughnour, R.B. 2008. Host plant use by and new host records of

apple maggot, western cherry fruit fly, and other Rhagoletis species (Diptera:

Tephritidae) in western Washington state. The Pan-Pacific Entomologist, 84:

179-~193.

Yee, W.L. and Norrbom, A.L. 2017. Provisional list of suitable host plants of the

apple maggot fly, Rhagoletis pomonella (Walsh) (Diptera: Tephritidae). In USDA

Compendium of Fruit Fly Host Information (CoFFHI). Edition 2.0. https://

coffhi.cphst.org [accessed November 2019].

Yee, W.L., Klaus, M.W., Cha, D.H., Linn, Jr., C.E., Goughnour, R.B., and Feder, J.L.

2012. Abundance of apple maggot, Rhagoletis pomonella, across different areas

in central Washington, with special reference to black-fruited hawthorns. Journal

of Insect Science, 12: 1-14.

Zar, J.H. 1999. Biostatistical analysis, Fourth edition. Prentice Hall, Upper Saddle

River, New Jersey.

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 58

Appendix 1. Fly-managed (insecticide-treated) and unmanaged sites in central

Washington State, U.S., surveyed for Rhagoletis pomonella infestation in Crataegus

douglasii and feral Malus domestica, 2010-2017, and temperate fruit sites in

southwestern Washington State, U.S., surveyed for Rhagoletis pomonella infestation,

2015-2018.

Site name Coordinates, elevation Ecosystem?

Union Gap 46°33'49""N, 120°27'5S7"W, 301 m Sagebrush

West Valley 46°32'25"N, 120°49'17”"W, 636 m Bunchgrass

Ellensburg 46°59'05"N, 120°32'27”"W, 460 m Bunchgrass

Klickitat Co. 45°56'14"N, 121°07'02”"W, 271 m Ponderosa Pine

Nile 46°49'47"N, 120°56'43” W, 621 m Ponderosa Pine

Cle Elum AT LL'49"'N, 120°55'08"W, 586 m Ponderosa Pine

Roslyn 47°13'10"N, 120°59'19”"W, 669 m Ponderosa Pine

Goldendale 45°49'31"N, 120°48'54”"W, 499 m Ponderosa Pine

Brooks Memorial Park 45°56'60"N, 120°39/59”"W, 797 m Ponderosa Pine

Wenas 46°50'11""N, 120°43'11"W, 683 m Bunchgrass

Devine 45°37'S6"N, 122°37'05"W, 55 m Coast Forest

Woodland 45°56'24’'N, 122°40'21"W, 52 m Coast Forest

Woodland (town) 45°54'12"N, 122°44'49"W, 9 m Coast Forest

FN oll (H.0.) 45°40'38"'N, 122°39'04"W, 73 m Coast Forest

Terrace Garden (T.G.) 45°40'31"N, 122°38'59"W, 82 m Coast Forest

Leverage Park 45°39'03"N, 122°39'27"W, 41 m Coast Forest

Salmon Creek 45°42'44"N, 122°40'41”"W, 6m Coast Forest

Cherry Grove 45°40'33"N, 122°38'54"W, 75 m Coast Forest

Vancouver Orchard 45°38'05"N, 122°33'18”"W, 95 m Coast Forest

H.O., Upper Field 45°40'29"N, 122°38'47"W, 95 m Coast Forest

‘Ecosystem classification based on Lyons and Merilees (1995).

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 59

SCIENTIFIC NOTE

Promachus dimidiatus Curran (Diptera:

Asilidae): a robber fly genus and species

new to British Columbia

R. A. CANNINGS!, T. EHLER 82,

A. MANWEILER?, T. KOHLER), En HAYES°*, AND

D. KNOP P4

Promachus dimidiatus Curran (Figs. 1, 2) is a large grassland robber fly

native to western North America, and ranging from southern Manitoba and

northern Saskatchewan west to Alberta and south to Utah, New Mexico, Kansas

and Wisconsin (Fisher and Wilcox 1997; Cannings 2014). This note records the

genus and species for the first time in British Columbia (BC).

Figure 1. Promachus dimidiatus, male. Photographed by Denis Knopp, Vernon, BC

(50.22976°N, 119.2986°W), 21 June 2018.

\Corresponding author: Royal British Columbia Museum, 675 Belleville Street, Victoria, BC V8W 9W2;

rcannings@royalbcmuseum.be.ca

2Masse Environmental Consultants Ltd., 812 Vernon Street, Nelson, BC, VOG 2J0

3Department of National Defence, Government of Canada, 7 -5535 Korea Road, Chilliwack, BC, V2R

5P2

447330 Extrom Road, Chilliwack, BC, V2R 4V1

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 60

During a survey of arthropods at risk at the Vernon Military Camp,

Department of National Defence (DND) in Vernon, BC (Masse Environmental

Consultants 2019), Tyson Ehlers and a team of biologists discovered a

population of P dimidiatus at a grassland site (50.22974°N, 119.29874°W)

approximately 3.9 km southwest of the centre of downtown Vernon, BC (Figs. 3,

4). This location is not accessible to the general public. Seven males and one

female were recorded, although probably some of the same males were reported

by more than one person at different times. On 21 June 2018, Denis Knopp

photographed a male (Fig. 1); a female was also captured, photographed and

released. Ehlers captured, photographed (Fig. 2) and released two males on 21

June 2018 at about 12:45 PDT (https://www.inaturalist.org/observations/

18905860). Both RAC and Eric Fisher (El Dorado Hills, CA, in /itt. 11

December 2018) identified the flies in the photos as Promachus dimidiatus.

Figure 2, Promachus dimidiatus, male. Photographed by Tyson Ehlers, Vernon, BC

(50.22976°N, 119.2986°W), 21 June 2018.

To better document the occurrence of this species in BC, DND employees

Angela Manweiler, Todd Kohler and Erik Hayes collected two males and a

female on 11 June 2019 at the same location. RAC examined the specimens and

confirmed that they keyed to P. dimidiatus. Comparison of genitalic dissections

of one of these Vernon males and a male from southern Manitoba (Spruce Woods

Provincial Park, 3 July 1986, T.D. Galloway; RBCM collection) show slight

differences that are likely within the range of normal variation. Further

comparison of the BC and Great Plains populations is planned.

Promachus is a large genus of more than 200 species worldwide (Pape and

Evenhuis 2019), with most of these living in the Northern Hemisphere. In the

New World, the genus ranges south to Venezuela (Fisher 2009); 21 species are

recorded in North America, most in the West (Fisher and Wilcox 1997). No

usable published identification key to the species of the Americas exists (Fisher

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 61

2009). The most often cited key to the genus in North America (Hine 1911) is out

of date; the most useful key is an unpublished manuscript prepared by Joseph

Wilcox, but it remains incomplete.

Promachus dimidiatus is a relatively large and robust robber fly, about 20-25

mm long, in the Tribe Apocleini (anatergite without setae) of the subfamily

Asilinae. It belongs to the large group of Promachus species in which the male

has a dense white pad of setae on the epandrium, resulting in a distinctive, showy

abdominal tip. The eighth sternite is triangular and strongly directed ventrally. In

the female, the ovipositor is typical of Promachus — long, including abdominal

segment 8, and with segments 6—7 partially modified into ovipositor segments.

In this species, these latter segments are laterally compressed but retain the

brown tomentum. Much of the tomentum on the thorax and abdomen of the fly is

light brown; the dorsum of the abdominal tergites is black. The legs are mainly

red with black anterior faces on the femora; the front tarsi of the male have

denser white setae than those of the other legs. Most of the setae throughout are

white, including the setae of the abdomen base, the katatergite and scutellum; the

most robust bristles of the legs, dorsum of abdominal segment 1, and thorax

dorsum (including the scutellum) are largely black. The mystax and beard are

pale yellow; the eyes in life are mostly green.

The records come from the geographic area known as the Vernon

Commonage, a geologic region composed of rolling hills, benchland terrain, and

talus slopes amidst a series of ridges. It is in the Southern Interior ecoprovince

and the North Okanagan Basin ecosection. Ecosystems fall within the Okanagan

Very Dry Hot Interior Douglas-fir Variant (IDFxh1) biogeoclimatic subzone,

which is characterized by warm dry summers, a long growing season, and cool

winters (Meidinger and Pojar 1991). Elevations range from 460 m to 680 m.

The site is at the base of a south-facing slope (Fig. 4). It is a gravelly flat in a

Pseudoroegneria spicata (Pursh) A. Léve (Poacaeae) grassland that ranges from

greatly to slightly disturbed. There is a drainage ditch to the south, beyond which

is an old, disturbed field. The field is recovering with grasses and young trees of

Populus trichocarpa Torr. & A. Gray ex Hook. (Salicaceae). Other common

plants in the area are Bromus tectorum L., B. mollis L., Poa bulbosa L., P.

pratensis L. (all Poaceae), Lupinus sericeus Pursh (Fabaceae), Potentilla recta L.

(Rosaceae), and Achillea millefolium L. (Asteraceae).

The robber flies perched on the riparian cottonwood shrubs and hunted in the

open gravelly area (Fig. 4). Other insects active at the site included Machimus

occidentalis Hine (Diptera: Asilidae), Hemipenthes sinuosa (Wiedemann)

(Diptera: Bombyliidae), Cicindela pupurea Olivier (Coleoptera: Carabidae), and

Ochlodes sylvanoides Boisduval (Lepidoptera: Hesperiidae).

One of us (RAC) has searched for Promachus in southern BC grasslands for

many years, hoping to find one of the two northwestern US species (Promachus

aldrichii Hine and Promachus princeps Williston), which range from eastern

Washington State southward (Fisher and Wilcox 1997). These species were

thought to be the most likely to be recorded in BC because of the significant

biogeographic connections between Washington State and BC Interior grasslands

and the fact that other insects have apparently moved northwards into Canada

through the Okanagan Valley in recent years (e.g., Cannings and Scudder 2009;

Cannings and Pym 2017), possibly because of climate warming. The absence of

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 62

the genus in asilid lists for the region (Cannings 2011, 2012) shows that the

search for Promachus was unsuccessful. Thus, it was a surprise to discover this

distinctive prairie species in the Okanagan Valley; much robber fly collecting has

occurred in this area — including on the Vernon Commonage — over the past

century: how was Promachus missed?

Figure 3. Promachus dimidiatus site Figure 4. Promachus dimidiatus habitat,

(circle); Vernon city centre (square); photographed by Tyson Ehlers, Vernon,

north end of Okanagan Lake (left); BC (50.22976°N, 119.2986°W), 21 June

arrow indicates north; horizontal bar= 2018.

5km.

Three Promachus species range north into Canada. Promachus dimidiatus

was first described from grasslands at Aweme, Manitoba, by Curran (1927), and

it meets the closely related, northeastern Promachus bastardii (Macquart) in the

forest—grassland interface of southeastern Manitoba (Cannings 2014). The latter

species is fairly common across southern and central Ontario (e.g., Skevington

1999), whereas the third species, Promachus vertebratus (Say), is rare in extreme

southwestern Ontario (Paiero ef al. 2010).

ACKNOWLEDGEMENTS

We thank the Department of National Defence for access to the study site;

Rachel McDonald (Senior Environmental Advisor) gave us permission to publish

DND data. We thank Eric Fisher for his opinion on the identity of the flies in the

2018 photographs and for commenting on the manuscript.

REFERENCES

Cannings, R.A. 2011. Robber Flies (Insecta: Diptera: Asilidae). Jn Assessment of

species diversity in the Montane Cordillera Ecozone (version 2). Edited by

G.G.E. Scudder and I.M. Smith. Pp. 461-483. Available from https://

royalbcmuseum.bc.ca/assets/Montane-Cordillera-Ecozone.pdf?

_ga=2.72458963.1785280708.1562859516- 661879581.1562270990 [accessed

12 August 2019].

Cannings, R.A. 2012. Checklist of the robber flies (Diptera: Asilidae) of British

Columbia. Jn E-Fauna BC: Electronic Atlas of the Fauna of British Columbia.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 63

2018. Edited by B. Klinkenberg. Lab for Advanced Spatial Analysis, Department

of Geography, University of British Columbia, Vancouver, British Columbia,

Canada. Available from http://ibis.geog.ubc.ca/biodiversity/efauna/documents/

RobberFliesofBCApril2012.pdf [accessed 12 August 2019].

Cannings, R.A. 2014. The Robber Flies (Diptera: Asilidae) of Western Canadian

Grasslands. Jn Arthropods of Canadian Grasslands (Volume 4): Biodiversity and

Systematics Part 2. Edited by D.J. Giberson and H.A. Carcamo. Biological

Survey of Canada, Ottawa, Ontario, Canada. Pp. 269-297. Available from http://

staff.royalbcmuseum.bc.ca/wp-content/uploads/2014/06/Cannings-Asilidae-

BSC-003-Vol4-Ch7-2014-.pdf [accessed 12 August 2019].Cannings, R.A. and

Pym, R.V. 2017. Archilestes californicus McLachlan (Odonata: Zygoptera:

Lestidae): a damselfly new to Canada. Journal of the Entomological Society of

British Columbia, 114: 77—82.

Cannings, R.A. and Scudder, G.G.E. 2009. Eleodes obscurus (Coleoptera:

Tenebrionidae): confirmation of a Canadian population and possible northward

expansion from Washington State into British Columbia in the Okanagan Valley.

Journal of the Entomological Society of British Columbia, 106: 81-82.

Curran, C.H. 1927. Descriptions of Nearctic Diptera. The Canadian Entomologist,

59: 79-92.

Fisher, E.M. 2009. Asilidae (robber flies, assassin flies, moscas cazadoras, moscas

ladronas). Jn Manual of Central American Diptera (Volume 1). Edited by B.V.

Brown, A. Borkent, J.M. Cumming, D.M. Wood, N.E. Woodley, and M.A.

Zumbado. NRC Research Press, Ottawa, Ontario, Canada. Pp. 585-632.

Fisher, E.M. and Wilcox, J. 1997. Catalogue of the robber flies (Diptera: Asilidae) of

the Nearctic Region. Unpublished preliminary draft. Sacramento, California,

United States of America.

Hine, J.S. 1911. Robberflies of the genera Promachus and Proctacanthus. Annals of

the Entomological Society of America, 4: 153-172.

Meidinger, D. and Pojar, J. 1991. Ecosystems of British Columbia. British Columbia

Ministry of Forests and Crown Publications Inc., Victoria, British Columbia,

Canada.

Masse Environmental Consultants. 2019. 2018 Surveys for Arthropods at Risk,

Vernon Military Camp, British Columbia. Prepared for Defence Construction

Canada on behalf of Department of National Defence Canada, Natural Resources

Program, 1262-12-4.

Paiero, S.M., Marshall, S.A., Pratt, P.D., and Buck, M. 2010. Insects of Ojibway

Prairie, a southern Ontario tallgrass prairie. Jn Arthropods of Canadian

Grasslands (Volume 1): ecology and interactions in grassland habitats. Edited by

J.D. Shorthouse and K.D. Floate. Biological Survey of Canada. Pp. 199-225.

Pape, T. and Evenhuis, N.L. 2019. Systema Dipterorum (version 2.3). Available from

sd.zoobank.org/ [accessed 12 Aug 2019].

Skevington, J.H. 1999. New Canadian records of Asilidae (Diptera) from an

endangered Ontario ecosystem. Great Lakes Entomologist, 32: 257-265.

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 64

SCIENTIFIC NOTE

Toxonevra muliebris (Harris) (Diptera:

Pallopteridae): a European fly new to

North America

R.A. CANNINGS! AND J.F. GIBSON

The Pallopteridae are acalypterate Diptera classified in the superfamily

Tephritoidea along with families such as Piophilidae, Lonchaeidae, and

Tephritidae (Woodley et al. 2009). They are usually called flutter flies, because

the males of some species extend their wings and vibrate them (Marshall 2012;

Rotheray 2014). Pallopterid flies are 3 to 6 mm long, with a presutural

dorsocentral seta, and are usually grey or yellow with brown-patterned wings

that are conspicuously longer than the abdomen (Fig. 1). The proboscis is short,

and the ovipositor is prominent with a non-retractile sheath (McAlpine 1987).

Figure 1. Zoxonevra muliebris, Victoria, BC, 23 August 2018. Photo: Thomas

Barbin

',2Royal British Columbia Museum, Victoria, BC, V8W 9W2

2Corresponding author: jgibson@royalbcmuseum.be.ca

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 65

There are 12 genera containing about 70 extant species of Pallopteridae

worldwide (Pape ef al. 2011), distributed in the northern temperate region,

temperate South America, and New Zealand (Marshall 2012). In North America,

!,2Royal British Columbia Museum, Victoria, BC, V8W 9W2

2Corresponding author: jgibson@royalbcmuseum.bce.ca

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 66

three native genera and nine species are recorded (Shewell 1965; McAlpine

1987). British Columbia (BC) has at least six species, with more that are

undescribed (Cannings and Scudder 2005).

Flutter fly larvae can apparently be saprophagous, phytophagous, or

carnivorous (McAlpine 1987; Rotheray 2014). Some have been found in the

flower buds and stems of plants in the aster and carrot families; others live under

the bark of dead trees and prey on the larvae of long-horned and bark beetles

(Rotheray 2014). On Vancouver Island, Palloptera claripennis Malloch has been

reared from the cones of Douglas-fir, where the larvae fed on the larvae of

Contarinia midges (Cecidomyiidae) (McAlpine 1987). Adults are usually found

on flowers or on the lower branches of trees and shrubs (Teskey 1976; McAlpine

1987).

Species of Zoxonevra have been included in the genus Palloptera Fallen (e.g.,

Shewell 1965; Watson and Dallwitz 2003; Jones 2014) and the spelling

Toxoneura Macquart has been commonly used (e.g., McAlpine 1987; Ozerov

1999, 2009). The correct spelling is Zoxonevra (Marshall 2012; Pape and

Thompson 2018).

This note reports the first records in North America of a European pallopterid,

Toxonevra muliebris (Harris), in Victoria, BC, Canada. Morge (1984) states that

this fly ranges in Europe from Spain and Italy to Great Britain, France, and

Austria. It has also been recorded once in Turkey (Ozerov 2009) and sporadically

in Ireland (Speight 1986; Wallace and O’Connor 1997), but it is not known to

occur in western Russia (Ozerov 2009) or eastern Asia (Ozerov 1999). The wing

pattern is distinctive (Fig. 1), making the species one of the easiest pallopterids

to identify.

The first record is a photograph taken in Victoria on 2 June 2016 by Eriko

Yamamoto and posted on BugGuide by Talmage Bachman (https://bugguide.net/

node/view/1544782). Bachman and Yamamoto subsequently collected two

specimens that are now deposited in the RBCM: one female (label data: Canada,

BC, Victoria, 3946 Quadra St., indoors, 48°28’00"N, 123°21'46”"W, 17.viii.2018,

T. Bachman, E. Yamamoto, ENT018-004842) and one male (label data: Canada,

BC, Victoria, 536 Herald St., indoors, 48°25’50"N, 123°22'05”W, 23.viii.2018, T.

Bachman, E. Yamamoto, ENT018-004854). Andrew Simon and Lauren Magner

reported a dead individual discovered in a house cupboard at 281 Highland

Road, Galiano Island (48°52'8.5"N, 123°20'45"W), 9 September 2017 (https://

bugguide.net/node/view/1469157/bgimage; https://www.inaturalist.org/

observations/18366976). Two additional female specimens were collected on

indoor windows; they are also housed in the RBCM. The label data read:

Canada, British Columbia, Victoria, 1909 Shotbolt Road, at indoor window,

urban garden, 48°24'52.3"N, 123°19'34.8"W, 31m asl, 3.ix.2017, Joan C. Kerik

and Robert A. Cannings, ENT017-012001; ibid.,18.x.2017, ENT017-012000.

Thomas Barbin photographed a male in Victoria on the corner of Raynor Ave.

and Catherine St., 48°26'12"N, 123°23'05"W on 23 August 2018 (https://

bugguide.net/node/view/1595875) (Fig. 1). Finally, on 4 August 2019, Scott

Gilmore collected two on a screen mesh on his house deck at 7494 Andrea

Crescent, Lantzville (49°14'46"N, 124°05'31"W). These specimens are now in

the RBCM: male, ENT019-003779; female, ENT019-003780.

J. ENTOMOL. Soc. BRIT. COLUMBIA 116, DECEMBER 2019 67

Although 7: muliebris occurs in outdoor habitats in Europe (where larvae are

often found under tree bark, where they probably feed on beetle larvae and other

insects), the fly is also reported indoors (Wallace and O’Connor 1997), where the

larvae are thought to prey on carpet beetles (Coleoptera: Dermestidae) (Jones

2014). All but two of the BC specimens have been found inside buildings, where

perhaps the fly has adopted the same behaviour. The two found outside (the

Lantzville record) were captured on a house deck.

Toxonevra muliebris may already be widespread on southeastern Vancouver

Island and the Gulf Islands, given the time frame of the records (2016 through

2019) and the six different locations. The Lantzville record is approximately 108

kilometres north of the southernmost Victoria site; the Galiano Island site is

about 50 kilometres northeast of the same Victoria location. The species likely

has the potential to spread over a much larger area and, if it establishes in North

America as a predator of Anthrenus and other dermestid beetles, it might even

help control those species that are household pests.

Although many historical alien insect introductions from Europe to the

Pacific Coast of Canada came directly via shipping (Spence and Spence 1988),

many later ones probably arrived indirectly via eastern North America (Copley

and Cannings 2005; Cannings ef al. 2007). Other recent introductions have come

directly from eastern Asia in part because of the plentiful marine traffic between

the two continents (e.g., Cannings 1989; Canadian Forest Service 1999; deWaard

et al. 2010). It is most likely that the southwestern BC population of Zoxonevra

muliebris arrived directly from Europe because the fly is not known to be

introduced anywhere else, and is distinctive enough that if it had it would likely

have been noted. We speculate that the species may have arrived in household

goods, such as clothing or luggage.

The discovery of Yoxonevra muliebris in North America is an excellent

example of citizen science and the value of online species identification and

documentation sites such as BugGuide and iNaturalist. Postings of photographs

of this fly on these sites by naturalists and other members of the public brought

the records to the attention of entomologists who recognized the unusual nature

of the observations.

ACKNOWLEDGEMENTS

We appreciate the help and information given by John Carr, Martin Hauser,

Morgan Jackson, Joel Kits, Owen Lonsdale, and Al Norrbom. Talmage

Bachman, Thomas Barbin, Scott Gilmore, Joan Kerik, Lauren Magner, Andrew

Simon, and Eriko Yamamoto provided photos and/or specimens and associated

data. Tristan McKnight translated some of the Russian literature and Claudia

Copley commented on the manuscript. We thank them all.

REFERENCES

Canadian Forest Service, Natural Resources Canada. 1999. Alien forest pests:

context for the Canadian Forest Service’s Science Program (Science Program

context paper). Available from http://www.cfs.nrcan.gc.ca/pubwarehouse/pdfs/

10465_e.pdf.

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 68

Cannings, R.A. 1989. An Asian Hornet, Vespa simillima xanthoptera (Hymenoptera:

Vespidae) in North America. Journal of the Entomological Society of British

Columbia, 86: 91.

Cannings, R.A., Miskelly, J.W., Schiffer, C.A.H., Lau, K.L.A., and Needham, K.M.

2007. Meconema thalassinum (Orthoptera: Tettigoniidae), a foreign katydid

established in British Columbia. Journal of the Entomological Society of British

Columbia, 104: 91-92.

Cannings, R.A. and Scudder, G.G.E. 2005. The true flies (Diptera) of British

Columbia. Jn E-Fauna BC: Electronic Atlas of the Fauna of British Columbia

[www.efauna.bc.ca]. Edited by B. Klinkenberg, 2018. Lab for Advanced Spatial

Analysis, Department of Geography, University of British Columbia, Vancouver,

British Columbia, Canada.

Copley, C. and Cannings, R.A. 2005. Notes on the status of the Eurasian moths

Noctua pronuba and N. comes (Lepidoptera: Noctuidae): on Vancouver Island,

British Columbia. Journal of the Entomological Society of British Columbia,

102: 83-84.

deWaard, J.R., Humble, L.M., and Schmidt, B.C. 2010. DNA barcoding identifies

the first North American records of the Eurasian moth, Eupithecia pusillata

(Lepidoptera: Geometridae). Journal of the Entomological Society of British

Columbia, 107: 25—32.

Jones, R. 2014. The Womanly Bow-wing. https://bugmanjones.com/tag/palloptera-

muliebris/

Marshall, S.A. 2012. Flies: the natural history and diversity of Diptera. Firefly

Books, Richmond Hill, Ontario, Canada.

McAlpine, J.F. 1987. Pallopteridae [chapter 68]. J: Manual of Nearctic Diptera. Vol.

2. Edited by J.F. McAlpine, B.V. Peterson, G.E. Shewell, H.J. Teskey, J.R.

Vockeroth, and D.M. Wood. Agriculture Canada Monograph 28. Pp. 839-843.

Morge, G. 1984. Family Pallopteridae. Jn Catalogue of Palaearctic Diptera. Edited by

A. Sods and L. Papp. Akadémiai Kiad6, Budapest, Hungary. Pp. 242-246.

Ozerov, A.L. 1999. Family Pallopteridae [chapter 73]. Jn Key to the Insects of the

Russian Far East. Vol. VI. Diptera and Siphonaptera. Part 1. Edited by P.A. Lehr.

Vladivostok. Dal’nauka. Pp. 531-534. (In Russian).

Ozerov, A.L. 2009. Review of the family Pallopteridae (Diptera) of the fauna of

Russia. Russian Entomological Journal, 8(2): 128-146. (In Russian).

Pape, T., Blagoderov, V., and Mostovski, M.B. 2011. Order Diptera Linnaeus, 1758.

In Animal biodiversity: An outline of higher-level classification and survey of

taxonomic richness. Edited by Z.-Q. Zhang. Zootaxa, 3148: 222—229.

Pape, T. and Thompson, F.C., editors. 2018. Systema Dipterorum (version 2.0, Jan

2011). In Species 2000 & ITIS Catalogue of Life, 30th October 2018. Edited by

Y. Roskov et a/. Digital resource available from www.catalogueoflife.org/col.

Species 2000: Naturalis, Leiden, The Netherlands.

Rotheray, G.E. 2014. Development sites, feeding modes and early stages of seven

European Palloptera species (Diptera, Pallopteridae). Zootaxa, 3900: 50—76.

Shewell, G.E. 1965. Family Pallopteridae. Jn A catalogue of the Diptera of North

America north of Mexico. Edited by A. Stone, C.W. Sabrosky, W.W. Wirth, R.H.

Foote, and J.R. Coulson United States Department of Agriculture, Washington,

DC, USA. Pp. 714-715.

Speight, M.C.D. 1986. Cheilosia argentifrons (Diptera: Syrphidae) new to Ireland:

Donacia cinerea (Coleoptera: Chrysomelidae) and Palloptera muliebris (Diptera:

Pallopteridae), presence confirmed in Ireland. Irish Naturalists’ Journal, 22: 159-

160.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 69

Spence, J. and Spence, D. 1988. Of ground beetles and men: introduced species and

the synanthropic fauna of western Canada. Memoirs of the Entomological

Society of Canada, 120 (S144): 151-168. doi:10.4039/entm120144151-1.

Teskey, H.J. 1976. Diptera larvae associated with trees in North America. Memoirs

of the Entomological Society of Canada, 100: 1-83.

Watson, L. and Dallwitz, M.J. 2003 onwards. Insects of Britain and Ireland: the

families of Diptera. Version: 7 October 2018. Available from http://www.delta-

intkey.com/ [accessed 19 April 2019]

Wallace, P.F. and O'Connor, J.P. 1997. Palloptera muliebris (Harris) (Dipt.,

Pallopteridae) discovered in Dublin City. Entomologist’s Monthly Magazine,

133: 114.

Woodley, N.E., Borkent, A., and Wheeler, T.A. 2009. Phylogeny of Diptera. Jn

Manual of Central American Diptera: Volume 1. Edited by B.V. Brown, A.

Borkent, J.M. Cumming, D.M. Wood, N.E. Woodley, and M.A. Zumbado. NRC

Research Press, Ottawa, Ontario, Canada. Pp. 79-94.

J. ENTOMOL. SOc. BRIT. COLUMBIA 116, DECEMBER 2019 70

OBITUARY

Peter Belton

(6 September 1930 — 1 April 2019)

Peter Belton passed away the morning of | April 2019 at the age of 88. Peter

led a full life, and touched the lives of many. He moved with Bryan Beirne’s lab

to Simon Fraser University in 1967, and became one of the founding faculty

members of the Master of Pest Management Program, where he stayed until his

retirement in 1994.

I met Peter at my MSc defence. In the audience, filled largely with people I

knew, sat an elderly gentleman genuinely interested in my research. After

meeting Peter, I knew he had to be on my PhD committee; I am forever grateful

that he accepted. In the over 12 years that I knew Peter, we spoke a lot about

research, of course, but also about his life, the family he loved so very much, his

graduate students, and everything from murder mysteries, to how the Vancouver

Canucks were doing. On a drive down to a conference in the United States in

2008, Peter let me record his story. Selections of that, with information he

provided since, follow.

Peter was born, an only child, in Driffield, a small Yorkshire town, on 6

September 1930. His dad, who worked for the civil service, was a keen gardener

and beekeeper. On weekends, the family would pack up an old Morris Ten, and

go to the Yorkshire moors to tend their bees. These weekends sparked Peter’s

interest in all things entomological.

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 71

When his dad was promoted, the family moved to London, during the Blitz.

Always one to find a silver lining, when recounting this move, Peter commented

that it meant the family was able to get a fantastic deal on a house, and that going

to the basement during the raids was rather exciting. The gardening and

beekeeping continued, and to those hobbies, Peter added trainspotting, stamp

collecting, and rearing of any Lepidoptera larvae he could find. It was during this

time that Peter knew he would become an entomologist.

Peter’s national service was deferred for a year after his dad’s untimely death

in 1948, and his mum’s poor health. He took the opportunity to work for the

Ministry of Food at Sheffield Market, in the meat and egg sector. He enjoyed

candling the eggs as a form of quality control.

After working for the Ministry, Peter did his 2 years of national service in the

RAF. He specialized in wireless, radio-work, which fueled his interest in

electronics. He was put in charge of a transmitter station in Norfolk, and worked

his way up to Corporal. Peter’s experience playing rugby during his school days

gave him an ‘in’ with the officers, and landed him a spot on their rugby team.

The two years passed quickly.

Peter entered the Applied Entomology Program at Imperial College. A strong

student, Peter’s success in his physical chemistry course saw the Chemistry

Department try to convince him to change his program. Luckily for us, Peter

stayed with entomology. He went back briefly for some work in civil service, but

finished his degree, doing an honours project in his last year. Under the

supervision of Peter Haskell, Peter’s research on tiger moth tympanal organs

yielded recordings from moth tympanal nerves. His first journal article, detailing

these groundbreaking recordings, was published in Nature.

During this final year, Peter won a grant to attend a prestigious conference in

Oxford. There, he met Elizabeth, a young entomologist studying the taxonomy of

weevils. When Peter started his PhD in Glasgow, he was surprised to find her in

the zoology department; Elspeth, as most of us know her, was doing her degree

there. Peter quickly realized he would have to brush up his Scottish country

dancing skills, as Elspeth was a keen dancer. Dressed in a kilt given to him by a

fellow graduate student, Peter impressed Elspeth, and the two of them became

part of the university’s dance team. They married in February 1957.

Peter’s supervisor, neurophysiologist Graham Hoyle, convinced Peter to work

on the nerves and muscles of the Lepidoptera leg. Peter had originally gone to

Glasgow to study moth tympanal organs, but Hoyle worried that an American lab

was further ahead in the research. When it came time to defend, although Hoyle

felt Peter’s research was complete, external examiner John Pringle thought more

work was needed. Most of this work was completed under neurologist Harry

Grundfest’s supervision, because Peter and Elspeth moved to New York for Peter

to work with Grundfest at Columbia University. His thesis was completed in

1960.

Peter loved his brief time in New York. The Grundfest lab spent the summers

at Woods Hole, studying the nervous system of lobster walking legs; the meaty

claws were always spared, but never wasted! Elspeth worked for Asher Treat as a

lab instructor for his general biology course, and Peter and Elspeth would visit

Treat’s summer cottage on the occasional weekend. Unfortunately for Peter,

moths were not available for purchase, so he spent his time working on

J. ENTOMOL. SOC. BRIT. COLUMBIA 116, DECEMBER 2019 72

mealworm muscles, because the larvae were readily available from local pet

stores. He also did some work on the slow muscles of frogs, which were

somewhat similar to insect muscles. His breakthrough research was showing that

action potentials in the mealworm muscles could be produced with potassium.

Thanks to Asher Treat meeting George Wishart at an international conference,

Wishart learned of Peter, and convinced Bryan Beirne to interview him for a job

in Belleville. When Peter and Elspeth went to Ontario that June for the interview,

they were caught off guard by a late snowfall. Despite that cold introduction to

Ontario, they moved to Belleville. Peter taught electrophysiology as an adjunct at

Queen’s University, and was a research scientist and group leader at the

Agriculture Canada Research station. They welcomed a daughter and son while

in Belleville, then had another son after they moved to Vancouver.

In terms of research, Peter is best known for his work on mosquitoes, and his

1983 book on the mosquitoes of British Columbia. However, Peter remained

fascinated by all things bioacoustic, and participated in many projects, right until

his death. He treated everyone as equal, was a patient teacher, encouraged

original research, and never spoke poorly of others. This brief piece only touches

on a few aspects of Peter’s life, and cannot do justice to the amazing person he

was.

Before Peter died, I was fortunate enough to visit and say my goodbyes.

During that last visit, Peter said he felt very lucky to have lived such an

incredible life; and, in the fashion suiting a gentleman, he asked that I let people

know he was doing well. He didn’t want anyone to worry, or to cause a bother.

The world is a little dimmer for the loss of such a brilliant person. Father to

three children he adored, grandfather to seven, and academic father to years of

students, both undergraduate and graduate, Peter will live on in the fond

memories each of us has of him.

Written by

MEL HART