H I D D E N C R E A T U R E S – S Y S T E M A T I C S O F T H E E U P H O R I N A E

Julia Stigenberg

Hidden creatures – systematics of the Euphorinae (Hymenoptera)

Julia Stigenberg

©Julia Stigenberg, Stockholm 2013 Front cover: Illustration modified from Curtis 1840. Zele albiditarsus (left) with Smyrinum olusatrum (Alexanders, Alexanderloka), Leiophron apicalis (right) with Lysimachia nummularia (Money wort, Penningblad) ISBN 978-91-7447-605-7 Printed in Sweden by US-AB, City 2013 Distributor: Department of Zoology, Stockholm University

ABSTRACT

Parasitic wasps constitute one of the last remaining frontiers in the charting of animal diversity. For instance, parasitic wasps comprise about one third of those Swedish animal species that are so poorly known that they cannot be evaluated against the red-list criteria established by the IUCN (International Union for Conservation of Nature). The Braconidae is the second most spe-cies-rich family of parasitic wasps; the world fauna has been estimated at 40 000 species and the Swedish fauna is believed to include a little more than 2 000 species, 1 200 of which are currently documented. This thesis is a con-tribution to the rapidly increasing knowledge of braconid diversity. In paper I, a new gregarious parasitoid, Meteorus acerbiavorus sp. nov. (Braconidae: Eupohrinae), is described from specimens reared from the cocoons of the butterfly Acerbia alpina (Quensel) (Lepidoptera, Arctiidae) in northwestern Finnish Lapland. Based on a molecular phylogenetic analysis, the new spe-cies is shown to belong to the M. rubens species group. In the second paper, the scope is expanded to the two genera included in the euphorine tribe Me-teorini, Meteorus and Zele. More than 300 species are known, about one fifth of which occur in the Western Palearctic. The Western Palearctic fauna of the tribe is revised, seven new species are described from Fennoscandian material, and a key to the Western Palearctic species is presented. Two mo-lecular markers, 28S and COI, are used to study phylogenetic relationships in the tribe and test previously proposed species groups. The Bayesian anal-ysis of the combined molecular data showed that the Meteorini fall into four well supported clades. The results confirm the monophyly of some species groups, while others are shown to be assemblages of unrelated lineages. The results also reveal considerable cryptic species diversity in the Fennoscandi-an material. Because we were not able to find consistent morphological dif-ferences separating the 12 cryptic molecular species from their closest rela-tives, they are not formally described. The third paper deals with distribu-tional, phenological and in many cases rearing data from nearly 2 500 spec-imens (44 species) of the Meteorini in the collection of the National Museums of Scotland (NMS), Edinburgh. Patterns in the breadth of host ranges are discussed in relation to a reiterated speciation hypothesis and in the light of the molecular phylogeny presented in paper II. Paper IV exam-ines the phylogenetic relationships of the entire subfamily Euphorinae based upon four gene regions (18S, CAD, 28S D2, and COI). The data set included representatives from 43 of the 48 previously recognized extant genera. Based

on the results, a revised classification of the Euphorinae is proposed that recognizes 55 genera and 14 tribes. We propose that four lineages, whose relationships have been contentious previously, should all be included as tribes in the Euphorinae: Meteorini, Planitorini, Cenocoelini and Ecnomiini. Our results necessitate the recognition of a new euphorine tribe, Pygostolini, and we show that the much-discussed genus Myiocephalus belongs in the Syntretini. With respect to host preferences, our analysis shows that early members of the Euphorinae were parasitoids of coleopteran larvae, with a host shift to larval Lepidoptera occurring early in the evolution of the Mete-orini. The sister lineage of Meteorini, producing the bulk of euphorine spe-cies, started parasitizing adult coleopterans. Some descendant lineages later shifted to adult or immature hosts in the Hemiptera, Hymenoptera, Neurop-tera, Orthoptera and Psocoptera. Keywords: Parasitic wasps, Braconidae, Euphorinae, systematics, phyloge-ny, molecular taxonomy, cryptic species, evolution, host preferences, identi-fication keys.

LIST OF PAPERS

This thesis is based on the following papers, referred to in the text by their roman numerals: I Stigenberg, J., Vikberg, V., Belokobylskij, S.A. 2011 Meteorus acer-biavorus sp. nov. (Hymenoptera, Braconidae), a gregarious parasitoid of Acerbia alpina (Quensel) (Lepidoptera, Arctiidae) in North Finland. Journal of Natural History, 45:1275–1294. II Stigenberg, J., Ronquist, F. 2011 Revision of the Western Palearctic Meteorini (Hymenoptera, Braconidae), with a molecular characterization of hidden Fennoscandian species diversity. ZOOTAXA, 3084:1-95. III Stigenberg, J., Shaw, M.R. 2013 Western Palaearctic Meteorinae (Hymenoptera; Braconidae) in the National Museums of Scotland, with rear-ing and distributional data, including six species new to Britain, and a dis-cussion of a potential route to speciation. Entomologist’s Gazette, (In press) IV Stigenberg, J., Boring, C.A. Phylogeny of the parasitic wasp subfami-ly Euphorinae (Braconidae) and evolution of its host preferences. Manu-script. Reprints were made with permission from the respective publishers. Disclaimer: The nomenclatural acts presented in papers III and IV herein are not intended to have any effect on zoological nomenclature (§8.2, ICZN, 1999). Paper I © Thomson Reuters Paper II © Magnolia Press

CONTENTS

INTRODUCTION ........................................................................................... 1  Important footprints - early Swedish Hymenopterists ..................................................... 2  Evolution of the Hymenoptera and the origin of hymenopteran parasitoids ................... 3  The parasitic wasp family Braconidae ............................................................................ 6  General outline of Braconidae ........................................................................................ 7  

Early stages ............................................................................................................ 10  Host ranges and speciation events amongst Braconidae ....................................... 12  Adult parasitism: evolutionary origin and adaptations ............................................. 13  

The Euphorinae ............................................................................................................ 14  Evolution of adult parasitism in the Euphorinae ...................................................... 14  

Collecting ...................................................................................................................... 17  The Swedish Taxonomy Initiative ........................................................................... 18  

Aims of the Thesis Research ........................................................................................ 19  

MATERIAL AND METHODS ....................................................................... 20  DNA-workflow ............................................................................................................... 20  Phylogenetic Analyses ................................................................................................. 21  Morphological Study and Illustrations ........................................................................... 23  

SUMMARY OF PAPERS ............................................................................. 24  Paper I .......................................................................................................................... 24  Paper II ......................................................................................................................... 25  Paper III ........................................................................................................................ 28  Paper IV ........................................................................................................................ 28  

DISCUSSION ............................................................................................... 30  

LITTERATURE CITED ................................................................................. 32  

ACKNOWLEDGEMENTS ............................................................................ 36  

SVENSK SAMMANFATTNING .................................................................... 38  

1

INTRODUCTION

All life on earth is connected through common ancestry, commonly illustrat-ed as the tree of life. This implies that we, humans, share genes not only with our closest relatives, such as the gorilla, but also with the most distant of life forms, such as the bacteria (Futuyma 2009). Learning how to climb the tree of life is the difficult part. First we are required to study the tree and try and understand the relationships along the branches – this is systematics. Sys-tematics includes phylogenetics, which aids us in finding out how species are related to each other, and taxonomy, which helps us describe what is out there. Without the taxonomy (names on taxa) and systematics (the classifica-tion of taxa) we wouldn’t be able to understand the relationships of organ-isms, everything would be a profound disorder. In short, taxonomy is the naming of organisms, the identification, what classifies them. Taxonomy is governed by the rules of nomenclature. For animals, the rules are found in the International Code of Zoological Nomen-clature (ICZN 1999). Phylogenetics produces phylogenies, which are tree-like diagrams illustrating hypothesized evolutionary relationships. The data (shared characters) originate from molecular sequences, morphological char-acters or behavioural data, and they are typically analysed using statistical methods. The phylogenies form the basis for the classification of organisms. This thesis is about the taxonomy, systematics and biology of parasitic Hymenoptera. Almost anyone recognizes ants, bees, yellow jackets, hornets and bumble bees, which are all hymenopterans that menace or frighten peo-ple during their summer vacation, when people are more outdoors than in-doors. But what people don’t know is that these vernacular names apply to hundreds of species and that there are tens of thousands more hymenopteran species quietly going around minding their business unnoticed, many of them still unnamed. In fact, members of the insect order Hymenoptera form one of the most abundant groups of insects, about 10% of all known species on our planet (Austin et al. 2005) and have ruled successfully on this earth for millions of years. The most successful hymenopteran group is the para-sitic wasps. They are often labelled parasitoids, meaning that they are organ-isms that feed on a single host individual, which is eventually killed. Thus, they are ecologically intermediate between parasites, which usually do not kill their host, and predators, which feed on several individuals (prey). Para-sitoids occur in the insect orders Diptera, Coleoptera, Lepidoptera, Trichop-tera, Neuroptera and Strepsiptera but the vast majority belong to the Hyme-

2

noptera (Quicke 1997). In this short introduction I hope to give you a glimpse of the lives of these hidden creatures that surround us. A teaser, so that you also will start looking closer at the insects -those that really rule the world. With other words: let’s start climbing!

Important footprints - early Swedish Hymenopterists Carl von Linné (1707–1778) is undoubtedly Sweden’s most important and famous naturalist. Apart from being the father of modern taxonomy, he also described 70 species of parasitic wasps in the superfamily Ichneumonoidea, 14 of them now considered to belong to the family Braconidae, the group studied in this thesis (Linnaeus 1758, Yu et al. 2011). Other Swedish ento-mologists that also made important contributions to Hymenoptera research include DeGeer (1720–1778), Thunberg (1743–1828), Fallén (1764–1830), Dalman (1778–1828), Boheman (1796–1868), Dahlbom (1806–1859), Thomson (1824–1899), who set the standard for “modern” Hymenoptera collections in Sweden, Holmgren (1829–1888), Aurivillius (1853–1928), Adlerz (1858–1918), Roman (1872–1943), Malaise (1892–1978), Erlands-son (1902–1989), Hedqvist (1917–2009), and Huggert (1942–2003). In total, roughly 40 Swedish taxonomists have worked with Hymenoptera through the years (M. Forshage pers. corr.).

The Swedish entomologists were not only taxonomists, some also studied and contributed important information on the biology of parasitic wasps. The foremost among these was August E. A. Holmgren (1829–1888), who is called the father of Swedish applied entomology. During the summers of the 1850’s, he collected insects from all over Sweden and described 39 new parasitic wasps. He published literature that was used for 50 years at “Skog-sinstitutet”, (nowadays incorporated in the Swedish University of Agricul-tural Science, SLU). One of these books was ‘For trees and bushes beneficial and harmful insects’ (original title: De för träd och buskar nyttiga och skadliga insekterna). He also made birdhouses common in the 1870’s by turning the opinion on so-called non-useful small passerine birds. He was mostly interested in the ecology of insects and blamed other entomologists for only studying dry specimens and arguing over the rule of priority. His expedition to Svalbard and Bjørnøya led not only to findings of new species, but also to new insights into the relationships between plants and insects in arctic areas. His research even attracted the interest of the Frenchmen, who wanted him to find a cure for the “Phylloxera plague”, a grape wine pest that destroyed most vineyards in Europe. Though he was unable to help them in their despair, back in Sweden he became hired as an expert on insect biolog-ical control for agriculture and forestry (Franzén 1973). Another man worth mentioning is Dr and Associate Professor Simon Fredrik Bengtsson (1860–1939). Like Holmgren, Simon Bengtsson travelled through Sweden to make

3

new zoological findings. Bengtsson was more tied to the academic world, especially to Lund University, where he had shifting positions such as teach-er and as curator and manager of the Entomological Department of the Zoo-logical Institution. Bengtsson was most interested in the insect order Ephem-eroptera (May-flies), where he tripled the number of described species, but he also studied other orders such as Hymenoptera. Besides anatomical and morphological studies he was very interested in the life cycles of insects and issues concerning pest control. He was hired to do biological research to investigate the severe outbreak of the Black Arches Moth (or Nun) (Lymantria monacha) in Sweden during 1898-1902 (Aurivillius 1922). An-other great researcher focusing on pests, parasitic wasps, and forestry recent-ly passed away: Karl-Johan Hedqvist (1917–2009). He mostly travelled around Sweden as a field entomologist, studying pests such as the Pine Saw-fly and the Black Arches Moth. During these field expeditions he collected a great deal of other insects. He was known to disappear from his office at “Skogsvårdsstyrelsen”, (nowadays incorporated in the Swedish University of Agricultural Science, SLU) into the field during spring and reappear again come autumn. He was later hired as a taxonomist at the Swedish Museum of Natural History (NRM) mainly focusing his research on parasitic wasps (Hansson 2010). Just a couple of months before his death I had the oppor-tunity to meet him on one of his few visits to NRM. He then told me that he was currently working on a Swedish catalogue of the Braconidae. Unfortu-nately, the catalogue was not finished before his death.

Evolution of the Hymenoptera and the origin of hymenopteran parasitoids The Hymenoptera are holometabolous insects (also known as endopterygote insects). This means that their development involves four distinct life stages: egg, larva, pupa and adult. The transition from pupa to adult is particularly striking and is referred to as metamorphosis. It has been clear for some time that the Hymenoptera are among the basal lineages of the Holometabola but their exact position is somewhat unclear. The prevailing hypothesis, based primarily on morphological evidence, is that they are the sister to the Mecop-terida containing all Holometabola except the Coleoptera (beetles) and Neu-ropterida (ant lions and related forms) (Kristensen 1975, 1999). However, more recent molecular analyses place them as the sister group to all other holometabolous insects, including the Coleoptera and Neuropterida (Whit-field & Kjaer 2008, Wiegmann et al. 2009, Meusemann et al. 2010) (Fig. 1).

4

Mollusca

Annelida

Tardigrada

Nematoda

Onychophora

Myriapoda

Pycnogonida

Ixodida

Xiphosura + AraneaeAstigmata

Amphipoda

Decapoda

CirripediaCopepoda

Branchiopoda

Nonoculata

Collembola

ArchaeognathaOdonata

Ephemeroptera

Hemiptera

Phtiraptera“Orthoptera”

Isoptera + “Blattaria”

Hymenoptera

Coleoptera

Lepidoptera

Diptera0,2

Pterygota

EntognathaHexapoda

Pancrustacea

Euarthropoda

Ectognatha

Hex

apod

aCr

usta

cea

Chel

icer

ata

Endopterygota

Figure 1. Results from a maximum likelihood analysis of 117 taxa and 129 genes to resolve the arthropod tree of life (simplified from Meusemann et al. 2010). Note the placement of the Hymenoptera, at the base of the Holometabola (Endopterygota).

5

Traditionally, the Hymenoptera have been divided into the suborders Symphyta (sawflies) and Apocrita (wasps, ants, bees, parasitic wasps). Ex-cept for the unusual family Orussidae, the Symphyta are all phytophagous (vegetarians) as larvae, while the vast majority of the Apocrita have carnivo-rous larvae. It is now considered firmly established that the Symphyta are paraphyletic with respect to the Apocrita and that the phytophagous lifestyle is ancestral in the Hymenoptera. The first Hymenoptera appeared in the fossil record in the Triassic (225 million years ago). The earliest known hymenopteran fossils are from the still extant family Xyelidae, which belongs to the Symphyta (Gauld 1988). During the Jurassic, when the vegetation only consisted of non-flowering plants (conifers, ferns, etc), the hymenopteran fauna was dominated by the sawflies. Many of the extant sawflies are still associated with these non-flowering plants. As mentioned above, the Apocrita apparently arose from a symphytan lineage. The symphytan families Orussidae and Cephidae have both been proposed to be the sister group of the Apocrita (Gibson 1985, Withfield 1998, Viitasaari 2002), with more recent analyses favouring the Orussidae, particularly those based on morphology (Gibson 1985, Withfield 1998, Heraty et al. 2011, Sharkey et al. 2011). The Orussidae are unique among symphytans in being parasitoids, typically of wood-boring beetle larvae, but they lack the wasp waist that is characteristic of the Apocrita. The Apocrita includes familiar groups, such as the ants, wasps and bees but also parasitic wasps. Members of the Apocrita occupy a vast array of feeding niches, in-cluding gall inducers, predators, and pollen and nectar feeders, among oth-ers. However, the Apocrita are clearly primitively parasitoids and most cur-rent species belong to one of the lineages of parasitic wasps, most of which are still parasitoids. The hypothesis of a sister-group relationship between the Apocrita and the Orussidae thus suggests that the parasitoid life-style evolved just once within the Hymenoptera. Gauld (1988) names two features that are likely to have played a key role in hymenopteran evolution: (1) the retention and refinement of the ancestral holometabolan structure of the ovipositor, allowing the eggs to be injected into the food substrate; and (2) the endophytic larvae, common in the saw-flies and presumably facilitating the shift to a parasitoid life style in the Ap-ocrita. A key innovation that may have contributed to the spectacular radia-tion of the Apocrita is the wasp waist, permitting greater mobility of the ovipositor. The ectoparasitic idiobiont strategy (paralysing the host with venom and placing an egg on the cuticle of the host) is the ground plan of the Apocrita and is still known among the members of the group (Whitfield 1998). Venom is widely used amongst extant Apocrita and is produced in glands associated with the ovipositor in all Hymenoptera. The fine-tuning of the venom to being a means to modify the physiology of the host clearly is an important evolutionary step in the evolution of hymenopteran parasitoids.

6

The parasitic wasp family Braconidae Parasitic wasps comprise about one third of the Swedish animal species that are so poorly known that it is impossible to apply the red-listing criteria es-tablished by the IUCN. The Braconidae are the second most species-rich family of parasitic wasps; the world fauna has been estimated at 40 000 spe-cies and the Swedish fauna is believed to include a little more than 2 000 species; about 1 000 are currently known. Linné described the first valid braconid names in his 10th edition of Systema Naturae in 1758. In 1763 Sco-poli described 125 hymenopterans of which he categorized 28 as belonging to the genus Ichneumon, including several species that are now placed in the Braconidae. During 1700–1800, most braconid authors published a single, major book containing multiple descriptions of species belonging to several unrelated insect groups. During 1800–1900 the publications increased slight-ly and the number of published works increased to at most 10 per year in 1898. During the 20th century, until the end of the world wars (1945), the number of published braconid works averaged around 13 publications a year. After the wars, publications increased with a peak in 2004, when 56 publica-tions containing descriptions of new braconid species appeared. Nowadays, a description is not contained in a hardcover book as in the 18th or 19th century; the majority are single articles with a narrow focus field, published in scien-tific journals. On-line publication is becoming increasingly common. Though it seems that the number of publications has been decreasing in re-cent years (Fig. 2), there are still thousands of undescribed species and it is unlikely that the task is nearing completion.

Figure 2. Number of publications of original descriptions of braconids through time.

Data from Yu et al. (2011).

It is difficult for braconid workers today to handle the vast and rapidly in-creasing literature. Fortunately, there are several catalogues updated with

7

regular intervals that make their life a little easier, especially that of taxono-mists. In 1965 Shenefelt published a catalogue with over 8 000 titles and references to Braconidae literature, which was a milestone in the biblio-graphical knowledge of the family. In 2006, Ghahari, Yu and van Achter-berg presented an up-dated catalogue of the superfamily Ichneumonoidea in the form of a database, “Taxapad”. The database today contains 18 089 ref-erences concerning Braconidae (Yu et al. 2011). Hopefully, it will be possi-ble to continue updating Taxapad to fulfil the needs of researchers, making it easier for them to search and retrieve relevant published data. This would improve the quality of scientific studies, the foundation of political deci-sions.

As pointed out above, analysis of the Taxapad database shows that the number of published papers describing new braconid species has decreased in recent years (Fig. 2). This may be due to a general decline in political interest and funding of taxonomic research, which would be extremely un-fortunate. When analysing the overall number of published works on Braco-nidae from 1964 to 2003 (Ghahari et al. 2006), a similar decline can be noted but only in the last few years preceding 2003, and this could possibly be due to incomplete data sampling for that time period (Fig. 3). Unfortunately, similar data are not available for the time period after 2003.

5 Ghahari et al. Bibliography of the family Braconidae (Hymenoptera). NNM Tech. Bull. 8 (2006)

& Mason, 1994). Several subfamilies (e.g., Aphidiinae, Alysiinae, Apozyginae) have at one time or another been recognized as separate families. In general there is agreement on the basic subfamily or tribal groupings, with the exception of the “hormiine” and “exothecine” groups of genera (Whit!eld, 1992; Quicke, 1993; Wharton, 1993), but disagreement on the ranking or inclusiveness of some groups, since the phylogenies of the subfamilies are not yet robust. There is a perception that the number of recognized subfamilies has become in"ated, while the tribal rank has been underutilized. An attempt has been made to address this problem (Wharton, 2001), but requires further phylogenetic testing.

Identi!cation guides

Keys to the world subfamilies and general introductions to the literature are available in van Achterberg (1993) and Sharkey (1993). A new manual to the genera of the New World has been produced (Wharton et al., 1997 - Spanish version 1998), including individual subfamily chapters contributed by a number of the world's braconid specialists. Interactive versions of the keys in this manual are available online; van Achterberg (1997) has published the CD-ROM Braconidae: An illustrated key to all subfamilies.

Bibliography of Braconidae

The publication of Roy Shenefelt's catalogue (Shenefelt, R.D., 1965. A contribution towards knowledge of the world literature regarding Braconidae (Hymenoptera: Braconidae). Beiträge zur Entomologie 15: 243-500) was a milestone in the bibliographical knowledge of the family Braconidae. Over 8,000 titles and references to the litera-ture on Braconidae between 1758 till 1963 were included. The first author attempted to make an up-date; this was augmented by additional references present in the TAXAPAD database made by the second author. It resulted in a list of 10,431 titles concerning Braconidae for the period 1964-2003. Peculiar is the rapid growth of the amount of papers in the period covered by this bibliography. It quadrupled from about 100 papers per year in the sixties to about 400 papers in the nineties (fig. 2). This list is published on CD-ROM to make it accessible to all workers on Braconidae, also those who can not afford to buy the strongly recommended interactive TAXAPAD world catalogue on Ichneumonoidea (Yu, D.S., van Achterberg, K. (Braconidae) & Horstmann, K. (Ichneumonidae), 2005. Ichneumonoidea. Taxapad Biological Taxo-nomical Interactive CD-ROM), which appeared this year. The bibliography on Braconidae published here is useful also when the interactive catalogue is available because it makes the titles available in another format and can be searched for taxa and authors in another way. The authors hope that this bibliography will facilitate and probably augment the research on Braconidae.

Fig. 2. Publications on Braconidae 1964-2003.

500

450

400

350

300

250

200

150

100

50

0 1964 68 72 76 80 84 88 92 96 2000 Figure 3. The total number of published works on Braconidae from 1964–2003

(Ghahari et al. 2006).

General outline of Braconidae The Braconidae are one of the most diverse families of parasitic wasps. In terms of species richness, braconids are second only to the Ichneumonidae among families of animals. The Braconidae are divided into 46 subfamilies,

8

about 25 of which occur in Sweden (Yu et al. 2011). Braconids attack most insect orders and form an important component of most terrestrial and aquat-ic (Bennett 2007) food webs where they also provide important ecosystem services, making them interesting as biological control agents. They are also excellent objects for studies on biodiversity, conservation, evolution of para-sitism and host-parasitoid relations.

Figure 4. Two species of Euphorinae: Zele albiditarsus (left) and Leiophron apicalis

(right). Modified from Curtis (1840).

Braconids are free living/flying as adults and develop as larvae on or in-side of another insect – a host. For braconids this host is usually immature, a larva or a nymph, though the adult stage is attacked by at least two subfami-lies (Aphidiinae and Euphorinae). Braconids have been divided into cyclo-stomes and non-cyclostomes based on the morphology of the clypeal area of the lower face. Recently Sharanowski et al. (2011) presented an up-dated molecular phylogeny of the Braconidae (Fig. 5) showing more detail than previous phylogenetic analyses (Dowton et al. 2002, Shi et al. 2005).

9

Andesipolis sp.Aspilodemon sp.

Hydrangeocola sp.Ephedrus sp.

Pseudopraon sp.Maxfischeria ovumancoraMaxfischeria anic Maxfischeria folkertsorumMaxfischeria ameliaeMaxfischeria tricolor

Acrisis sp.Oncophanes sp.

Histeromerus canadensisDolopsidea sp.

Liobracon sp.Pambolus sp.Pseudorhysipolis sp.

Rhysipolis sp.1Rhysipolis sp.2

Hormius sp.Conobregma sp.

Clinocentrus sp.1Clinocentrus sp.2

Aleiodes sp.Cystomastax sp.

Macrostomion sp.Doryctes anatolikus

Doryctes sp.Notiospathius sp.

Leluthia sp.Heterospilus sp.2

Heterospilus sp.1Polystenidea sp.

Bracon sp.Vipio texanus

Hemibracon sp.Cyanopterus sp.

Pseudognaptodon sp.Shawiana sp.

Hoplitalysia slossonaeOpius sp.

Colastes sp.Meteoridea sp.1

Meteoridea sp.2Minanga serrata

Earinus limitarisCremnops montrealensis

Bassus annulipesIchneutes sp.

Proterops nigripennisMuesonia straminea

Phanerotoma sp.Epsilogaster sp.

Khoikhoia sp.Miracinae Gen. sp.

Cardiochiles sp.Fornicia sp.

Microplitis sp.Snellenius sp.

Capitonius chontalensisPerilitus sp.

Ecnomios sp.Leiophron sp.

Meteorus sp.1Meteorus sp.2

Euphorinae Gen. sp.1Meteorus sp.3

Kollasmosoma sp.Planitorus sp.

Mannokeraia sp.3Mannokeraia sp.1

Mannokeraia sp.2Urosigalphus sp.1

Urosigalphus sp.2Urosigalphus sp.3

Stantonia sp.Orgilus sp.

Homolobus sp.Microtypus wesmaelii

Charmon cruentatusAmicrocentrum concolorXiphozele sp.

Macrocentrus sp.Hymenochaonia sp.

Topaldios sp.Helconini Gen. sp.1

Helcon sp.7Helcon sp.5

Calohelcon sp.Ussurohelcon nigricornis

Helcon sp.6Austrohelcon inornatus

Helcon sp.3Helcon sp.4

Eumacrocentrus americanusWroughtonia sp.1

Helcon texanusHelcon tardatorWroughtonia ligatorWroughtonia ferruginea

Wroughtonia sp.4Vadumasonium sp.1 Diospilini Gen. sp.1

Apoblacus sp.Taphaeus sp.1

Brulleia sp.Grypokeros sp.1Grypokeros sp.2

Blacus sp.1Blacus sp.21.00

Diospilini Gen. sp.3Flavihelcon distanti

Schauinslandia sp.2Baeacis sp.1

Baeacis sp.2Diospilus sp.5

Diospilus sp.2Diospilus sp.3Diospilus sp.4

Eubazus (Allodorus) sp.4Eubazus (Calyptus) sp.2

Eubazus (Aliolus) sp.9Eubazus sp.1

Nealiolus sp.Eubazus (Brachistes) sp.5 1.00

Eubazus (Aliolus) sp.7Eubazus (Aliolus) sp.8

Eubazus (Brachistes) sp.6Eubazus (Aliolus) sp.3

Schizoprymnus sp.1Schizoprymnus sp.3Triaspis sp.1Schizoprymnus sp.2

Triaspis sp.20.01

*

**

*

*

*

*

*

***

*

***

*

*

***

**

*

**

**

**

**

**

*

**

**

**

***

*

*

*

* *

**

**

**

**

***

*

*

*

*

*

***

**

***

*

*

***

*

*

*

*

*

*

***

* **

* *

* * *

*

*

**

*

*

*

*

*

*

**

***

*

* *

**

*

* *

*

*

*

*

***

*

*

*

*

**

* *

*

**

* *

*

**

*

** *

*

**

*****

*

**

*

**

** * **

*

Aphidioid

Rogadinae*

Non-

Cyclostomes

Sigalphoid

Meteorideinae

Microgastroid

Euphoroid

Helconoid

Rogadinae*

Brachistini

Helconini

Pambolinae

Doryctinae*

Doryctinae*

Braconinae

Rhyssalinae

Rhysipolinae

Acampsohelconinae

Blacinae*

Diospilini*

Brulleiini*

Maxfischeriinae

Mesostoinae

Ichneutinae*

Aphidiinae

Cenocoeliinae

Macrocentroid

subcomplex

Alysioid

subcomplex

Cyclostomes

Dusona  sp.

Zagryphus  nasutus* *

*

* *

*

Figure 5. Most recent Bayesian phylogeny of the braconid subfamilies based on a

concatenated gene dataset (Sharanowski et al. 2011).

Braconids attack hosts from more than 120 insect families (Wharton

1993). Even though the majority of the braconid subfamilies are restricted to only one host order (usually Lepidoptera), there are exceptions that attack several host orders, such as the Euphorinae, Braconinae and Doryctinae. Braconid larvae can develop inside the host (endoparasitic species) or they

10

can feed on the host from the outside (ectoparasitic species). Examples of braconids that are endoparasitic parasitoids of Lepidoptera are: Adeliinae, Agathidinae, Cheloninae, Gnamptodontinae, Homolobinae, Macrocentrinae, Microcentrinae, Miracinae and Orgilinae. Examples of more unusual host order associations include the subfamily Aphidiinae that use aphids, the Alysiinae and Opiinae that are parasitoids of Diptera, and the Cenocoeliinae and Helconinae that are endoparasitoids of Coleoptera. There are also phy-tophagous (plant-eating) braconids. Phytophagy does not occur so frequently among the braconids though a number of species utilizing plant hosts can be found in the species-rich subfamilies Braconinae and Doryctinae (Perioto et al. 2011).

Almost all ectoparasitic braconids attack concealed hosts; endoparasitoids on the other hand are equally divided between exposed and concealed hosts (Wharton, 1993). Long ovipositors seem to be an attribute of endoparasitoids that attack concealed hosts. Several species within these subfamilies have long ovipositors: Braconinae, Doryctinae, Orgilinae, Helconinae, Agathi-inae, Microgastrinae, Opiinae and Alysiinae (Wharton 1993). Some females drill or push the ovipositor through the substrate whilst others probe existing holes and tunnels to reach a host. When drilling through hard wood, the pro-cedure can take up to three hours (Ryan 1962). Braconids attack different host stages: eggs, larvae/nymphs and adults. Egg-larval parasitism is known in several subfamilies, such as the Cheloninae, Alysiinae, Helconinae, Ich-neutinae, Microgastrinae and Opiinae. Larval parasitism occurs amongst for instance Braconinae, Doryctinae, Rogadinae, Homolobinae and Euphorinae. Adult parasitism occurs amongst the Braconidae only within the subfamilies Aphidiinae and the Euphorinae.

There are many important factors that determine the success of the parasi-toid-host relationship. These have been grouped according to five separate events or steps in the life history of a parasitoid: host habitat location, host location, host acceptance, host suitability and host regulation (Vinson 1980). The fifth step, host regulation, concerns whether the parasitoid has the ability to regulate the physiology of the host, so that the parasitoid can develop suc-cessfully within the host. Both chemical and vibratory cues are important to various degrees in several of the above-mentioned steps. Much research has been focused on these events and this literature is summarized in the paper by Wharton (1993) and the handbook by Shaw & Huddleston (1991).

Early stages Parasitic wasps are either pro-ovigenic, (the female has almost all her eggs ready for oviposition at emergence) or synovigenic (the eggs mature succes-sively during the female adult life) (Shaw & Huddleston 1991, Wharton 1993, Quicke 1997, Shaw 2006). In pro-ovigenic species, the eggs are small because of little or no yolk and are placed inside a living host. Already in the

11

egg stage, the parasitoid takes advantage of its host. The parasitoid egg ab-sorbs nutrients in the haemolymph of the host body, through the eggshell. After hatching, the first larval stage can start to actively feed on the host.

Pro-ovigenic species are usually koinobionts and often have a relatively short but active adult life. Therefore, sugar is very important as an energy resource to them (Shaw & Huddleston 1991, Wharton 1993). Synovigenic species are usually long-lived and require protein and sugar: protein to fill the typically large eggs with yolk and sugar to sustain the longevity. Protein sources sometimes include pollen but often the female wasps get their pro-tein by feeding on the haemolymph from their hosts while ovipositing (Jervis & Kidd 1986). The parasitoid larvae can be either feeding on the host inter-nally (an endoparasitoid) or externally (an ectoparasitoid). The host is al-ways killed one way or the other. Species that kill or paralyze the host during the initial attack are called idiobionts. Idiobionts are often considered as generalists. The host of a koinobiont parasitoid, in contrast, continues its development after being attacked (Askew & Shaw 1986). Koinobionts are considered as highly specialized and usually overwinter as larvae inside the diapausing host, being physiologically adapted to the life cycle of the host. Without an appropriate host at the appropriate time the parasitic wasp may not be able to complete its life cycle (Shaw, 2006). The parasitoid larvae can be either gregarious, developing together with many siblings on the same host, or solitary, not needing to compete with its siblings for the food.

An endoparasitoid may be nicely tucked away inside the host, with the scary world left outside, but it has to evade several physiological defences of the host to develop successfully (Vinson & Iwantch 1980, Quicke 1997). The majority of Braconidae are koinobiont endoparasitoids, though several members of the subfamily end their last development stage as ectoparasi-toids (Shaw & Huddleston 1991). The final-instar larva emerges from the body of the host and completes its feeding externally. After the emergence from the host, the parasitoid larva spins a protective cocoon, within which it pupates. In paper I we discuss the life history of a gregarious parasitoid, which attacks a moth (Acerbia alpina) that may spend years of alternately feeding and hibernating/diapausing before it pupates. The wasp Meteorus acerbiavorus emerges from the host larva after the host has made a protec-tive cocoon around itself and the parasitoid larva remains in the host cocoon for its metamorphosis into adult wasp. In Sweden as well as in other north-ern countries with shorter seasons of insect activity, parasitoids seem to be mostly univoltine whilst in warmer countries they may be plurivoltine, fit-ting in two generations (or more) in one season. Paper III gives detailed phenological information on the species of the genera Meteorus and Zele found in the collections at the National Museums of Scotland (NMS).

12

Host ranges and speciation events amongst Braconidae Finding out the host ranges of a parasitoid is an important but difficult task. If the host range of a parasitoid wasp is known, one can both predict their current role in the ecosystem but also understand the speciation processes that defined them as a species. Host range is defined as the set of potential hosts that the parasitoid usually attacks successfully, following a pattern of searching behaviour enabling it to encounter them regularly (Shaw 1994). This implies that quantitative rearing data needs to be used for assessing a host range. The breadth of host range varies remarkably; some species ap-pear to be monophagous (e.g. Meteorus acerbiavorus on Acerbia alpina) while some attack large numbers of host species (e.g. M. pulchrichornis, Table 2).

Table 2. Hosts of Meteorus pulchrichornis, records from paper III.

Host family Reared specimens (species) Choreutidae 1 Gelechiidae 3 Pyralidae 1 Nymphalidae 1 Lasiocampidae 1 Geometridae 9 (5) Noctuidae 4 (10)

In his paper on parasitoid host ranges, Shaw (1994) summarized the main problems with literature records of host data: unrepresentative data from simplified environments and from single sites, host records presented as accumulated data without quantitative assessment of the frequency of the individual hosts, numerous misidentifications of parasitoids, and uncon-trolled rearings. These objections, and others, render many literature sources nearly useless. Minimally, second-hand sources should be viewed with scep-ticism. Shaw (1994) hypothesizes that in koinobiont parasitoids, new species always originate (by speciation) through the separation of an extreme spe-cialist from a more generalist mother lineage. Broad host ranges are seen as the result of a subsequent recruitment process. Some parasitoids are special-ists with a limited host range (limited to a single species or a bit broader, encompassing several ecologically similar or phylogenetically closely relat-ed hosts). Species with a broader host range are postulated to have a tenden-cy to diverge, to split into several lineages, through specialization. This pro-duces initially monophagous ‘young’ species that are most closely related to ‘old’ species having wider host ranges. According to this hypothesis, the host range should vary depending on the placement of the species in a phy-logenetic tree. The longer time the species has had opportunity to add new

13

hosts to its repertoire, the broader its expected host range (Shaw & Horst-mann 1997). Among braconids, Shaw’s (1994) hypothesis is supported by the results presented in paper III as well as by an earlier study of Aleiodes species (Rogadinae) (Shaw 1994, Zaldívar-Riverón et al. 2008).

Adult parasitism: evolutionary origin and adaptations Parasitism on fully-grown individuals, adults, is rather unusual amongst parasitic wasps. However, in the family Braconidae there are two subfami-lies specialized on such hosts, namely the Euphorinae, attacking mostly bee-tles but also many other groups of insects, and the Aphidiinae, attacking aphids. All Euphorinae are koinobiont endoparasitoids, and while the majori-ty of species attack adult hosts, there are also some lineages that attack lar-vae. All other braconids, as well as all other members of their sister group, the ichneumonids, oviposit in immature stages of their insect hosts, such as egg or larva, emerging from the larval or pupal host stage. Amongst the Ich-neumonidae, adult parasitoids are only found in the Polysphincta -group (Ichneumoninae), members of which are external parasitoids of spiders. In other groups of parasitic wasps, adult parasitism is also found in Aphelinidae and Encyrtidae (Chalcidoidea). Both Aphelinidae and Encyrtidae parasitize different groups of Hemiptera.

Early morphological analyses of relationships among the subfamilies within Braconidae and the genera within Euphorinae (van Achterberg 1984, Shaw 1985, 1988) indicated that the Neoneurini and Aphidiinae were not closely related to each other or to the Euphorinae. Thus, adult parasitoidism was believed to have evolved on three separate occasions in the Braconidae. Today it is widely accepted, based on molecular and morphological evi-dence, that the Neoneurini form a tribe deeply nested within Euphorinae (Belshaw & Quicke 2002, Pitz et al. 2007, Sharanowski et al. 2011). How-ever, recent studies have confirmed that the Euphorinae and Aphidiinae are not closely related (Fig. 5) (Sharanowski et al. 2011). The Aphidiinae actual-ly utilize hosts from a wide range of host ages, from embryos and nymphs to adults. This may be due to the fact that there is not much of an age difference in the physiology of an aphid. Thus, unlike the Euphorinae, the Aphidiinae are not as clear-cut imagobionts (parasitoids of adults). If Aphidiinae are indeed regarded as parasitoids of adults, it is clear that adult parasitism must have evolved twice independently in the Braconidae (Shaw 1988, Pitz et al. 2007). While the Aphidiinae are narrowly restricted to aphids, the shift to adult host stages opened up for significant diversification amongst host in-sect orders in the Euphorinae.

14

The Euphorinae The Euphorinae form one of the larger subfamilies within the Braconidae with approximately 1 200 species (Yu et al. 2011). The species within the Euphorinae differ greatly in morphology. They can be minute or measure over one centimetre in length, and the sculpture can be very smooth or coarsely rugose. They can also vary in colour from almost translucent to pitch black (Fig. 4). Identifying a euphorine may be difficult for a novice but there is a combination of characters that mostly gives a euphorine specimen away: the curved radius and the two submarginal cells of the fore-wing in combination with the petiolate second abdominal tergite (Fig. 4.). Not all species have these characters but the majority. All euphorines are koinobiont endoparasitoids, and as most koinobionts they have a rather narrow host range at the species level. Remarkable for the subfamily is that as a whole it has a very broad host range. Euphorinae attack seven insect orders: Coleop-tera, Hemiptera, Hymenoptera, Neuroptera, Orthoptera, Psocoptera, and Lepidoptera. Another remarkable thing is that they attack several different host stages: larvae, nymphs and adults. Most euphorines are solitary but there are some genera that include gregarious species, such as Microctonus (Loan 1967), Meteorus (Huddleston 1980), Perilitus (Waloff 1961) and Syntretus (Alford 1968). Some of the genera are noted for their rather ex-treme morphological attributes such as the raptorial antennae of the genus Streblocera and the huge mandibles of the genus Cosmophorus. These at-tributes are evolutionary modifications to facilitate oviposition: both the antennae and the mandibles are believed to be tools used to hold on to the host while inserting the ovipositor.

Generally the Euphorinae are among the fastest parasitoids among the Hymenoptera when it comes to oviposition behaviour. Quick oviposition may be a necessary trait for koinobiont parasitoids attacking adult hosts that can run away or fight back. When the egg is laid inside the host it floats around in the haemolymph, swelling to considerable size. If the egg was not placed in the preferred abdominal cavity, the first instar larva will move there and the following developmental stages occur in the right place (Loan 1964). The moult to the fifth instar takes place when the larva leaves the host. Most hosts are by then so weakened that they die shortly thereafter. The parasitoid larva spins its cocoon in the soil, on twigs or hang them by a thin thread, as some of the Meteorus species do. The cocoon has a circular cap at one end, serving as the exit for the adult wasp.

Evolution of adult parasitism in the Euphorinae The phylogenetic relationships among the genera of Euphorinae have been studied by several authors (Marshall 1897, Tobias 1966, Shaw 1985, 1988) but there are still remaining questions to resolve. According to the most used

15

classification before this study was initiated, the subfamily Euphorinae was divided into 14 tribes with 53 genera and a total of 1 198 species worldwide (Yu et al. 2011). The subfamily is clearly of significant age. In a Baltic am-ber fossil from the Eocene, one can clearly see a parasitoid larva emerging from an adult ant worker. The larva has been assigned to the tribe Neoneu-rini (Euphorinae) based on the biology and the larval cephalic structures (Poinar and Miller 2002). This piece of fossil evidence clearly shows that adult-parasitic Euphorinae existed already 40 million years ago. The Euphorinae are exclusively koinobiont endoparasitoids and predomi-nantly attack adult insects, but they also include lineages that parasitize lar-vae or nymphs (Meteorini and Euphorini). Most of the Euphorinae genera are parasitoids of Coleoptera, though other insect orders, including Hyme-noptera, Lepidoptera, Orthoptera, Heteroptera, Psocoptera, and Neuroptera, act as hosts for many species. The Euphorinae attack both concealed and exposed hosts (Loan 1974). The phylogenetic analysis of ichneumonoid lineages by Belshaw and Quicke (2002) strongly suggested that the ancestral life history of the Euphorinae was that of attacking exposed hosts – implying that within the Euphorinae, there have been several transitions to attacking concealed hosts. Imagobiosis requires several specific adaptations. Fully grown insects can evade pursuers either by running away or by taking flight; attacking them thus requires speed and agility of the ovipositing female. Adult insects also tend to have a much stronger outer shield, the exoskeleton, than immature insects. In beetles, the favoured hosts of euphorine imagobionts, this extra hard outer shield covers most of the body. It only leaves a few weak spots where oviposition can occur, as in the joints. This demands extreme preci-sion in the insertion of the ovipositor of the attacking parasitoid. Apparent euphorine adaptations to increase precision during ovipoisiton include the ventrally or anteriorly extended large compound eyes, the specialized peti-ole, the relatively curved ovipositor and the quick oviposition (Tobias 1965, Shaw 1985). The number of adult host specimens per any given area or vol-ume of habitat is much smaller than the number of host eggs or immatures, constraining the potential host biomass available to imagobionts. The amount of food (biomass) in each adult host is also less than in the larval stage (Shaw 2004). However, adult specimens may also be less well protect-ed than immature stages. For instance, Shaw (2004) points out that adult leaf beetles (Chrysomelidae) are probably not as toxic as the younger specimens, which accumulate plant toxins to deter parasitoids and predators. A striking feature of the Euphorinae is the great variation in the shape of the second abdominal tergite (petiole) and the ovipositor. This variation is probably the result of extreme requirements on speed, agility and precision during oviposition leading to diversifying selection on female morphologies when combined with the different habits of the attacked hosts. It is also easy to find correlations between female morphologies and attacked hosts. For

16

instance, Muesebeck (1936) noted that the ovipositor of nymphal parasitoids (Leiophron, Aridelus) were very short (subexserted) while the ovipositor of Coleoptera parasitoids were longer. The species Meteorus corax has one of the longest ovipositors within the subfamily, and they parasitize hidden Col-eoptera larvae, such as larvae of Scolytidae. The females of Leiophron and Peristenus have a short petiole and a short, strongly curved ovipositor. They chase the host nymph, lift it up, and insert the ovipositor ventrally in the abdomen. Another special feature for the Euphorinae and the Aphidiinae is their way of bending the abdomen with the ovipositor anteriorly, under their own body, thus facing the host right on with the ovipositor. This mode of attack gives the female better control during the oviposition attack. Other hymenopterans usually sting or drill downward into a host, with the main action happening behind their back or underneath them. The Euphorinae way of manoeuvring the ovipositor might make them able to chase more effec-tively after a host trying to run away. Also the amount of time spent at each egg-laying event is very short. Shaw (2004) writes that only a second or two are needed for oviposition in euphorines, in contrast to other braconids that spend more time laying each egg. The Aphidiinae also lay their eggs very quickly (Clausen 1940). Better manoeuvrability, speed and accuracy in aim when placing eggs in a host are clearly adaptations to more mobile hosts and hence to imagobiosis. To understand the evolution of life histories in the Euphorinae, it is neces-sary to understand the phylogeny of the subfamily. Marshall (1897) was first to attempt a phylogenetic tree, and Tobias (1966) continued further with analyses of the subfamily, using morphology and biology to build an intui-tive phylogenetic tree. Almost two decades later, Shaw (1985) did a formal re-evaluation based on morphology using a character matrix and computer analysis tools. A few years later, Shaw (1988) published the latest available paper on Euphorinae phylogeny, in which he presented his hypothesized tree of the relationships among the Euphorinae based on morphological data. The focus in the paper was on the implications of the phylogeny for our under-standing of the evolution of host utilization in the subfamily. Li et al. (2003) were the first to present a molecular phylogenetic analysis of the Euphori-nae, which was based on the D2 region of 28S rRNA. They used sequences from 21 species belonging to 18 genera from 8 tribes of Euphorinae, and analysed them using two different methods, neighbour-joining and maxi-mum parsimony. Their analyses remarkably showed that all tribes included in the analysis, except two, were paraphyletic (the paraphlyletic tribes were Euphorini, Dinocampini, Perilitini, Syntretini, Cosmophorini and Centistini). The only tribes for which their analysis provided some evidence for mon-ophyly were Meteorini and Microctonini. Additional results relevant to eu-phorine phylogeny were presented in the larger braconid analysis by Shi et al. (2005), which was later reassessed by Pitz et al. (2007). The analysis included the genes 16S, 18S and 28S. Unfortunately, however, only ten Eu-

17

phorinae genera were included, making the results less illuminating with respect to Euphorinae relationships. Sharanowski et al. (2011) published the most recent braconid phylogeny, a combined dataset using 4 kb of sequence data for 139 taxa with 4 genes (18S, 28S, CAD and ACC), unfortunately with only 13 Euphorinae taxa. Still today the Euphorinae phylogeny is poor-ly understood. The current state of the art is primarily (with modifications) based on Shaw’s analyses (1985, 1988), but a new phylogeny based on mo-lecular data is presented as part of this thesis (paper IV) The original ideas on the evolution of host utilization in euphorines go back to Tobias (1966). In the “Chrysomelidae-hypothesis”, termed so by Shaw (1985), Tobias stated that adult parasitism in Euphorinae originally evolved from parasitism of Chrysomelidae beetles. He based this conclusion upon observations of Euphorinae females ovipositing in both larvae and adults of Chrysomelidae beetles, along with his understanding of the phylog-eny of Euphorinae. He placed Meteorus basally in the Euphorinae based on their parasitism of larvae, the fore wing venation (lacking the submedian cross-vein) and the structure of the female genital apparatus, amongst other traits. Chrysomelidae parasitism is rather rare among the Braconidae (Shaw 2004), but exhibited by the euphorine tribes Perilitini, Centistini and Townesilitini (Shaw 1988, Yu et al. 2011), which were believed to be basal among the core Euphorinae (Euphorinae excluding Meteorini). The evidence presented in paper IV suggest that the Euphorinae originally attacked freely mobile Coleoptera larvae, like other basal Braconidae, and that the core Eu-phorinae shifted to attacking Chrysomelidae early in their evolution.

Collecting Euphorines can be collected in several different ways. The traditional meth-od is to use a sweep net but putting up traps or beating the foliage of bushes and trees are also popular collecting methods. Most specimens from paper II were collected by the use of a so-called Malaise Trap. The Malaise Trap is by far the best way to collect large quantities of insects. It is a tent-like con-struction that intercepts flying insects and leads them into a collecting jar filled with alcohol. The data presented in paper III are based on material assembled by collecting and rearing of the immature host insect, preferably infected by a parasitoid. This method gives an understanding of both the host life cycle as well as the parasitoid life cycle. The best way of implementing this method depends on whether the interest is focused on finding the differ-ent parasitoids attacking a certain host species or finding the hosts of a cer-tain parasitoid. Both approaches require a great deal of studying prior to the implementation, which always involves some degree of trial and error. Find-ing a presumed parasitized insect larva requires knowledge of where to search for the species of interest. Upon finding a potential host larva, it is

18

almost impossible, except for a specialist, to say beforehand whether the potential host it is parasitized or not. Rearing immatures in quantities is es-sential to get good sample sizes, but it requires quite an effort. The rearing equipment and materials need to be absolutely free of mould and bacteria, and they must be easy to handle and suit the larvae throughout their life cy-cle. Rearing success depends on the prior knowledge of the species ecology. Any parasitoids reared from the hosts must be associated with informative labels, or the crucial information is almost inevitably lost. A label should to a minimum contain locality, date and collector. If the specimen is reared there should be additional information about the host, emergence date, and the host plant (Shaw 1997). Tedious rearing efforts by several enthusiasts and amateur entomologists have contributed greatly to making paper III possi-ble. If your sole interest is to collect adult wasps, then host-rich, flower-rich localities with abundant sources of sugar are in my experience good habitats, though you are bound to get at least some parasitic wasps wherever you use a sweep net.

The Swedish Taxonomy Initiative In 2002, Sweden’s government launched the largest biodiversity inventory project in the country, the Swedish Taxonomy Initiative (STI, ‘Svenska Artprojektet’). The quest was to describe all multicellular animals and plants in Sweden in twenty years (Ronquist & Gärdenfors 2003). The STI is divid-ed into three major activities: inventories, taxonomic research and the pro-duction of a biodiversity encyclopedia for the general public. This disserta-tion is the result of funding from STI and is involved in two of these catego-ries, namely inventories and taxonomic research. One of the largest STI in-ventories is the Swedish Malaise Trap Project (SMTP), where insects, especially the poorly known wasps and flies, have been caught in so-called Malaise Traps. These traps (more than 70) were placed in known Swedish “hot spots” as well as in localities thought to be good for the biodiversity of flies and wasps. They were run for three years (2003-2006), collecting in-sects during all seasons, summer as well as winter (Karlsson et al. 2005). Volunteers operated these traps and many students have started their ento-mological research careers sorting this material. The insect material is sorted in three steps into fractions that are suitable for further study by experts. Last year (2012) the SMTP had sorted 50% of the material through the first step, with more than half of that material in turn also sorted through the additional two steps and ready to send out to experts for taxonomic research. In total, the material is estimated to consist of about 80 million insect specimens. Research based on the SMTP material has so far resulted in the discovery of roughly 750 species new to science and 1490 species new to Sweden (Karls-son pers. comm.).

19

Apart from the terrestrial fauna inventory, STI has also carried out a ma-rine benthic inventory of the Swedish west coast, part of the North Sea. This project started in 2006 and the inventories ended in 2009. So far, this inven-tory has resulted in the discovery of 15 new species to science and 55 new species to Sweden (Sundin & Gärdenfors 2012), and there is most probably a lot more to come as additional specialists work through the material. The systematic research based on the material of both the SMTP and the marine inventory will probably last several lifetimes. Specialists from all parts of the world currently work on the material and the results help increase the knowledge of the Swedish fauna. In total, the STI has contributed to the discovery of 2 400 new species to Sweden and roughly 900 new species to science thus far. The production of a national encyclopedia for the general public has resulted in 15 volumes. Nine of these volumes treat insect groups: Lepidoptera (4 volumes), Syrphidae (2 volumes), Cerambycidae (1 volume), Psocoptera (1 volume), and Hymenoptera (1 volume).

Aims of the Thesis Research The aim of this thesis was to contribute to the knowledge of the taxonomy, systematics, and biology of euphorine braconids, with particular reference to the Swedish fauna. The thesis includes a revision of the tribe Meteorini (pa-per II), a phylogenetic analysis covering the entire Euphorinae (paper IV), and two papers centred on morphology and biology of the wasps (papers I and III). The thesis work was initially based to a large extent on the study of Swedish material, but addressing the taxonomic and phylogenetic problems in the group required expansion of the geographic scope, first to the Western Palearctic (paper I–III) and finally to the whole world fauna (paper IV). The work has necessitated worldwide collaboration with curators, collectors and amateurs, and study of material from all corners of the world. Phyloge-netic and taxonomic questions were addressed using a combined morpholog-ical and molecular approach. Results include the discovery of a number of new Fennoscandian species, comprehensive phylogeny-based revisions of the higher classification of the group, and new insights into their biology and evolution.

20

MATERIAL AND METHODS

Much of the material studied during this dissertation is from the SMTP. All material from the SMTP was originally preserved in 80% ethanol. For exact measurements, illustrations and for addition to the reference material in the Entomology Collection of the Swedish Museum of Natural History (NHRS), some specimens were dry-pinned, including all holotypes and some para-types. These specimens were treated in a mixture of alcohol and xylene (60:40) for 24 hours and then glued to a piece of cardboard. Some paratypes were left in 80 % ethanol for future molecular or morphological study. Most of the material, the voucher specimens and the DNA extractions are deposit-ed in NHRS. All Euphorinae material in NHRS was studied as well as large parts of the collections at the following institutions: the Lund Zoological Museum (MZLU), the Uppsala University Department of Entomology (UUZM), the Natural History Museum in London (BMNH), the Nationaal Natuurhistorisch Museum in Leiden (RMNH), the Zoological Institute in St. Petersburg (ZIN), the Entomological laboratory, the Hokkaido University Museum (EIHU), the Entomological laboratory, the University of Wyoming Insect Museum (UWIM), the University of Kentucky Entomological De-partment (UKED), the Naturhistorisches Museum Bern (NMBE), and the University of Oslo, Zoological Museum (ZMUN). Additional material, mostly type material, was borrowed from the Institut Royal des Sciences naturelles de Belgique (IRSNB), the Naturhistorisches Museum Wien (NHMW), the Museum für Naturkunde der Humboldt-Universität Berlin (ZMHB), the Zoological Museum of the University of Copenhagen (ZMUC) and the Faculty of Agriculture at Kyushu University (KUEC).

DNA-workflow Prior to dry-pinning of specimens, one hind leg was removed from all spec-imens intended for DNA sampling. If needed, the leg can be reconnected to the specimen again with glue. The leg is placed in a sterile tube, to which fluids are added that digest the tissue and dissolve the DNA. At the Molecu-lar Systematics Lab (MSL) of the Swedish Museum of Natural History, there is a “GeneMole” robot that does the rest of the dirty work, adding ethanol and buffers to stabilize and clean the DNA. Left in the vial is now pure

21

DNA. Since we are only interested in a small part of the DNA, and the amount of that little part is rather low, we need to amplify it so that the tube contains mainly this tiny part of the DNA. A process called the Polymerace Chain Reaction (PCR) produces millions of copies of the DNA sequence of interest through cycles of thermal heating, as follows. Step 1) The tempera-ture is raised until DNA strands are denatured into single strands. Step 2) The temperature is lowered to the annealing temperature, when the so-called primers anneal to the DNA strands at the complimentary positions. The pri-mers are short strands of nucleic acid designed as starting points for the pro-duction of new copies of the DNA sequence of interest. Step 3) The tem-perature is increased slightly for the ultimate elongation step, where the pol-ymerase enzyme builds a complimentary strand (new DNA copy) with nu-cleotides as building blocks. The three-step cycle is repeated until enough short DNA strands of the specific part of interest are contained within the vials. To check if the PCR worked, the product is run through a gel electro-phoresis machine. This step allows one to detect if the PCR amplification has succeeded. The next and last step is called cycle sequencing and it is also a PCR, the difference being that stained dideoxynucleotides (ddNTP) are added. The polymerase adds nucleotides to the sequence but when a ddNTP is added, the DNA synthesis is stopped. A ddNTP emits light and when all these billions of DNA molecules of different length are led through a DNA sequencer, the wavelength of the ddNTP’s is recorded and the result is a four-coloured electropherogram, from which the DNA sequence can be read. After thorough inspection of the resulting sequences, they are ready for phy-logenetic analysis. For every voucher specimen, DNA is only extracted once, whilst a PCR and cycle sequencing reaction is required for every gene of interest. For the molecular analyses in all papers, the DNA was extracted and se-quenced using standardized methods as described in the papers. Two genes were used in the papers I and II; a mitochondrial gene (COI) and a nuclear gene (28S rRNA). In paper IV two additional nuclear genes (18S and CAD) were also used.

Phylogenetic Analyses Throughout this thesis, molecular data has been the backbone that we have relied on to give support to our phylogenetic hypotheses. With the molecular data, phylogenies are constructed using different methods of analysis: parsi-mony, maximum likelihood and Bayesian inference. In all four papers we have relied on Bayesian inference as the main approach, but the other meth-ods were also used when appropriate. In paper II and paper IV, we used Fitch parsimony to reconstruct ancestral host preferences. The parsimony approach explains the evolution of characters, such as host preference, by

22

minimizing the number of transitions between states (steps). The Fitch vari-ant treats character states as unordered, in contrast to the Wagner variant, where states are ordered into a hypothesized transformation series that must be followed when counting steps. In paper IV we compared the Bayesian inference (BI) results to those from the maximum likelihood (ML) method. Both are statistical methods based on an evolutionary model. ML maximizes the likelihood of the data given the model parameters, while BI gives the posterior probability of the model parameters given the data. ML gives you a point estimate, one tree with highest likelihood. To assess the strength of support for different parts of the phylogeny, ML is often combined with a non-parametric method, bootstrapping. The bootstrap procedure is based on numerous, slightly perturbed versions of the original data set, each of which is subjected to the same statistical analysis. In the phylogenetic case, the main result is a set of bootstrap support values showing how many times a certain clade (branch or group in the tree) occurs in the bootstrapped repli-cates. BI on the other hand returns a probability distribution of trees, often summarized in a majority-rule consensus tree with posterior probability val-ues (sample frequencies) of clades, representing the Bayesian assessment of support. Before running ML or BI analyses we make some prior assumptions in the model we apply, such as the number of rate parameters of nucleotide substitutions. The more complicated the model, the more computational time it requires and the higher the risk that some model parameters are estimated incorrectly. However, the model needs to be sufficiently complex to capture the essential properties of the evolutionary process. Otherwise, the results will not be accurate. In our analyses we have used the General Time Re-versible (GTR) model and the Mixed substitution rate model. The GTR model assumes that all pairwise substitution rates are different, and that the substitution matrix is symmetric, meaning that changes involving a particu-lar pair of nucleotides occur with the same rate in both directions at equilib-rium (for instance, the rate of going from A to T is the same as the rate of going from T to A, given that the equilibrium state frequencies may be dif-ferent). The Mixed substitution rate model is only available in BI and uses reversible-jump MCMC to integrate over all 203 possible reversible 4x4 nucleotide substitution models (Huelsenbeck et al. 2004). The sequence data in paper I, II and IV were run as single gene analyses and as combined analyses under the BI method. In paper IV the Mixed+G model (18S, 28S) and Mixed+I+G model with positions (1+2 vs. 3) were used for the protein coding genes CAD and COI. The BI concatenated anal-yses used the same codon-based partitioning scheme for the protein coding genes, and also allowed all substitution model parameters to vary across partitions. The concatenated ML analysis used GTR+G without the estima-tion of invariable sites for all three codon positions in the CAD and COI genes. Under the ML method for concatenated analyses using the RAxML

23

web service, there is no available function to partition sequences based on codon position.

Morphological Study and Illustrations The illustrations in paper II were constructed in several steps. First the spec-imen was photographed using an Olympus SZX12 stereomicroscope with a top-mounted Olympus DP70 camera. The pictures were initially captured with the program AnalySIS docu version 5.0 from Olympus Soft Imaging Solutions GmbH. The pictures were then further processed (stacked) using CombineZ 5.3, Adobe Photoshop and Adobe Illustrator. Drawings were prepared using a Bamboo pen tablet from Wacom. For detailed morphologi-cal imaging, the SEM (Scanning Electron Microscopy) instrument at the Swedish Museum of Natural History, a Hitachi S-4300, was used.

24

SUMMARY OF PAPERS

Paper I Stigenberg, J., Vikberg, V., Belokobylskij, S.A. 2011. Meteorus acer-biavorus sp. nov. (Hymenoptera, Braconidae), a gregarious parasitoid of Acerbia alpina (Quensel) (Lepidoptera, Arctiidae) in North Finland. Journal of Natural History, 45:1275–1294. In our first paper a new species of parasitoid wasp, Meteorus acerbiavorus Belokobylskij et al. (Hymenoptera; Braconidae) (Fig. 6), is described from the Northern parts of Finland. The new species is diagnosed and compared both morphologically and molecularly to closely related species in the genus Meteorus. The molecular phylogenetic analyses show that the new species belongs to the M. rubens species group. It was reared from the larvae of Acerbia alpina (Quensel) (Lepidoptera, Arctiidae), a rare Holarctic moth occurring, in Europe, in the northernmost parts of Scandinavia. The moth Acerbia alpina lives in Scandinavia on a limited number of south-facing mountaintops with very sparse vegetation. The moth A. alpina has a lifecycle that in harsh climates encompasses several years as larva. This is the first record of a gregarious parasitoid of the genus Meteorus reared from a mem-ber of the Arctiidae. Rich illustrations, descriptions and discussions of both the gregarious parasitoid Meteorus acerbiavorus and its host Acerbia alpina are presented, along with discussions on the phylogenetic relations and the remarkable morphological variation of the parasitoid.

25

Figure 6. Illustration of the parasitoid Meteorus acerbiavorus (A), its host Acerbia

alpina (B) and an opened cocoon of the moth with cocoons of M. acerbiavorus. Illus-tration modified from Paper 1 (Stigenberg et al. 2011), photo: J. Stigenberg (A),

Harry Lonka (B) and Pekka Malinen (C).

Paper II Stigenberg, J., Ronquist, F. 2011. Revision of the Western Palearctic Me-teorini (Hymenoptera, Braconidae), with a molecular characterization of hidden Fennoscandian species diversity. ZOOTAXA, 3084:1–95. In this paper, we expand the scope to the tribe Meteorini, to which Meteorus acerbiavorus belongs. Before our study, phylogenetic relationships within the tribe were poorly understood and had only been based on morphological analyses. The tribe Meteorini includes two genera, Meteorus and Zele. They are koinobiont endoparasitoids of larval Lepidoptera and Coleoptera. More than 300 species are known in the world and about one fifth occur in West-ern Palearctic. A large material stemming from the Swedish Malaise Trap project (SMTP) and a few other collections was used to revise the tribe. Al-most 2 000 specimens representing 54 species of Meteorus Haliday and 5 species of Zele Curtis were studied. Of these, 177 specimens representing 41 species were sequenced. Seven new species are described in the paper, and the molecular analysis indicated the presence of at least 12 additional cryptic species, morphologically inseparable but molecularly distinct. The phyloge-netic analysis presented in the paper (Fig. 7) helped define the major line-ages within the tribe and the morphological and biological differences be-

26

tween them. By mapping host preferences onto the phylogenetic tree (Fig. 7), we concluded that parasitism of Coleoptera is a derived state in Meteorini and that the forms feeding on Lepidoptera are more ancestral. This paper also includes illustrations of all Western Palearctic species and an easy-to-use key to the species.

27

Figure 7. Illustration from Paper II (Stigenberg & Ronquist, 2011) showing the

reconstruction of host preferences.

28

Paper III Stigenberg, J., Shaw, M.R. Western Palaearctic Meteorinae (Hymenoptera; Braconidae) in the National Museums of Scotland, with rearing and distribu-tional data, including six species new to Britain, and a discussion of a poten-tial route to speciation. Entomologist’s Gazette, (In press). Distributional, phenological and rearing data are crucial information for un-derstanding the biology of species and their role in ecosystems. In this paper, we present distributional, phenological and in many cases rearing data for 44 species of western Palaearctic Meteorini in the genera Meteorus and Zele, based on the nearly 2 500 specimens of the group present in the collection of the National Museums of Scotland, Edinburgh. Six species are recorded as new to Britain (M. alborossicus, M. eklundi, M. limbatus, M. longipilosus, M. oculatus, M. sibyllae). The rearing data show that both Meteorus and Zele contain species with quite broad host ranges. Patterns in the breadth of host ranges are discussed in the light of the molecular phylogeny presented in paper II and in relation to a reiterated speciation hypothesis proposed by Shaw (1994) for koinobiont braconids. The hypothesis states that all new species start out as narrow specialists, a nascent/daughter species being a specialist on just part of the ancestral/parental host range. Over time, a new species widens its host range by recruiting new hosts from the potential hosts encountered regularly in its searching environment. We find that phylogenet-ic and host-range patterns in the Meteorini fit the hypothesis well. In the paper we also stress the importance of obtaining correct host records, and not to rely on unverified and sometimes erroneous records based on poorly con-trolled rearings.

Paper IV Stigenberg, J., Boring, C. A. Phylogeny of the parasitic wasp subfamily Euphorinae (Braconidae) and evolution of its host preferences. Manuscript. In the final paper, we expand the scope even further to include the entire braconid subfamily Euphorinae, a large, cosmopolitan group of koinobiont endoparasitoids. Most species attack adult hosts, a rare thing among parasitic wasps, but there are also many species that develop in nymphs and larval stages of their hosts. Euphorine hosts include members of a surprisingly large variety of insect orders (Coleoptera, Hemiptera, Hymenoptera, Neurop-

29

tera, Orthoptera, Psocoptera and Lepidoptera) although the majority of eu-phorine lineages are confined to Coleoptera (including the tribes Centistini, Cosmophorini, Dinocampini, Perilitini, Planitorini, Pygostolini, Townesiliti-ni and partially Meteorini). In this paper we revise the higher-level classifi-cation and investigate phylogenetic relationships in the Euphorinae based on morphological and molecular characters (37 morphological characters and 3 kb of nucleotide data from four markers: 18S, 28S, CAD, and COI). We also study the host association patterns, and discuss the evolutionary diversifica-tion of the Euphorinae. Results from the analyses show support for the mon-ophyly of the subfamily Euphorinae including the tribes Meteorini and Planitorini (Mannokeraia + Planitorius). Based on the phylogenetic results, a revised classification of the Euphorinae is proposed, in which 55 extant genera and 14 tribes are recognized. We reinstate the genus Microctonus belonging to the tribe Perilitini, and propose the following tribal rearrange-ments: Spathicopis and Stenothremma are transferred to Perilitini; Tuberide-lus, Sinuatophorus and Plynops to Cosmophorini; Ecclitura and Napo to Dinocampini; Chrysopophthorus and Wesmaelia to Helorimorphini; and Townesilitus to Proclithoporini.

30

DISCUSSION

To improve our understanding of the evolution of parasitoids, it is necessary to acquire information from several different sources. First, we need a more complete picture of parasitoid diversity, including correct circumscription of the extant species. Second, we need more complete information on the life histories of the species, especially on their host preferences. Finally, we need to be able to infer phylogenetic relationships among parasitoids and use this knowledge to reconstruct the evolution of important life-history traits. In all papers (I–IV), we used an integrated approach, based on morpholo-gy or biology and genetic information, to the systematics and phylogeny of the studied taxa. The papers show the value of the genetic information in particular. In Paper I, the morphological variation indicated the presence of two species in the studied material. Some of the morphological features sep-arating them, such as the structure of the dorsope, are traditionally consid-ered informative and reliable species-distinguishing characters. Neverthe-less, the molecular evidence showed that the variation was intraspecific. In Paper II, the molecular information helped show that some morphological variation that is generally not considered reliable at the species level, such as colour differences, actually provides good diagnostic characters for some closely related species. The molecular information from Paper II also helped reveal a surprising amount of cryptic species diversity. Even after the cryptic diversity had been identified, it was still extremely difficult or im-possible to find consistent morphological differences between the cryptic species and their closest relatives. In Paper III, we integrated distributional, phenological, molecular and rearing data to show that species of Meteorus and Zele have quite broad host ranges and added evidence supporting a spe-ciation hypothesis that states that new species start out as narrow specialists, widening the host range over time. Sometimes, phenological and host data implied that the morphological identification of a species was erroneous and we could then (if the material was suitable) include molecular data to verify species identification. The results from Paper IV not only shed light on host associations of the Euphorinae but also suggested that adult parasitism is not ancestral in the subfamily. The main focus of Paper IV was the higher-level classification of the Euphorinae, and the phylogenetic results led to a rear-rangement of the tribal classification. We also used the phylogenetic results to study the host association and diversification.

31

As mentioned above, it is critical for our understanding of parasitoid biodi-versity and evolution that we collect more life-history data on the extant species. Most of these data are collected by ecologists, amateur entomolo-gists, and other non-specialists. To facilitate the recovery of parasitoid life-history data, it is crucial to provide user-friendly and well-illustrated keys to the parasitoid species as well as collaborative tools, user-friendly databases, on-line publication services and other ‘cyber taxonomy tools’ that make it more feasible to reach the goal of describing all species of parasitoids on Earth. It is my hope that this thesis has contributed in some small way to this endeavour.

32

LITTERATURE CITED

Alford, D.V. 1968. The biology and immature stages of Syntretus splendidus (Mar-shall) (Hymenoptera: Braconidae), a parasite of adult bumblebees. Transactions of the Royal Entomological Society of London 120:375-393.

Askew, R.R., Shaw, M.R. 1986. Parasitoid communities: their size, structure and development. In: Waage, J. & Greathead, D. (Eds.) Insect parasitoids, 225–264. Academic Press, London. (Symposium of the Royal Entomological Society of London 13).

Aurivillius, C. 1922. Simon Fredrik Bengtsson. Svenskt Biografiskt Lexikon, Band 03, p 205

Austen, A.D., Johnson, N.F., Dowton, M. 2005. Systematics, evolution, and biology of Scelionid and Platygastrid wasps. Annual Reviews of Entomology 50:553–582. doi:10.1146/annurev.ento.50.071803.130500

Belshaw, R. & Quicke, D.L.J. 2002. Robustness in Ancestral State Estimates: Evo-lution of Life History Strategy in Ichneumonid Parasitoids. Systematic Biology 51:450–477.

Bengtsson, S. 1900. C. G. Thomson [Nekrolog]. Entomologisk Tidskrift 21:1–16. Bennett, A. 2007. Global diversity of hymenopterans (Hymenoptera; Insecta) in

freshwater. Hydrobiologia 595:529–534. Chen, X. & van Achterberg, C. 1997. Revision of the subfamily Euphorinae (ex-

cluding the tribe Meteorini Cresson) (Hymenoptera: Braconidae) from China. Zoologische Verhandelingen 313:104–116.

Clausen, C.P. 1940. Entomophagous Insects. McGraw-Hill Book Company, Inc. New York & London.

Curtis, J. 1840. British entomology; being illustrations and descriptions of the gene-ra of insects found in Great Britain and Ireland: containing coloured figures from nature of the most rare and beautiful species, and in many instancesand in many instances of the plants upon which they are found. Printed for the author, 1823–1840, London.

Dowton, M. Belshaw, R., Austin, A.D., Quicke, D.L.J. 2002. Simultaneous molecu-lar and morphological analysis of braconid relationships (Insecta: Hymenoptera: Braconidae) indicates independent mt-tRNA gene inversions within a single wasp family. Journal of Molecular Evolution 54:210–226.

Efil, L., Güçlü, C., Belokobylskij, S.A. 2009. Leiophron (Euphorus) deficiens Ruthe (Hymenoptera, Braconidae, Euphorinae), A Parasitoid of Campylomma diversi-cornis (Reuter) (Heteroptera, Miridae) in Turkey. Journal of the Entomological Research Society 11:65–73.

Franzén, O. 1973. August E A Holmgren, Svenskt Biografiskt Lexikon, Band 19, p 256.

Futuyma, D.J. 2009. Evolution 2nd ed. Sinauer Associates, Inc. USA. Gauld, I.D. 1988. Evolutionary patterns of host utilization by ichneumonid parasi-

toids (Hymenoptera: Ichneumonidae and Braconidae). Biological Journal of the Linnean Society 35:351–377.

33

Ghahari, H., Yu, D.S., van Achterberg, C. 2006. Bibliography of the family Braco-nidae (Hymenoptera: Ichneumonoidea) (1964–2003). NNM Technical Bulletin 8.

Gibson, G.A.P 1985. Some pro-and mesothoracic structures important for phyloge-netic analysis of Hymenoptera, with a review of terms used for the structures. Canadian Entomologist 117:1395–1443.

Hansson, C. 2010 Till minne av Karl-Johan Hedqvist, Entomologisk Tidskrift 131:137–144.

Heraty, J., Ronquist, F., Carpenter, J.M., Hawks, D., Schulmeister, S., Dowling, A.P., Murray, D., Munro, J., Wheeler, W.C., Schiff, N., Sharkey, M.J. (2011) Evolution of the Hymenoptera megaradiation. Molecular Phylogenetics and Evolution 60:73–88.

Huddleston, T. 1980. A revision of the western Palaearctic species of the genus Meteorus (Hymenoptera: Braconidae). Bulletin of the British Museum (Natural History), Entomology series 41:1–58.

International Commission of Zoological Nomenclature [ICZN] (1999) International code of zoological nomenclature [the Code]. Fourth edition. The International Trust for Zoological Nomenclature, c/o Natural History Museum, London, i–xxix,+ 306 pp. [online version at http://www.nhm.ac.uk/hosted-sites/iczn/code/]

Jervis, M.A. & Kidd, N.A.C. 1986. Host-feeding strategies in hymenopteran parasi-toids. Biological Reviews 61:395–434.

Karlsson, D., Papae, T., Johanson, K.A., Liljeblad, J. & Ronquist, F. (2005) Svenska Malaisefälleprojektet, eller hur många arter steklar, flugor och myggor finns i Sverige? Entomologisk Tidskrift 126:43–53.

Kristensen, N.P. 1975. The phylogeny of hexapod ‘orders’. A critical review of recent accounts. Zeitschrift für die Zoologische Systematik und Evolu-tionsforschung 13:1–44.

Kristensen, N.P. 1999. Phylogeny of endopterygote insects, the most successful lineage of living organisms. European Journal of Entomology 96:237–253.

Li, F.F., Chen, X.X., Piao, M.H., He, J.H. & Ma, Y. 2003. Phylogenetic relation-ships of the Euphorinae (Hymenoptera: Braconidae) based on the D2 variable region of 28S ribosomal RNA. Entomotaxonomia 25:217–225.

Linnaeus, C. 1758. Systema Naturae, Ed. 10, vol. 1. Loan, C.C. 1974. Hosts and generic relations in the Euphorini (Hymenoptera,

Braconidae). Contributions of the American Entomological Institute 20:388–397 Loan, C.C. 1964. Observations on the biology of Centistes excrucians Haliday.

Proceedings of the Entomological Society of Ontario 94:56–61. Loan, C.C. 1967 Studies on the taxonomy and biology of the Euphorinae (Hyme-

noptera: Braconidae). II. Host relations of six Microctonus species. Annals of the Entomological Society of America 60:236–240.

Marshall, T.A. 1897. Les Braconides. In: André E. (Ed.) Species des Hymenopteres d'Europe et d'Algerie. Tome 5, 635 pp.

Meusemann, K., von Reumont, B.M., Simon, S., Roeding, F., Strauss, S., Kück, P., Ebersberger, I., Walzl, M., Pass, G., Breuers, S., Achter, V., von Haeseler, A., Burmester, T., Hadrys, H., Wägele, J.W., Misof, B. 2010. A Phylogenomic Ap-proach to Resolve the Arthropod Tree of Life. Molecular Biology and Evolution 27:2451–2462. doi:10.1093/molbev/msq130

Muesebeck, C.F.W. 1936. The genera of parasitic wasps of the braconid subfamily Euphorinae, with a review of the Nearctic species. USDA Miscellaneous Publi-cations 241:1–38.

Perioto, N.W., Lara, R.I.R., Fernandes, D.R.R., Pedroso, E.D.C., Volpe, H.X.L., Nais, J., Correa, L.R.B., Viel, S.R. 2011. A new phytophagous Bracon Fabricius

34

(Hymenoptera, Braconidae) associated with Protium ovatum Engl. (Burserace-ae) fruits from Brazilian savannah. ZOOTAXA 3000:59–65.

Pitz, K.M., Dowling, A.P.G., Sharanowski, B.J., Boring, C.A., Seltmann K.C., Sharkey, M.J. 2007. Phylogenetic relationships among the Braconidae (Hyme-noptera: Ichneumonoidea): A reassessment of Shi et al. (2005). Molecular Phy-logenetics and Evolution 43:338–343.

Poinar, G. Jr. & Miller, J.C. 2002. First fossil record of endoparasitism of adult ants (Formicidae: Hymenoptera) by Braconidae (Hymenoptera). Annals of the Ento-mological Society of America 95:41–43.

Quicke, D.L.J. 1997. Parasitic Wasps. Chapman & Hall, London. 470p. Ronquist, F. and Gärdenfors, U. 2003. Taxonomy and biodiversity inventories: Time

to deliver. Trends in Ecology and Evolution 18:269–270. Rousse, P. & Braet, Y. 2012. Braconid wasps (Hymenoptera) of Reunion. 1. Eu-

phorinae (including Meteorini): key to species and description of six new spe-cies. ZOOTAXA 3449:26–46.

Ryan, R.B. Rudynsky, J.A. 1962. Biology and habits of the douglas-fir beetle para-site, Coeloides brunneri Viereck (Hymenoptera: Braconidae) in western Oregon. The Canadian Entomologist 94:748–763.

Sharanowski, B.J., Dowling, A.P.G., Sharkey, M.J. 2011. Molecular phylogenetics of Braconidae (Hymenoptera: Ichneumonoidea), based on multiple nuclear genes, and implications for classification. Systematic Entomology 36:549–572.

Sharkey, M.J., Carpenter, J.M., Vilhelmsen, L., Heraty, J., Liljeblad, J., Dowling, A.P.G., Shulmeister, S., Murray, D., Deans, A.R., Ronquist, F., Krogmann, L.,Weeler, W.C. 2011. Phylogenetic relationships among superfamilies of Hy-menoptera. Cladistics 27:1–33.

Shaw, M.R. 1988. Meteorus brevicauda Thomson (Hymenoptera: Braconidae) reared from larvae of Zeugophora subspinosa (Fabricius) (Coleoptera: Chrysomelidae) Entomologist’s Gazette 39:205–206.

Shaw, M.R. & Huddleston, T. 1991. Classification and Biology of Braconid Wasps (Hymenoptera: Braconidae). Handbooks for the Identification of British Insects 7:1–126.

Shaw, M.R. 1994. Parasitoid host ranges In: Hawkins BA, Sheehan W, (Eds). Para-sitoid Community Ecology. Oxford University Press 111–144p .

Shaw, M.R. 1997. Rearing Parasitic Hymenoptera. The Amateur Entomologist 25. The Amateur Entomologist's Society, Orpington. 46pp.

Shaw, M. R. & Horstmann, K. 1997. An analysis of host range in the Diadegma nanus group of parasitoids in western Europe, with a key to species (Hymenop-tera: Ichneumonidae: Campopleginae). Journal of Hymenoptera Research 6:273–296.

Shaw, M.R. 2006. Habitat considerations for parasitic wasps (Hymenoptera). Jour-nal of Insect Conversation 10:117–127.

Shaw, S.R. 1985. A phylogenic study of the subfamilies Meteorinae and Euphorinae (Hymenoptera: Braconidae) Entomography 3:277–370.

Shaw, S.R. 2004. Essay on the evolution of adult-parasitism in the subfamily Eu-phorinae (Hymenoptera: Braconidae). Proceedings of the Russian Entomologi-cal Society 75:82–95.

Shi, M., Chen, X.X. & van Achterberg, C. 2005. Phylogenetic relationships among the Braconidae (Hymenoptera: Ichneumonoidea) inferred from partial 16S rDNA, 28S rDNA D2, 18S rDNA gene sequences and morphological charac-ters. Molecular Phylogenetics and Evolution 37:104–116.

Simbolotti, G., Villemant, C., Andreï-Ruiz, M-C. 2002. Leiophron Nees (Hymenop-tera, Braconidae : Euphorinae): a debated genus. Provisional key for West Pal-

35

aearctic species, additional data on related genera and newly collected material from Corsica. Annales de la Société Entomologique de France 38:339–350.

Stigenberg, J. & Ronquist, F. 2011. Revision of the Western Palearctic Meteorini (Hymenoptera, Braconidae), with a molecular characterization of hidden Fen-noscandian species diversity. ZOOTAXA 3084:1–95

Sundin, R. & Gärdenfors, U. 2012. Svenska artprojektets vetenskapliga del – de första tio åren (2002−2011). ArtDatabanken rapporterar 12. ArtDatabanken, SLU, Uppsala.

Tobias, V.I. 1965. Generic groupings and evolution of parasitic Hymenoptera of the subfamily Euphorinae (Hymenoptera: Braconidae). I. - Entomologicheskoye Obozreniye 44, 841–865 [In Russian; translation in Entomological Review Washington 44:494–508].

Tobias, V.I 1966. Generic groupings and evolution of parasitic Hymenoptera of the subfamily Euphorinae (Hymenoptera, Braconidae). II. Entomological Review (Washington) 45:348–358.

Tobias, V.I. 1967. A review of the classification, phylogeny and evolution of the family Braconidae (Hymenoptera). Entomologicheskoye Obozreniye 46:645-669. [In Russian; translation in Entomological Review 46:387–399].

van Achterberg, C. 1984. Essay on the phylogeny of Braconidae (Hymenoptera: Ichneumonoidea). Entomologisk Tidskrift 105:41–58.

Viitasaari, M. 2002. Sawflies (Hymenoptera, Symphyta). A review of the suborder, the Western Palearctic taxa of Xyeloide and Pamphilioidea. 516p. Helsinki.

Vinson,S.B., Iwantsch, G.F. 1980. Host suitability for insect parasitoids. Annual Review of Entomology 25:397–419.

Waloff, N. 1961. Observations on the biology of Perilitus dubius (Wesmael) (Hy-menoptera: Braconidae), a parasite of the chrysomelid beetle Phytodecta oliva-cea (Forster). Proceedings of the Royal Entomological Society of London (A) 36:96–102.

Wharton, R.A. 1993. Bionomics of the Braconidae. Annual Review of Entomology 38:121–143.

Whitfield, J.B. 1998. Phylogeny and The Evolution of Host-parasitoid Interactions in Hymenoptera. Annual Reviews of Entomology 43:129–51.

Whitfield, J.B., & Kjaer, K.M. 2008. Ancient Rapid Radiations of Insects: Chal-lenges for Phylogenetic Analysis. Annual Reviews of Entomology 53:449–72.

Wiegmann, B.M., Trautwein, M.D., Kim, J.W., Cassel, B.K., Bertone, M.A., Win-terton, S.L., Yeates, D.K. 2009. Single-copy nuclear genes resolve the phyloge-ny of the holometabolus insects. BMC Biology 7:34.

Yu, D.S., van Achterberg, C., Horstmann, K. 2011. World Ichneumonoidea 2004 - Taxonomy, Biology, Morphology and Distribution. DVD/CD. Taxapad. Van-couver, Canada. www.taxapad.com.

Zaldívar-Riverón, A., Shaw, M.R., Sáez, A.G., Mori, M., Belokobylskij, S.A., Shaw, S.R., Quicke, D.J.L. 2008. Evolution of the parasitic wasp subfamily Rogadinae (Braconidae): phylogeny and evolution of lepidopteran host ranges and mummy characteristics. BMC Evolutionary Biology 8:329

Zitani, N.M. (2003) The evolution and adaptive significance of silk use in the Mete-orinae (Hymenoptera, Braconidae). Unpublished PhD. Dissertation, University of Wyoming, 126 pp.

36

ACKNOWLEDGEMENTS

Now this intense journey is over and it feels rather unreal that I made it through. It has been a rollercoaster ride both workwise and emotionally. But now the load on my shoulders is gone and I am curiously excited of what the future will hold. Wherever I end up I will always fondly remember this peri-od in my life, the museum, the wasps, my dear colleagues and especially Niclas Eklund who has put up with me during this journey but also partaken in it.

My mother often tells me that everything always falls into my path so eas-ily, that I have had such luck with the choices in my life. She is damn right about that! I have always had a rather good knowledge about ‘things’ around us, but the first time I “really” got my eyes up for insects was when I attended a zoological systematics course at Stockholm University. The teacher for the day, Kjell Arne Johanson, made us draw the mouthparts of a beetle, through a microscope. This was the door opener for me – to have a close-up look at these splendid creatures! After this course I attended all the insect courses I could find in Sweden, that is, two in Uppsala. During those courses I got a wider insight in what insects really are and this amazed me even more. The diversity, the life histories, strategies, how they look and behave. The knowledge of two teachers (Mattias Forshage and Dave Karlsson) spurred me onwards. Through Dave’s and Mattias’s help I got a biology project as a sorter for the Swedish Malaise Trap Project (SMTP). I learned to pick out the different insect orders and then pick out the different groups of Hyme-noptera. When the time was up for the project I had learned the different braconid subfamilies. I wanted to dig deeper and with Fredrik Ronquist as my supervisor I did my master thesis on the tribe Meteorini. With Fredrik’s ambitions and efforts we got funding so that I could continue as a PhD stu-dent. I am very grateful for the SMTP, for giving me summer jobs, material for my research, for the people that I have met there: Project leader Kajsa Glemhorn, my birding and insect collecting friends Svante Martinsson and Viktor Nilsson, Dave Karlsson, Pelle Magnusson with his wonderful family, Camilla Sarin, Eje Rosén (the master of both plants and giving hugs), Sibylle Häggqvist and loads of more people, students, guests that I don’t remember the names but enjoy the memory of.

37

I also wish to thank my teachers at the University of Gotland for teaching me the secrets of biology; Kjell Larsson, Karin Bengtsson, Bertil Ståhl and Bertil Widbom. All my friends from that time – chasing butterflies during botany class to the teachers despair and doing all sorts of other stupid but fun things – let’s keep doing stupid, fun stuff! My dear friends Barbro and Hans Pettersson, their hospitality and ‘little cottage’ harboured me during those Gotland years and Hans still asks me every time we meet; “have you become a professor yet?” The museum - what a great workplace! Freezing cold during winter and way too hot during summer. Luckily it is filled with the most wonderful, warm-hearted people. Some that have had a direct, important role in helping me along the way with my thesis are Seraina Klopfstein, Mattias Forshage and Tobias Malm. I am utterly grateful and forever indebted, without your help I would never have come this far.

My closest colleagues and best room-mates: Hege Vårdal, Gunvi Lind-berg and Niklas Apelqvist – you give me support and laughs every day, pull me up when I’m down and give direction when I’m lost, you are certainly the best!

The people at the molecular systematics laboratory –thank you for all help, especially Keyvan Mirbakhsh for helping me out in the beginning and teaching me tricks in the lab.

All the nice PhD students and Post-doc’s that have shared both troubles and laughs: Tobias K, Michel C, Inga M-W, Jonas S, Sibylle H, Jeanette N, and especially Rasa B – a million thanks for doing the tedious lab work for me. All the tricky administrative stuff became easy when I had Marie Svens-son and later Jane Sagindykova to take care of it. The rest of the pack at the Entomology department: Julio Ferrer, Bert Gustafsson, Torbjörn Kronestedt, Bert Viklund, Johannes Bergsten, Yngve Brodin, Marika Källsen and Kjell Arne Johanson, your doors have always been open for my questions, just talking, but most important the laughs – thank you. Not to forget Ellen R, Rasmus H, James B and especially Marianne E for teaching me a great deal, friends in field and at the office. At the BI department (which I administra-tively belong to) I would like to thank Kevin Holston and Markus Englund for irrelevant and relevant exchange of thoughts, tips and tricks and a good portion of humour. Not to forget are my mentors/teachers abroad that I owe so much, for their hospitality, patience, friendship and knowledge; Kees van Achterberg, Mark Shaw and Scott Shaw– thank you for taking me in.

Finally, I wish to thank my supervisors Johannes Bergsten and especially Fredrik Ronquist. Fredrik for presenting this possibility and putting up with me, both of Johannes and Fredrik for correcting and immensely improving my papers, giving me feedback, but most of all keeping me on track and giving me a friendly push whenever I have needed it – thank you.

38

SVENSK SAMMANFATTNING

Parasitsteklar finns praktiskt taget överallt. De är oftast små och svarta och vid närmare granskning visade sig I regel ha två par vingar och långa anten-ner. Alla har säkerligen någon gång stött på dessa små insekter men förväxlat dem med mygg eller andra småkryp. Ta en titt på fönsterbrädan en sommardag och du kommer säkert hitta några parasitsteklar bland alla andra insekter som fångats av fönsterrutans falska budskap om frihet. Plocka upp dem och studera dem närmare under en lupp och du kommer att häpna!

Att rekonstruera parasitsteklars fylogenetiska (evolutionära) historia har länge varit ett intresse bland biologer inom olika fält. Bortsett från deras värde inom biologisk bekämpning så är parasitsteklarna ett ypperligt objekt för studier av bland annat biodiversitet, morfologiska mönster, evolutionen av parasiter och samevolutionen mellan parasiter och deras värdar.

Tidigare fylogenetiska studier har ofta begränsats av begränsad insamling (få representerade grupper), begränsad molekylär data (få/dåliga gener) och/eller användandet av “dåliga” morfologiska karaktärer. Att försöka förstå parasitsteklarnas evolution kräver en robust fylogeni (släktträd) att bygga på, men rekonstruktionen försvåras av parasitsteklarnas enorma ar-trikedom och det faktum att de generellt är så dåligt kända.

Parasitsteklar utgör ungefär en tredjedel av de svenska djurarter som är så dåligt kända att de ännu inte kunnat bedömas enligt rödlistekriterierna. Den näst artrikaste familjen av parasitsteklar är Braconidae (bracksteklar); där världsfaunan uppskattas till runt 40 000 arter och den svenska faunan tros bestå av något över 2 000 arter, varav 1 200 är kända idag. Bracksteklar par-asiterar på de flesta insektsordningar och utgör en viktig del i de flesta terres-tra och några akvatiska (!) näringskedjor, vilket bland annat gör dem intres-santa för biologisk bekämpning. Avhandlingen koncentreras på en svår och dåligt känd underfamilj bland bracksteklarna, Euphorinae (skalbaggsteklar). Vi uppskattar att den svenska faunan av Euphorinae uppgår till drygt 200 arter, varav endast 102 (Yu et al. 2011) noterats hittills. Euphorinerna kan som sagt hittas praktiskt taget överallt, än så länge inga akvatiska. Man fångar dem med fjärilshåv eller med diverse insektsfällor och bevarar dem antingen i etanol eller så limmar man fast dem på en bit kartong. När man studerar dessa små steklar krävs ett mikroskop för artbestämning. För dju-pare studier av det molekylära slaget krävs (i bästa fall) endast ett bakben som plockas av stekeln och processas vidare i ett DNA-labb. För framtida forskning och för egen dokumentation är det av yttersta vikt att alla delar

39

(DNA, sekvens, individ och etikett) kan kopplas till varandra, annars är in-formationen endast av begränsad nytta.

I denna avhandling undersöks bland annat hur många arter av Euphorinae som finns i Sverige, främst med fokus på släktena Meteorus och Zele. Vidare undersöks släktskapsförhållandena mellan grupperna inom Euphori-nae genom att jämföra sekvenser från några gener. Avhandlingen tar också upp värdrelationer och hur dessa kan tänkas ha evolverat, kryptiska arter samt några arters biologi. Avhandlingen består av fyra så kallade “papper”. I papper I presenteras och beskrivs en ny art för vetenskapen, nämligen Me-teorus acerbiavorus Belokobylskij et al. M. acerbiavorus parasiterar på den sällsynta fjärilen Acerbia alpina Quensel (Nordlig igelkottspinnare). I artikeln behandlar vi de dramatiska morfologiska skillnaderna inom arten, men jämför den även med närliggande och morfologiskt liknande artfränder. En molekylär analys baserad på delar av en nukleär gen (28S) och en mi-tokondriell gen (COI) visar att M. acerbiavorus hör till artgruppen runt M. rubens. I papper II studerar vi hela tribusen Meteorini med ca 30 tidigare kända arter i Sverige. Studien baserades på 17 morfologiska karaktärer och data från två molekylära gener (28S och COI; totalt ungefär 1 000 baspar). Materialet till studien kom huvudsakligen från Svenska Malaisefälleprojke-tet, en nyligen genomförd systematisk inventering av den svenska in-sektsfaunan med hjälp av ett stort antal tältliknande Malaisefällor utplacer-ade över hela landet. I materialet kunde vi påvisa tio nya arter för Sverige, varav sju nya för vetenskapen. Den molekylära analysen visade att det fanns ytterligare ett tiotal kryptiska arter, men eftersom vi inte hittade några säkra morfologiska kännetecken för dessa så lämnade vi dem obeskrivna. Vi försökte också rekontruera evolutionen av värdpreferenser, då vissa arter parasiterar på Coleoptera (skalbaggar) och vissa på Lepidoptera (fjärilar). Våra analyser tyder på att de former som parasiterar på Lepidoptera är mer basala än de som parasiterar på Coleoptera. Papper III är en studie av det material av Meteorus och Zele (totalt 2 500 individer av 44 arter) som finns i samlingarna på National Museums of Scotland i Edinburgh. Vi presenterar en mängd nya data över arternas fenologi (uppträdande under året), geo-grafiska utbredning, samt biotop- och värdpreferenser. Vi rapporterar också sex nya arter för de Brittiska öarna. Våra data ger ytterligare stöd till den tidigare framförda hypotesen att nya arter troligtvis uppkommer ur general-ister. I papper IV försöker vi reda ut och befästa släktskapen bland de olika släktena inom hela underfamiljen Euphorinae med hjälp av fyra gener, varav tre nukleära gener (18S, 28S, CAD) och en mitokondriell gen (COI). Vi har lyckats analysera material från 44 av de 48 släktena som räknas som giltiga idag. Baserat på dessa resultat reviderar vi klassifikationen för underfamiljen Euphorinae och föreslår att världsfaunan indelas i 55 släkten och 14 triber. Våra resultat visar att tre omstridda grupper lämpligen inkluderas som triber inom Euphorinae, nämligen Meteorini, Planitorini och Ecnomiini. Vi finner det också nödvändigt att skapa en ny tribus, Pygostolini. Tribusen Meteorini

40

parasiterar även den på larver men av både Lepidoptera och Coleoptera, här är dock parasiterandet av Coleoptera det ursprungliga läget och parasiter-andet på Lepidoptera är en modern händelse i det fylogenetiska trädet. Grundstommen bland euphorinernas värdregister tenderar att vara Coleop-tera och parasitism på såväl adulta individer som på Hemiptera, Hymenop-tera, Psocoptera, Orthoptera och Neuroptera är moderna händelser.

Sammantaget har arbetena i denna avhandling avsevärt ökat vår förståelse och kunskap om parasitsteklarnas biodiversitet och spridning i Sverige, Eu-ropa och i världen. Studier av släktskapen inom gruppen har gett viktiga evolutionära såväl som taxonomiska insikter.