Evolutionary and mechanistic aspects of insect host plant

preference

Alexander Schäpers

Department of Zoology, 2016

© Alexander Schäpers Cover picture: Photos by Vlad Dinca and Alexander Schäpers ISBN: 978-91-7649-381-6 Printed in Sweden: Holmbergs, Malmö 2016 Distributor: Department of Zoology, Stockholm University

List of Papers

I Carlsson MA, Bisch-Knaden S, Schäpers A, Mozuraitis R, Hansson BS & Janz N; (2011) Odor maps in the brain of butterflies with divergent host-plant preferences, PLoS ONE 6(8) e24025

II Schäpers A, Carlsson MA, Gamberale-Stille G & Janz N; (2015) The role of olfactory cues for the search behavior of a specialist and generalist butterfly, Journal of Insect Behavior, 28:77-87

III Schäpers A, Carlsson MA, Nylin S & Janz N; (2016) Specialist and generalist oviposition stratiegies in butterflies: maternal care or precocious young?, Oecologia, 180(2): 335-343

IV Schäpers A, Nylin S, Gonzalez-Voyer A, Wang H, Wahlberg N & Janz N; Clutch size and host use in butterflies – and how to measure diet breadth, Manuscript

Reprints were made with the permission of the publishers

II

Content

GLOSSARY .......................................................................................... III

INTRODUCTION ................................................................................. 1

AN OBSERVATION .................................................................................................... 1 SENSORY INPUT ........................................................................................................ 2 INTERNAL/MOTIVATIONAL STATES ..................................................................... 3 LARVAL ABILITIES .................................................................................................... 4 EVOLUTIONARY CONTEXT ..................................................................................... 4 AIMS OF THE THESIS ................................................................................................ 5

MATERIAL AND METHODS .............................................................. 6

THE STUDY SYSTEM.................................................................................................. 6 RESPONSES TO OLFACTORY CUES ......................................................................... 8 FEMALE PREFERENCE – LARVAL ABILITIES AND PERFORMANCE ................. 10 HOST RANGE AND CLUTCH SIZE: A PHYLOGENETIC COMPARISON .............. 11

RESULTS AND DISCUSSION .............................................................12

PHYSIOLOGICAL RESPONSES TO ODORS ............................................................ 12 BEHAVIORAL RESPONSES ...................................................................................... 13 A PERSPECTIVE OF EVOLUTIONARY CONTEXT................................................. 16

CONCLUDING REMARKS ..................................................................18

SVENSK SAMMANFATTNING ...........................................................19

ACKNOWLEDGEMENTS................................................................... 20

REFERENCES ......................................................................................21

III

Glossary

Oviposition: Laying eggs

Herbivore: an animal eating plant material

Lepidoptera: Scientific name for the insect order that contains butterflies and moths

Nymphalidae: The brush-footed butterflies, a family within the order of Lepidoptera, containing about 6000 butterfly species

Aglais urticae: Scientific name for the small tortoiseshell. Aglais is a genus of butterflies within Nymphalidae and contains several species of tortoiseshell butterflies. “urticae” is the species name

Olfaction: The sense of smell

Generalist: In herbivorous insects this describes a species that can use a relatively broad range of plants as hosts

Specialist: In herbivorous insects a specialist utilizes a narrow set of plants as hosts

1

Introduction

An observation

In the cool springtime in Sweden, when the sun is just strong enough to

warm the day to 10°C around midday, the first butterflies can be sighted

fluttering about. It is then, in mid-April, when female small tortoiseshell

butterflies (Aglais urticae, figure 1a) can be observed in their search for their

only host plant, the stinging nettle (Urtica dioica). If following a female in such a

search, one can see her flying across many nettle individuals seemingly

ignoring most of them. Sometimes she is landing on a potential host plant

evaluating the leaf with antennae and fore-tarsi (Ilse, 1955; Wiklund, 1977), but

in most cases continuing her flight after only a brief contact with the plant.

However, on rare occasions will the female not fly away after landing, but

instead curl her abdomen, exploring the leaf surface further until she finds a

suitable spot on the leaf and starts to deposit a clutch of 50-150 eggs (Figure

1b). The larvae emerge from the eggs after a few weeks and will then feed and

grow on the meticulously chosen host plant until they develop into pupae

Fig 1: (a) Ovipositing small tortoiseshells and (b) the egg clutches (indicated by arrow); Photos by Christer Wiklund

2

from which new butterflies emerge (Eliasson et al., 2005).

The simple behavioral observation of a female in search for a host plant

is a beautiful and fascinating thing to observe, but also opens up for a flood of

questions: How does she locate the plants? What makes her accept a plant as

host? Why does she discriminate so strongly between different plants? Why is

the tortoiseshell butterfly only feeding on stinging nettles? Why is she laying so

many eggs at the same time?

More questions could be asked, but generally the search flight and host

plant preference are results of processes on different organizational levels.

How processes on these different levels are connected and integrated with

each other and translate into behaviors is, however, poorly known. A

simplified scheme for host plant selection would be: (1) sensory input from the

environment (e.g. visual or olfactory), that the small tortoiseshell female

perceives and processes; (2) this sensory information is combined and

influenced by internal/motivational states (e.g. hunger or mating status) before

translated into an observable behavioral output, such as the host plant

selection and oviposition processes; (3) the set of abilities to perceive, process

and modify information and the host plant use are a result of her evolutionary

history.

Sensory input

In each evaluation and navigational task, such as the described host plant

selection process, the female takes advantage of information provided by the

environment. Plants and other sources emit a large variety of volatile

compounds that insects can use as cues to navigate towards targets. Antennae

and labial palps are the insect “nose”, where the majority of olfactory receptor

neurons are located and with which volatile molecules are detected from the

3

ambient air (Hansson, 1995; Lee et al., 1985). The signal from an activated

receptor neuron is projected via its axon directly into the primary olfactory

center in the brain, the antennal lobe (Hansson, 1999; Stocker, 1994). There it

is processed and subsequently sent to higher brain centers, such as mushroom

bodies and suboesophageal ganglion (Hansson, 1999). In higher brain centers

the collected and modified olfactory information is integrated with information

from other senses and internal states of the organism (Balkenius et al., 2009;

Hansson, 1999). This mix of interacting information is then influencing the

behavioral output of the egg-laying female. Studying the ability to find food or

host plant targets can improve our understanding of evaluation and selection

processes in general and the role of mechanistic abilities for processes of host

plant selection in particular.

Internal/motivational states

Without information from the environment any task would be

impossible to perform. However, all information is perceived in a context and

incoming information is not always directly translated into a behavior. For

example, a smell of host plant and a smell of food can be perceived at the

same time, but the tasks of egg-laying and feeding are usually not performed

simultaneously. Instead, reaction to stimuli is context-dependent, leading to

one type of stimulus taking priority over another. Stimulants that can evoke

different behaviors can prompt a female to choose host plants in the proximity

of food sources and neglect those farther away (Janz, 2005; Murphy et al.,

1984). A more direct influence can be observed when preference for host plant

odors is dependent on the mating status of a female (Saveer et al., 2012, Paper

II). Also, prior experience of a plant (i.e. learning) can modify preference for

that plant in the future (Cunningham et al., 1998; Snell-Rood and Papaj, 2009;

Thöming et al., 2013). A case of state dependency is when a female has not

4

had the opportunity to lay eggs for several days she is more prone to accept a

low quality plant as a host that would not have been acceptable otherwise

(Courtney et al., 1989; Fitt, 1986; Rosenheim and Rosen, 1991).

Larval abilities

It should be noted that the offspring are directly exposed to the female

choice, but that larval feeding patterns also play a role in the process. A meta-

analysis found a general correlation between female host preference and larval

performance (Gripenberg et al., 2010), although several cases of invasive plant

species disrupting this relationship have been reported (Feldman and Haber,

1998; Keeler and Chew, 2008; Nakajima et al., 2012). Furthermore, in many

cases larvae seem to be able to feed on a broader range of host plants than the

females would oviposit on (Friberg et al., 2015; Janz et al., 2001; Wiklund,

1975) and in some tree-feeding moths, larvae have the ability to disperse to

neighboring host plants, probably mitigating the impact of female oviposition

site (Cox and Potter, 1990; McDonald, 1991; Zalucki et al., 2002). Hence,

female selection of oviposition site is not an ultimate choice of offspring host

plant, but especially for immobile offspring, her choices largely limit the set of

plants the larva will encounter.

Evolutionary context

The small tortoiseshell (A. urticae), mentioned in the first paragraph, has

closely related species in North America (A. milberti), Europe (A. io) or Asia

(A. caschmirensis). All Aglais species have several traits in common: they are

specialists feeding on Urtica, are distributed in the northern hemisphere and

deposit their eggs in clutches (Ebert, 1993; James and Nunnallee, 2011; Reed,

2005). When zooming out to relatives in the related genera Nymphalis, Polygonia

and Vanessa, we can observe a larger variation in life-history traits. For example

5

Nymphalis and Aglais lay their eggs in clutches, while Polygonia and Vanessa

mainly lay their eggs singly (Ebert, 1993; Eliasson et al., 2005). However, while

Aglais has a specialized host plant use of Urtica, the ability to utilize plants from

the family Urticaceae and the order Rosales is shared with many other species

in the subfamily Nymphalini (Janz et al., 2001). Through understanding of the

evolutionary context of a species, insights on geographical distribution, life-

history traits and host plant associations can be gained (Ehrlich and Raven,

1964; Graves and Shapiro, 2003; Janz, 2011; Janz and Nylin, 2008;

Matsubayashi et al., 2010; Singer et al., 2008).

Aims of the Thesis

Host plant use is an important driver of diversification in herbivorous

insects (Futuyma and Moreno, 1988; Janz, 2011; Janz and Nylin, 2008;

Matsubayashi et al., 2010; Singer and McBride, 2012). Underlying mechanisms

of the search flight and the choice of a plant as a host are incoming stimuli,

internal integration of information and the translation into behavior. Somehow

these factors are connected to the evolutionary history of a species, but to

what extent each factor influences or how they are linked with each other is at

best little understood. This Thesis studied female host plant preference

through investigating female sensory abilities (Paper I) and the role of these

sensory abilities for the behavioral output (Paper II). Further, I explored how

the female oviposition behavior is connected with larval performance (Paper

III) and the evolutionary context of species (Paper III+IV).

6

Material and Methods

The study system

Butterflies in general, and those from the Nymphalidae family in particular, are

well-studied insect herbivores with respect to distribution, taxonomy and

knowledge of host plant use. Such abilities qualify them as an excellent study

system when studying evolutionary and mechanistic aspects of insect-plant

interactions. In this thesis I used the nymphalid butterflies Aglais urticae, A. io,

Vanessa atalanta, V. cardui and Polygonia c-album to varying degrees in the

experimental studies (Paper I-III) and the clade “nymphalines” after Wahlberg

(2009) in Paper IV.

Aglais urticae (small tortoiseshell, Fig 2a) and A. io (peacock, Fig 2b)

are closely related specialist butterflies feeding on stinging nettles (Urtica dioica,

Urticaceae) (Eliasson et al., 2005). Both species can be seen flying in spring

and autumn. A. urticae has 1-2 generations in the Stockholm area with egg-

laying adults flying in April and June. A. io fly and lay their eggs in May, and

the new generation emerges in July/August. Both deposit their eggs in

clutches. The search for a suitable host plant includes landings on many nettles

before they deposit their eggs on fresh leaves. In spring, adults can be seen

feeding on flowers of Salix, Tussilago and Taraxacum or tree-sap. In late

summer/ autumn they also include flowers of many other plants, such as

Cirsium or Buddleja, but can also include rotten fruit (Eliasson et al., 2005).

7

Vanessa atalanta (red admiral, Fig 2c) also has a narrow host plant range in

Europe, including Urtica and Parietaria (both Urticaceae) (Stefanescu, 2001),

although the latter is rare in Sweden and more commonly used in Central and

Southern Europe. V. atalanta is a migratory butterfly tracking favorable

conditions in Europe, visiting Sweden to reproduce in one generation only in

the summer months (Stefanescu, 2001). Adults can be observed feeding on

rotten fruit and various flowers. In the field, larvae can be easily found by

searching for characteristic tents that the larvae weave from the leaves of their

host plants (Thomas and Lewington, 2010).

Polygonia c-album (comma butterfly, Fig 2d) utilizes a broader range of

host plants, including trees, bushes and herbs from the plant orders Rosales

(Urtica [nettle], Humulus [hop], Ulmus [elm]), Saxifragales (Ribes [currants]),

Malphigiales (Salix [willows]) and Fagales (Betula [birch], Corylus [hazel])

Fig 2: The studied specialists (a) small tortoiseshell Aglais urticae, (b) peacock A. io, (c) red admiral Vanessa atalanta and generalists (d) comma butterfly (Polygonia c-album), (e) painted lady V. cardui and their host plant (f) stinging nettle (Urtica dioica); Photos a, c, d and e by Vlad Dinca.

8

(Eliasson et al., 2005; Nylin, 1988, Appendix Paper IV). They are robust, good

flyers, easily dispersing between patches and lay their eggs singly during a

period of several weeks. They have one generation in Sweden, lay their eggs in

spring (end of April through May until June) and emerge in July/August

before overwintering as adults. They primarily feed on tree-sap and rotten

fruits, but can in spring also be seen to visit flowers of Salix, Tussilago or

Taraxacum (personal observations).

Vanessa cardui (painted lady, Fig 2e) is one of the most polyphagous

butterflies reported to utilize more than 100 host plant species, although they

seem to prefer plants from the Asteraceae (Cirsium, Carduus, Arctium, Artemisia,

Centaurea, Cynara) and Malvaceae (Malva, Althaea), but also Boraginaceae

(Cynoglossum, Echium), Urticaceae (Urtica) and Plantaginaceae (Plantago) (Ebert,

1993; Stefanescu, 1997, Appendix Paper IV). They are almost globally

distributed. Similarly to V. atalanta they migrate to Sweden in the summer

months. They lay their eggs singly, sometimes several eggs on the same plant.

Adults feed on nectar from flowers (Stefanescu, 1997).

The host plant Urtica dioica (stinging nettle, Fig 2f) is a common,

widespread, perennial herb growing predominantly in ruderal areas. Nettles are

rather rich in nitrogen and often grow in dense stands (Taylor, 2009).

Responses to olfactory cues

In Lepidoptera, the primary olfactory processing center is the antennal

lobe, which consists of 60-65 spherical glomeruli (Hansson, 1995; Rospars,

1983, Paper I). Each glomerulus is innervated by the neurons housing one

specific receptor type so that the number of olfactory receptor types on the

antennae and the number of glomeruli are identical (Vosshall et al., 2000).

Several types of receptors can be activated by the same compounds and hence

9

activate several glomeruli (Carlsson et al., 2002, Paper I). The way of how

olfactory receptor neurons and glomeruli are connected with each other,

generate glomerular activations patterns that are unique for each odor

(Carlsson et al., 2002; Joerges et al., 1997).

Through application of a fluorescent Ca2+-sensitive marker to brain

tissue of an opened head capsule, odor-evoked neural activity patterns can be

visualized by means of a light-sensitive CCD camera. As there is an increase of

Ca2+ concentration in active cells, those cells or glomeruli in the antennal lobe

that are activated by an odorant can be visualized in a microscope and used to

study variation in Ca2+ activity patterns between individuals or species (Bisch-

Knaden et al., 2012, Paper I). In Paper I, the abilities of the specialist butterfly

A. urticae and the generalist P. c-album to respond to odorants in general were

investigated first. Then, the two species were compared in their responses to a

set of single odorants and to more complex host plant odors, to detect

possible differences of qualitative and quantitative responses of the tested

butterflies.

Fig 3: Setup of the Y-tube olfactometer in Paper II. Plants were invisible for the tested animals, decision lines are indicated in yellow

10

The observations of the physiological abilities (Paper I) were then

followed-up in a behavioral setup (Paper II). Here, the aim was to verify if the

documented detection of odors is translated into a behavioral response, but

also how good the resolution of this response is and if this response varies

with the mating status of the females. The experiments were performed on

adult butterflies in a classical Y-tube setup (Figure 3), where the only source of

information was olfactory. During the experiment butterflies did not come in

contact with the plants and could not see them.

Female preference – larval abilities and performance

In experiments of host plant choice, often leaves of plants are provided.

This gives the choosing female the possibility to obtain, in addition to

olfactory information, information from visual, contact-chemical and tactile

cues, which then should elicit the most accurate possible response. In Paper

III, leaves of ‘high’ and ‘low’ quality stinging nettle (U. dioica) were offered to

the specialists Aglais io, A. urticae and Vanessa atalanta and to the generalists V.

cardui and Polygonia c-album. Species’ oviposition choice between the leaves of

different qualities was compared and followed up by larval experiments. In

particular, larval mobility and ability to survive in unfavorable situations was

assessed in a series of experiments (see methods Paper III). Based on the

finding that generalist species are less accurate in their oviposition choice

(Bernays, 1998; Egan and Funk, 2006; Janz and Nylin, 1997) specialist species

were expected to distinguish between the different host plant qualities while

generalists were not. Similarly, I hypothesized larvae of generalist species to be

better at coping with unfavorable situations either through better survival on

poor plant tissue or the ability to mitigate female choice by moving away.

11

Host range and clutch size: a phylogenetic comparison

For species with rather immobile larvae the female oviposition site choice

is a major determinant of larval host plant. Some species aggregate their eggs

and the hatching larvae forage gregariously together (Bryant et al., 2000;

Denno and Benrey, 1997; Stamp, 1980). This life-style could limit host plant

range for these species in two ways: First, a host plant needs to have sufficient

foliage to sustain the larvae, so unless a plant species is not growing aggregated

together, small plant species can probably not be used. Second, oviposition on

poor quality or non-host tissue will entail a large cost for a clutch-laying

female, as a large portion of her potential realized fitness would die without

completing development. As generalist species are known to be less accurate in

their oviposition decision (Bernays, 1998; Bernays and Wcislo, 1994; Egan and

Funk, 2006; Janz and Nylin, 1997), this could favor a more specialized host

plant use among clutch laying species. Paper IV tests in a phylogenetic

comparative study, whether there is a relationship between diet breadth and

clutch size in a clade of Nymphalidae butterflies and whether this relationship

is influenced by the growth form of the host plant (as indicated by Janz and

Nylin, 1998). However, as results seem to vary depending on the taxonomic

level from which the measure of diet breadth has been derived (Beccaloni and

Symons, 2000; Symons and Beccaloni, 1999), a phylogeny of the host plants

was created to calculate phylogenetic measures of diet breadth. Additional

measures of diet breadth were obtained by counting the number of higher taxa

that were utilized, i.e. the number of host plant genera, families and orders.

These measures were then related to clutch size.

12

Results and discussion

Physiological responses to odors

Within Lepidoptera, moths are well known to use olfactory cues in their

search for targets and there have been numerous investigations of their

olfactory processing abilities (e.g. Balkenius et al., 2009; Carlsson et al., 2002;

Hansson, 1995; Saveer et al., 2012). In contrast, while the related day-active

butterflies often have been demonstrated to employ visual cues (Bergman et

al., 2015; Goulson and Cory, 1993; Kelber, 1999; Rausher, 1978),

investigations of their olfactory senses have been less thorough (but see for

recent studies Carlsson et al., 2013; Ikeura et al., 2010).

In Paper I we show that the olfactory system of the investigated

butterflies can detect different odors and process these in the antennal lobe.

The related generalist and specialist species had in general similar response

patterns both for the tested single compounds and the plant extracts (Figure

Fig 4: Responses to plant extracts in a similarity matrix. The color of each field indicates the median correlation coefficient between pairs of plant odors. Red color depicts high correlation between an odor pair, blue low correlation. Responses are significantly correlated between the two species (R2=0.64, p<0.001).

13

4). However, species-specific variation in sensitivity and differences in

responses to two common green leaf volatiles were detected. Among others

the pattern in the antennal lobe caused by the host plant Urtica dioica correlated

less with the pattern of the other tested plants in A. urticae than in P. c-album

indicating that the specialist has a more specific response pattern to the host

plant than the generalist.

Behavioral responses

The behavioral study in the Y-tube confirmed the functionality of the

butterfly olfactory system (Paper I) in that both tested species navigated

towards food and host plant targets with the help of olfactory cues alone

(Paper II, Figure 5). Unmated females of both species did not show a

preference for host plant odors, whereas mated ones did (Fig 5, see also Saveer

et al 2012). However, even for mated females olfactory information seemed to

be insufficient to distinguish between the odors of two host plants differently

Fig 5: Behavioral responses to odors in a choice between host and non-host plant by mated and unmated female Aglais urticae and Polygonia c-album. Sample size and p-values shown within bars.

14

ranked in the preference hierarchy or between host plant individuals of

different quality (Paper II). Neither could any preference be found in the

generalist butterfly to a blend of several host plants over a single host plant.

Interestingly, the specialist did not differentiate between a ‘high’ and a ‘low’

quality host plant individual with olfactory information alone (Paper II),

despite that they had a strong preference for the ‘high’ over ‘low’ quality host

plant when they had access to the leaves (Paper III). These findings suggest

that olfactory information provides the tested butterflies with a binary signal of

‘host’ or ‘non-host’, which then requires the females to approach the plant for

further evaluation (see also Mozūraitis et al., 2016). The generalist P. c-album

has host plants of high, intermediate and poor quality that differ strongly in

larval suitability (Nylin, 1988), so this low olfactory resolution makes the

search for good host plants a time-consuming business as each potential host

needs to be evaluated from close range. The specialist A. urticae, on the other

hand, has only one host plant, so it could appear that they do not suffer the

same time-cost as the generalist. However, as A. urticae deposits an egg-clutch

only every 1-2 days and have rather immobile larvae that perform poorly in

unfavorable situations (Paper III), they probably also need to invest much time

to find and select a suitable host plant.

A match between developmental stages (a co-adaptation of female

oviposition accuracy with larval capabilities), could be found in all five

butterfly species tested in Paper III. Females that discriminated between the

leaf qualities had larvae that were less mobile and with low performance on

poor quality leaves (Figure 6 & 7). On the other hand, less choosy females had

mobile, precocious larvae that better survived on poor leaves (Figure 6 & 7).

15

Fig 6: (A) Larval survival and (B) weight on “poor” and “high” quality host plant leaves (***p<0.001, **p<0.01, n.s: p>0.05)

Fig 7: (A) Female oviposition choice between “poor” and “high” quality host plant leaves (***p<0.001, n.s: p>0.05); A. urticae and A. io lay their eggs in clutches, thus no standard error; (B) Larval mobility on wooden rod; Box-plots capped with different letters differ significantly (p<0.05)

16

It should be noted, though, that female choosiness did not correspond

exactly to the generalist and specialist axis. The clutch laying specialist Aglais

responded as expected from a specialist, whereas the specialist V. atalanta did

not differentiate between leaf qualities and had mobile and persisting larvae

(Figure 6 & 7), thus rather fitting with the pattern expected by a generalist.

Notably, their larvae weave characteristic tents from the leaves of its host plant

(Thomas and Lewington, 2010). To do that the larvae need to be mobile and

probably have to select the leaf they are weaving themselves, which in turn

requires less accurate females.

A perspective of evolutionary context

A comparison of the two generalist species revealed that females of P. c-

album laid in average 25 eggs during a 3h time-interval whereas V. cardui laid

about 120. Intriguingly, the two genera differ in general from each other in egg

weight (Paper III, Janz & Nylin, unpublished data) and fecundity, as Polygonia

butterflies lay relatively large eggs whereas Vanessa’s eggs are rather small

(Paper III). Despite these differences in life-history, within the genera Polygonia

and Vanessa both, species with a broad and a narrow diet can be found (Braby,

2000; Scott, 1986). This indicates that the two species realized their broad host

plant ranges differently and that different life-history traits may allow for

pathways to a wider diet breadth.

The species that fitted best the expectation of a specialist in Paper III

were the two clutch laying Aglais species. A simple conclusion from that would

be to assume that most clutch layers have a rather specialized diet. Resource

specialization has been shown to allow for better adaptations to the host plant

in various ways (Bernays, 2001; Egan and Funk, 2006; Janz and Nylin, 1997).

Hence, if staking the life of many offspring in one oviposition event by “laying

17

all eggs in the same basket” the female should certainly avoid ovipositing in

the wrong place. Such a selective pressure could then favor a specialized host

plant use among clutch laying species.

Deposition of eggs in clutches and the subsequent gregarious larval

feeding is a life-history trait largely affecting a species’ life-style, through for

example female realized fecundity (Courtney 1984), predator interactions

(Clark and Faeth, 1997; Sillén-Tullberg, 1988; Stamp, 1980), facilitation of

larval feeding (Denno and Benrey, 1997; Kawasaki et al., 2009, Paper III), and

limitations of risk-spreading possibilities (Hopper, 1999; Root and Kareiva,

1984). The trait clutch-laying is a phylogenetically rather conserved trait in the

investigated nymphalid butterflies (Paper IV). Moreover, when taking a look at

the raw data of Paper IV, it seems to have arisen several times in different

genera, but is usually either present or absent in all species of a genus (see

Supplementary Paper IV). Host range, on the other hand seems to be an

evolutionary transient trait that is scattered in the phylogeny of nymphalid

butterflies (Nylin et al., 2014). Since host use is believed to be an essential

aspect of diversification among herbivorous insects (Ehrlich and Raven, 1964;

Jaenike, 1990; Janz, 2011; Janz and Nylin, 2008; Matsubayashi et al., 2010), it is

as important to understand the factors that can influence the dynamics of host

plant use.

Returning to the question if clutch-laying species have a more specialized

host use, it is evident that the relationship between clutch size and diet breadth

is not straightforward. While analyses from continuous data cannot detect a

significant relationship, discrete analyses of ‘generalist/specialist’ and ‘single-

egg layer/clutch layer’ find a significant relationship (Paper IV). The data

contains many species with a rather narrow diet and a few species with a

relatively broad host range. The influence of this skewness of the data is

18

avoided by transforming it into discrete states. It then indicates that host range

and clutch size varies relatively independently for most species but may be

connected in the most generalist species: no clutch layers can be found among

the most polyphagous species. This dichotomy of results from continuous and

discrete data is consistent among several measures of diet breadth (Paper IV).

The clutch size-diet breadth relationship does not seem to interact with host

plant growth form even though we found plant growth form to be correlated

with both clutch size and diet breadth (Paper IV). In detail, woody plants

receive somewhat larger clutches and species feeding on woody host plants

have a narrower diet breadth, but only 3-4% of the variation is explained. At

the same time it would be puzzling if a single or a few factors could explain a

complex trait such as diet breadth. It seems more realistic that a multitude of

factors acting on different aspects of the insect and the plant together have

shaped and shape the dynamics of host plant use in phytophagous insects.

Concluding remarks

The beauty of butterflies has caused many researchers and laymen to

focus their attention on this group of plant feeding insects. This interest since

the late 19th century has provided a relatively good record of abundance,

distribution and host plant use. The richness of biogeographical information is

contrasted by relatively little knowledge of the mechanistic factors behind

these patterns. In this Thesis I have shown that the investigated butterflies

have a functioning olfactory system and that they are capable to use it to locate

food and host plants. However, the ability to select host plants seems to be

limited if based on olfactory cues alone, but even direct access to plants

allowing the full set of mechanistic evaluation tools does not result in optimal

19

female oviposition choices. This paradox of suboptimality can be better

understood if taking into account how the larvae handle the female choice of

host plant and how the evolutionary history of a species looks like. Hence,

embedding mechanistic findings in an ecological and evolutionary context

improves our ability to explain the dynamics of insect-plant interactions.

Svensk sammanfattning

Ungefär 25% av alla djurarter i världen är växtätande insekter. De flesta

växtätande insekter har dock ett mycket begränsat antal värdväxter och kan

bara livnära sig på ett fåtal arter. Sambandet mellan växtätande insekter och

växter spelar en nyckelroll för evolutionen av insekter och har bland annat

resulterat i en överväldigande artrikedom. Tyvärr finns det dock stora luckor i

vår kunskap kring hur interaktionen mellan insekter och växter har

evolverat.Den föreliggande avhandlingen har studerat praktfjärilar

(Nymphalidae), och utgår från problemet hur en fjärils-hona väljer ut växter

för äggläggning. Larverna kan oftast inte röra sig mycket, så honans val av växt

är avgörande. Det är allmänt känt att såväl luktsinne, syn och smak används för

att känna igen och välja ut en växt för äggläggning. Att lukten av en värdväxt

upptäcks betyder dock inte nödvändigtvis att insekten navigerar direkt till

källan – även om de har förmågan att göra så. Lukt-informationen integreras

med intryck från andra sinnen men hänsyn tas även till om honan är hungrig,

vilka erfarenheter man har eller mer allmänt hur motiverad man är. I denna

avhandling visar jag att fjärilar kan använda sig av dofter för att hitta växter

eller mat, men också att lukten i sig inte är tillräckligt för de undersökta

fjärilarna för att göra finjusterade val. Intressant nog är inte ens direkt tillgång

20

till växten, där fjärilen’s alla sinnen kan komma till användning, tillräckligt för

att fatta optimala beslut. Detta suboptimala beteende blir mer logiskt när man

tar hänsyn till larvernas beteende, värdväxtbredden och den specifika

evolutionära bakgrunden en art har. Studierna i avhandlingen visar att de

mekanistiska processer som ligger bakom valet av värdväxt blir mer begripliga

om man ser dem i ett ekologiskt och evolutionärt perspektiv. Genom en

kombination av forskning kring mekanistiska och ekologiska processer, kan

frågan ”vad gillar en fjäril?” främja vår förståelse av den dynamik som finns i

samspelet mellan insekter och växter, och som har skapat den enorma

biodiversiteten vi finner idag.

Acknowledgements

I would like to express my gratitude to my supervisors Niklas Janz, Mikael

Carlsson and Gabriella Gamberale-Stille. You three all come from different

research fields and having taught me a lot of things. I thank you for having an

open ear to my many questions, supporting and guiding me, freedom of

actions and for being good company.

Especially Niklas I have plagued with emails, questions and discussions, but he

was always happy (?) to discuss and to give an objective opinion. While

checking that I was on track you let me test my own ideas and projects and

gave me much freedom to go my own way. A big thank you for that!

Christer Wiklund generously shared his extremely broad knowledge on

butterflies and plants and was always helpful. Thank you!

Sören, although you were not my formal supervisor you were involved in

several studies and had to endure your share of questions. Thank you for your

always friendly input!

21

I also would like to thank my other collaborators for their contributions to our

scientific work, you know who you are!

I have spent about 6 years at the Department of Zoology and had contact with

too many people to acknowledge them all here that all gave me a good time at

the Department. However, I would especially thank Alexandra B, Anna F,

Hélène A, Marianne H, Martin O, Philipp L and Sandra S for nice

conversations, support and being good company.

Last but not least I want to thank my parents for encouraging me to do what I

wanted (whatever it was) during my studies. Thank you!

My wife Anna for being the best support that I can imagine, for tolerating and

even encouraging me to pursue the unpredictable career of scientific research.

My deepest thanks for that!

Hannes and Marja, my beloved ones: being imaginative, full of energy and by

your presence always reminding me of the important things in life.

References

Balkenius A, Bisch-Knaden S, Hansson BS (2009). Interaction of visual and odour cues in the mushroom body of the hawkmoth Manduca sexta. J Exp Biol 212:535-541.

Beccaloni GW, Symons FB (2000). Variation of butterfly diet breadth in relation to host-plant predictability: results from two faunas. Oikos 90:50-66.

Bergman M, Lessios N, Seymoure BM, Rutowski RL (2015). Mate detection in a territorial butterfly—the effect of background and luminance contrast. Behav Ecol.

Bernays EA (1998). The value of being a resource specialist: Behavioral support for a neural hypothesis. Am Nat 151:451-464.

22

Bernays EA (2001). Neural limitations in phytophagous insects: Implications for diet breadth and evolution of host affiliation. Annu Rev Entomol 46:703-727.

Bernays EA, Wcislo WT (1994). Sensory capabilities, information processing, and resource specialization. Q Rev Biol 69:187-204.

Bisch-Knaden S, Carlsson MA, Sugimoto Y, Schubert M, Mißbach C, Sachse S, Hansson BS (2012). Olfactory coding in five moth species from two families. J Exp Biol 215:1542-1551.

Braby MF (2000). Butterflies of Australia: their identification, biology and distribution. Collingwood, Great Britain: CSIRO Pub.

Bryant SR, Thomas CD, Bale JS (2000). Thermal ecology of gregarious and solitary nettle-feeding nymphalid butterfly larvae. Oecologia 122:1-10.

Carlsson MA, Galizia CG, Hansson BS (2002). Spatial representation of odours in the antennal lobe of the moth Spodoptera littoralis (Lepidoptera : Noctuidae). Chem Senses 27:231-244.

Carlsson MA, Schäpers A, Nässel DR, Janz N (2013). Organization of the olfactory system of Nymphalidae butterflies. Chem Senses 38:355-367.

Clark BR, Faeth SH (1997). The consequences of larval aggregation in the butterfly Chlosyne lacinia. Ecol Entomol 22:408-415.

Courtney SP, Chen GK, Gardner A (1989). A general model for individual host selection. Oikos 55:55-65.

Cox DL, Potter DA (1990). Aerial dispersal behavior of the bagworm. J Arboric 16:242-243.

Cunningham JP, Jallow MFA, Wright DJ, Zalucki MP (1998). Learning in host selection in Helicoverpa armigera (Hübner) (Lepidoptera : Noctuidae). Anim Behav 55:227-234.

Denno RF, Benrey B (1997). Aggregation facilitates larval growth in the neotropical nymphalid butterfly Chlosyne janais. Ecol Entomol 22:133-141.

Ebert G (1993). Die Schmetterlinge Baden-Württembergs Bd. 1, Tagfalter 1. Stuttgart (Hohenheim): Ulmer.

Egan SP, Funk DJ (2006). Individual advantages to ecological specialization: insights on cognitive constraints from three conspecific taxa. Proc R Soc B - Biol Sci 273:843-848.

Ehrlich PR, Raven PH (1964). Butterflies and plants: a study in coexolution. Evolution 18:586-608.

Eliasson CU, Ryrholm N, Gärdenfors U (2005). Nationalnyckeln till Sveriges flora och fauna. Fjärilar: Dagfjärilar. Hesperiidae-Nymphalidae. Uppsala: ArtDatabanken, Sveriges lantbruksuniversitet.

Feldman TS, Haber WA (1998). Oviposition behavior, host plant use, and diet breadth of Anthanassa butterflies (Lepidoptera: Nymphalidae) using plants in the Acanthaceae in a Costa Rican community. Fla Entomol 81:396-406.

Fitt GP (1986). The influence of a shortage of hosts on the specificity of oviposition behaviour in species of Dacus (Diptera, Tephritidae). Physiol Entomol 11:133-143.

Friberg M, Posledovich D, Wiklund C (2015). Decoupling of female host plant preference and offspring performance in relative specialist and generalist butterflies. Oecologia 178:1181-1192.

23

Futuyma DJ, Moreno G (1988). The evolution of ecological specialization. Annu Rev Ecol Syst 19:207-233.

Goulson D, Cory JS (1993). Flower constancy and learning in foraging preferences of the green-veined white butterfly Pieris napi. Ecol Entomol 18:315-320.

Graves SD, Shapiro AM (2003). Exotics as host plants of the California butterfly fauna. Biol Conserv 110:413-433.

Gripenberg S, Mayhew PJ, Parnell M, Roslin T (2010). A meta-analysis of preference-performance relationships in phytophagous insects. Ecol Lett 13:383-393.

Hansson BS (1995). Olfaction in Lepidoptera. Experientia 51:1003-1027. Hansson BS (1999). Insect olfaction. Berlin: Springer-Verlag. Hopper KR (1999). Risk spreading and bet-hedging in insect population biology. Annu

Rev Entomol 44:535-560. Ikeura H, Kobayashi F, Hayata Y (2010). How do Pieris rapae search for Brassicaceae

host plants? Biochem Sys Ecol 38:1199-1203. Ilse D (1955). Behaviour of butterflies before oviposition. J Bombay Nat Hist Soc

53:486-488. Jaenike J (1990). Host specialization in phytophagous insects. Annu Rev Ecol Syst

21:243-273. James DG, Nunnallee D (2011). Life histories of Cascadia butterflies. Corvallis,

Oregon: Oregon State University Press. Janz N (2005). The relationship between habitat selection and preference for adult and

larval food resources in the polyphagous butterfly Vanessa cardui (Lepidoptera : Nymphalidae). J Insec Behav 18:767-780.

Janz N (2011). Ehrlich and Raven revisited: mechanisms underlying codiversification of plants and enemies. Annu Revi Ecol Evol Syst 42:71-89.

Janz N, Nyblom K, Nylin S (2001). Evolutionary dynamics of host-plant specialization: a case study of the tribe Nymphalini. Evolution 55:783-796.

Janz N, Nylin S (1997). The role of female search behaviour in determining host plant range in plant feeding insects: a test of the information processing hypothesis. Proc R Soc B - Biol Sci 264:701-707.

Janz N, Nylin S (1998). Butterflies and plants: a phylogenetic study. Evolution 52:486-502.

Janz N, Nylin S (2008). The oscillation hypothesis of host plant-range and specialization. In: Tilmon KJ, editor. Specialization, speciation, and radiation: the evolutionary biology of herbivorous insects Berkeley, California: University of California Press. p. 203-215.

Joerges J, Kuttner A, Galizia CG, Menzel R (1997). Representations of odours and odour mixtures visualized in the honeybee brain. Nature 387:285-288.

Kawasaki N, Miyashita T, Kato Y (2009). Leaf toughness changes the effectiveness of larval aggregation in the butterfly Byasa alcinous bradanus (Lepidoptera: Papilionidae). Entomol Sci 12:135-140.

Keeler MS, Chew FS (2008). Escaping an evolutionary trap: preference and performance of a native insect on an exotic invasive host. Oecologia 156:559-568.

24

Kelber A (1999). Ovipositing butterflies use a red receptor to see green. J Exp Biol 202:2619-2630.

Lee JK, Selzer R, Altner H (1985). Lamellated outer dendritic segments of a chemoreceptor within wall-pore sensilla in the labial palp-pit organ of the butterfly, Pieris rapae L. (Insecta, Lepidoptera). Cell Tissue Res 240:333-342.

Matsubayashi KW, Ohshima I, Nosil P (2010). Ecological speciation in phytophagous insects. Entomol Exp Appl 134:1-27.

McDonald G (1991). Oviposition and larval dispersal of the common armyworm, Mythimna convecta (Walker) (Lepidoptera, Noctuidae). Austral J Ecol 16:385-393.

Mozūraitis R, Radžiutė S, Apšegaitė V, Cravcenco A, Būda V, Nylin S (2016). Volatiles released from foliar extract of host plant enhance landing rates of gravid Polygonia c-album females, but do not stimulate oviposition. Entomol Exp Appl in press.

Murphy DD, Menninger MS, Ehrlich PR (1984). Nectar source distribution as a determinant of oviposition host species in Euphydyras chalcedona. Oecologia 62:269-271.

Nakajima M, Boggs CL, Bailey S, Reithel J, Paape T (2012). Fitness costs of butterfly oviposition on a lethal non-native plant in a mixed native and non-native plant community. Oecologia 172:823-832.

Nylin S (1988). Host plant specialization and seasonality in a polyphagous butterfly, Polygonia c-album (Nymphalidae) Oikos 53:381-386.

Nylin S, Slove J, Janz N (2014). Host plant utilization, host range oscillations and diversification in nymphalid butterflies: a phylogenetic investigation. Evolution 68:105-124.

Rausher MD (1978). Search image for leaf shape in a butterfly. Science 200:1071-1073. Reed RD (2005). Gregarious oviposition in butterflies. J Lep Soc 59:40-43. Root RB, Kareiva PM (1984). The search for resources by cabbage butterflies (Pieris

rapae): ecological consequences and adaptive significance of Markovian movements in a patchy environment. Ecology 65:147-165.

Rosenheim JA, Rosen D (1991). Foraging and oviposition decisions in the parasitoid Aphytis lingnanensis: distinguishing the influences of egg load and experience. J Anim Ecol 60:873-893.

Rospars JP (1983). Invariance and sex-specific variations of the glomerular organization in the antennal lobes of a moth, Mamestra brassicae, and a butterfly, Pieris brassicae. J Comp Neurol 220:80-96.

Saveer AM, Kromann SH, Birgersson G, Bengtsson M, Lindblom T, Balkenius A, Hansson BS, Witzgall P, Becher PG, Ignell R (2012). Floral to green: mating switches moth olfactory coding and preference. Proc R Soc B - Biol Sci 279:2314-2322.

Scott JA (1986). The butterflies of North America: a natural history and field guide. Stanford, California: Stanford University Press.

Sillén-Tullberg B (1988). Evolution of gregariousness in aposematic butterfly larvae: a phylogenetic analysis. Evolution:293-305.

25

Singer MC, McBride CS (2012). Geographic mosaics of species' association: a definition and an example driven by plant–insect phenological synchrony. Ecology 93:2658-2673.

Singer MC, Wee B, Hawkins S, Butcher M (2008). Rapid natural and anthropogenic diet evolution: three examples from checkerspot butterflies. In: Tilmon KJ, editor. Specialization, Speciation, and Radiation - The evolutionary biology of herbivorous insects Berkeley: University of California. p. 311-324.

Snell-Rood EC, Papaj DR (2009). Patterns of phenotypic plasticity in common and rare environments: a study of host use and color learning in the cabbage white butterfly Pieris rapae. Am Nat 173:615-631.

Stamp NE (1980). Egg deposition patterns in butterflies: why do some species cluster their eggs rather than deposit them singly? Am Nat 115:367-380.

Stefanescu C (1997). Migration patterns and feeding resources of the painted lady butterfly, Cynthia cardui (L.) (Lepidoptera, Nymphalidae) in the northeast of the Iberian peninsula. Misc Zool 20:31-48.

Stefanescu C (2001). The nature of migration in the red admiral butterfly Vanessa atalanta: evidence from the population ecology in its southern range. Ecol Entomol 26:525-536.

Stocker RF (1994). The organization of the chemosensory system in Drosophila melanogaster: a review. Cell Tissue Res 275:3-26.

Symons FB, Beccaloni GW (1999). Phylogenetic indices for measuring the diet breadths of phytophagous insects. Oecologia 119:427-434.

Taylor K (2009). Biological Flora of the British Isles: Urtica dioica L. J of Ecol 97:1436-1458.

Thomas J, Lewington R (2010). The butterflies of Britain and Ireland. Dorset: British Wildlife Publishing Ltd.

Thöming G, Larsson MC, Hansson BS, Anderson P (2013). Comparison of plant preference hierarchies of male and female moths and the impact of larval rearing hosts. Ecology 94:1744-1752.

Wahlberg N, Leneveu J, Kodandaramaiah U, Peña C, Nylin S, Freitas AVL, Brower AVZ (2009). Nymphalid butterflies diversify following near demise at the Cretaceous/Tertiary boundary. Proc R Soc Lond B Biol Sci 276:4295-4302.

Wiklund C (1975). The evolutionary relationship between adult oviposition preferences and larval host plant range in Papilio machaon L. Oecologia 18:185-197.

Wiklund C (1977). Oviposition, feeding and spatial separation of breeding and foraging habitats in a population of Leptidea sinapsis (Lepidoptera). Oikos 28:56-68.

Vosshall LB, Wong AM, Axel R (2000). An olfactory sensory map in the fly brain. Cell 102:147-159.

Zalucki MP, Clarke AR, Malcolm SB (2002). Ecology and behavior of first instar larval Lepidoptera. Annu Rev Entomol 47:361-393.