sign in sign up
<<
Home , Agulla (snakefly), Blattoptera, Chresmoda, Chrysopa perla, Chrysopida, Conwentzia

NEUROPTERIDA OF THE LOWER OF SOUTHERN , WITH A STUDY ON AND EXTANT RAPHIDIOPTERA

A thesis submitted to The University of Manchester for the degree of PhD in the Faculty of Engineering and Physical Sciences

2010

JAMES EDWARD JEPSON

SCHOOL OF EARTH, ATMOSPHERIC AND ENVIRONMENTAL SCIENCES

TABLE OF CONTENTS

FIGURES...... 8

TABLES...... 13

ABSTRACT...... 14

LAY ABSTRACT...... 15

DECLARATION...... 16

COPYRIGHT STATEMENT...... 17

ABOUT THE AUTHOR...... 18

ACKNOWLEDGEMENTS...... 19

FRONTISPIECE...... 20

1. INTRODUCTION...... 21

1.1. The Project...... 21

1.2. The Importance of Studying Fossil ...... 22

1.3. Thesis Layout...... 23

2. LITERATURE REVIEW...... 24

2.1. What are the ?...... 25

2.1.1. Neuropterida’s place within Holometabola...... 28

2.1.2. Relationships of the three Orders...... 29

2.1.3. Relationships within the Orders...... 31

2.2. Developmental History of Neuropterida...... 39

2.2.1. ...... 39

2.2.2. Raphidioptera...... 39

2.2.3. ...... 39

2

2.3. Extant British Neuropterida...... 40

2.3.1. Neuroptera...... 40

2.3.2. Raphidioptera...... 41

2.3.3. Megaloptera...... 41

2.4. Fossil Record of Neuropterida...... 41

2.4.1. Neuroptera...... 41

2.4.2. Raphidioptera...... 44

2.4.3. Megaloptera...... 47

2.4.4. Fossil British Neuropterida...... 48

2.5. Environment and Ecology...... 49

2.5.1. Neuropterida...... 50

2.5.2. Neuroptera...... 51

2.5.3. Raphidioptera...... 53

2.5.4. Megaloptera...... 54

2.5.5. The Percentage of Neuropterida in Environments...... 55

2.6. Geological Settings...... 56

2.6.1. flora, a brief note...... 56

2.7. Purbeck Group...... 56

2.7.1. Palaeogeography...... 56

2.7.2. Palaeoclimate...... 58

2.7.3. Palaeoenvironments...... 58

2.7.4. Lulworth Formation...... 59

2.7.5. Durlston Formation...... 60

3

2.76. Palaeoflora...... 61

2.77. Palaeofauna...... 62

2.8. Wealden Supergroup (Weald Sub-Basin)...... 62

2.8.1. Palaeogeography...... 62

2.8.2. Palaeoclimate...... 63

2.8.3. Palaeoenvironments...... 63

2.8.4. Palaeoflora...... 65

2.8.5. Palaeofauna...... 66

2.9. Climate and Environmental Change in the Purbeck and Wealden...... 67

2.9.1. Sedimentology...... 67

2.9.2. Palaeontological evidence...... 69

3. MATERIAL AND METHODS...... 70

3.1. Fossil Acquisition...... 70

3.2. Study of the ...... 70

3.3. Wing Venation...... 70

3.4. Problems Associated with Palaeoentomology...... 72

3.4.1. Fragmentary Specimens...... 72

3.4.2. The Wing Problem...... 72

PART ONE A – NEUROPTERIDA OF THE LOWER CRETACEOUS OF SOUTHERN ENGLAND: PURBECK...... 75

4. NEW LACEWINGS (INSECTA: NEUROPTERA) FROM THE LOWER CRETACEOUS

PURBECK LIMESTONE GROUP OF SOUTHERN ENGLAND...... 76

4.1. Abstract...... 76

4

4.2. Geological Setting...... 77

4.3. Material and Methods...... 78

4.4. Systematic Palaeontology...... 78

4.5. Discussion...... 109

5. RAPHIDIOPTERA (INSECTA: NEUROPTERIDA) FROM THE LOWER CRETACEOUS

PURBECK LIMESTONE GROUP, DORSET, UK...... 111

PART ONE B – NEUROPTERIDA OF THE LOWER CRETACEOUS OF SOUTHERN ENGLAND: WEALDEN...... 112

6. NEUROPTERAN (INSECTA) PALAEODIVERSITY WITH PREDICTIONS FOR THE

CRETACEOUS FAUNA OF THE WEALDEN...... 113

7. NEW LACEWINGS (INSECTA: NEUROPTERA) FROM THE LOWER CRETACEOUS

WEALDEN SUPERGROUP OF SOUTHERN ENGLAND...... 114

8. TWO NEW OF (INSECTA: RAPHIDIOPTERA) FROM THE LOWER

CRETACEOUS OF ENGLAND AND WITH A REVIEW OF OTHER FOSSIL

RAPHIDIOPTERANS FROM THE /CRETACEOUS

TRANSITION...... 115

9. NEW (INSECTA: RAPHIDIOPTERA) FROM THE LOWER CRETACEOUS OF

BRITAIN, SPAIN AND BRAZIL...... 116

10. PALAEOECOLOGY OF THE LOWER CRETACEOUS PURBECK AND WEALDEN

NEUROPTERIDS (INSECTA: NEUROPTERA, RAPHIDIOPTERA)...... 117

10.1. Abstract...... 117

10.2. Introduction...... 117

10.3. Material and Methods...... 118

10.4. Geological Settings...... 118

10.4.1. Purbeck Limestone Group...... 119

10.4.2. Wealden Supergroup, Weald Sub-basin...... 120

5

10.5. Taphonomy...... 120

10.6. Lower Cretaceous Neuropterida from Britain...... 124

10.6.1. Raphidioptera...... 129

10.6.2. Neuroptera...... 131

10.7. Neuropterid Palaeoenvironment and Palaeoecology...... 134

10.8. Faunal Change...... 135

10.9. Lower Cretaceous Neuropterid Faunas World-wide...... 136

10.9.1. Palaeobiogeography...... 138

10.10. Discussion...... 138

11. CONCLUSIONS...... 142

12. FURTHER WORK...... 144

PART TWO - A STUDY ON FOSSIL AND EXTANT RAPHIDIOPTERA...... 145

13. A STUDY ON FOSSIL AND EXTANT RAPHIDIOPTERA...... 146

13.1. Introduction...... 146

13.2. Literature Review...... 146

13.2.1. Raphidioptera...... 146

13.2.2. Raphidioptera Morphology...... 147

13.2.3. Potential problems of using wings for phylogenetic studies...... 150

13.2.4. Raphidioptera Phylogeny...... 150

13.3. Keys To The Families And Genera Of Fossil Snakeflies (Insecta, Raphidioptera)...... 163

13.3.1. Abstract...... 163

6

13.3.2. Introduction...... 163

13.3.3. Key to Families...... 164

13.3.4. Keys to Genera...... 167

13.4. A Cladistic Study Of Fossil Snakeflies (Insecta, Neuropterida, Raphidioptera)...... 177

13.4.1. Abstract...... 177

13.4.2. Introduction...... 177

13.4.3. Material and Methods...... 178

13.4.3.1. Body and Wing Morphology...... 178

13.4.3.2. Taxa...... 178

13.4.3.3. Potential Problems...... 181

13.4.4. Cladistic Study...... 182

13.4.4.1. Characters...... 182

13.4.4.2. Analysis...... 186

13.4.4.3. Synapomorphies...... 187

13.5. Discussion...... 190

13.6. Conclusion...... 193

13.6.1. Further Work...... 194

REFERENCES...... 195

WORD COUNT = 79 169

7

FIGURES

Figure 2.1. Photographs of extant Neuropterida (Megaloptera, Raphidioptera, Neuroptera)...... 25

Figure 2.2. Schematic drawing of neuropterid lifecycle ...... 26

Figure 2.3. Schematic drawing of holometabolan larvae with the imaginal discs highlighted ...... 27

Figure 2.4. Pie chart displaying present day diversity...... 28

Figure 2.5. Phylogeny of the Holometabola...... 29

Figure 2.6. A cladogram of neuropterid relationships...... 30

Figure 2.7. A cladogram of the relationships of Raphidioptera and Megaloptera...... 32

Figure 2.8. A cladogram of the suborders of Neuroptera...... 33

Figure 2.9. A cladogram of the relationships within ...... 34

Figure 2.10. A cladogram of the relationships within ...... 35

Figure 2.11. A cladogram displaying the disrupted relationships within Hemerobiiformia...... 36

Figure 2.12. Phylogenetic tree of neuropterid relationships...... 38

Figure 2.13. Palaeogeographic map of the Lower Cretaceous of southern England...... 57

Figure 2.14. Stratigraphic column of the Purbeck Limestone Group...... 59

Figure 2.15. Stratigraphic column of the Wealden Supergroup...... 64

Figure 3.1. Comstock (1918) wing venation interpretation...... 71

Figure 3.2. Modified Comstock (1918) wing venation interpretation...... 71

8

Figure 4.1. Sophogramma wimbledoni sp. nov. holotype (No. 15)...... 79

Figure 4.2. gen. et sp. indet. (BMNH Pal. I 3958)...... 81

Figure 4.3. Pterinoblattina penna holotype (BMNH Pal. I. 3968)...... 83

Figure 4.4. Pterinoblattina pluma holotype (BMNH Pal. I. 12324)...... 84

Figure 4.5. Pterinoblattina purbeckensis sp. nov. holotype (DB175/NEUR11)...... 85

Figure 4.6. Pterinoblattina fasciata sp. nov. holotype (DB175/NEUR1)...... 86

Figure 4.7. Pterinoblattina sp. B. (BMNH Pal. I. 12311)...... 87

Figure 4.8. Purbepsychops parallela gen. et sp. nov. holotype (CAMSM J53807)...... 89

Figure 4.9. Ovalorobius purbeckensis gen. et sp. nov. holotype (DB175/NEUR4)...... 90

Figure 4.10. Mesosmylidus vulgaris gen. et sp. nov. holotype (DB175/NEUR9)...... 92

Figure 4.11. Osmylochrysa minima gen. et sp. nov. holotype (DB175/TM1096)...... 94

Figure 4.12. Osmylochrysa anomala gen. et sp. nov. holotype (DB175/NEUR18)...... 95

Figure 4.13. Stenochrysa gradata gen. et sp. nov. holotype (DB175/NEUR12)...... 96

Figure 4.14. Osmylopsis duplicata holotype (BMNH Pal. I. 3954)...... 98

Figure 4.15. Mesypochrysa minuta sp.nov. holotype (DB175/NEUR 28)...... 100

9

Figure 4.16. Purbemerobius medialis gen. et sp. nov. (DB175/NEUR5)...... 102

Figure 4.17. Epimesoberothus osmyloides gen. et sp. nov. holotype (USCDB/TM432) and paratype (CAMSM TN272)...... 104

Figure 4.18. Sialium sipylus holotype (BMNH Pal. I. 13951)...... 106

Figure 4.19. Paraksneura alleni gen. et sp. nov. holotype (HCSD NEUR1)...... 108

Figure 10.1. Palaeogeography of the Lower Cretaceous of southern England...... 119

Figure 10.2. Schematic reconstruction of the Wealden environment...... 122

Figure 10.3. Position of Londinia with respect to the Weald Sub- basin...... 123

Figure 10.4. Pie charts showing the percentage of insects described from the Purbeck and Wealden and the percentage of neuropterid families in the Lower Cretaceous of southern England...... 127

Figure 10.5. Spindle diagrams of the number of neuropterid specimens throughout the Purbeck and Wealden...... 128

Figure 10.6. Geological ranges of the neuropterid families in the Lower Cretaceous of southern England...... 129

Figure 10.7. Raphidiopteran wing size and climate change...... 131

Figure 13.1. Life stages of Raphidioptera...... 147

Figure 13.2. Labelled body morphology of adult raphidiopteran...... 148

Figure 13.3. Genital morphology of raphidiopterans...... 149

Figure 13.4. Labelled raphidiopteran wing venation...... 150

10

Figure 13.5. Raphidiopteran apomorphies...... 152

Figure 13.6. and apomorphies...... 153

Figure 13.7. Raphidiidae phylogeny...... 154

Figure 13.8. Inocelliidae phylogeny...... 158

Figure 13.9. Ren and Hong‟s (1994) raphidiopteran cladogram...... 160

Figure 13.10. Willmann‟s (1994) raphidiopteran phylogeny...... 162

Figure 13.11. Labelled wing venation of a raphidiopteran...... 164

Figure 13.12. Termination of Sc on C...... 165

Figure 13.13. Raphidiopteran wing shapes...... 165

Figure 13.14. Cells between Rs and M...... 165

Figure 13.15. Trifurcation of R1, M and Cu...... 166

Figure 13.16. Fusion of Ma with R1 and the free analis in the hind wing...... 166

Figure 13.17. Medial cells and pterostigma...... 167

Figure 13.18. The position of the free branch of MA...... 167

Figure 13.19. cua-cup crossveins in hind wing...... 168

Figure 13.20. First cu crossvein relative to the CuA – MP fork...... 168

Figure 13.21. Sub-horizontal branch of M...... 169

Figure 13.22. Enlarged medial and discal cells...... 171

Figure 13.23. Cubital crossveins...... 171

Figure 13.24. rs-c crossveins and A3 not fused...... 172

Figure 13.25. MA fused with R and rs-r1 and r-sc crossveins...... 173

11

Figure 13.26. r-sc and cup-a1 crossveins in forewings and medial cell in the hind wing...... 173

Figure 13.27. Stem of M in hind wing...... 174

Figure 13.28. Proximal radial area and apical terminus of R...... 175

Figure 13.29. Features of Electrinocellia...... 175

Figure 13.30. Second sc-r crossvein...... 176

Figure 13.31. The wing venation of Permoberotha...... 181

Figure 13.32. Labelled raphidiopteran wing venation...... 183

Figure 13.33. 50% majority rule strict consensus tree of raphidiopteran relationships...... 187

Figure 13.34. Raphidiopteran tree combining the consensus tree and extant relationships...... 189

12

TABLES

Table 2.1. Fossil record of Neuropterida in Britain...... 50

Table 2.2. Clay mineralogy and the environment...... 69

Table 2.3. Clay mineralogy and the environment...... 69

Table 10.1. Neuropterida specimens of the Purbeck...... 125

Table 10.2. Neuropterida specimens of the Wealden...... 126

Table 13.1. Raphidiopteran genera...... 179

Table 13.2. Data matrix for the raphidiopteran cladistic study...... 186

13

ABSTRACT

The study of Neuropterida from the Lower Cretaceous Purbeck Limestone Group and Wealden Supergroup of southern England has shown that 13 families, 24 genera and 33 species are present. The majority of the taxa described are new to science with the exception of six genera and four species, which are redescribed. The Purbeck fauna consists of the families Kailligrammatidae, , , Prohemerobiidae, , ?, , , Mesoraphidiidae and one . The Wealden contains Kalligrammatidae, Psychopsidae, Osmylidae, , Mesoraphidiidae and two incertae sedis families. The psychopsoid families are the most dominant neuropterid in the Purbeck and Wealden neuropterid fauna. The discovery of the families Chrysopidae, ?Hemerobiidae, Osmylidae, Ithonidae, Berothidae represent the first record in the Mesozoic of Britain. The ithonid also represents the oldest fossil record of that family. Megaloptera have not been found in these deposits. The neuropterids in both deposits would have lived in a distant wooded area based on modern analogues. On death they would have undergone disarticulation, decomposition/sorting and then deposition of wings only in the depositional basin. No body parts have been found associated with any neuropterid from these deposits. The climate change through the English Lower Cretaceous may have affected the neuropterid fauna in size (Raphidioptera) and family turnover (Neuroptera). Biogeographical links have been hinted at with Asia and South America with the occurrence of some of the same genera.

A preliminary key to the fossil raphidiopterans has been presented aiding identification to family and generic level. This is the first time such a key has been presented on fossil Raphidioptera. A cladistic study was undertaken (21 characters and 28 taxa) on the genera of fossil Raphidioptera with the inclusion of extant genera. Fifteen synapomorphies were found, but, relationships are largely unresolved, with few well-supported . Baissopteridae and Mesoraphidiidae were found to be paraphyletic. Baissopteridae is to all raphidiopterans. Both the key and cladistic study highlighted the need for a major taxonomic revision of fossil raphidiopterans.

14

LAY ABSTRACT

Over thirty new species of fossil lacewings and snakeflies (insects) have been described from 145-125 million year old rocks from southern England. More species are now known to be present than previously thought and the older rocks in Dorset have yielded a higher number of lacewings than the younger rocks from the south east of England (Kent, Surrey and Sussex). The snakeflies are consistent across the whole of southern England. and have yet to be discovered in England at this time. The lacewings and snakeflies would have lived in a distant forested environment far from the place where they were deposited. On death they would have decomposed and disarticulated before being washed into a watercourse. In the watercourse they would have become more disarticulated and at the place of deposition only their wings have been preserved. All of the wings show damage from the transportation process. Between 145-125 million years ago the climate changed from hot and dry to more humid and wet. This change had an effect on the snakeflies, which were smaller in the hot periods, and the lacewings, many of which became extinct when the climate became wetter.

An identification key for fossil snakeflies is presented and a study of the evolutionary relationships within snakeflies has been undertaken. The evolutionary relationships were largely unresolved. This work on fossil snakeflies has highlighted that this group of fossil insects has been poorly described over the years and a revision of the entire group needs to be undertaken before any other evolutionary study is undertaken.

15

DECLARATION

No portion of the work referred to in this thesis was submitted to support an application towards the obtainment of another degree of this or any other school or institute of learning.

James E. Jepson

2010

16

COPYRIGHT STATEMENT

i. The author of this thesis (including any appendices and/or schedules to this thesis) owns certain copyright or related rights in it (the “Copyright”) and s/he has given The University of Manchester certain rights to use such Copyright, including for administrative purposes. ii. Copies of this thesis, either in full or in extracts and whether in hard or electronic copy, may be made only in accordance with the Copyright, Designs and Patents Act 1988 (as amended) and regulations issued under it or, where appropriate, in accordance with licensing agreements which the University has from time to time. This page must form part f any such copies made. iii. The ownership of certain Copyright, patents, designs, trademarks and other intellectual property (the “Intellectual Property”) and any reproductions of copyright works in the thesis, for example graphs and tables (“Reproductions”), which may be described in this thesis, may not be owned by the author and may be owned by third parties. Such Intellectual Property and Reproductions cannot and must not be made available for use without the prior written permission of the owner(s) of the relevant Intellectual Property and/or Reproductions. iv. Further information on the conditions under which disclosure, publication and commercialisation of this thesis, the Copyright and any Intellectual Property and/or Reproductions described in it may take place is available in the University IP Policy (see http://www.campus.manchester.ac.uk/medialibrary/policies/intellectual- property.pdf), in any relevant Thesis restriction declarations deposited in the University Library, The University Library‟s regulations (see http://www.manchester.ac.uk/library/aboutus/regulations) and in The University‟s policy on presentation of Theses.

17

ABOUT THE AUTHOR

As a child I was fascinated by the natural world. This fascination continued throughout my academic life. In 2005 I graduated from the University of Manchester with a B.Sc (Hons) in Geology, after which I found myself in the right place at the right time and ended up deciding to self fund a Ph.D. on neuropterids from the Purbeck and Wealden. A decision that has never been regretted.

18

ACKNOWLEDGEMENTS

I wish to thank John Nudds (University of Manchester) for supervision, help and support throughout the Ph.D. and Dmitri Logunov and Phil Rispin (The Manchester Museum) for help, support and access to the collections. Thanks to Phil Manning and Tony Adams (University of Manchester) for their help sorting out my financial situation. Thanks to colleagues past and present from the Palaeontology Research Group (University of Manchester) Dave Penney, Federica Menon, Karl Bates, Peter Falkingham, Holly Barden and Nick Edwards for their help and support over the years. Very special thanks to Sam Heads (Illinois) for inspiration, help, support and encouragement throughout the Ph.D. Many thanks to the museums and curators that have loaned and allowed access to specimens over the years: John Cooper, Gerald Legg (Booth Museum of Natural History, Brighton), Andrew Ross, Claire Mellish (Natural History Museum, London), Dan Pemberton, Matt Lowe, Matt Riley (Sedgwick Museum of Earth Sciences), Stuart Ogilvy (York Museum), David Gelsthorpe, Dmitri Logunov, Phil Rispin (The Manchester Museum) and Günter Bechly (Staatlisches Museum für Naturkunde, Stuttgart). I would also like to thank those people who have collaborated with me on papers throughout the project: Vladimir Makarkin (Russian Academy of Sciences,Vladivostok), Ed Jarzembowski (Maidstone Museum), Rob Coram (British fossils) and Jörg Ansorge (Greifswald, ). Thanks to all the people who have sent me literature: Alexandr Ponomarenko (PIN, Moscow), Horst and Ulrike Aspöck (Wien), Ed Jarzembowksi, Vladimir Makarkin, Thomas Schlüter (UNESCO), Dong Ren (Capital Normal University, Beijing), Michael Engel (University of Kansas), George Poinar (Oregon State University), Rob Coram, Andrew Ross, Xavier Delclos (University of Barcelona) and to John Oswald (Texas A&M University) whose Bibliography of the Neuropterida has been a valuable resource for neuropterid data.

Thanks to my family and last but not least Lisa Abbott for putting up with me throughout the whole process!

I wish to also acknowledge the Palaeontographical Society Fund, Royal Entomological Society conference travel grant, Alkins Fund (University of Manchester) and the Wingate Scholarship for funding, without which this Ph.D. could not have been completed.

19

THE WEALDEN SNAKEFLY PRORAPHIDIA HOPKINSI HUNTING

THE PENAPHIS WOOLARDI. DRAWING BY NEIL WATSON

If it be now, 'tis not to come; if it be not to come, it will be now; if it be not now, yet it will come: the readiness is all.

William Shakespeare, from Hamlet (1600-1602)

20

CHAPTER 1. INTRODUCTION

This chapter gives an introduction to the aims of the project and why studying fossil insects (especially from the Mesozoic) is important.

1.1. The Project

Fossil Neuropterida (Megaloptera, Raphidioptera and Neuroptera) are an understudied group within the Lower Cretaceous rocks of southern England. With regards to Mesozoic British fossil neuropterids, only one major work has been completed on the Jurassic fauna (Whalley 1988). Prior to this study only four families and three species of Neuroptera had been recorded from the Purbeck and Wealden. There were no Megaloptera and only two Raphidioptera recorded, one from the Purbeck and one from the Wealden, which were mentioned but not described (Jarzembowski 1984, 2001; Clifford et al. 1994). After searching through many specimens, often uncatalogued in museum collections, many Neuropterida were discovered. The first aim of the project was to undertake a comprehensive survey of the Neuropterida from these deposits. From this study, the Neuropterida of southern England were able to be tracked across from the Purbeck into the Wealden in terms of evolution and adaptation with regards to the changing environment. This enabled a comparison with other neuropterid faunas from the Lower Cretaceous. There is also a study of the palaeoecology of these insects; how they interact and react to the flora and fauna in each deposit.

The second part of this thesis examines the fossil record of Raphidioptera, which includes an identification key to allow identification of fossil raphidiopterans to family and generic level. A cladistic study incorporating fossil raphidiopterans into a phylogeny with extant representatives; this study was undertaken at generic level. There have been few previous attempts at a cladistic analysis of fossil and extant raphidiopterans but these were not greatly successful (Ren & Hong 1994; Willmann 1994). It is hoped that this study will shed some light on the evolutionary relationships within fossil and extant raphidiopterans, or at least give a good starting point for future research.

21

1.2. The Importance of Studying Fossil Insects

Insects first appeared in the fossil record in the . In the winged insects such as the large -like Meganeura are recorded, while the early members of most modern day Orders are found preserved in the rocks of the and . The Cretaceous saw the beginnings of a modern biota evolving with many modern families appearing, in fact it has been estimated that over half of the Recent insect families evolved in the Cretaceous (Grimaldi & Engel 2005). The Cretaceous also has the advantage of having many fossil insect sites that have been subjected to fewer Earth processes, e.g. deformation and erosion. Also within the Cretaceous the evolution of the Angiosperms occurred, which had a major effect on insect evolution.

With regards to Neuropterida, the Cretaceous was a very important time. Effectively „the Golden Age‟ of Neuropterida. Although Neuroptera and Megaloptera evolved in the Permian, and Raphidioptera in the Jurassic, great diversity of families and species was observed in a burst of evolutionary activity in the Mesozoic, particularly the Cretaceous. Therefore, to understand how the modern fauna evolved one needs to look at this all-important time in the evolution of this group. In particular in the Purbeck and Wealden there is a gap in our knowledge of Neuropterida with very little work previously undertaken; this study aims to fill that gap.

With regards to the integration of fossil insects with the extant to create a phylogeny, this is very important to get a fuller picture of the evolutionary history of a group.

1.3. Thesis Layout

This thesis is submitted under the 'Alternative Format Regulations', a collection of scientific papers that have been written as a result of the investigation carried out during the past years of the PhD (2005-2010). According to the regulations, the chapters, excluding the introductory and conclusive ones, have to be a paper submitted, published or in preparation.

The thesis is subdivided into two main parts comprising papers published and submitted. All papers were conceived by me. For each paper I carried out the bulk of the work, for example, descriptions, figures and writing the manuscripts. Co-author

22 contributions are given below. The introductory chapter (chapter 1), literature review (chapter 2) of extant and fossil neuropterids, geological settings, material and methods (chapter 3) precede the two parts. Part one focuses on the neuropterid assemblages of the Purbeck and Wealden: it is mainly a collection of systematic papers, giving an account of which Raphidioptera, Neuroptera and Megaloptera are present. It comprises nine chapters: the first of these chapters (chapter 4) is in collaboration with Drs Vladimir Makarkin (Institute of Biology and Soil Sciences, Vladivostock) and Edmund A. Jarzembowski (Maidstone Museum) on the Neuroptera fauna of the Purbeck. Dr Vladimir Makarkin is a major authority on Neuroptera, especially fossil taxa; his input was to clarify and aid in identification of the specimens, while EAJ helped with the geological setting and environment of the specimens. This paper has been submitted to Palaeontology. The next chapter (chapter 5) is on the Purbeck raphidiopteran fauna, written in collaboration with Drs Robert A. Coram (British fossils) and Edmund A. Jarzembowski. RAC collected the specimen and gave information on the geology of the deposit, EAJ gave general advice on the paper. The paper was published in Cretaceous Research (Jepson et al. 2009b). The next set of chapters are on the Wealden Neuropterida; the chapter (chapter 6) on the palaeodiversity predictions of the Neuroptera of the Wealden was written in colloboration with Dr Dave Penney (University of Manchester). DP gave assistance in implementing his predictive methods previously used on fossil . This paper was published in Palaeogeography, Palaeoclimatology, Palaeoecology (Jepson & Penney 2007). The chapter (chapter 7) on the Neuroptera of the Wealden was written in collaboration with Drs Vladimir Makarkin and Edmund A. Jarzembowski. VN helped in clarifying and identification of the specimens and EAJ helped in the description of Psychopsites rolandi. The paper was published in Cretaceous Research (Jepson et al. 2009a). The next two chapters concentrate on the raphidiopteran fauna. The first of these (chapter 8) describes a new species from the Wealden, written in collaboration with Dr Edmund A. Jarzembowski. EAJ helped with the geological setting and partially with the description. The paper was published in Alavesia (Jepson & Jarzembowski 2008). The next chapter (chapter 9) describes more species from the Wealden in collaboration with Drs Jörg Ansorge and Edmund A. Jarzembowski. JA supplied and helped in the identification of the Spanish specimens, EAJ helped with geological setting. The paper is in press in Palaeontology. The next chapter (chapter 10) of this section is on the neuropterid

23 palaeoecology and palaeoenvironment of the Wealden and Purbeck. This is a sole author work submitted to Palaeogeography, Palaeoclimatology, Palaeoecology. Part one ends with a conclusion (chapter 11) and future work (chapter 12) summing up the work completed on the neuropterid fauna of the Lower Cretaceous of southern England.

Part two (chapter 13) concentrates on Raphidioptera, in particular the systematics and evolutionary history of the group. The first section comprises an introduction, literature review which includes an in-depth look at raphidiopteran morphology, previous work on raphidiopteran systematics and evolutionary history. The main part consists of two sections, the first (13.3) being a key to the fossil families and genera of Raphidioptera. This is a sole authored paper, which will be submitted to Zootaxa. The next section (13.4) is a sole author paper on a cladistic analysis on fossil and extant raphidioptera. A discussion, conclusion and ideas for future work summarizes this study on raphidiopteran evolutionary history.

24

CHAPTER 2. LITERATURE REVIEW

This chapter introduces the Neuropterida: their systematics, life habits, fossil record and geological settings of the Purbeck and Wealden

2.1. What are the Neuropterida?

Neuropterida is a superorder of insects, which includes the Orders Megaloptera, Raphidioptera and Neuroptera (Fig 2.1).

Figure 2.1. Top left: Megaloptera, lutaria from Delamere, UK. (Photograph: Lisa Abbott). Top right: Subilla confinis, (species also found in the UK) (Photograph: J.K. Lindsey). Bottom centre: Neuroptera, perla, UK (Photograph: www.glaucus.org.uk).

Neuropterida is a member of the holometabolous insects which are insects that undergo complete (dramatic change from one developmental stage to another); they have a distinct larval, pupal and adult stage (Fig. 2.2) (Aspöck 2002b; Grimaldi & Engel 2005). The larval stage is soft-bodied, wingless and morphologically reduced and the pupal stage is mostly quiescent (inactive) (Grimaldi & Engel 2005). The adults are those individuals which are reproductively mature.

25

Figure 2.2. Schematic diagram of the life-cycle of Neuropterida, using the example of Chrysopidae (Insecta: Neuroptera).

The defining feature of holometabolous insects is the larvae; the immature form is soft-bodied, slender, has six or no thoracic legs and no wing buds (Grimaldi & Engel 2005). The larval head is sclerotized with mandibles, palps, antennae and reduced eyes (or isolated ommatidia (conical structure of a ) separated into stemmata (single facet of a compound eye)). When the insect is in its pupal stage metamorphosis occurs – the majority (or all) of the larval cuticle is replaced by adult cuticle, the new cuticle for legs, eyes, genitalia and wings derived from the imaginal discs (Fig. 2.3). Imaginal discs are specialized epidermal cells (histoblasts) under the cuticle (in neuropteroids the imaginal discs develop in the last larval ); during the late larval and pupal stages the histoblasts increase and differentiate (Grimaldi & Engel 2005).

26

Figure 2.3. A schematic drawing showing the position of the imaginal discs (yellow) of the of Drosophila (Insecta: Diptera). A similar situation occurs in all holometabolous insects including the Neuropterida (redrawn from Grimaldi & Engel 2005 fig. 9.3 p. 334).

The Holometabola account for 85% of all Recent insects (Fig. 2.4) and the feature of complete metamorphosis is considered to be one of the two most important innovations of insect evolution (the other is the development of wings) (Grimaldi & Engel 2005). Holometabola is a considered to be a monophyletic group (Hennig 1981; Whiting et al. 1997; Whiting 2002b).

27

Figure 2.4. Pie chart showing the proportion of extant insects, based on described species. Although Holometabola are the dominant insect group, Neuropterida are only a very small proportion. Holometabola are dominated by , Diptera and Coleoptera. (Chart drawn with data from Grimaldi & Engel 2005).

2.1.1. Neuropterida’s place within Holometabola

Neuropterida is a monophyletic superorder comprising three Orders Raphidioptera, Megaloptera and Neuroptera. The monophyly of Neuropterida is based upon rather inconpicuous characters in thoracic and abdominal structures, the musculature of the ovipositor and characters of the larval stemmata (Aspöck et al. 2003). Neuropterida‟s relationships within Holometabola is as the sister-group to Coleoptera; this is based on the ovipositor structure and synapomorphies of the wing base (Mickoleit 1973; Hörnschmeyer 1998; Kristensen 1999; Aspöck et al. 2001; Aspöck et al. 2003; Grimaldi & Engel 2005). A mitochondrial genome phylogeny of the Neuropterida showed that Coleoptera and Neuropterida were sister-groups when run under a Bayseian hypothesis; however, when run under a parsimony analysis Neuropterida was found not to be monophyletic, Raphidioptera were placed as sister- group to all Holometabola, with the exception of , and Neuroptera were shown as sister-group to Diptera (Cameron et al. 2009). This is an interesting outcome; however, the vast majority of evidence at the present time, which includes morphological and most molecular studies, shows that Neuropterida are

28 monophyletic and sister-group to Coleoptera. Figure 2.5 shows a phylogenetic tree of the relationships within Holometabola; other than the basal Coleoptera and Neuropterida there are the Orders Glosseltyrodea, Hymenoptera (e.g. bees, wasps and ants), (scorpionflies), Siphonaptera (), Diptera (true ), (twisted wings), Trichoptera () and Lepidoptera ( and moths) (Grimaldi & Engel 2005).

Figure 2.5. Phylogenetic tree of holometabolous insects, showing the relationship of Neuropterida with the other holometabolous insects (modified from Grimaldi & Engel 2005). The solid black lines indicate ranges based on fossil evidence; dashed lines indicate extrapolated ranges based on phylogenetic relationships.

2.1.3. Relationships of the three Orders within Neuropterida

All three Orders of Neuropterida are considered as monophyletic; Neuroptera and Raphidioptera have been rarely considered otherwise, Megaloptera, however, has had its monophyly debated (see Aspöck et al. 2001), but is now considered to be monophyletic (Aspöck et al. 2001; Aspöck et al. 2003; Aspöck & Aspöck 2008).

29

Neuroptera monophyly is based upon the character of complex sucking tubes of the larva and the silken cocoon of the (Aspöck et al. 2003). Raphidioptera monophyly is based on the characters Sc running into C, third tarsale enlarged, tergite and sternite of abdominal segment 9 in male amalgamated to a ring, and ovipositor in female elongated to form a hose-like structure (Aspöck et al. 2003; Aspöck & Aspöck 2004). Monophyly of Megaloptera is supported by the organization of somatic ovarian tissues, tracheal gills of the larvae, and an often overlooked autapomorphy, the presence of eversible sacs with the fused gonocoxites 11 in and Sialidae (Aspöck et al. 2003; Aspöck & Aspöck 2008).

The proposed arrangement of the relationships within the superorder has changed over time. Traditionally it was thought that Neuroptera was sister-group to Raphidioptera + Megaloptera (Fig 2.6A). This hypothesis was based on symplesiomorphies, such as the gula, or on complex characters, for example, the postepimeron-complex, the polyintersegmental muscle between tergites 2 – 5 , the reduction of tergite 2, and various oviarole concepts (Kristensen 1999; Aspöck et al. 2003). Molecular studies also arrived at this result using 18S and 28S ribosomal DNA (Whiting et al. 1997; Whiting 2002a, b). The superficial similarity may have been the inducement for this hypothesis (Aspöck et al. 2003).

Figure 2.6. Relationships of the Orders within Neuropterida. A is the traditional hypothesis (e.g. Whiting et al. 1997; Kristensen 1999; Whiting 2002a, b), which has

30 been superseded by hypothesis B based on more robust evidence (e.g. Aspöck 1995; Aspöck et al. 2001; Aspöck 2002b; Aspöck & Aspöck 2008).

Recent studies, however, have moved towards a Raphidioptera + (Megaloptera + Neuroptera) hypothesis (Fig 2.6 B). This is a revival of Boudreaux‟s hypothesis (1979), which has corroboration mainly from the work of the Aspöcks (Aspöck 1995; Aspöck et al. 2001; Aspöck 2002b, c; Aspöck et al. 2003; Haring & Aspöck 2004; Aspöck & Aspöck 2008). The hypothesis is supported by characters from a morphological based study on larval characters, such as the head capsule, mouthparts, and primary aquatic lifestyle (Aspöck 1995), which was further corroborated by a cladistic approach based on numerous morphological characters (Aspöck et al. 2001; Aspöck & Aspöck 2008). It is additionally supported by molecular data (Haring & Aspöck 2004). The Raphidioptera are separated by autapomorphies, such as the tergite and sternite forming a ring in the male genitalia and the hose-like elongation of the ovipositor and the unpaired gonopophyses 8 of the same length (Aspöck & Aspöck 2008). Neuroptera and Megaloptera share the synapomorphies of an appendage-like gonocoxites and rosette arranged trichobothria on the ectoproct (in male and females) (Aspöck & Aspöck 2008). An autapomorphy of Neuroptera is the associated or amalgamated gonocoxites 9 with gonocoxites 11, and an autapomorphy of Megaloptera is the eversible sacs (Aspöck & Aspöck 2008). This relationship is the one accepted by most workers.

2.1.3. Relationships within the Orders

The relationships being reviewed in this section are at family level, while a more in depth review of Raphidioptera at generic level will be given in part 2 of this thesis.

There has been little doubt regarding the relationships within the Orders Raphidioptera and Megaloptera; both of these Orders have two families within, which comfortably form a sister-group relationship (Fig. 2.7). The evidence for the relationship Raphidioptera: Raphidiidae + Inocelliidae is based upon morphological evidence such as the absence of sternite 8 in the female terminalia Raphidiidae and fused sclerites in the parameres of the male terminalia (Aspöck et al. 1980; Aspöck et al. 2001). Molecular data also corroborate the sister-group relationship of these two families (Aspöck et al. 2003). The sister-group relationship of Megaloptera: Corydalidae + Sialidae is based on morphological evidence, for example, a paired

31 ectoproct in Corydalidae (unpaired in Sialidae), reduced trichobothria in Sialidae (organized into a rosette in Corydalidae) (Aspöck et al. 1980; Aspöck et al. 2001).

Figure 2.7. Relationships of the families in Raphidioptera and Megaloptera based on morphological and molecular evidence (e.g. Aspöck et al. 2001; Haring & Aspöck 2004; Aspöck & Aspöck 2008).

The phylogenetic relationships between Neuroptera have been long debated, with some relationships being better supported than others, based on weaker evidence. Recently U. Aspöck has done much work on the phylogeny of Neuroptera (Aspöck et al. 1980; Aspöck et al. 2001; Aspöck 2002b; Aspöck et al. 2003; Haring & Aspöck 2004; Aspöck & Aspöck 2008). Neuroptera can be divided into three suborders: Nevrorthiformia, Myrmeleontiformia and Hemerobiiformia based primarily on the larval head capsule (Aspöck 2002b). The larval head capsule found in Nevrorthiformia is considered archaic; it is compact with large gula, which is considered to be a pleisiomorphic state. The Nevrorthiformia archaic larval head capsule gave rise to the derived maxillary head (Hemerobioid head of MacLeod 1964) of the Hemerobiiformia, which is composed mainly of elements of the maxillae, and the more derived, heavy and large inflated head capsule (Myrmeleontoid head of MacLeod 1964) of Myrmeleontiformia, the inflated size of the head provided space for the musculature of the huge sucking tubes (Aspöck et al. 2001; Haring & Aspöck 2004).

32

Nevrorthiformia is composed of just one family (), Myrmeleontiformia has five families (, Psychopsidae, Nymphidae, Myrmeleontidae and ), while the largest of three suborders is Hemerobiiformia, with 11 families (Polystoechotidae, Ithonidae (=Rapismatidae), Chrysopidae, Osmylidae, Hemerobiidae, , Sisyridae, , , and Berothidae) (Aspöck et al. 2001; Aspöck 2002b; Grimaldi & Engel 2005). The monophyly of the suborders is based on the aforementioned structure of the larval head (Aspöck et al. 2001; Aspöck 2002b).

The relationships of the suborders are Nevrorthiformia + (Myrmeleontiformia + Hemerobiiformia) based on morphological evidence of the neuropteran larva (Aspöck et al. 2001; Aspöck 2002b) (Fig. 2.8A). The molecular and genital sclerite evidence, confirms the monophyly of Nevrorthiformia and Myrmeleontiformia, but splits Hemerobiiformia. The relationship of Nevrorthiformia + rest of Neuroptera still stands up to this new evidence, but, the division of Hemerobiiformia causes conflict (Aspöck et al. 2001; Aspöck & Aspöck 2008). The split of Hemerobiiformia puts Myrmeleontiformia as sister-group to part of Hemerobiiformia (Fig 2.8B)

Figure 2.8. Relationships of the the suborders of Neuroptera. A is based on morphological evidence (e.g. Aspöck et al. 2001) and B is based on morphological and in particular molecular evidence (e.g. Haring & Aspöck 2004; Aspöck & Aspöck 2008).

The relationships within Myrmeleontiformia have caused one major debate, which is: Psychopsidae + (Nemopteridae + (Nymphidae + (Myrmeleontidae +

33

Ascalaphidae)) (Fig. 2.9A) or (Psychopsidae + Nemopteridae) + (Nymphidae + (Myrmeleontidae + Ascalaphidae))? (Fig 2.9B). The former hypothesis is the traditional and the move is towards the latter based on the presence of a vena triplica in the hindwing (Aspöck et al. 2001) and gonocoxites connected by a sclerotized bridge (Aspöck & Aspöck 2008). However, the molecular data still support the former hypothesis (Haring & Aspöck 2004), therefore this relationship remains unresolved. The sister-group relationship of Ascalaphidae and Myrmeleontidae is corroborated by molecular data (Haring & Aspöck 2004) and morphological data including reduced trichobothria and a ventrolaterally divided tergite 9 (Aspöck et al. 1980; Aspöck et al. 2001; Aspöck 2002b). The sister-group relationship of Nymphidae + (Myrmeleontidae + Ascalaphidae) is supported by the presence of mandibles and grasping teeth (New 1989; Aspöck et al. 2001).

Figure 2.9. Relationships of the families within Myrmeleontiformia. The two hypotheses are highlighted A and B, the relationships are still unresolved (e.g. Aspöck et al. 2001).

Although the Myrmeleontiformia have unresolved relationships, it is nothing compared to the state of Hemerobiiformia. As mentioned before, conflicts have occurred between morphological characters, and morphological and molecular characters. It was previously thought that Hemerobiiformia was monophyletic, based primarily on the larval head capsule (Aspöck et al. 2001). The relationships of the

34

Orders from the first cladistic study are: (Ithonidae+Polystoechotidae) + ((Chrysopidae+Osmylidae) + ((Hemerobiidae + (Coniopterygidae + Sisyridae)) + (Dilaridae + (Mantispidae + (Rhachiberothidae + Berothidae))) (Fig 2.10) (Aspöck et al. 2001).

Figure 2.10. Relationships of the families within Hemerobiiformia. Based on morphological evidence of Aspöck et al. (2001).

The ithonid (Ithonidae + Polystoechotidae) is based on the synapomorphy of phytosuccivory of larva and the larval head morphology (MacLeod 1964; Aspöck et al. 2001; Aspöck & Aspöck 2008). The rest of the suborder is known as the „Higher Hemerobiiformia‟, defined by the loss of the larval gula and the terminal constriction of the cardines (Aspöck et al. 2001). Within the „Higher Hemerobiiformia‟ the dilarid clade (Dilaridae + (Mantispidae +(Rhachiberothidae + Berothidae))) receives strong support with the synapomorphies of the elongation of the postmentum, ventrally flat „maxillary head‟ and the unpaired mediuncus, reduction of the transverse sclerite in gonocoxites 11 to an arch shape and the potential character of a long ovipositor (Aspöck et al. 2001; Aspöck & Aspöck 2008) and the reduction of the transverse sclerite into an arch-shape (Aspöck & Aspöck 2008). The internal relationships within the dilarid clade, however, have been sufficient to debate. Rhachiberothidae has been considered as a subfamily, Rhachiberothinae, within the family Mantispidae (Willmann 1990), but now the consensus is that it is a family in its own right (Aspöck & Mansell 1994). The sister-group relationships of the families Coniopterygidae and Sisyridae are weakly supported with the terminal

35 segment of the labial palps being enlarged (although this may have evolved independantly) and narrowing of the cardines (Aspöck et al. 2001). Hemerobiidae is placed as sister-group to (Coniopterygidae + Sisyridae) + Dilarid clade based on the character of the cardines direction being parallel. The sister-group relationship of Osmylidae + Chrysopidae is based on the length of the cardines being shorter than half the length of the head capsule (Aspöck et al. 2001). The relationship of (Osmylidae + Chrysopidae) + rest of the „Higher Hemerobiiformia‟ is based on the characters of the gula being reduced to a small triangle and a terminal constriction of the cardines (Aspöck et al. 2001).

The molecular data on the whole disagrees with the study of Aspöck et al. (2001) causing a major disruption of the Hemerobiiformia (Fig 2.11).

Figure 2.11. Relationships of the families within Hemerobiiformia. Shows the disruption of the traditional view of the suborder with the use of molecular data (Haring & Aspöck 2004).

Haring and Aspöck‟s (2004) study separated Sisyridae and Osmylidae from the Hemerobiiformia. There is some corroboration for this with each having strong larval autapomorphies, however this arrangement is debatable. The dilarid clade, which gained strong support from morphological data is split with Coniopterygidae and Dilaridae forming a sister-group and the split of Berothidae and Mantispidae as a monophylum. Haring and Aspöck (2004) accept the split of the Dilaridae and they

36 reinterpret the former synapomorphies (secondary flat larval head, secondary broad insertion of the larval cardines) as sympleisiomorphies. The split of Berothidae and Mantispidae is still a problem as both have larval development is characterized by hypermetamorphosis (Haring & Aspöck 2004). Ithonidae came out as paraphyletic from the molecular study, which gave an arrangement of Australian Ithonidae + (American Ithonidae + Polystoechotidae); this may be due to the fact that the families are poorly. The placement of Hemerobiidae and Chrysopidae as a sister- group goes back to a traditional hypothesis based on larval similarities and the potential synapomorphy of the wing-like enlargement of the gonocoxites (Aspöck & Aspöck 2008).

The phylogeny of Neuroptera presently has many unresolved relationships, especially with the incorporation of molecular data and the Hemerobiiformia. Also, as yet no one has successfully created a phylogenetic tree comprising extant and extinct Neuropterida families; figure 2.12 incorporates some extinct families but not all.

37

Figure 2.12. Phylogenetic tree showing the relationship of the Neuropterida. All extant and some extinct families are shown (modified from Jepson & Penney 2007 fig. 2, using data from Grimaldi and Engel, 2005).

Other than the tree displayed in figure 2.12, little work has been done on incorporating fossil and extant neuropterans; the only other examples are Martynova (1949) and Schlüter (1986). However, these, like the tree from Grimaldi and Engel (2005), do not use all the extinct families, and unfortunately they do not give resolved relationships between the fossil and extant taxa. The reasons for this are most likely the lack of strong diagnostic characters within the fossil taxa, for example, the vast majority are recognized on wings only, whereas the extant forms have a wealth of characters, such as morphology, genitalia and the use of molecular data.

38

2.2. Developmental History of Neuropterids

Neuropterids can have varying time scales of development. This section reviews the current knowledge of development from egg to adult of each Order.

2.2.1. Neuroptera

Development of neuropterans from egg to adult lasts a few weeks (common) to a few years (rare). Most species undergo at least one complete generation each year; temperate species can have several generations in a limited time of the year, with a phenological break accompanied by a period of dormancy in winter (New 1989). Temperate Hemerobiidae tend to be univoltine (one generation), bivoltine (two generations) or multivoltine (many generations) with several overlapping generations during warm periods of the year. Neuropterans are generally multivoltine when conditions change, e.g. the environment (New 1989). Over-wintering occurs mainly as a pre-pupa within the cocoon (it is rare to have a pupa over-wintering). Adults and larva can overwinter and may even feed; fasting however, does occur (New 1989).

2.2.2. Raphidioptera

In Raphidioptera the egg stage lasts a few days to a few weeks and the pupal stage can last up to 10 months in some species. The number of larval is not fixed; it varies from 10–11 and may reach 15 (Aspöck 2002a). The larval period lasts from 1–6 years, but the majority are at the lower end of this range (Aspöck 2002a). Snakeflies hibernate in their last or penultimate larval stage or pupa (Aspöck 2002a). Larvae need low temperatures (near freezing) to induce pupation or hatching of the imago (final stage) (Aspöck 2002a; Grimaldi & Engel 2005).

2.2.3. Megaloptera

Few tropical megalopteran species have been studied in detail, and most of the present knowledge has been obtained from the study of Temperate region Megaloptera (New & Theischinger 1993). An aquatic larva is shared by all of the Order Megaloptera. Eggs of megalopterans tend to hatch in less than two weeks (New & Theischinger 1993). Adults are short-lived; however, this is not true for their larvae, which can last up to a few years, varying in different environments. For example, in the family Sialidae, has a one-year life-cycle in

39

(Iversen & Thorup 1987), a two-year life-cycle in Britain (Elliott 1977; Brooker 1979) and up to a three-year life-cycle in high altitudes in the Pyrenees (Giani & Laville 1973). Larvae of some species of Sialis may only last 10–11 months (New & Theischinger 1993). The family Corydalidae has larvae that can last in excess of three years, Dysmicohermes disjunctus larvae have been estimated to need four years to develop, and Orohermes crepusculus requires an estimated five years for development (New & Theischinger 1993).

Megalopterans live their life as larvae in streams, rivers and lakes. Their development after hatching may be delayed if the water body dries up. In this situation aestivation (period of inactivity/dormancy) occurs; for example in Protochauliodes aridus the larvae burrow into gravel of the dry streambed and wait until water returns. Some water, however, is required otherwise the larvae will expire (New & Theischinger 1993).

Larvae of the family Sialidae have 10 instars; in Corydalidae the number of instars is unknown due to the longevity of the larvae (New & Theischinger 1993). The last instar is the over-wintering stage that finally develops in the following spring (New & Theischinger 1993).

Pupation occurs terrestrially in a pit dug into the soil, or under rocks or mosses. The chamber can be many metres away from the water habitat of the larvae. The pupa stage is fairly short-lived, lasting under a month (New & Theischinger 1993).

2.3. Extant British Neuropterida

All three Orders of Neuropterida are present in Britain today, with Neuroptera being the most diverse. They are found throughout Britain in varying habitats, but are most numerous in woodland areas.

2.3.1. Neuroptera

There are six families of Recent Neuroptera in Britain: Chrysopidae, Coniopterygidae, Hemerobiidae, Myrmeleontidae, Osmylidae and Sisyridae. The first three mentioned families, Chrysopidae, Coniopterygidae and Hemerobiidae, are the more common families in the Order. In Chrysopidae there are 18 species in 7 genera: Chrysopa, Chrysoperla, Chrysopida, Cunctochrysa, Dichochrysa, Nineta

40 and Notochrysa (Plant 1997). Coniopterygidae has 12 species in 6 genera: Aleuropteryx, Coniopteryx, Conwentzia, Heliconis, Parasemidalis and Semidalis (Plant 1997). Hemerobiidae is the most diverse neuropteran in Britain with 30 species in 7 genera: Drepaneptertyx, Hemerobius, Megalomus, Micromus, Psectera, Sympherobius and Wesmaelius (Plant 1997). Myrmeleontidae, Osmylidae and Sisyridae are not as common in Britain as the other neuropterans. Myrmeleontidae is represented by one species in the . Osmylidae is also represented by one species in one genus: Osmylus. Sisyridae has three species in one genus: Sisyra (Plant 1997).

2.3.2. Raphidioptera

Raphidiopterans are fairly uncommon in Britain; however, they are quite diverse at the generic level. The raphidiopterans are from one family, Raphidiidae, and they comprise four species in four genera: Atlantaraphidia, Phaeostigma, Subilla and Xanthostigma (Plant 1997).

2.3.3. Megaloptera

Only three species are found in Britain in one genus: Sialis, which belong to the family Sialidae (Plant 1997).

2.4. Fossil Record of Neuropterida

The world-wide fossil record of Neuropterida is reviewed below.

2.4.1. Neuroptera

The first records of definitive Neuroptera date back to the Permian with the family Permithonidae, which are found in the former USSR (Martynova 1952), Siberia (Ponomarenko & Shcherbakov 2004), Urals (Novokschonov & Novokschonova 1997), Kansas (Tillyard 1937; Carpenter 1943a, 1976) and New South Wales, Australia (Tillyard 1926; Davis 1943). In the Triassic Neuroptera have been found in New South Wales, Australia: Prohemerobiidae, Psychopsidae, Osmylopsychopidae (Tillyard 1917, 1919, 1922, 1923), Queensland, Australia: Osmylidae (Lambkin 1988), Siberia: Archeosmylidae (Ponomarenko & Shcherbakov 2004) and Germany: Family incertae sedis (Brauckmann & Schlüter 1993). Neuroptera were widespread during the Permian and Triassic having been found in Russia, Australia and

41

America. However, it was during the Jurassic and especially the Cretaceous and into the „‟ that neuropterans had their „Golden Age‟.

In the Jurassic species of Neuroptera are found in the Lias, of Britain: Permithonidae and Prohemerobiidae (Whalley 1988), from the Lias of Dobbertin, Germany: Chrysopidae, Prohemerobiidae (Ansorge & Schlüter 1990), Luxembourg: Chrysopidae (Nel & Henrortay 1994; Ponomarenko 1995) in Solnhofen, Germany: Kalligrammatidae, Nymphitidae, Solenoptilidae, Mesochrysopidae, Myrmeleontidae, Hemerobiidae and Brongniartiellidae (Weyenbergh 1869; Meunier 1897; Carpenter 1929; Lambkin 1994; Ponomarenko 2003), Turkestan: Mesochrysopidae, Prohemerobiidae (Martynov 1927), Mongolia: Osmylidae, Prohemerobiidae, Mesithonidae, Kalligrammatidae and Mesochrysopidae (Ponomarenko 1992b), Karatau, Kazakhstan: Coniopterygidae, Mesithonidae, Promegalomidae, Kalligrammatidae, Psychopsidae, Mantispidae, Osmylidae, Epiosmylidae, Grammosmylidae, Mesochrysopidae and Nymphidae (Meinander 1975; Panfilov 1980; Engel 2005) and China: Kalligrammatidae, Polystoechotidae, Grammolingidae, „osmylid-like‟, Nymphidae, Osmylidae (Ren 2002; Ren et al. 2002; Ren & Oswald 2002; Ren & Yin 2003; Ren & Engel 2007).

There are many Cretaceous fossil insect localities, in both sediments and . Neuroptera were widespread and very diverse at this time, and many of the „modern‟ families appeared in the Cretaceous. Neuropterans are found in the Lower Cretaceous of Britain: Kalligrammatidae, Psychopsidae, Brongniartiellidae and Nymphitidae (Jarzembowski 1984, 2001; Clifford et al. 1994), Transbaikalia: Nymphidae, Babinskaiidae and (Ponomarenko 1992a), Israel: Palaeoleontidae (Dobruskina et al. 1997), Mongolia: Kalligrammatidae, Prohemerobiidae, Hemerobiidae, Grammostylidae, Mesochrysopidae, Myrmeleontoidea (Ponomarenko 1992b), Baissa, Siberia: Osmylidae, Mesithonidae, Nymphidae, Mantispidae, Psychopsidae, Chrysopidae (Makarkin 1990a, b, 1996, 1997a, b, 1999; Wedmann & Makarkin 2007), , China: Myrmeleontoidea, Myrmeleontidae, Aetheogrammatidae, Kalligrammatidae, Osmylidae, Chrysopidae, Berothidae (Ren & Guo 1996; Ren & Yin 2002; Ren & Engel 2008a, b), Laiyang Formation, China: Chrysopidae, Allopteridae (Yang & Hong 1990; Zhang 1991), , Brazil: Nemopteridae, Ascalaphidae, Nymphidae, Myrmeleontidae, Palaeoleontidae, Araripeneuridae, Babinskaiidae,

42

Makarkiniidae, Brongniartiellidae, Kalligrammatidae, Ithonidae, Osmylidae, Chrysopidae, Hemerobiidae, Sisyridae, Berothidae, Mesochrysopidae, Limaiidae, Allopteridae (Martins-Neto & Vulcano 1989a, b, 1990a, b, 1992, 1997; Martins- Neto 1990, 1992, 1994, 1997, 2000, 2002, 2005, 2006; Heads et al. 2005; Makarkin & Menon 2005; Menon et al. 2005; Makarkin & Menon 2007; Martins-Neto et al. 2007; Menon & Makarkin 2008; Martins-Neto & Rodrigues 2009), Japan: Osmylopsychopidae (Fujiyama 1978), Lebanese amber: Coniopterygidae, Rhachiberothidae (Whalley 1980; Azar et al. 2000; Nel et al. 2005a; Nel et al. 2005c; Engel & Grimaldi 2007), : Coniopterygidae, Rhachiberothidae (Engel 2004a, 2004c; Engel & Grimaldi 2007), Taimyr (Russia) amber: Coniopterygidae (Meinander 1975) and French amber: Rhachiberothidae (Nel et al. 2005c). The Upper Cretaceous yields neuropterans from New Jersey amber: Berothidae, Coniopterygidae, Psychopsidae, Mantispidae, Rhachiberothidae (Grimaldi et al. 2000; Engel 2002b; Engel & Grimaldi 2007), Canadian amber: Berothidae (Klimaszewski & Kevan 1986; Engel & Grimaldi 2007), French amber: Rhachiberothidae (Schlüter 1983; Engel & Grimaldi 2007), Siberian amber: Coniopterygidae (Engel & Grimaldi 2007), Russia: Psychopsidae, Osmylidae, Chrysopidae, Berothidae (Makarkin 1994) and Kazakhstan: Psychopsidae (Makarkin 1994).

Many Neuroptera are found in what was previously known as the „Tertiary‟, which is now the Palaeogene and Neogene (Gradstein et al. 2005). Most Neuroptera from this time are found in amber deposits, but there are also numerous sedimentary deposits that yield neuropterans. In the Palaeogene neuropterans are found in the Palaeocene of Queensland, Australia: Osmylidae (Lambkin 1987). Neuroptera are also found in the in the Fur Formation, Denmark: Polystoechotidae, Chrysopidae (Schlüter 1982; Willmann & Brooks 1991; Willmann 1993; Andersen 2001; Makarkin & Archibald 2003), British Columbia, Canada: Polystoechotidae (Makarkin & Archibald 2003), Parisian amber: Coniopterygidae, Rhachiberothidae, Sisyridae (Nel et al. 2003; Nel et al. 2005a; Nel et al. 2005c; Engel & Grimaldi 2007), Russian Far East: Ascalaphidae, Solenoptilidae, Chrysopidae (Makarkin, 1998), Baltic amber: Ascalaphidae, Nymphidae, Psychopsidae, Coniopterygidae, Dilaridae, Hemerobiidae, Nevrorthidae, Osmylidae, Rhachiberothidae, Sisyridae (MacLeod 1970; Meinander 1975; Engel 1999; Engel & Grimaldi 2007; Makarkin & Wedmann

43

2009), Hat Creek amber: Berothidae (Archibald & Makarkin 2004; Engel & Grimaldi 2007), Rovno amber, Ukraine: Coniopterygidae (Kupryjanowicz & Makarkin 2008), Germany: Mantispidae (Wedmann & Makarkin 2007), Okanagon Highlands, America: Ithonidae, Nymphidae (Archibald et al. 2009; Makarkin & Archibald 2009). Neuroptera have been recorded from the Oligocene Bembridge Marls in Britain: Mantispidae, Hemerobiidae (Cockerell 1921; Jarzembowski 1980), : Chrysopidae, Ascalaphidae, Nemopteridae, Myrmeleontidae?, Coniopterygidae, Hemerobiidae, Mantispidae (Nel 1988b, 1990; Nel & Séméria 1988), Mexican amber: Mantispidae (Engel & Grimaldi 2007), Dominican amber: Ascalaphidae, Chrysopidae, Coniopterygidae, Hemerobiidae, Mantipidae, Myrmeleontidae (Poinar 2006; Engel & Grimaldi 2007). The Neogene has a few insect deposits that yield Neuroptera; they include the Florissant: Hemerobiidae, Chrysopidae, Osmylidae, Nemopteridae (Cockerell 1907, 1908, 1909a, b, 1914; Carpenter 1935, 1943b, 1959; Adams 1967), Spain: Chrysopidae (Peñalver et al. 1995), and France: Sisyridae, Nevrorthidae (Nel & Jarzembowski 1997). The Upper Pliocene of Germany yields Chrysopidae (Schlüter 1982).

2.4.2. Raphidioptera

The earliest definitive snakeflies recorded are Lower Jurassic (Engel, 2002a). Older Palaeozoic specimens assigned to Raphidioptera have now been placed in other Orders (Carpenter 1962, 1992; Sharov 1968; Storozhenko & Novokshonov 1995; Storozhenko 1998) their placement within Raphidioptera has been thought suspect for many years (see Carpenter 1967). The Palaeozoic family Permoraphidiidae is now placed in (Carpenter 1967, 1992; Sharov 1968), Sojanoraphidiidae is now placed within Grylloblattaria (Storozhenko & Novokshonov 1995), and Letopalopteridae with (Novokshonov 1998); the family Fatjonopteridae was synonymized with the protorthopteran family Ampelipteridae (Scherbakov 1995). Aspöck and Aspöck (2004) debated the status of the family Priscaenigmatidae within Raphidioptera, based on the autapomorphies of the Order; they decided that this family was not raphidiopteran and called for the removal of it from the Order. Therefore the only true raphidiopteran families are Raphidiidae, Inoceliidae, Baissopteridae, Mesoraphidiidae and Alloraphidiidae.

44

Snakeflies were more diverse in the geological past than they are at present; the Mesozoic was their heyday and the Cretaceous itself was when great bursts of evolutionary activity took place. Mesoraphidiidae is the oldest family with species in the Lower Jurassic, while Baissopteridae, Alloraphidiidae and Inocelliidae are Cretaceous and Raphidiidae is „Tertiary‟ in origin.

The oldest snakeflies are from the Lower Jurassic (Lias) of Britain and Germany comprising the species Metaraphidia confusa Whalley, 1988 and Metaraphidia vahldieki Willmann, 1994 respectively. Two species, Ororaphidia megalocephala Engel and Ren, 2008 and Styporaphidia magia Engel and Ren, 2008, are found in Mongolia, China. In the Upper Jurassic snakeflies are found in Karatau, Kazakhstan: Proraphidia turkestanica Martynova, 1947, Mesoraphidia pterostigmalis Martynova, 1947, M. similis Martynova 1925b, M. parvula Martynov, 1925b, M. inaequalis Martynov, 1925b, M. elongata Martynova, 1925b and M. grandis Martynov, 1925a, and Mongolia M, gobiensis Ponomarenko, 1988.

In the Upper Jurassic–Lower Cretaceous boundary area of Gansu, China Sinoraphidia viridis Hong, 1982 is present.

The majority of fossil snakeflies originate from Cretaceous deposits. Baissa in Russia (Lower Cretaceous) is a very productive site for snakeflies with mesoraphidiids, Siboptera medialis Ponomarenko, 1993, S. eurydictyon Ponomarenko, 1993, Baissoraphidia orientalis Ponomarenko, 1993 and B. glossopteron Ponomarenko, 1993. The family Baissopteridae is first recorded in this deposit with the species Cretoraphidia reticulata Ponomarenko, 1993, C. magna Ponomarenko, 1993, C. macrocella Ponomarenko, 1993, Baissoptera sibrica Ponomarenko, 1993, B. minima Ponomarenko, 1993, B. kolosnitsynae Martynova, 1961, B. elongata Ponomarenko, 1993, B. cellulosa Ponomarenko, 1993 and B. martinsoni Martynova 1961. Alloraphidiidae also have their first appearance in Baissa with the species Pararaphidia vitimica (Martynova, 1961), Archeraphidia baisensis Ponomarenko, 1993 and Alloraphidia asiatica Ponomarenko, 1993. Cretoraphidia certa Ponomarenko, 1993 is found in another Lower Cretaceous deposit. Mongolia has a small fauna with three baissopteriids, Lugala longissima (Ponomarenko, 1988), Cretaraphidiopsis bonstagensis (Ponomarenko, 1988), and Cretinocellia cellulosa Ponomarenko, 1988, and four alloraphidiids, Pararaphidia

45 deserta (Ponomarenko, 1988), Archeraphidia yakowlewi Ponomarenko, 1988, A. hosbayari Ponomarenko, 1988, and Alloraphidia petrosa Ponomareko, 1988. The oldest member of the family Inocellidae, Sinoinocellia liaoxiensis Wang, 1987, is from Xiaoershijiazi, China. The Purbeck and Wealden deposits prior to this study had only one specimen from each deposit mentioned but not described (Jarzembowski 1984; Clifford et al. 1994). The Lower Cretaceous of Montsec, Spain has a limited fauna including Proraphidia gomezi Jepson and Jarzembowski, 2008. Chinese Cretaceous deposits have yielded a diverse fauna of mesoraphidiids including Jilinoraphidia dalazensis Hong and Chang, 1989 (Zhixian, Jilin province), Kezuoraphidia kezuoensis (Hong 1992b), Xuraphidia liaoxiensis Hong, 1992b (Kezuo. Liaoning), Siboptera fornicata (Ren, 1994), Mesoraphidia sinica Ren, 1997, M. shangyuanensis (Ren, 1994), M. polyphlebia (Ren, 1994), M. myrioneura (Ren, 1997), M. obliquivenatica (Ren, 1994), M. heteroneura Ren, 1997, M. longistimosa (Ren, 1994), M. glossophylla (Ren, 1997), M. gaoi (Ren in Ren et al., 1995), M. furcivenata Ren in Ren et al., 1995, M. amoena Ren, 1997, Huaxiaraphidia shandongensis Hong, 1992a, H. sinensis Hong, 1992a, and baissopterids Baissoptera liaoningensis Ren, 1994, B. grandis Ren in Ren et al., 1995, B. euneura Ren, 1997 and one alloraphidiid Alloraphidia anomola Ren, 1997 (Beipiao, Liaoning). Two mesoraphidiid species have been discovered in the Cretaceous of South Korea, Mesoraphidia koreenensis Engel et al., 2006 and M. phantosma Engel et al., 2006. The Crato Formation, Brazil has yielded four species of snakefly, three baissopterids, Baissoptera pulchra (Martins-Neto and Nel, 1992), B. brasiliensis Oswald, 1990 and Austroraphidia brasiliensis (Nel et al., 1990), and one familia incertae sedis, Arariperaphidia rochai Martins-Neto and Vulcano, 1989.

The middle Cretaceous has only yielded one species, an alloraphid, Alloraphidia dorfi Carpenter, 1967 from Labrador, Canada. Two species are found in the Upper Cretaceous, both are mesoraphidiids, Nanoraphidia electroburmica Engel, 2002a, from Burmese amber, which is the smallest snakefly ever recorded, and Mesoraphidia luzzi Grimaldi, 2000 from New Jersey amber.

The Cenozoic raphidiopterans are represented by Raphidiidae and Inocellidae. The Mesozoic families Mesoraphidiidae, Alloraphidiidae and Baissopteridae became extinct at the Cretaceous-„Tertiary‟ boundary, and it has been hypothesized by Aspöck (1998) that these families died out for the same reasons as the other

46 unfortunate at the Cretaceous-„Tertiary‟ mass . A small number of snakeflies have been discovered in the Eocene Baltic amber, the raphidiid species are baltica Carpenter, 1956, and the inocellid species Electroinocellia peculiaris (Carpenter, 1956) and carpenteri Engel, 1995. The main deposit in the Cenozoic for snakeflies is Florissant at the Eocene-Oligocene boundary with the raphidiid species Raphidia creedei Carpenter, 1936, R. exhumata Cockerell, 1909, R. mortua Rohwer, 1909, R. tranquilla Scudder, 1890, R. tumulata (Scudder, 1890) and R. veterana (Scudder, 1890). These species have been tentatively kept in Raphidia in Engel‟s (2002a) review of fossil snakeflies, but after revision he suggests that many will be taken out of this genus and reassigned to others within Raphidiidae. Only two inocelliid are known from Florissant, Fibla exusta (Cockerell and Custer, 1925) and Fibla erigena (Menge, 1856). A few Upper Miocene deposits have yielded snakefly species including the raphidiid Ohmella coffini Nel, 1993 from France and the inocelliid Fibla cerdanica (Nel, 1993) from Spain.

Modern snakeflies have approximately 206 species in the families Inocelliidae (21 species) and Raphidiidae (185 species); there are 25 recognized genera in Raphidiidae and 6 genera in Inocelliidae (Aspöck 2002b). They are restricted to the Northern Hemisphere and in temperate latitudes, in areas where there is a cold winter to help development (Aspöck 1998).

Raphidiopteran history spans the Jurassic to the present day. They are represented throughout geological time as both larvae and adults. The earliest snakeflies from the Jurassic already possessed synapomorphies of the Order suggesting an earlier origin, perhaps Triassic or late Permain. The age of related clades, e.g. Neuroptera and Megaloptera, would also suggest this, but fossil evidence is needed to support this conjecture (Engel 2002a).

2.4.3. Megaloptera

Megaloptera have a sparse fossil record with very few species. The oldest megalopterans are Permian in age and comprise Parasialis dissedens Ponomarenko, 1977, P. latipennis Ponomarenko, 1997, Sojanosialis longipennis Ponomarenko, 1977 from Soyana River, Parasialis ovata Ponomarenko, 2000 (family Parasialidae) and Tychtodelopterum archaicum Ponomarenko, 1977 from Kuznetsk Basin, Russia (family Tychtodelopteridae).

47

No megalopterans have been found in the Triassic. In the Jurassic one species in the family Sialidae, Dobbertina reticulata Handlirsch, 1920, is found in Dobbertin Germany. The Lower Cretaceous has two sites where megalopterans have been found: Cretochaulus lacustris Ponomarenko, 1976 (family Corydalidae) from Baissa, Russia and unnamed specimens from the Crato Formation, Brazil (Martins- Neto et al. 2007). The Upper Cretaceous has one site at E. Taymyr, Russia with one species, Chauliosalis sukatshevae Ponomarenko, 1976, in the family Corydalidae.

The Cenozoic has the most fossil megalopteran species. The Palaeocene has one species, Austrosialis sp., in the family Sialidae from Queensland (Lambkin 1992), Australia. The Eocene Baltic amber has three species in the family Corydalidae, Corydasialis inexpectatus Wichard, et al., 2005, Chauliodes carsteni Wichard, 2003 and C. prisca Pictet, 1854 , and four species from the family Sialidae, Sialis groehni Wichard, 1997, S. (Protosialis) baltica Wichard, 1997, S. (Protosialis) herrlingi Wichard, 2002 and S. (Protosialis) voigti Wichard and Engel, 2006. Another amber deposit, the Eocene Parisian amber, has one species of Megaloptera in the family Sialidae, Eosialis dorsi Nel et al. 2002. The Miocene has five species in the family Sialidae, four from France: Proindosialis cantelensis Nel, 1988, Sialis muratensis Nel, 1988, S. sp., and one undetermened species (Nel 1990), and one species from Turkey, Indosialis beskonakensis Nel, 1988. One species in the family Sialidae, Sialis strausi Illes, 1967, is found in the Pliocene rocks of Germany.

There are approximately 270 species of extant Megaloptera belonging to two families, Corydalidae and Sialidae (Engel 2004b).

2.4.4. Fossil British Neuropterida

Fossil Neuropterida have been found in the „Tertiary‟, Cretaceous, Jurassic and Triassic rocks in Britain. Little work has been done on these insects, and much of what has been done was undertaken in the 19th and early 20th Century. A note of warning is needed when referring to 19th Century literature, because terms such as „Neuroptera‟, „Neuropterous‟ and „Sialis‟ were used as bucket terms for many insects with a dense venation pattern, for example, ( and ) would be regarded as Neuropterous/Neuroptera. A notable example of this is in Brodie‟s (1845) work on fossil insects in Britain.

48

There has yet to be any fossil representatives of Megaloptera found in Britain. Table 2.1 lists the localities with taxonomic data (at the family level) included when known.

Age Order Family Locality Reference

Triassic Neuroptera Prohemerobiidae Strensham (Tillyard 1933)

Jurassic Neuroptera Leicestershire (Woodward 1892)

Neuroptera Prohemerobiidae Alderton (Scudder 1886; Handlirsch 1906– 1908)

Neuroptera Prohemerobiidae Dumbleton (Tillyard 1933)

Raphidioptera Mesoraphidiidae Charmouth (Whalley 1985, 1988)

Cretaceous Neuroptera Prohemerobiidae Dorset (Purbeck) (Whalley, 1988)

Uncertain Dorset (Purbeck) (Giebel 1856)

Brongniartiellidae Dorset (Purbeck) (Westwood 1854; Giebel 1856; Scudder 1885; Whalley 1988)

Nymphitidae Dorset (Purbeck) (Westwood 1854; Giebel 1856; Handlirsch 1906– 1908)

Raphidioptera Mesoraphidiidae Dorset (Purbeck) (Clifford et al. 1994)

Neuroptera Kalligrammatidae (Jarzembowski 1984, 2001)

Psychopsidae (Jarzembowski 1984)

Brongniartiellidae (Jarzembowski 1984)

Palaeogene Neuroptera Mantispidae Isle of Wight (Cockerell 1921; Jarzembowski 1980)

Hemerobiidae Isle of Wight (Jarzembowski 1980) Table 2.1. The fossil record of neuropterids in Britain.

49

2.5. Environment and Ecology

This section comprises a review of the environment, ecology and plant and prey associations. The literature is reviewed with a view to using modern analogues for the fossil Neuropterida of the Purbeck and Wealden. With regards to Raphidioptera and Megaloptera the whole Orders will be investigated to gather modern analogues, because the fossils of these Orders are from families that are extinct. Some neuropteran families from the Purbeck and Wealden have persisted to the modern day; these include Chrysopidae, Berothidae, Nymphidae, Ithonidae, Psychopsidae, Osmylidae and Hemerobiidae, and these therefore have been the focus in the literature review.

2.5.1. Neuropterida

Many neuropterid taxa are associated with specific plants and vegetation. They use the plants and vegetation in a variety of ways, whether it is for a platform to lay eggs, protection for the larvae, food source or hunting ground (New 1989; Plant 1997; Grimaldi & Engel 2005). Prey plant specificity or preference can be a likely influence on habitat selection (New 1989).

Neuropterida are predators of small invertebrates making them of economic use; they are often used as pest control on crops (e.g. Wise 1995). Both larvae and adults are predators with the exception of the adults of some families who have taken to feeding upon pollen (Megaloptera, Nemopteridae, Sisyridae) (New 1989; New & Theischinger 1993; Grimaldi & Engel 2005). Some Neuropterida are generalist predators, feeding on any type of prey, while some are more specialized predators, restricted to one type. The neuropterids which are restricted to one type of prey tend to be found living in the same environments (New 1989).

Duelli et al. (2002) investigated Neuroptera and Raphidioptera habitats in forests. They found that it was the edges of forests rather than the forest interior that was the main habitat for many of them. The study, however, was based on a managed forest, and this therefore may not be true in a wild situation.

Microhabitats have scarcely been investigated, especially with regards to Neuroptera (New 1989). The little work done shows that different species of Neuropterida live in different areas on the same tree, with some being trunk dwellers and some being

50 canopy dwellers. The canopy dwellers themselves can live in different layers within the canopy (Barnard et al. 1986; New 1989). Trunk dwellers tend to be brown and mottled and the canopy dwellers are pale, demonstrating their different camouflage adaptations (Barnard et al. 1986).

2.5.2. Neuroptera

The life-cycle of most extant Neuroptera is in some way associated with vegetation. Below are some examples of the different ways in which the neuropteran families utilize this resource. The examples are illustrated with varied ; this has been done because of the great variation of plant utilization seen within families, genera and species.

Psychopsidae have predominantly arboreal larvae; living under or within deep crevices in the bark (Grimaldi & Engel 2005). Some subcortical larvae appear to be limited to Eucalyptus (New 1989). Immatures of the family Nymphidae live in litter and under logs. Some larvae are arboreal and have become flattened to live under the bark or in crevices on the bark (Grimaldi & Engel 2005). Ithonidae (=Rapismatidae) larvae are associated with decaying bark and vegetation, which may form a source of food for them (Grimaldi & Engel 2005).

The majority of species in the family Hemerobiidae and Chrysopidae are arboreal (Grimaldi & Engel 2005). A great majority occur on vegetation, some having host- plant specificity (New 1989). The hemerobiid Wesmaelius is often associated with low-lying vegetation, others occur on all types of vegetation, with some being specialized on various trees, for example, Abies (Fir), Pinus (Pine), Quercus (Oak) and Populus (Poplar) (Monserrat & Marin 1996). European species of Wesmaelius are clearly associated with (New 1989). Some hemerobiids and chrysopids feed on pollen and other plant substances (Monserrat & Marin 1996).

The family Osmylidae may possibly have an aquatic larva, although this has not been studied in detail. The adult osmylids have been consistently found near bodies of water; this behaviour is observed in most insects that have an aquatic larva, for example, megalopterans (New 1989; Grimaldi & Engel 2005).

The majority of neuropterans, both adults and larvae, are predators (New 1989; Grimaldi & Engel 2005). Hemerobiidae and Chrysopidae are found around prey

51 colonies such as . They usually hunt by walking, but other hunting strategies are used such as „sit and wait‟ and with the competent fliers, hunting on the wing (New 1989). The osmylid Osmylus fulvicephalus probes wet mud to feed on dipteran larvae. Cannibalism has been observed and of the first instar by siblings could be due to scarcity of prey (Duelli 1981). Berothidae is seen to hunt on , and this does seem to be a specialism (New 1989).

Little evidence of prey specificity within Neuroptera has been found and the prey is usually limited to what physically can be caught (New 1989). Most neuropterans have a varied diet, for example, the larvae of hemerobiids and chrysopids, which consume many small such as , aphids and other scale insects (Homoptera). Prey specificity has not been studied in great detail within Neuroptera. It may in fact be preferred prey of an insect as part of a varied diet (New 1989).

Some species of Neuroptera do seem to have prey specificity and this can be induced from the first meal the larvae feeds on; if changed after it could be fatal (Babrikova 1979). The work of Hydorn and Whitcomb (1979) and Canard (1970, 1973) on chrysopids has shown developmental abnormalities resulting from feeding on certain types of prey, e.g. failure to spin cocoons, death as pupa within the cocoon and males becoming sterile. Less fatal changes may occur such as a change in the weight of the cocoon (New 1989).

Miller et al. (2004) have undertaken work on the predator/prey relation of Neuropterida and Coccoidea (). This work lists numerous species of Hemerobiidae (Neuroptera), Inocelliidae and Raphidiidae (Raphidioptera) from the Afrotropical, Australasian, Nearctic (Greenland, highlands of Mexico and most of ), Oriental and Palearctic (, Asia north of Himalayas, northern Africa) regions and their Coccoidea prey. The study, however, was biased towards managed agricultural ecosystems.

Prey tends to be discovered at random by neuropterans. Some prey can be rejected by neuropterans on the basis of the prey‟s physical characters, e.g. the hard cuticle of a gall wasp (Thorne 1972), the waxy cuticle of some aphids (Ru et al. 1975) and aphid cornicle excretions (Pupedis 1987).

52

Neuroptera are prey for many insects, for example, Odonata; the adults are caught on the wing by many species of dragonfly. The dipterans Empididae and Asilidae also feed on neuropterans, as do Hymenoptera (Bembix). Mites are known to eat neuropteran eggs (Killington 1925; Evans 1978; Evans et al. 1982; New 1989). Vertebrate predators of Neuroptera include birds and bats (New 1989; Szentkirályi & Kristín 2002)

2.5.3. Raphidioptera

Snakeflies (adults and larvae) live almost exclusively on trees, shrubs and bushes (Aspöck 1998, 2002a; Grimaldi & Engel 2005). They are found from sea level up to the timberline, in areas which have a cold winter period (Aspöck 1998, 2002a). The extant British fauna (Raphidiidae) have been recorded on various types of trees such as conifers, oak and willow (Plant 1997). Snakeflies use their long ovipositor to lay eggs in crevices in the bark or under it, and the larvae then live on or under the bark (Grimaldi & Engel 2005). Other species larvae live in the soil and leaf litter at the base of vegetation, and also in rock crevices (Aspöck 1998, 2002a; Grimaldi & Engel 2005). Their whole life is confined to the aboreal environment; they are poor fliers and often will not leave „their‟ shrub or tree (Aspöck 1998).

The larvae of Raphidioptera are active predators hunting at night and feeding on a variety of invertebrates; their prey can include the larvae and eggs of other arthropods in particular Lepidoptera, Coleoptera and Hymenoptera. They will also feed on the larvae and adults of insects (e.g. , , , Collembola, mites) and even spiders (Aspöck 2002a). There has been little field study on raphidiopterans to determine their true diet and to investigate differences, if any, between the tree dwelling larvae and the leaf litter dwelling larvae; however, experiments in laboratory conditions have yielded no differences (Aspöck 2002a).

The adults of Raphidiidae are predacious with their main diet being aphids and other small arthropods; however, pollen has been eaten when in captivity and also has been found in the gut contents of wild raphidiids. The diet of Inocelliidae is virtually unknown; however, in captivity, as with Raphidiidae, pollen is taken as food, but pollen has not been found in the gut of wild inocelliids. (However, no extensive field research has been undertaken to discover the diet of the family (Aspöck 2002a)).

53

Raphidiopterans form food for predaceous insects, and other arthropods, for example, spiders, and they are often taken by birds (Szentkirályi & Kristín 2002).

2.5.4. Megaloptera

Adults of this Order utilize vegetation overhanging water to lay eggs, so that when the larva emerges it drops directly into the underlying water (Engel 2004b; Grimaldi & Engel 2005). Feeding strategies of adult Megaloptera is on the whole unknown (New & Theischinger 1993). It was thought that adult megalopterans did not feed, instead living on the reserves stored when feeding in the larval stage. However, adults of some species have been observed feeding on pollen and nectar, therefore suggesting that they must visit flowers to feed (Dubois & Geigy 1935; New & Theischinger 1993; Engel 2004b). Megalopterans in captivity have readily fed on honey-water solutions (Parfin 1952), fermented molasses and bananas with water and sugar (New & Theischinger 1993). Protohermes grandis has been observed feeding on tree sap (Yoshida et al. 1985). The feeding strategies above may be linked to the sex of the insect; females in laboratory conditions readily went for the sweet liquids, whereas males were content just drinking water. The large mandibles of the male corydalids also inhibit feeding as an adult (New & Theischinger 1993).

The larva of megalopterans are aquatic. Larva hatch from overhanging vegetation and drop into the underlying water where they live until they are ready to pupate, when they then leave the water for good (Grimaldi & Engel 2005). The aquatic habitat of Sialidae is diverse; they are found in freshwater, in small ponds and streams, lakes, and large slow-flowing rivers. Corydalidae are also found in diverse aquatic environments, but, the majority are to be found in streams (Grimaldi & Engel 2005). Some larvae (e.g. Sialis lutaria) when small live in the littoral zones (shallow water), moving when larger in their second year to the sub-littoral or profundal zone (deeper water) (New & Theischinger 1993).

The larvae can also inhabit microhabitats, for example, this has been observed when there are coexisting sialids and corydalids. The work of Evans (1972) on Neohermes californicus and Protochauliodes aridus in one stream in California, demonstrated the microhabitats of these two corydalids. N. californicus was always found in the deeper faster water, whereas P. aridus was found in the shallower water. Evans also observed another microhabitat choice for N. californicus and another species of

54 corydalid P. spenceri. These two species coexist in two small, connected intermittent streams, one with a soft substrate and the other with hard substrate. N. californicus was observed in the firm substrate stream and P. spenceri the soft substrate stream. This work, however, has not been verified in other streams inhabited by corydalids. Other factors can influence where various species are found; these include size of stream, water temperature, water pH and the level of pollution in the water (New & Theischinger 1993).

The larvae of Megaloptera are voracious predators of various aquatic invertebrates and vertebrates, and cannibalism is frequently observed (New & Theischinger 1993). They prey on larvae of , caddisflies, damselflies, dipterans (chironomids, Simulium), adult isopods, tadpoles and tubificid worms (New & Theischinger 1993). There are no links with prey interaction with plants and therefore they are not tied to any particular plant because of their prey.

Megaloptera are prey for other insects, for example, Trichoptera and Odonata, and their eggs form part of the diet of larvae, the neuropteran family Sisyridae and slugs. Vertebrate predators include rodents and birds (New & Theischinger 1993).

2.5.5. The Percentage of Neuropterida in Environments

Neuropterida do not make up a large percentage of the insect fauna in specific habitats, neither extant nor fossil. This is probably due to the fact that they are predators and there are always fewer predators than prey. Below is the percentage of Neuropterida in various insect faunas.

In the Wealden there are 0.1–0.9% Neuroptera present (Jarzembowski 1995), while in the Purbeck Neuroptera and Raphidioptera are not as rare (Coram & Jarzembowski 2002). Less than 1% is normal in both extant and fossil habitats, higher proportions have been known, but they are rare. Examples of these are the Okanagan Highlands, British Columbia (1.5%) and the Fur Formation, Denmark (3.8–5%). In extant communities Neuroptera make up less than 1% of the insect fauna, for example, they are less than 1% in the tropical forests of Borneo; 0.45% in Queensland, Australia; and 0.25% in Primorye, Russia (Archibald & Makarkin 2006; Jepson & Penney 2007).

55

2.6. Geological Settings

The insects studied herein are from two Lower Cretaceous deposits in Britain, the Purbeck Limestone Group, and the Wealden Supergroup (Weald Sub-basin). This section gives the geological settings and palaeoenvironments of these two deposits. The climate change observed through the Purbeck into the Wealden will also be reviewed.

2.6.1. Mesozoic flora, a brief note

The flora of the Lower Cretaceous (Purbeck and Wealden), southern England is dominated by gymnosperms and pteridophytes (e.g. Francis 1983; Hill 1996). One putative Angiosperm has been identified from the Wealden (Hill 1996). Angiosperms first appeared around the Jurassic/Cretaceous boundary, but, their main radiation did not occur until ~115MA (phylogenetic tree in Grimaldi & Engel 2005 page 612). The main radiation took place ~5–10MA after the end of the Wealden, therefore not having a great effect on the entomofauna. Gymnosperms and pteridophytes, however, were present long before Purbeck and Wealden times (Thomas 1981).

2.7. Purbeck Limestone Group

The Purbeck Limestone Group is most dominant in present day Dorset and Wiltshire; this section will be looking at the deposit‟s palaeogeography, palaeoclimate and palaeoenvironment. The majority of the neuropterid specimens are from the type area for the Purbeck Limestone Group, Durlston Bay, Swanage, Dorset.

2.7.1. Palaeogeography

The latitude of the Purbeck during deposition has been suggested as being 30–35°N (Ogg et al. 1994) (where the present day Mediterranean is situated). It lay between a complex of massifs (Fig. 2.13); Cornubia, Armorica, Londinia and the Irish and Welsh Massifs (Allen 1998). The massifs were bordered by the widening Protoatlantic to the west, the Boreal Sea to the northeast and the Tethys Sea to the south (Allen 1998). The sediment was deposited between the massifs in the Wessex sub-basin with Cornubia to the west and Armorica to the south (Hancock & Rawson

56

1992; Allen 1998). Purbeck sediments were also deposited in the Weald sub-basin (which yield little insect material) and the Midlands basin (very insectiferous deposits) (Brodie 1845; Coram 2005).

Figure 2.13. Palaeogeographic map at the time of the deposition of Purbeck and Wealden sediments (redrawn from Allen 1998).

The environment of deposition was originally thought to be estuarine (Brodie 1845; Fisher 1856), but has now been widely accepted (based on more complete geological evidence) to be lagoonal with intermittent connection to the sea. (Clements 1973). Lagoonal deposition dominates the Lower Purbeck, with little or no influence from the sea. The Middle and Upper Purbeck have much greater influence from the sea, observed by salinity changes and differences in the aquatic fauna (Clements 1973; Coram 2005). The lagoon was bordered (or broken up) by low-lying marshes, mudflats, shell beaches and sabkhas, and was also subjected to inundation, with subareial exposure observed by the presence of dinosaur footprints and desiccation cracks (Ensom 2002a). The exposed flats would have been several 10s of kilometres wide (limited in size in the north due to the South Dorset Swell) and they were probably unvegetated (scarcity of roots and plant remains) (Coram 2005). The present day Trucial Coast (Persian Gulf) and Laguna Madre (Texas/Mexico) have been cited as modern examples of such a landscape (Schreiber 1986; Eisma 1998; Coram 2005).

57

A marine regression is observed at the start of the Purbeck, going from the marine deposits of the Portland Beds (Jurassic) to the non-marine Purbeck Limestone Group (Lower Cretaceous). During this transition the region was gradually cut off from the open-sea (Batten 2002).

2.7.2. Palaeoclimate

The lower and middle Purbeck are now widely accepted as having a semi-arid, Mediterranean type climate, gradually becoming wetter towards the end of the (Sladen & Batten 1984; Allen 1998; Batten 2002). This is reflected in the facies, with hypersaline deposits common in the lower Purbeck, and freshwater clastics dominating the upper part (Clements 1993; Batten 2002).

2.7.3. Palaeoenvironments

The Berriasian Purbeck Limestone Group consists of the Lulworth and Durlston formations (Fig 2.14). Below is an in-depth look at the stratigraphy and environments of deposition.

58

Figure 2.14. Stratigraphic column of the Purbeck Limestone Group, modified from Rasnitsyn et al. (1998), showing neuropteran localities.

2.7.4. Lulworth Formation (Fig. 2.14):

The lower part of the Lulworth Formation is the Broken Beds Member, which was deposited in a hypersaline lagoon. The upper part of the member was deposited in a more shallow water environment (Batten 2002). Moving up through the succession the next member is the ‘Cypris’ Freestone (named after a genus of ostracod). This member is the oldest part of the Purbeck Limestone Group where fossil insects have been found (Coram & Jarzembowski 2002). It was deposited in a lagoonal environment in shallow water with varying salinity, which is now preserved as

59 limestone (West 1975). The lagoon partly dried up in droughts in the summer months leaving evaporites such as halite within the fine carbonate sand (Batten 2002). The next unit, the Hard Cockle Member was deposited in a hypersaline environment, indicated by the presence of numerous evaporites. The overlying Soft Cockle Member has a similar environment, but with marls and shales in the upper part, with a sabkha facies common (Batten 2002). Moving into the Middle Purbeck from the Lower, one finds the penultimate member of the Lulworth Formation, the Marly Freshwater Member. This member was deposited in a shallow lake; and contains the well-known Mammal Bed (Batten 2002). The last member of the Lulworth Formation, the Cherty Freshwater Bed, was deposited in slight saline conditions with „seasonal‟ droughts causing changes in salinity (Allen 1998).

2.75. Durlston Formation (Fig. 2.14):

The Durlston Formation begins in the Middle Purbeck with the Cinder Member the base of which formerly marked the Jurassic/Cretaceous boundary. (Coram & Jarzembowski 2002; Coram 2005). The member has been suggested to have been deposited in a fully marine environment (Batten 2002). Overlying the Cinder Bed is the Intermarine Member; the lower part of this member was deposited in a freshwater-slightly brackish lake, while the upper part was deposited in more saline conditions with some shell beaches (El-Shahat & West 1983; Batten 2002). Neither of these deposits have yielded any fossil insects (Coram 2005). Next is the Scallop Member, deposited in a marine environment, also devoid of insects. Overlying the Scallop Member is the Corbula Member (named after a genus of bivalve). This is one of the main insect-bearing units of the Purbeck (Batten 2002; Coram & Jarzembowski 2002), and was deposited in mainly near marine conditions (Batten 2002). Above the Corbula Member is the Chief Beef Member, which yields no fossil insects, and was deposited in brackish conditions (Batten 2002; Coram & Jarzembowski 2002). Overlying the Chief Beef Member is the Broken Limestone Member and Unio Member deposited in freshwater (Batten 2002). The transition from the Chief Beef to the Broken Limestone Member marks the change from Middle to the Lower Purbeck. The youngest unit of the Purbeck is the Upper ‘Cypris’ Clays and Shales Member (including the Purbeck Marble) which is overall non-marine and similar to the overlying Wealden Group (Batten 2002).

60

Coram‟s (2003) work on the taphonomy and ecology of Purbeck insects splits the insects into an aquatic and terrestrial fauna. The articulated aquatic fauna lived in or quite near to the major bodies of water; being articulated shows evidence of little travel or little time to decay and disarticulate. The terrestrial fauna is on the whole disarticulated, which suggests that there has been an element of transport or period of decay. Distant woodland has been suggested for the disarticulated insects and for the better-preserved ones it is suggested that they may have lived on the sparsely vegetated lagoon margins (Coram 2003).

The environment proposed by Coram (Coram 2003, p. 316) is one of a brackish lagoon bordered by a shell beach or sabkha, with freshwater lake/marsh a damp forest/water margins and open glades with a distant forested region.

2.7.6. Palaeoflora

Algae, spores and pollen grains are present within the Purbeck. These are useful environment indicators (e.g. Batten 1982, 2002), but, it is difficult to link dispersed pollen and spores to their growth communities and parent plants, especially in the absence of macrofossils (Batten 2002).

There are few plant macrofossils in the Purbeck (Batten 2002). Cycadophyte remains have been found, as have conifers. Silicified pieces of coniferous trees have been found higher up in the succession, which are the same type as those that dominate the Lower Purbeck soil beds (Batten 2002). Foliage of different species have been discovered, but not yet studied, only an „Equisetum‟ stem, coniferalean foliage and Araucarites cone have been described (Barker et al. 1975; Batten 2002). The palynological evidence from the Lower Purbeck suggests that gymnosperms dominated. Charcoal is present within the Purbeck indicating that wildfires burnt the vegetation. However, the fires do not seem to have affected the tree stumps, and therefore, they must have been of low intensity affecting the foliage only (Batten 2002).

The Purbeck vegetation is not very diverse; it is dominated by conifers (one or two species) and has less common benittitaleans, ferns, lycopods, mosses and liverworts. No angiosperms are present (Francis 1984).

61

2.7.7. Palaeofauna

The entomological fauna consists of Coleoptera, Hemiptera, Diptera, Orthoptera, Mecoptera, Heteroptera, , Hymenoptera, Trichoptera, Odonata, Neuroptera, Raphidioptera and Megaloptera (Coram & Jarzembowski 2002).

The entomofauna is split into aquatic and terrestrial; the aquatic fauna can be fully preserved or disarticulated, reflecting transport, the disarticulated representing an allochthonous fauna and the articulated being autochthonous (Coram 2003). The terrestrial entomofauna are disarticulated, reflecting transport, e.g. Coram‟s (2003) disarticulation on sabkha/shell beach, then after disarticulation the insect wings and sclerites being washed into pools or collected by the rising lagoon.

Bivalves, gastropods, ostracods, crustaceans, chelicerates, crocodiles, turtles, fish, dinosaurs, amphibians and mammals make up the rest of the fauna of the Purbeck Limestone Group (see papers in Milner & Batten 2002).

2.8. Wealden Supergroup (Weald Sub-Basin)

The Wealden Supergroup of southern England is a succession of Lower Cretaceous rocks deposited in two sub-basins the Weald and Wessex. This section concentrates on only one of these sub-basins, the Weald, because to date no Neuropterida have been discovered in the Wessex Sub-basin.

2.8.1. Palaeogeography

The distribution of land and sea from the Purbeck (Berriasian) to the Lower Weald Clay (mid-) did not change (Fig. 2.13); however, a gradual retreat of shorelines did occur (Hancock & Rawson 1992).

The Weald Sub-basin, as with the Purbeck Wessex Sub-basin was surrounded by a number of massifs: Londinia to the north, Cornubia to the west, Armorica to the south and the Welsh massif to the northeast (Hancock & Rawson 1992; Radley 2005).

The sediment (and derived fossils) in the Weald Sub-basin is largely from the Londinia massif (Allen 1975; Radley 2005), but, during Weald Clay times, sediment

62 especially sand, was occasionally washed in from the Cornubia massif (Radley 2005).

2.8.2. Palaeoclimate

Much work has been done on the climate of the Wealden, using evidence from various sources: palaeontology (e.g. fossil insects: Jarzembowski 1995), sedimentology (Allen & Wimbledon 1991; Allen 1998) and climate models (Haywood et al. 2005). This multi-disciplinary work suggests a sub-tropical/warm temperate „Mediterranean-like‟ climate with the occurrence of wet and dry (commonly with wild fires) seasons. Also proposed is year-round precipitation with little evidence for prolonged drought; the apparent dry season is due to higher evaporation rates in the summer months (Batten 1998; Haywood et al. 2005). Average temperatures proposed by Haywood et al. (2005) are 4–8°C for cold months and 36–40°C for the warm months. Evidence for wildfires exists in the form of fusainized beetle elytra and plant remains (Watson & Alvin 1996; Batten 1998; Jarzembowski 2003).

2.8.3. Palaeoenvironment

The Wealden Supergroup consists of the Hastings Beds Group and the Weald Clay Group (Fig. 2.15). Below is an in depth look at the environments of the Wealden.

63

Figure 2.15. Stratigraphic column of the Wealden Supergroup modified from Rasnitsyn et al. (1998), with insect localities indicated.

The Hastings Beds and Lower Weald Clay groups (Fig. 2.15) represent three „megacycles‟ (Allen 1975; Allen & Wimbledon 1991; Radley 2005). A megacycle consists of an alluvial deposit, Allen‟s (1981) arenaceous formations. These deposits are observed in the Ashdown Beds, and the Lower and Upper Tunbridge Wells Sand. The alluvial formations are overlain by the second part of a cycle, the lake-lagoon clays, Allen‟s (1981) argillaceous formations. These are observed in the Wadhurst, Grinstead and basal Weald Clays (Allen 1981; Radley 2005). The upper part of the

64

Weald Clays represent the bulk of the Weald Clay Group, and are dominated by lacustrine-lagoonal with distal alluvial facies (Allen & Wimbledon 1991).

Jarzembowski‟s (1995) study on insect palaeoenvironment discusses a restricted freshwater fauna for insects dominated by Odonata adults with rare Trichoptera and . The restricted fauna ties in with sedimentological and palynological evidence, which suggests the water may have been brackish at times (Allen 1975; Batten 1982). This is further emphasized by the absence of larval Odonata and freshwater insects in the Wealden, suggesting that the aquatic fauna was not living in the main Wealden waterbody. (The majority of insects are not able to survive in water of over 0.2% salt, especially Odonata, Trichoptera and some Diptera.) Also the muddy water would not have favoured Ephemeroptera, which are absent from the Wealden; these favour clear, aerated water (Jarzembowski 1995).

The insect evidence suggests that the Wealden watercourses were bordered by woodland with occasional glades (Jarzembowski 1995). The evidence for this is a high number of wood associated insects, for example, Isoptera, Coleoptera and Hemiptera. The presence of Diptera suggests a damp woodland environment and the rarity of Orthoptera hints at a lack of open spaces (Jarzembowski 1995).

The insects are disarticulated with wings dominating the fossils and body parts being rare; the wings are also on the whole incomplete/damaged. The Wealden, as mentioned previously, has sediment derived mainly from Londinia (in the north) and it could be that the insects lived on the emergent ground of the south side of the Londinia massif, being washed into the area of deposition (the incompleteness of the wings is possibly due to damage during transport) (Jarzembowski 1995).

2.8.4. Palaeoflora

Much work has been done on the Wealden flora, most notably works by Watson and Alvin (1996 and references therein); their work demonstrated that the flora was dominated by gymnosperms and ferns (pteridophytes) with a possible eoangiosperm present (Hill 1996). Watson and Alvin (1996) give a floral list of all published flora up to 1996.

The ferns are dominated by Weichselia reticulata with fragmentary Flicopsida (Jarzembowski 1991; Ross & Cook 1995; Watson & Alvin 1996). The gymnosperms

65 from the Wealden are represented by horsetails, club mosses, twigs and cones (Ross & Cook 1995; Watson & Alvin 1996).

The potential early angiosperm (Bevhalstia pebja) has been interpreted as being from a marshy or aquatic environment. The plant, having delicate structures preserved, means it has not been transported far from place of life to the place of fossilization (Hill 1996). Bevhalstia pebja has been found in abundance in the Weald Sub-basin in both Upper and Lower Weald Clay (Hill 1996).

2.8.5. Palaeofauna

The entomological fauna of the Wealden is diverse with the majority of extant Orders present. The Orders that have been recorded to date are Coleoptera, Odonata, Blattodea, Orthoptera, Hemiptera, Mecoptera, Diptera, Trichoptera, Hymenoptera, Raphidioptera, Neuroptera, Isoptera and Plecoptera (Jarzembowski 1984, 1995; Sinitshenkova 1998).

The woodland environment would have been dominated by the plant-feeding insects, the xylophagous (wood-eating) insects; the Isoptera would live in damp, fallen or standing trees and cupedoid Coleoptera would feed on dead wood. The coleopteran Paleoscolytus sussexensis, the aphid Penaphis woolardi and the fulgoroid Hemiptera would have fed on living trees. The lower vegetation was likely food for Orthoptera, Hymenoptera, some Coleoptera, Hemiptera, Psocoptera and some Diptera (Jarzembowski 1995). The Auchenorryncha (Homoptera) may have been associated with the numerous ferns at that time.

The insect predators (entomophagous) are Odonata, lower aculeate Hymenoptera, notonectoid Hemiptera, some Diptera, Raphidioptera and Neuroptera. Predators (with chewing mouthparts) include Odonata, Raphidioptera, adult Neuroptera and Hymennoptera, and predators (with sucking mouthparts) include Neuroptera larvae and some dipterans and hemipterans (Jarzembowski 1995). Aerial predators would be Odonata; arboreal predators would be Raphidioptera and Neuroptera; leaf litter predators would be some larval Raphidioptera, empidid and rhagionid Diptera (Jarzembowski 1995).

66

Jarzembowski (1995) summarized the anti-predation strategies of the insects, such as cryptic patterns on the wings of kalligrammatids (Neuroptera), wing shedding in Isoptera and a „click‟ escape mechanism in elateriform Coleoptera.

Bivalves, gastropods, ostracods, conchostrachens, , foraminiferans, pterosaurs, isopods, echinoderms, fish, dinosaurs, sharks and reptiles are also present in the Wealden fauna (Gallois 1978; Jarzembowski 1991).

2.9. Climate and Evironmental Change in the Purbeck and Wealden

The Purbeck and Wealden combined lasted for approximately 21 Ma and throughout that time the environment and climate underwent many changes. The main trend of change is from a semi-arid condition in the Lower Purbeck, becoming increasingly wetter through the Middle Purbeck onwards into the Wealden. This section reviews the published evidence for the Purbeck/Wealden climate change, with the divisions of Sedimentology and Palaeontological evidence.

2.9.1. Sedimentology

On a large scale the sedimentological evidence points to a lagoonal environment in the Purbeck, displayed by the characteristic lagoonal . Evaporites are observed in the Lower Purbeck becoming less frequent and rare from the base of the Marly Freshwater Beds upwards (Coram 2005). The sedimentology of the Wealden indicates a fluvial/lacustrine dominated environment with cycles of sandstone and mudstone, with the mudstone becoming more dominant toward the end of Wealden times (the Weald Clays) (Allen 1998).

On a smaller scale clay minerals have highlighted a possible change in climate (tables 2.2 and 2.3).

67

Mineral Inferred Environment

Kaolinite Indicative of high temperature and high precipitation

Smectite Indicative of warm conditions with seasonal variation in humidity (also suggests a less pronounced wet and more pronounced dry season)

Palygorskite Indicative of arid/semi arid conditions

Illite No environment indication

Table 2.2. Clay minerals with their environment preference (Hallam et al. 1991; Coram 2005).

Stratigraphy Clay Minerals Environment Lulworth Throughout High smectite, Indicates an formation some palygorskite arid/semi arid and a lack of environment kaolinite

Durlston Cinder Beds Clay mineralogy changes mid- Marly Palygorskite is not Freshwater Beds seen any more

Intermarine Beds Kaolinite reappears A change to a more Rest of Purbeck into the Wealden Kaolinite is humid climate significant (wetter) is throughout (at the indicated. This expense of leads to an increase smectite) in weathering of the source massif.

Late Wealden (late ) Reappearance of Indicates an palygorskite increase in aridity

Table 2.3. Summarizing the clay mineralogy and the interpretation of climate (Sladen & Batten 1984; Ruffell & Batten 1990; Hallam et al. 1991; Coram 2005).

Other evidence suggesting a moistening of climate includes an increase of pyrite, sand, glauconite and reworked clastics in the higher Purbeck rocks (West 1979;

68

Coram 2005), the presence of siderite in the Wealden (Hallam 1984) and the character of palaeosols (Allen 1998).

2.9.2. Palaeontological Evidence

The types of aquatic biota are dependant on the water conditions; in the Purbeck the ostracods and molluscs have been used to demonstrate palaeosalinity (Horne 2002; Radley 2005). The palaeosalinity information obtained from these organisms reflects a change from a drier to a more humid climate.

Evidence from bioturbation, which appears at the base of the Marly Freshwater Beds, would suggest that the water was much fresher than previously in the Purbeck (Coram 2005).

Within the vertebrate fauna, the greatest diversity is seen in the Marly and Cherty Freshwater beds (Coram 2005). In the Middle and Upper Purbeck, dinosaur tracks, other vertebrate traces, crocodile and turtle remains are observed; these are rare or absent in the Lower Purbeck (see papers in Milner & Batten 2002). This may suggest a greater abundance of fresh water and food, reflecting a more tolerable „wetter‟ climate (Coram 2005).

The evidence from palaeobotany is very limited with regard to plant macrofossils (Batten 2002). However, the miospore (spore or pollen grain less than 200 μm) record is very useful. For much of the Lower and Middle Purbeck the pteridophytes, characteristic of a „wetland‟, were restricted and indeed confined to small damp habitats; later on they become more widespread and diverse up into lower Wealden times (change to a wetter climate) and then declined towards the end of Wealden times (return to a more arid environment) (Allen 1998; Batten 2002; Coram 2005).

It is difficult, however, to pinpoint the exact time of the change in climate in the Purbeck and Wealden, because the many studies on sedimentology and palaeontology give conflicting results (Norris 1969; Sladen & Batten 1984; Hunt 1985; Coram 2005).

69

CHAPTER 3. MATERIAL AND METHODS

3.1. Fossil acquisition

All the fossils were acquired from prior collected material such as museum collections and one private collection (where all described material has been, or will be, deposited in museum collections). The museums whose collections were used are the Booth Museum of Natural History, Brighton (BMB); Natural History Museum of London (BMNH Pal.), Maidstone Museum and Bentliff Art Gallery (MNEMG), Sedgwick Museum of Earth Sciences, Cambridge (CAMSM) and Devizes Museum, Wiltshire. The private collection used was that of Dr Robert Coram (Swanage), representing over 10 years of collecting.

3.2. Study of the fossils

The fossils were studied immersed under a 10% solution of ethanol to enhance fine detail. The microscope used was an Olympus SZH stereozoom microscope. An aeroneedle (Selden, 2003) was used when needed, to remove dust and small particles that obscured the wing. Line drawings were created using a camera lucida attachment. Where necessary the part and counterpart drawings were merged to form a composite image using a light-box, which gave a more complete venation drawing. The drawings were scanned into a computer to create a digital image, which was then labelled using Adobe Illustrator. Digital photographs were taken with a Nikon D1X SLR digital camera, and a Canon 450D SLR digital camera; the cameras were fitted with a macro lens and mounted on a rostrum stand.

3.3. Wing venation

Descriptions of the specimens were made using a modified version of the terminology of Comstock (1918) (Figs 3.1 and 3.2), with a few exceptions in accordance with current usage in neuropterology (e.g. Oswald 1993a; Archibald & Makarkin 2006).

70

Figure 3.1. The schematic wing venation diagram (Comstock 1918), which is used to name veins on the wings of all insects.

Figure 3.2. Modified venation diagram (modified from Comstock, 1918). The brackets show an alternative venation pattern sometimes used.

Wing vein abbreviations are as follows: C, costa; Sc, subcosta; R, radius; Rs, radial sector; Rs1, first branch of Rs; Rs2, second branch of Rs, M, media; MA, media anterior; MP, media posterior; MP1, first branch of MP; MP2, second branch of MP; Cu, cubitus; CuA, cubitus anterior; CuP, cubitus posterior; A, analis; r, radial cell; dc, discal cell; mc, medial cell; ac, anal cell; p, pterostigma.

71

3.3. Problems Associated with Palaeoentomology

3.3.1. Fragmentary specimens

The majority of the specimens used in this study are of a fragmentary nature. The insects are represented by disarticulated wings with no body parts preserved. On rare occasions both fore and hind wings have been preserved, but, in the vast majority of cases only one wing is present. The wings are on the whole incomplete, for example, a wing may have a portion missing, or the specimen may be represented by a small fragment of the wing. The problems in identification that result from only having a wing of an insect will be discussed below in detail. The problem of having wing fragments are that one is unable to identify the specimen to lower taxonomic ranks such as family, genus or species and if the specimen is in a particularly bad state it may be impossible to say in which Order the specimen may belong. Therefore, many specimens were discarded and only the important/better-preserved specimens have been illustrated in this thesis.

3.3.2. The wing problem

The majority of extant Neuropterida species are described and diagnosed on the basis of morphology, in particular genital morphology (Aspöck & Aspöck 2008). As mentioned previously (Chapter 2.1.) the relationships of Neuropterida families/Orders have been studied using morphological and molecular data. The wings of Neuropterida are useful; they can be used to distinguish between Orders/families/genera and in some cases species (Plant 1997). Venation, however, is not just different between species, it can be different within species; this is called intraspecific variation, and examples of this and what can cause it are given below.

Sexual dimorphism (differences between the male and female of a species) can cause differences in the wing of the same species; the major difference observed in most species of Neuropterida is size, the female being larger than the male (Aspöck & Aspöck 1972).

Species of Raphidioptera can be identified on the basis of venation, for example, in Plant‟s (1997) key for the differentiation of species of British Raphidioptera, venational features are used to distinguish between the four species (Atlantoraphidia maculicollis, Phaeostigma notata, Xanthostigma xanthostigma and Subilla confinis).

72

The keys use specific features to identify the species. A warning, however, has to be made; while some features are fixed, others are variable. Through my own observations of specimens in The Manchester Museum and through published literature that figures numerous specimens of the same species, I have seen that variations do occur that are intraspecific such as:

 Small variation is observed in the terminal forking of veins. Differences are seen in the male and female and even in the right and left wing in the same specimen (Aspöck et al. 1983)  Angles of some veins also change between specimens from 1˚ to ~25˚  Extra crossveins may be observed between specimens e.g. in the pterostigmal area

These differences can be ignored when one has the whole ; they can be disregarded when, for example, genitalia between the specimens is the same. However, when one has only a fossil wing, this is more difficult.

Extant species of Megaloptera are not identified on the basis of venation, but are usually separated by the examination of genitalia. For example, the British species Sialis lutaria, S. fulginosa and S. nigripes are separated on genital morphology or in females on the anal plates. The venation of these three species of Sialis is very similar and unreliable as a method of identification (Plant, 1997).

A non-biological factor that can alter venation is post mortem deformation, for example tectonic (heat and pressure) distortion can occur. This is observed in Carboniferous insects (Jarzembowski, pers. com. 2008) and possibly in the Purbeck (Coram, pers. com. 2008). Softening of the cuticle causing wing distortion and wings settling at an angle onto soft sediment and being vertically compressed on the bedding planes may also have an effect on venation.

However, if one is cautious and takes these factors into account, and tries to be as conservative (without being over conservative) as possible when naming taxa, then no problems should arise. The fact is that true biological species cannot be described (as there is no way of knowing if fossil insects could have successfully mated, producing fertile offspring, especially when looking at just wings). Therefore one is

73 dealing with morphospecies, species that are described solely on morphological characters.

74

PART ONE A:

NEUROPTERIDA OF THE LOWER

CRETACEOUS OF SOUTHERN ENGLAND

PURBECK LIMESTONE GROUP

And dead wings carried like a paper kite

Robert Frost, from Design (1936)

75

CHAPTER 4. LACEWINGS (INSECTA: NEUROPTERA) FROM THE LOWER

CRETACEOUS PURBECK LIMESTONE GROUP OF SOUTHERN ENGLAND

(Jepson, J.E., Makarkin, V.N. and Jarzembowski, E.A. Lacewings (Insecta: Neuroptera) from the Lower Cretaceous Purbeck Limestone Group of southern

England. Submitted to Palaeontology)

4.1. Abstract

Several new lacewing taxa are described from the Purbeck Limestone Group of southern England. Sophogramma wimbledoni sp. nov. (Kalligrammatidae), Pterinoblattina purbeckensis sp. nov., P. fasciata sp. nov., Purbepsychops parrallela gen. et sp. nov. (Psychopsidae), Ovalorobius purbeckensis gen. et sp. nov. (Prohemerobiidae), Mesosmylidus vulgaris gen. et sp. nov., Osmylochrysa minima gen. et sp. nov., O. anomala gen. et sp. nov., Stenochrysa gradata gen. et sp. nov. (Osmylidae), Mesypochrysa minuta sp. nov. (Chrysopidae), Purbemerobius medialis gen. et sp. nov. (?Hemerobiidae), Epimesoberothus osmylodes gen. et sp. nov. (Berothidae) and Paraksneura alleni gen. et sp. nov. (Incertae sedis) are described and figured. Pterinoblattina pluma, P. penna (Psychopsidae), Osmylopsis duplicata (Osmylidae) and Sialium sipylus (Nymphidae) have been redescribed and refigured with their family affinity being ascertained. The genus Valdipsychops has been synonymized with Pterinoblattina after the redescription of P. pluma.

Key words: Psychopsidae, Chrysopidae, Osmylidae, ?Hemerobiidae, Nymphidiae, Kalligrammatidae, Dorset

NEUROPTERA along with the Orders Raphidioptera and Megaloptera make up the superorder Neuropterida. Neuropterida are holometabolous insects, undergoing complete metamorphosis (Grimaldi & Engel 2005). Neuropteran (lacewing) adults are mostly predators with biting mouthparts; some also consume pollen or honeydew. The generally terrestrial larvae are sucking predators in foliage, leaf litter and soil or, rarely, parasitic.

There are approximately 6000 described species of extant Neuroptera in 17 families (Aspöck 2002b). In Britain today there are six families: Chrysopidae, Osmylidae, Coniopterygidae, Hemerobiidae, Myrmeleontidae and Sisyridae, and sixty-six species (Plant 1997). British Mesozoic Neuroptera are known from the Upper

76

Triassic (Tillyard 1932), from the Lower Jurassic, with genera attributed the families Prohemerobiidae, Mesopolystoechotidae [=Polystoechotidae], Brongniartiellidae, and Kalligrammatidae (Whalley 1988) and the Lower Cretaceous Wealden Supergroup, represented by the families Kalligrammatidae, Psychopsidae, Osmylidae, Ithonidae and two incertae sedis (Jarzembowski 2001; Jepson et al. 2009a (chapter 7)). The older Purbeck Limestone Group underlies the Wealden Supergroup; previously only four species of Neuroptera had been formally described: Sialium sipylus Westwood, Pterinoblattina pluma Giebel, P. penna Scudder and Osmylopsis duplicata Giebel (Westwood 1854; Giebel 1856; Scudder 1885; Whalley 1988). Herein we revise these species and describe 8 new genera and 12 new species.

4.2. Geological Setting

The Purbeck Limestone Group of southern England is a predominantly lagoonal succession currently considered to be basal Cretaceous (Berriasian) in age (c. 146 – 140 ma: Ogg et al. 1994). The Group is divided into a lower Lulworth Formation and upper Durlston Formation, attributed here to the Lower and Upper Berriasian respectively, although the precise position of the Jurassic/Cretaceous boundary is still in dispute, and may lie within the Lulworth Formation (Allen & Wimbledon 1991; Ensom 2002b; W. A. Wimbledon, pers. comm. 2009). Insect fossils are common and diverse in fine-grained fissile micrites in both the Lulworth and Durlston formations, comprising mainly disarticulated remains of terrestrial taxa.

With the exception of one specimen collected from a quarry near Teffont, in the Vale of Wardour, Wiltshire (National Grid Reference ST 991 310), all material described herein was collected from the coastal outcrop at Durlston Bay, near Swanage, Dorset (National Grid Reference SZ 035 780), the type section of the Purbeck Limestone Group and most prolific source of Purbeck fossil insect material. Fossil insects from this site were first studied and named in the mid 19th Century, principally by Prof. J. O. Westwood (1854). It remains highly productive and has provided the holotypes of approximately 200 insect species to date.

Fossils of Neuropterida are relatively uncommon in the Purbeck Limestone Group, comprising 1.4 per cent of identifiable insect wings from Durlston Bay (Coram

77

2005). Approximately three-quarters of these are Neuroptera, described below; the remainder are Raphidioptera, recently described by Jepson et al. (2009b (chapter 5)).

4.3. Material and Methods

The neuropteran material is deposited in the following collections: Devizes Museum, Wiltshire; MNEMG, Maidstone Museum and Bentlif Art Gallery, Kent; BMNHM, the Natural History Museum, London; and CAMSM, the Sedgwick Museum of Earth Sciences, Cambridge.

The material was studied using an Olympus SZH stereomicroscope with a camera lucida attachment for producing drawings. Some drawings are composite, merging the part and counterpart to establish a more complete venation; this was achieved using a light box. 10% ethanol was used to enhance vein detail. Photographs were taken using a Canon EOS 450D digital SLR camera mounted on a rostrum stand, except the Natural History Museum, London specimens, which were photographed by Phil Crabb (NHM).

We follow the traditional (sensu Wootton 2002) venational terminology of Comstock (1918) with the recent interpretation of Oswald (1993b) and Archibald and Makarkin (2006). Wing vein abbreviations are as follows: C, costa; Sc, subcosta; R, radius; R1, first branch of R; Rs, radial sector; Rs1, most proximal branch of Rs; M, media; MA, media anterior; MP, media posterior; Cu, cubitus; CuA, cubitus anterior; CuP, cubitus posterior; 1A–3A, first to third anal veins.

4.4. Systematic Palaeontology

Order NEUROPTERA Linnaeus, 1758

Family KALLIGRAMMATIDAE Handlirsch, 1906

Genus Sophogramma Ren et Guo, 1996

1996 Sophogramma Ren et Guo, 461.

Type species. Sophogramma papilionaceum Ren et Guo, 1996, by original designation.

78

Diagnosis. Eye-spots absent in both wings; in forewing, Rs1 profusely branched distally; both MA and MP arched, not forked until terminal branching; in hind wing, costal space very broad; CuA dividing near wing base into 2 long branches; CuP appears to originating proximal to fork of CuA (Ren and Guo, 1996).

Included species. Six species from Lower Cretaceous of Eurasia: Lower Berriasian of England (new record), / Barremian of Transbaikalia, Russia, and Barremian of the Yixian Formation, NE China (Ren & Guo 1996; Yang et al. 2009; Makarkin 2010; herein).

Sophogramma wimbledoni sp. nov. Figure 4.1

1996 Kalligrammatidae; Ross and Jarzembowski, p. 113, text-fig. 5.

Figure 4.1. Sophogramma wimbledoni sp. nov. holotype No. 15. A. Forewing venation drawing. B. photograph. Scale bar = 1 mm.

Derivation of name. After William Wimbledon

Holotype. Devizes Museum (No.15 of Revd. W.R. Andrews‟ colln). A fragmentary hind wing. Type locality and horizon. Near Teffont, Wiltshire; Lulworth Formation; Lower Berriasian.

Diagnosis. Proximal branching of anterior branch of CuA in hind wing absent.

Description. Hind wing, 44 mm preserved length (estimated complete length about 60 mm), 36.8 mm preserved width. C not preserved, costal space incomplete, 11-12 subcostal veinlets (including perhaps humeral veinlet) and several crossveins

79 between them preserved. Sc, R1 incomplete with many, rather regularly spaced crossveins preserved between them. Space between R1 and Rs nearly equal in width, with 16 r1-rs crossveins preserved. Rs origins very near to wing base, with 10 branches preserved, some forked distally. MA and MP appear originated separately from Rs (or M dividing into MA and MP very close to wing base). MA concave, simple for most length. MP with 3 anteriorly directed branches. Cu origin not preserved. CuA forked near to wing base, its anterior branch slightly concave, posterior branch strongly convex; both running parallel to each other, not branched before terminal branching. CuP strongly convex, pectinately branched distally. 1A very long, few branched distally. 2A long, running very close to 1A, pectinately branched. 3A poorly determinable, perhaps pectinate. Rather closely spaced numerous crossveins throughout wing. Wing membrane appears brown; eye-spot not detected.

Remarks. Four species of the genus Sophogramma are known only from forewings (Ren & Guo 1996; Makarkin 2010), but the fifth (Sophogramma lii Yang et al.) is a beautifully preserved complete specimen with open wings (Yang et al. 2009). The preserved hind wing venation of the Purbeck species is very similar to that of S. lii, particularly in the structure of the M and Cu systems. Cu of the new species has 3 long branches beginning near the wing base, and CuA is deeply forked and CuP not. Previously, only two genera were known whose Cu has 3 long branches, these were, Limnogramma Ren, 2003 (Makarkin et al. 2009) and Kalligrammula Handlirsch, 1919 (i.e., Kalligrammula senckenbergiana Handlirsch, 1919). The latter requires restudy, and the venation as figured in Handlirsch (1919, Fig. 1) does not allow the basal forking of Cu to be determined. In Limnogramma, CuP is deeply forked, and CuA is not (Makarkin et al. 2009). However, the assignment of this species to Sophogramma is somewhat tentative because of incompleteness and poor preservation of the holotype of S. wimbledoni sp. nov., but if our interpretation of the venation is correct, we can say with some confidence that it belongs to this genus.

80

Kalligrammatidae gen et sp. indet Figure 4.2

Figure 4.2. Kalligrammatidae gen et sp. indet BMNH Pal. I. 3958. A. venation drawing. B. photograph. Scale bar = 1 mm.

Material. BMNH Pal. I. 3958. An incomplete wing.

Locality and horizon. Purbeck Limestone Group, Durlston Bay, Swanage.

Description. Wing 18.7 mm preserved length, 13.8 mm preserved width. Apparently only 6 branches of Rs partially preserved, two forked; numerous crossveins present throughout.

Remarks. A small fragment of a kalligrammatid wing, which unfortunately has no features preserved that allow identification to a lower taxonomic rank. It differs, however, from the other Purbeck kalligrammatid, particularly in its larger size.

Family PSYCHOPSIDAE Handlirsch, 1906

Genus PTERINOBLATTINA Scudder, 1885, sit. nov.

1885b Pterinoblattina; Scudder, p. 105. 2009b (chapter 7) Valdipsychops Jepson et al., p. 1330, syn. nov.

Type species. Blatta pluma Giebel, 1856, by subsequent designation (Handlirsch 1906–08: 607).

Diagnosis. In forewing, Rs dichotomously branched; several short gradate series of crossveins in radial space; CuA concave with relation to MP, pectinately branched; CuP non-pectinate.

81

Included species. Between 9 and 11 species from the Lower Cretaceous of the Purbeck Limestone Group [Lower Berriasian] to the Wealden Supergroup [Hauterivian and Barremian] of England: P. pluma (Giebel), P. penna Scudder, P. purbeckensis sp. nov., P. fasciata sp. nov., Pterinoblattina sp. B (all from the Purbeck), P. logunovi Jepson et al., comb. nov., P. brigidae Jepson et al., comb. nov., P. maculosus Jepson et al., comb. nov., P. proudlovei Jepson et al., comb. nov., P. minimus Jepson et al., comb. nov., Pterinoblattina sp. A [=Valdipsychops sp. of Jepson et al., 2009b] (all from the Wealden) (Jepson et al. 2009a (chapter 7); herein).

Remarks. Examination of the type specimens of the Purbeck species of this genus (P. pluma and P. penna) shows with confidence that the genera Pterinoblattina and Valdipsychops are synonyms, and therefore belong to the family Psychopsidae. Unfortunately, the features of the type species are too vague to determine the family assignment with certainty. But other species of the genus having similar size, colour pattern and venation allow a confident placement within Psychopsidae. Previously, this genus had been placed within the family Brongniartiellidae. However, examination of photographs of the type species of the type genus of the family (i.e., Brongniartiella gigas (Weyenbergh)) shows that the latter has quite different venation from that of Pterinoblattina (Makarkin & Archibald 2005; Makarkin 2010). The assignment of Pterinoblattina kobdoensis Ponomarenko, 1986 (Lower Cretaceous of Mongolia) to this genus is doubtful.

Pterinoblattina pluma (Giebel, 1856) Figure 4.3

1854 Unnamed neuropteran; Westwood, p. 394, pl. 15, fig. 14. 1856 Blatta pluma Giebel, p. 322 (). 1864 Blattidium pluma; Heer, p. 289 (Blattoptera). 1880 Blatta pluma; Geinitz, p. 520 (Blattoptera). 1885a Pterinoblattina pluma; Scudder, p. 755, fig. 935 (Blattoptera). 1885b Pterinoblattina pluma; Scudder, p. 105, 106 (Blattoptera). 1886 Pterinoblattina pluma; Scudder, p. 469, 470, pl. 48, figs 7, 8 (Blattoptera). 1906–08 Pterinoblattina pluma; Handlirsch, p. 608, pl. 48, fig. 7 (Prohemerobiidae).

82

1988 Pterinoblattina pluma; Whalley, p. 48, 56, 58, fig. 9 (Brongniartiellidae). 1993 Pterinoblattina pluma; Jarzembowski, p. 178 (Neuroptera). 1993 Pterinoblattina pluma; Ross and Jarzembowski, p. 418 (Brongniartiellidae).

Figure 4.3. Pterinoblattina pluma holotype, BMNH Pal. I. 3968. A. ?Forewing venation drawing. B. photograph. Scale bar = 1 mm.

Holotype. BMNH Pal. I.3968 (Brodie colln, 1898). An incomplete ?forewing.

Type locality and horizon. Durlston [= Durdlestone] Bay, Swanage, Dorset; Durlston Formation; Upper Berriasian.

Diagnosis. Several transverse costal gradate series.

Description. ?Forewing 8.8 mm preserved length (estimated length 20 mm), 5.2 mm preserved width. Costal space incomplete; 26 subcostal veinlets preserved, some forked near margin; costal gradate series present. Sc, R1, Rs incomplete. Subcostal and R1 spaces equal in width. Nine sc-r1 crossveins preserved. Rs with ten branches preserved, majority dichotomously forked, forming 30 branches distally. Few crossveins preserved in radial space. M, Cu, A not preserved.

Remarks. Pterinoblattina pluma is represented by a very fragmentary specimen and as such there are very few diagnostic characters preserved to allow a confident separation from the other species of Pterinoblattina. The main diagnostic character of P. pluma is the presence of several gradate series of costal crossveins.

83

Pterinoblattina penna Scudder, 1885b Figure 4.4

1885b Pterinoblattina penna Scudder, p. 106, 108 (Blattoptera). 1886 Pterinoblattina penna; Scudder, p. 470, pl. 48, fig. 14 (Blattoptera). 1906–08 Pterinoblattina penna; Handlirsch, p. 608 (Prohemerobiidae). 1949 Pterinoblattina penna; Martynova, p. 168 (Brongniartiellidae). 1988 Pterinoblattina penna; Whalley, p. 48, 58, fig. 10 (Brongniartiellidae). 1993 Pterinoblattina penna; Jarzembowski, fig. 4 (Brongniartiellidae).

Figure 4.4. Pterinoblattina penna holotype, BMNH Pal. I. 12324. A. Forewing venation drawing. B. photograph. Scale bar = 1 mm.

Holotype. BMNH Pal. I.12324 (Brodie colln.). An incomplete forewing.

Type locality and horizon. Durlston [=Durdlestone] Bay, Swanage, Dorset; Lulworth Formation; Lower Berriasian.

Diagnosis. Absence of costal gradate series; reduced colour pattern.

Description. Forewing, preserved length 10.4 mm (estimated length 19 mm), preserved width 7.7 mm. C not preserved. Costal space wide, about 3 times wider than „vena triplica’, with 38 subcostal veinlets preserved, many forking towards anterior wing margin. ‘Vena triplica’, formed by Sc, R1 and Rs, runs parallel and straight, with 12 sc-r1 crossveins and three r1-rs crossveins preserved. Rs with 18 branches, many forked, some dichotomously, forking to 31 distally. Several crossveins preserved in radial space, forming two small gradate series. Fork of M not preserved; MA and MP (convex) appear simple before terminal branching; Cu incomplete, CuA pectinately branched, CuP and Anal veins not preserved.

84

Remarks. The wing of Pterinoblattina penna is interpreted as a forewing because of the extremely broad costal space, a posterior pectinate vein (CuA) and a convex vein anterior to it (MP). However, in this interpretation CuA is concave in relation to MP which is more characteristic of hind wings. But, in P. purbeckensis, sp. nov. the concavity/convexity relationships are the same (see below). Therefore, it is possible that these relationships are diagnostic of the genus, or these wings maybe hind despite their broad costal space. In P. logunovi (Jepson et al. 2009a (chapter 7)), the type species of Valdipsychops from the Wealden, these relationships appear the same: MP is clearly convex and CuA is slightly concave (Jepson et al. 2009b (chapter 7): fig. 7; anterior vein designated as „MA‟ is MA, posterior is MP). Unfortunately, all psychopsid wings, from both the Purbeck and Wealden, are very incomplete with the basal forking of M and Cu not being preserved, which does not allow confident identification of the veins.

Pterinoblattina purbeckensis sp. nov.

Figure 4.5

Figure 4.5. Pterinoblattina purbeckensis sp. nov. holotype, DB175/NEUR 11. A. Forewing venation drawing. B. photograph. Scale bar = 1 mm.

Derivation of name. After the Isle of Purbeck, Dorset.

Holotype. MNEMG [DB175/NEUR 11]. An incomplete forewing, collected by R.A. Coram.

Type locality and horizon. Durlston Bay, Swanage, Dorset; Clements' (1993) Bed DB175, Corbula beds, Durlston Formation; Upper Berriasian.

Diagnosis. Large size; spotted colour pattern.

85

Description. Forewing, 15 mm preserved length (estimated length 19 mm), 10 mm preserved width. C not preserved, costal area with 35 subcostal branches preserved and one gradate series of crossveins preserved. Sc, R1 and Rs forming „vena triplica‟. Sc long curving, terminating on R1 distally. R1 long with eight preserved branches distally. Eight sc-r1 branches. Rs pectinately branched with 14 branches, many branches heavily forked. Crossveins preserved in radial space forming a few gradate series. M deeply forked (one-third of vein length), both branches appear simple. CuA pectinately branched, with seven branches. CuP non-pectinately branched with two primary branches, forking to four at posterior wing margin. 1A, forked with two primary branches, forking to four at posterior wing margin. 2A incomplete, with two branches. Six gradate series of crossveins preserved throughout radial, medial, cubital sector. Colour pattern preserved on wing.

Pterinoblattina fasciata sp. nov.

Figure 4.6

Figure 4.6. Pterinoblattina fasciata sp. nov. holotype, DB175/NEUR 1. A. Forewing venation drawing. B. photograph. Scale bar = 1 mm.

Derivation of name. From Latin fascia, band, in reference to the presence of pale band over fuscous forewing.

Holotype. MNEMG [DB59/NEUR 1]. An incomplete wing, collected by R.A. Coram.

Type locality and horizon. Durlston Bay, Swanage, Dorset; Clements' (1993) Bed DB59, Soft Cockle beds, Lulworth Formation; Lower Berriasian.

86

Diagnosis. One costal gradate series; colour pattern: thick stripe basally and near apex.

Description. Forewing, preserved length 12.5 mm (estimated length about 18 mm), preserved width 11 mm. C not preserved, costal space with 40 subcostal veinlets preserved, majority forked, some dichotomously, closely spaced. One longitudinal costal gradate series of crossveins. Sc, R1 and Rs partially preserved; origin and termination not seen. Seven sc-r1 crossveins and three r1-rs crossveins preserved. Rs pectinately branched, many-forked dichotomously. Fork of M not preserved; configuration of MA unknown; MP convex, with terminal forking. CuA incomplete, pectinately branched. CuP and A not preserved. Colour pattern: thick stripes near base and apex.

Remarks. This species is placed in Pterinoblattina on the basis of the dichotomously branched Rs, pectinately branched CuA, and colour pattern. The colour pattern is similar to that of the Wealden species P. logunovi and P. proudlovei (Jepson et al. 2009b (chapter 7)).

Pterinoblattina sp. B Figure 4.7

Figure 4.7. Pterinoblattina sp. B, BMNH Pal. I. 12311. A. forewing venation drawing. B. photograph. Scale bar = 1 mm.

Material. NHM I. 12311. A fragmentary forewing.

Locality and horizon. Purbeck Limestone Group, Durlston Bay, Swanage, Berriasian.

87

Description. Forewing preserved length 10.5 mm (estimated length about 20 mm), preserved width 7.1 mm. C, Sc, R1 not preserved. Rs partially preserved, branches, some forked. Crossveins in radial space randomly arranged. M apparently two branched, origin not preserved; MA, MP fork not preserved. CuA convex, pectinately branched with six? branches. CuP partially preserved (two-four branches), apparently not pectinate. Anal veins not preserved.

Remarks. This species is assigned to Pterinoblattina because of the concave pectinately branched CuA, the Rs branching similar to that of some other species (e.g., P. penna, but not as multi-forked as in the majority of species placed in Pterinoblattina) and the radial crossveins are numerous and forming an irregular short series (with additional randomly arranged crossveins).

Genus PURBEPSYCHOPSIS gen. nov.

Type and only species. Purbepsychopsis parallela gen. et sp. nov.

Derivation of name. From Purbeck and Psychopsis, a genus-group name. Gender feminine.

Diagnosis. Elongate wing; parallel mostly simple branches of Rs; heavy distal forking of MA and MP; CuA dichotomous and CuP pectinate.

Remarks. The genus has been tentatively placed in Psychopsidae. The venation of this genus is mostly similar to that of Epactinophlebia (Upper Jurassic of Karatau, Kazakhstan) and several undescribed species from the Lower Cretaceous locality of Baissa (East Siberia) in general venational configuration and in the absence of crossveins, particularly in the radial space. The latter feature may be an artefact of preservation; but many Baissa specimens have no crossveins as well; therefore this is probably an autapomorphy of this genus (or genera group).

88

Purbepsychopsis parallela gen. et sp. nov.

Figure 4.8

Figure 4.8. Purbepsychopsis parallela gen. et sp. nov., CAMSM J58307. A. forewing venation drawing. B. photograph. Scale bar = 1 mm

Derivation of name. From Latin parallelus, parallel, in reference to the parallel branches of Rs.

Holotype. CAMSM [J58307]. A nearly complete forewing.

Type locality and horizon. Durlston Bay, Swanage, Dorset; Soft Cockle Beds, Lulworth Formation (interpretated from sediment by RAC), Lower Berriasian.

Diagnosis. As for genus.

Description. Forewing preserved length 15 mm (estimated length about 16-17 mm), preserved width 6.8 mm. C not preserved, costal space with 30 subcostal veinlets preserved. Subcostal and R1 spaces relatively narrow, equal in width. Rs with 25 parallel branches, vast majority not forked until end-twigging. M fork not preserved. MA concave, pectinately branched distally with five branches. MP convex, pectinately branched distally with four branches. Cu forked near wing base. CuA strongly convex proximal to distal dichotomous branching. CuP concave, pectinately branched with nine simple branches. 1A pectinately branched with four branches. 2A dicthotomously forked. 3A incomplete, pectinately branched, with three branches preserved.

89

Family PROHEMEROBIIDAE Handlirsch, 1906

Genus OVALOROBIUS gen. nov.

Type and only species. Ovalorobius purbeckensis gen. et sp. nov.

Derivation of name. From Latin ovalis, oval, and –robius, a traditional ending of hemerobiid-like genera (from Hemerobius, a genus-group name), in reference to the oval shape of the type species forewing. Gender masculine.

Diagnosis. Large size; CuA dichotomously branched.

Remarks. The assignment of the genera Liassopsychops Bode and Stenoleuta Bode to Prohemerobiidae (Ponomarenko 1996) is very doubtful. O. purbeckensis gen. et sp. nov. has been placed within Prohemerobiidae on the basis of Sc not being fused distally.

Ovalorobius purbeckensis gen. et sp. nov.

Figure 4.9

Figure 4.9. Ovalorobius purbeckensis gen. et sp. nov., DB175/ NEUR 4. A. forewing venation drawing. B. photograph. Scale bar = 1 mm

Derivation of name. After the Isle of Purbeck, Dorset.

Holotype. MNEMG [DB175/NEUR4]. An almost complete forewing, collected by R.A. Coram.

Type locality and horizon. Durlston Bay, Swanage, Dorset; Clements' (1993) Bed DB175, Corbula beds, Durlston Formation; Upper Berriasian.

90

Diagnosis. As for genus.

Description. Forewing preserved length 10.9 mm (estimated length 12.2 mm), width 7.3 mm. C preserved, costal area with over 26 subcostal crossveins preserved, many forked near anterior wing margin. Sc preserved, termination not preserved. R1 preserved, forking distally. Rs pectinately branched, with 12 primary branches, many of which are forked, producing over 20 branches. M deeply forked 0.75 mm from origin. MA and MP both deeply forked, forking again near posterior wing margin. Cu appears deeply forked, fork not preserved. CuA simple for most of length, non- pectinate, forking to give seven branches at posterior wing margin. CuP heavily branched, with three primary branches, forking to give 17 branches, each with end- twigging. 1A deeply forked, with both branches forking. 2A forked, with both branches forking again giving seven branches at posterior wing margin. 3A incomplete.

Family OSMYLIDAE Leach, 1815

Subfamily ?MESOSMYLININAE Bode, 1953

Genus MESOSMYLIDUS gen. nov.

Type and only species. Mesosmylidus vulgaris gen. et sp. nov

Derivation of name. From the Mesozoic and Osmylidae. Gender masculine.

Diagnosis. Long, pectinately branched CuP, with up to ten branches.

Remarks. The subfamily classification of the fossil Osmylidae has not been studied. Mesosmylininae was created to include Mesosmylina exornata Bode, 1953 from the Lower Jurassic of Germany. In later work, Ponomarenko (1996) adequately redescribed this species. Three other species were included in this genus: M. mongolica Ponomarenko, 1984 (Lower/ Middle Jurassic of Mongolia), M. sibirica Ponomarenko, 1985 (Middle Jurassic of Siberia) and M. falcifera Ansorge, 1996 (Lower Jurassic of Germany). M. sibirica is based on a poorly preserved specimen; its congeneric relationship with other species is impossible to confirm. The other three species may indeed belong to one genus. In particular, they share the few- pectinate CuP. The genus Jurosmylus [=Mesosmylus Panfilov, 1980] from the Upper Jurassic of Kazakhstan has similar venation, such as the structure of CuP (this genus

91 is probably not synonymous with Mesosmylina, being distinguished by other characters). These two genera may constitute the subfamily Mesosmylininae. It is unclear if Mesosmylina, as well as Stenosmylina Jepson et al., 2009b from the Wealden Supergroup, belong to it, because they both possess a very long CuP.

Mesosmylidus vulgaris gen. et sp. nov.

Figure 4.10

Figure 4.10. Mesosmylidus vulgaris gen. et sp. nov., DB175/NEUR9. A. forewing venation drawing. B. photograph. Scale bar = 1 mm

Derivation of name. From Latin vulgaris, usual, in reference to generalised venation.

Holotype. MNEMG [DB175/NEUR 9]. An incomplete forewing, collected by R.A. Coram.

Type locality and horizon. Durlston Bay, Swanage, Dorset; Clements' (1993) Bed DB175, Corbula beds, Durlston Formation; Upper Berriasian.

Description. Forewing preserved wing length 8.5 mm (estimated length 11-12 mm), preserved width 3.2 mm. Costal space rather broad, with 15 simple subcostal simple veinlets preserved, rather closely spaced. Subcostal space narrow, no sc-r1 crossveins detected. Rs originating from R near wing base; six r1-rs crossveins in proximal R1 space. Rs zigzagged, with two branches preserved; branches of Rs zigzagged. M deeply branched; MA simple for most of length, forked distally; MP forked many times distally, forming seven branches. Cu forked near wing base. CuA simple for most of its length, forking to four branches distally. CuP pectinately branched with 10 branches, some distal branches forked. Numerous crossveins

92 throughout radial to cubital spaces. One cup-1a crossvein preserved. 1A pectinately branched with six branches. Colour pattern consists of fuscous patches at ends of some crossveins proximally and fully dark-shaded crossveins distally.

Subfamily ?GUMILLINAE Navás, 1912

(=EPIOSMYLINAE Panfilov, 1980)

Genus OSMYLOCHRYSA gen. nov.

Type species. Osmylochrysa minima gen. et sp. nov.

Included species. Type species and Osmylochrysa anomala gen. et sp. nov.

Derivation of name. Osmylo- (from Osmylus, a genus group name), and –chrysa, a traditional ending of chrysopoid genera (from Chrysopa, a genus group name [from Greek chrysos, gold, and opa, eyes]), in reference to osmylid affinity and superficial resemblance to chrysopids. Gender feminine.

Diagnosis. Presence of only one crossvein between R1 and Rs before origin of Rs1 in forewing.

Included species. Two species described below.

Remarks. The venation and wing size of this genus are most similar to those of the genus Epiosmylus Panfilov, 1980 (the subfamily Gumillinae). Unfortunately, the diagnosis of Gumillinae includes such features as very long antennae and enlarged scapus, which are not preserved in the specimen. Therefore, the subfamily affinity of the genus is preliminary.

93

Osmylochrysa minima gen. et sp. nov.

Figure 4.11

Figure 4.11. Osmylochrysa minima gen. et sp. nov., DB175/TM1096. A. forewing venation drawing. B. photograph. Scale bar = 1 mm

Derivation of name. From Latin minimus, smallest, in reference to smallest size of the holotype among gumillines.

Holotype. MNEMG [DB175/TM1096]. An incomplete forewing, collected by Mr A. A. Mitchell.

Type locality and horizon. Durlston Bay, Swanage, Dorset; Clements' (1993) Bed DB175, Corbula beds, Durlston Formation; Upper Berriasian.

Diagnosis. More distal branching of 1A [distal to the origin of Rs].

Description. Forewing, 12.7 mm preserved length (estimated length 13 mm), 3.3 mm preserved width. Costal space typical for osmylids; 35 subcostal veinlets, majority simple, more closely spaced and forked towards the apex. Trichosors present on anterior wing margin, towards apex. Sc and R1 fused distally; veinlets of Sc+R1 forked. Seven r1-rs crossveins. Rs pectinately branched, partially preserved; Rs1 originated far distal to origin of Rs. M partially preserved; its fork not visible. Four m-cua crossveins preserved. Cu forked near wing base; CuA and CuP incomplete, forking not preserved. 1A pectinately branched, with simple branches. One 1a-2a crossvein. 2A incomplete, probably pectinately branched, with one branch preserved.

94

Osmylochrysa anomala gen. et sp. nov.

Figure 4.12

Figure 4.12. Osmylochrysa anomala gen. et sp. nov., DB175/NEUR18. A. forewing venation drawing. B. photograph. Scale bar = 1 mm

Derivation of name. From Latin anomalus, anomalous, in reference to the presence of subcostal crossveins in the forewing.

Holotype. MNEMG [DB175/NEUR18]; a proximal half of the forewing, collected by R. A. Coram.

Type locality and horizon. Durlston Bay, Swanage, Dorset; Clements' (1993) Bed DB175, Corbula beds, Durlston Formation; Upper Berriasian.

Diagnosis. Presence of subcostal crossveins and more proximal branching of 1A [proximal to the origin of Rs].

Description. Forewing, preserved length 5.4 mm (estimated length 11-12 mm) preserved width 3.1 mm. Costal space relative broad, dilated at basal c. one-fifth, narrowed basally, with 12 simple, widely spaced subcostal veinlets preserved. Subcostal space narrow, with four crossveins preserved. Three widely spaced rs-r1 crossveins preserved. Rs zigzagged, apparently pectinate with two branches partially preserved; one crossvein preserved between these. Origin of Rs1 distant from origin of Rs. One rs-m crossvein and two rs-ma crossveins preserved. M deeply forked; MA and MP simple for most of length, distal part of vein not preserved. Two ma-mp crossveins, four each m-cua and mp-cua crossveins. Cu dividing into CuA and CuP close to base. CuA pectinately branched distally with four simple branches. Four cua-cup crossveins. CuP pectinately branched with two branches. 1A pectinately branched with three branches.

95

Remarks. The venation of this species is somewhat „chrysopid-like”, especially in posterior portions of the wing (e.g., short CuA with simple, widely-spaced branches). In particular, it slightly resembles the chrysopoid family Mesochrysopidae, but the configuration of the venation is very different between them.

Subfamily UNCERTAIN

Genus STENOCHRYSA gen. nov.

Type and only species. Stenochrysa gradata gen. et sp. nov.

Derivation of name. From Greek stenos, narrow, and –chrysa, a traditional ending of chrysopoid genera (from Chrysopa, a genus group name), in reference to [presumable] narrow forewing of the type species. Gender feminine.

Diagnosis. Narrower R1 space [almost as wide as subcostal space] with widely- spaced crossveins.

Remarks. This genus may belong to Gumillinae, but the fragmentary nature of the wing does not allow a confident placement.

Stenochrysa gradata gen. et sp. nov.

Figure 4.13

Figure 4.13. Stenochrysa gradata gen. et sp. nov., DB175/NEUR12. A. forewing venation drawing. B. photograph. Scale bar = 1 mm

Derivation of name. From Latin gradatus, gradate, in reference to the presence of gradate series of crossveins in the radial space of the forewing.

Holotype. MNEMG [DB175/NEUR12], a fragmentary anterior portion of the forewing. Collected by R. A. Coram.

96

Type locality and horizon. Durlston Bay, Swanage, Dorset; Clements' (1993) Bed DB175, Corbula beds, Durlston Formation; Upper Berriasian.

Diagnosis. As for genus.

Description. Forewing, preserved length 27 mm (estimated length about 35 mm); preserved width 6.2 mm. Costal space relatively narrow; 25 simple subcostal veinlets preserved. Subcostal space comparatively broad, no subcostal crossveins detected. R1 space very narrow; six very widely spaced r1-rs crossveins preserved. Rs pectinately branched, with seven preserved branches. Twelve crossveins in radial space. Two rs-ma crossveins. M deeply forked, MA and MP incomplete, one ma-mp crossvein preserved. Two mp-cua crossveins preserved. Cu deeply forked, CuA and CuP incomplete; CuP pectinately branched, with 2 branches presaerved. Three cua- cup crossveins; one cup-a1 crossveins. Anal veins not preserved.

Genus OSMYLOPSIS Handlirsch, 1906, sit. nov.

1906–08 Osmylopsis Handlirsch, p. 614.

Type and only species. Abia duplicata Giebel, 1856, by monotypy.

Revised diagnosis. Falcate hind wing; distal-most veinlets of Sc+R1 forked.

Remarks. The family affinity of this genus was hitherto unclear because of an inadequate original description. It was usually considered as Neuroptera incertae sedis (Handlirsch 1906–1908; Martynova 1949, 1962; Carpenter 1992). Handlirsch (1939) thought that Osmylopsis might be assigned to Epigambriidae, a very poorly- defined family of Neuroptera. The hind wing of O. duplicata is poorly preserved but its examination shows with confidence that it belongs to the family Osmylidae (unknown subfamily).

97

Osmylopsis duplicata (Giebel, 1856)

Figure 4.14

1854 Unnamed orthopteran; Westwood, p. 390, 396; pl. 18, fig. 42. 1856 Abia duplicata Giebel, p. 264, 416 (Sialidae). 1886 Pterinoblattina? binneyi Scudder, p. 472, 473 (Blattoptera). 1887 Pterinoblattina? binneyi; Geinitz, p. 200 (Blattoptera; as a synonym of Abia buplicata). 1906–08 ?Osmylopsis duplicata; Handlirsch, p. 614; pl. 48, fig. 15 (Neuroptera incertae sedis). 1939 Osmylopsis duplicata; Handlirsch, p. 76 (?Epigambriidae). 1949 Osmylopsis duplicata; Martynova, p. 169 (Neuroptera incertae sedis). 1988 Osmylopsis duplicata; Lambkin, p. 450, 453 („osmylid-like‟ neuropteran). 1993 Osmylopsis duplicata; Jarzembowski, p. 178 (Neuroptera). 2003 Osmylopsis duplicata; Makarkin and Archibald, 176 (Neuroptera familia incertae sedis).

Figure 4.14. Osmylopsis duplicata, holotype, BMNH Pal. I. 3954. A. forewing venation drawing. B. photograph. Scale bar = 1 mm

Holotype. BMNH Pal. I. 3954. An incomplete hind wing.

Type locality and horizon. Durlston [=Durdlestone] Bay, Swanage, Dorset; Lulworth Formation; Upper Berriasian.

Diagnosis. As for genus.

Description. Hind wing 10.4 mm preserved length (estimated length about 14 mm), width about 4.0 mm. C preserved; trichosors not detected. Costal space narrow, slightly dilated apically. Subcostal veinlets widely spaced proximally, more closely spaced distally; Sc+R1 veinlets closely spaced, four distal-most forking. Sc fused

98 with R1 far distal to wing apex; Sc+R1 entering margin well before wing apex. Subcostal space rather broad. Origin of Rs not preserved. Rs strongly zigzagged medially, smoother distally, with 12 branches, majority with terminal forking. R1 space (between R1 and Rs) rather narrow medially, very narrow distally; 6-7 r1-rs crossveins preserved. Crossveins in radial space poorly preserved, apparently rare. M incomplete, origin and fork of MA and MP not preserved. MA strongly concave, pectinately branched, with 4-5 branches, most (or all) with terminal forking. MP slightly convex, pectinately branched, with 5 branches, some with terminal forking. Two ma-mp, 3 mp-cua crossveins preserved. CuA pectinately branched, with 5 simple branches preserved. CuP, Anals veins not preserved.

Remarks. Pterinoblattina? binneyi is based on the same specimen as Abia duplicata (see Scudder 1886: 473).

Family CHRYSOPIDAE Schneider, 1851

Subfamily LIMAIINAE Martins-Neto and Vulcano, 1989

Genus MESYPOCHRYSA Martynov, 1927

Type species. Mesypochrysa latipennis Martynov, 1927 ( of Karatau, Kazakhstan), by monotypy.

Diagnosis. Complete fusion of basal portion of MA and R, then stem of Rs; MA appears as proximal branch of Rs.

Included species. Sixteen species from Late Jurassic (/ ) to (Late ) of Europe (1 species, herein), Kazakhstan (5 species), Russia (Transbaikalia; 6 species), Mongolia (1 species), China (1 species) and Brazil (2 species) (Makarkin 1997a; Nel et al. 2005b).

99

Mesypochrysa minuta sp. nov.

Figure 4.15

Figure 4.15. Mesypochrysa minuta sp. nov.,holotype, DB175/NEUR 28. A. hind wing venation drawing. B. photograph. Scale bar = 1 mm

Derivation of name. From the Latin minutus, meaning minute, small.

Diagnosis. Broad mediocubital space in hind wing; small size.

Holotype. MNEMG [DB175/NEUR28]. A small poorly preserved hind wing, collected by R. A. Coram.

Type locality and horizon. Durlston Bay, Swanage, Dorset; Clements' (1993) Bed DB175, Corbula beds, Durlston Formation; Upper Berriasian.

Description. Hind wing preserved length 6.7 mm (estimated length about 7 mm), width 2.7 mm. Costal space narrow, with seven preserved subcostal veinlets. Sc termination not observed. R1 long, termination not preserved. Two distal r1-rs crossveins preserved. Rs with six branches, most with terminal forks. Two gradate series of crossveins in radial to mediocubital spaces: inner with 4 crossveins and outer with 8 crossveins, both nearly parallel to each other and hind margin. M fork absent; MA appears from Rs like its proximal branch; MP appears fused with R basally; both simple most of length, forking terminally. Cu origin and fork not preserved. CuA pectinately branched with three simple branches. Distal portion of CuP partially preserved with one pectinate branch preserved. Mediocubital space (between MP and CuA) very broad. Distal cua-cup crossvein short. Anal veins not preserved.

Remarks. The specimen is very poorly preserved, with veins just visible under ethanol. Some features, however, are not observable; for example, the dense cross-

100 venation near the wing apex. This however, may be due to the poor preservation of the specimen, because other crossveins are not preserved, and the ones that are preserved are very faint. It has been confidently placed within the genus Mesypochrysa on the basis of the basal fusion of MA with R and then Rs in the hind wing. This is the smallest known fossil species of Chrysopidae, and first record of the family from the Mesozoic of Europe.

Family ?HEMEROBIIDAE Leach, 1815

Genus PURBEMEROBIUS gen. nov.

Type and only species: Purbemerobius medialis gen. et sp. nov.

Derivation of name. From Purbeck, and Hemerobius, a genus-group name. Gender masculine.

Diagnosis. Larg size; structure of Rs; CuA pectinately forked relatively close to posterior wing margin; CuP simple for most of length.

Remarks. The placement of this genus in the family Hemerobiidae is tentative due to the poor preservation of the wing. The character that suggests placement in this family is the anterior radial trace bearing 3 “radial-sectors”. Fossil hemerobiids are rare in the Mesozoic with only two confirmed taxa: Promegalomus anomolous from the Jurassic of Karatau, Kazakhstan, and Cretomerobius disjunctus from the early Cretaceous of Mongolia (Panfilov 1980; Ponomarenko 1992b). A hemerobiid-like species, Protohemerobius perexiguus, that has been described from the Wealden most likely represents an undescribed family (Jepson et al. 2009a (chapter 7)). Mesohemerobius jeholensis, from the Lower Cretaceous of China, was previously placed in Hemerobiidae, but was removed from the family by Makarkin et al. (2003) and placed as Neuroptera incertae sedis. Two species of hemerobiid have been described from the Lower Cretaceous Crato Formation of Brazil; however these are in need of redescription to confirm their placement in the family (Makarkin et al. 2003).

101

Purbemerobius medialis gen. et sp. nov.

Figure 4.16

Figure 4.16. Purbemerobius medialis gen. et sp. nov., holotype, DB175/NEUR5. A. hind wing venation drawing. B. photograph. Scale bar = 1 mm

Derivation of name. From Media, a median vein, in reference to the structure of the Media of the hind wing.

Holotype. MNEMG [DB175/NEUR 5]. An incomplete hind wing, collected by R. A. Coram.

Diagnosis. As for genus.

Type locality and horizon. Durlston Bay, Swanage, Dorset; Clements' (1993) Bed DB175, Corbula beds, Durlston Formation; Upper Berriasian.

Description. Hind wing preserved length 8.8 mm (estimated length 9.5 mm), width 4.5 mm. C preserved. Costal space narrow, 44 costal veinlets, majority simple. Sc long, forking before termination on wing margin. R1 long dichotomously forked distally 1.3 mm before wing apex, one r1-rs crossvein. Rs with three branches; Rs1 deeply forked, one crossvein preserved between branches, both branches forking distally, splitting to 12 branches at termination on posterior wing margin. Three rs1- rs2 crossveins. Rs2 branch simple most of length, dichotomously forked distally. Two rs2-rs3 crossveins preserved. Rs3 pectinately branched with seven branches, heavily damaged, simple most of length, forking distally. One rs1-ma crossvein. M deeply forked, origin not preserved. MA forks just before mid-point of vein, dichotomously branched, heavily damaged, one crossvein preserved between the branches. MP deeply forked, first branch simple most of length, forking before

102 termination at posterior wing margin, one crossvein preserved between branches. Second branch deeply forked, both branches simple for most of length, forking near termination, one crossvein preserved between branches. CuA origin not preserved, pectinately branched, 11 branches, mostly simple, some forked. One crossvein preserved within CuA. CuP simple most of its length, forking near termination at posterior wing margin. 1A pectinately branched, six branches preserved, one forked. 2A partially preserved, appears pectinately branched, three branches partially preserved.

Remarks. The wing appears to have been damaged prior to fossilization, with many veins being broken. This is observed especially around the mid-point of the wing, affecting mainly the lower half, towards the posterior wing margin. Purbemerobius medialis gen. et sp. nov. possibly represents the first Mesozoic fossil hemerobiid in Britain.

Family BEROTHIDAE Handlirsch, 1906

Genus EPIMESOBEROTHUS gen. nov.

Type and only species. Epimesoberothus osmylodes gen. et sp. nov.

Derivation of name. From Greek epi, near, and Mesoberotha, a genus-group name, in reference to mesoberothid appearance. Gender masculine.

Diagnosis. Simple subcostal veinlets; lack of recurrent humeral vein; forewing size (<4.5 mm); Rs with 6-7 branches; pectinately branched CuA; wide costal area.

Remarks. The forewing venation and size of the new genus are rather similar to those of the Burmese amber berothids (Engel & Grimaldi 2008), except for the pectinately branched CuA. In the latter genera, CuA is rather few-branched. Epimesoberothus gen. nov. is similar by the pectinately branched CuA to Sinosmylites Hong, 1983 from the Middle Jurassic of China, which was preliminarily assigned to the family Prohemerobiidae by Makarkin and Archibald (2005).

103

Epimesoberothus osmylodes gen. et sp. nov.

Figure 4.17

Figure 4.17. Epimesoberothus osmylodes gen. et sp. nov., holotype, USCDB/TM432. A. forewing venation drawing. B. photograph. Paratype, CAMSM TN272. C. forewing venation. D. photograph. Scale bar = 1 mm

Derivation of name. From Osmylus, a genus-group name, and Greek -odes, like or resembling, in reference to the superficial resemblance of the type species to minute osmylids.

Holotype. MNEMG [USCDB/TM432]. A nearly complete forewing, collected by Mr A. A. Mitchell.

Paratype. CAMSM TN272 and CAMSM TN273 (part and counterpart). An incomplete forewing.

Type locality and horizon. Holotype and paratype: Durlston Bay, Swanage, Dorset; Upper Soft Cockle beds, Lulworth Formation; Lower Berriasian.

Diagnosis. As for genus.

104

Description of the holotype. Forewing, length 3.7 mm, width 1.7 mm. C complete. 23 subcostal veinlets. Trichosors present on distal part of anterior margin. Sc incomplete, terminating on R1. R1 long with six branches distally. Two r1-rs crossveins. Rs with seven branches, all forking distally past mid-point of vein, forming 22 branches going into posterior wing margin. Seven crossveins in radio- media area, six forming gradate series. M deeply forked, both MA and MP fork distally 0.4 mm from posterior wing margin; MA forking to three branches, MP dichotomously branched. Cu deeply forked. CuA pectinately branched, with six branches, majority simple. CuP deeply forked. Anal veins not preserved.

Description of the paratype. Forewing, preserved length 3.8 mm (estimated length 4.5 mm), preserved width 2.1 mm. C incomplete, 20 subcostal veinlets preserved. Trichosors present on anterior wing margin. Sc terminates on R1. R1 with seven branches distally. One rs-rs1 crossvein past mid-point of wing. Rs pectinately branched, with eight branches forking distally after mid-point. Two crossveins preserved inn radial sector. One rs-m crossvein. M deeply forked (fork not preserved). MA and MP branching distally to give five branches going into posterior wing margin. CuA pectinately branched with six branches, majority simple. CuP incomplete forked to two branches. Anal veins incomplete.

Remarks. The paratype‟s locality has been inferred by RAC, from the characteristic nature of the sediment (the light 'chalky' nature of the matrix, the abundant charcoal fragments and the small square halite (salt crystal) moulds), because no detailed locality information was present with the specimen.

Family NYMPHIDAE Rambur, 1842

Genus SIALIUM Westwood, 1854, sit. nov.

Type and only species. Sialium sipylus Westwood, 1854, by monotypy.

Diagnosis. Several subcostal veinlets forked proximal to pterostigmal region; 8 MP branches in the hind wing.

Remarks. Examination of the type species shows that the hind wing venation is typical for the family Nymphidae. In general, it is most similar to that of Mesonymphes, two species of which are known from the Late Jurassic of Germany

105 and Kazakhstan (Carpenter 1929; Panfilov 1980). The latter genus is distinguished from Sialium as indicated in the diagnosis.

Sialium sipylus Westwood, 1854 Figure 4.18

1854 Sialium Sipylus Westwood, p. 390, 396, pl. 18, fig. 24 (Neuroptera, allied to Sialis). 1856 Abia sipylus: Giebel, p. 263, 416 (Sialidae). 1862a Sialium Sipylus; Hagen, p. 104 (Neuroptera). 1862b Sialium sipylus; Hagen, p. 10 (Neuroptera). 1884 Syalium [sic] Sipylus; Geinitz, p. 571 (?Orthoptera). 1886 Pterinoblattina ? Sipylus; Scudder, p. 472 (Blattoptera). 1887 Pterinoblattina ? Sipylus; Geinitz, p. 200 (Blattoptera). 1891 Sialium sipylus; Scudder, p. 130 (Blattoptera). 1906–08 Sialium sipylus; Handlirsch, p. 609, pl. 48, fig. 10 (Nymphitidae). 1949 Sialium siphlus [sic]; Martynova, p. 167(Nymphitidae). 1988 Sialium sipylus; Lambkin, p. 451, 454 (Nymphitidae). 1988 Sialium sipylus; Whalley, p. 46 (Nymphitidae). 1993 Sialium sipylus; Jarzembowski, p. 178 (Neuroptera).

Figure 4.18. Sialium sipylus, holotype, BMNH Pal. I. 13951ab. A. hind wing venation drawing. B. photograph. Scale bar = 1 mm.

Holotype. BMNH Pal. I. 13951ab. An almost complete hind wing.

Type locality and horizon. Durlston [=Durdlestone] Bay, Swanage, Dorset; Lulworth Formation; Lower Berriasian.

Diagnosis. As for genus.

106

Description. Hind wing, preserved length 21 mm (estimated length about 25 mm), preserved width 6.8 mm. C partially preserved, trichosors preserved apically. Subcostal space narrow, 21 subcostal veinlets preserved, becoming densely spaced distally, some forked. Sc curves towards R1 distally, (probably fused). R1 preserved only in distal part. Sc+R1 heavily forked with five preserved branches, forking to 11. Five r1-rs crossveins preserved. Rs pectinately branched with 12 branches, forking close to posterior wing margin. Numerous crossveins preserved in proximal part of radial sector in a random pattern. MA origin not preserved, forking towards posterior wing margin to give ten branches. Eight ma-rs crossveins preserved. MP simple for most of its length, dichotomously forking very close to posterior wing margin to give four branches. Six mp-ma crossveins preserved. Five mp-cua crossveins preserved. Cu origin not preserved. CuA pectinately branched, with nine branches, majority forked distally to give 22 branches at posterior wing margin. CuA terminates past mid-way point of wing. Four crossveins preserved within. Three cua-cup crossveins preserved. CuP pectinately branched with seven branches forking to 11 distally, no crossveins preserved within. Anal veins not preserved.

Family NEUROPTERA familia incertae sedis

Genus PARAKSENEURA gen. nov.

Type and only species. Parakseneura alleni gen. et sp. nov.

Derivation of name. From Greek paraksenos, strange, and Neuroptera, in reference to uncertain family affinity of the genus.

Diagnosis. Large size; broad proximal costal area; costal crossveins; dense arrangement of subcostal veinlets in distal part of wing.

Remarks. The systematic position of Parakseneura alleni gen. et sp. nov. is uncertain. However, it has similar venation to the younger Cratochrysa (Neuroptera familial incertae sedis) from the Crato Formation, Brazil, which was placed as basal with respect to Chrysopoidea (Nel et al. 2005b). Cratochrysa and P. alleni gen et sp. nov. also have a superficial similarity to the megalopteran family Corydasialidae Wichard et al., 2005, with regards to Sc and R1 not being fused, many crossveins between R1 and Rs, M and Cu not being fused near wing base, and CuA (in C. martinsnetoi, not preserved in P. alleni gen. et sp. nov.) and CuP being forked.

107

Therefore, it is possible that Cratochrysa and Parakseneura gen. nov. could be allied to Megaloptera, however, without body parts it is impossible to ascertain the true affinity of the fossil.

Parakseneura alleni sp. nov. Figure 4.19

Figure 4.19. Parakseneura alleni gen. et sp. nov., holotype, HCSD NEUR 1. A. forewing venation drawing. B. photograph. Scale bar = 1 mm.

Derivation of name. After the late Professor Perce Allen.

Holotype. MNEMG [HCSD NEUR 1]. An almost complete forewing, collected by R. A. Coram.

Diagnosis. As for genus

Type locality and horizon. Durlston Bay, Swanage, Dorset; probably Hard Cockle Beds, Lulworth Formation; Lower Berriasian.

Description. Forewing, preserved length 23.3 mm (estimated total length 24.5 mm), width 8.3 mm. Costal area widened basally, with numerous crossveins. Gradate series of crossveins (eight preserved) in costal area near base of wing. Sc and R1 not fused apically. Three sc-r1 crossveins preserved; one at mid-point of wing and two in distal part of wing. Rs has five pectinate branches, splitting into seven. Rs branches from R1 5.0 mm from base of the wing. Eight crossveins present between R1 and Rs. M and CuA not fused basally. CuA incomplete, CuP long forking approximately 4 mm from posterior wing margin. Three cua-cup crossveins preserved. Three cup- 1a crossveins. Anal veins straight; 1A forked 4.0 mm from posterior wing margin; 2A with four branches preserved. Two 1a-2a crossveins preserved. A small portion of the underlying other forewing is exposed, but provides no further information.

108

4.5. Discussion

Prior to this study only three species of neuropterans were known from the Purbeck Limestone Group of southern England, all of which were described in the 19th century and poorly illustrated. These species Osmylopsis duplicata, Sialium sipylus Pterinoblattina pluma and P. penna have now been thouroughly described and figured. The family affinity of these species has also been confirmed, with Osmylopsis duplicata being placed in Osmylidae, Sialium sipylus in Nymphidae and Pterinoblattina pluma and P. penna in Psychopsidae. The redescription of Pterinoblattina has shown the psychopsid genera Valdipsychops from the Wealden to be a synonym. The new families recorded within the Purbeck include Osmylidae, Chrysopidae, ?Hemerobiidae, Berothidae, Kalligrammatidae and Prohemerobiidae. A familia incertae sedis is also present. The psychopsoid neuropterans (Psychopsidae and Kalligrammatidae) make up the majority of the neuropteran fauna and the vast amounts of undescribed fragmentary material are also of psyhopsoid affinity (JEJ pers. obs. 2008–2010). Osmylidae are the next dominant group after the psychopsoids. The chrysopids, nymphids, ?hemerobiids, berothids and prohemerobiids are represented by a single species each.

The dominance of the psychopsoids may be due to them being widespread and successful within the Cretaceous, being found in many other fossil deposits at that time. Alternatively it could be that their many veined wings make them more resilient to breaking down during transport. Human bias could also be a factor, psychopsoids are easily identified as neuropteran whereas other families may be overlooked or misidentified, therefore showing a bias. The high number of osmylids may be due to their lifestyle, it is possible that osmylid larva have a semi-aquatic lifestyle based on modern analogues, therefore if they are closer to water they would have a higher chance of being washed into the depositional area and a better chance of preservation, as opposed to the fully terrestrial fauna.

The first record of Chrysopidae, Berothidae and possibly Hemerobiidae in the Mesozoic of England has been recorded. The chrysopid is the smallest representative of the family recorded with an estimated wing length of 7 mm. The presence of Berothidae within England in the Lower Cretaceous extends the families

109 geographical range, with them known previously from Lebanon, China, Brazil, Burma, Russia, USA, Canada and Baltic amber (Engel & Grimaldi 2008).

The geographical ranges of the genera Sophogramma and Mesypochrysa have been extended with the discovery of them in the Purbeck. Sophogramma was previously only known from Russia and China (Ren & Guo 1996; Makarkin 2010). Mesypochrysa is more globally widespread with previous occurrences documented in Kazakhstan, Russia, Mongolia, China and Brazil (Makarkin 1997a; Nel et al. 2005b).

The neuropteran fauna of the Purbeck is more diverse than that of the younger Wealden having a higher number of families present [Wealden 4 (Jepson et al. 2009a (chapter 7)) and Purbeck 7 (herein)]. There are however similarities in each fauna, for example, the psychopsoids dominate in each, with Osmylidae being the next dominant. The family Brongniartiellidae has not been found in both the Purbeck and Wealden, with previously thought brongniartiellids being shown to be of the family Psychopsidae. Both the Purbeck and Wealden have genera which are present in the younger Crato Formation, Brazil; Principiala (Ithonidae) in the Wealden (Jepson et al. 2009a (chapter 7)) and Mesypochrysa (Chrysopidae) in the Purbeck (herein).

The neuropteran fauna of the Purbeck would have lived in the distant woodland area proposed by Coram (2003). On death they would have undergone disarticulation and transport before deposition, leaving only their wings. The proposed climate change within the Purbeck to the Wealden which affected the raphidiopterans (Jepson et al. 2009b (chapter 5)), seems to have had little affect on the neuropteran fauna, with the exception of Nymphidae being present in the Lulworth Formation. Nymphids are usually associated with arid/semi-arid environments and their presence in the hotter and drier Lower Purbeck, Lulworth Formation and not in the younger Durlston Formation and Wealden could be related to the change in climate.

110

CHAPTER 5

1. RAPHIDIOPTERA (INSECTA: NEUROPTERIDA) FROM THE LOWER

CRETACEOUS PURBECK LIMESTONE GROUP, DORSET, UK.

111

Cretaceous Research 30 (2009) 527–532

Contents lists available at ScienceDirect

Cretaceous Research

journal homepage: www.elsevier.com/locate/CretRes

Raphidioptera (Insecta: Neuropterida) from the Lower Cretaceous Purbeck Limestone Group, Dorset, UK

James E. Jepson a,*, Robert A. Coram b, Edmund A. Jarzembowski c a School of Earth, Atmospheric and Environmental Sciences, University of Manchester, Williamson Building, Oxford Road, Manchester M13 9PL, UK b 6 Battlemead, Swanage, Dorset BH19 1PH, UK c Maidstone Museum and Bentlif Art Gallery, St. Faith’s St., Maidstone, Kent ME14 1LH, UK article info abstract

Article history: Four new species of snakefly (Raphidioptera: Mesoraphidiidae): Mesoraphidia durlstonensis sp. nov., Received 27 February 2008 Mesoraphidia purbeckensis sp. nov., Mesoraphidia websteri sp. nov. and Mesoraphidia mitchelli sp. nov., are Accepted in revised form described from the Lower Cretaceous Purbeck Limestone Group, southern England. This increases the 22 September 2008 number of described snakeflies from the Mesozoic deposits of Britain from two to five species, when the Available online 8 October 2008 doubtful family Priscaenigmatidae is excluded. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Raphidioptera Mesoraphidiidae Mesozoic Britain Purbeck Limestone Group New species

1. Introduction The Mesozoic raphidiopterans comprise three valid extinct families: Baissopteridae, Mesoraphidiidae and Alloraphidiidae, and The order Raphidioptera has approximately 210 extant species one doubtful extinct family, Priscaenigmatidae (see Aspo¨ck and in two families: Raphidiidae and Inocelliidae (Aspo¨ck, 2002; Gri- Aspo¨ck, 2004 for discussion). The Mesozoic raphidiopterans are maldi and Engel, 2005; Engel et al., 2006). Raphidioptera is the least very diverse at species level, supporting the idea that extant diverse order of the holometabolous insects, and is grouped within Raphidioptera are (Engel, 2002; Grimaldi and Engel, 2005). the superorder Neuropterida along with Neuroptera and Mega- Snakeflies from the Mesozoic of Britain are rare. Jarzembowski loptera (Aspo¨ck, 2002). The distinctive features of the order include (1984) recorded a specimen from the Weald Clay (Lower Cretaceous, an elongated prothorax, a long female ovipositor, and typical Hauterivian) of Surrey. Whalley (1985, 1988) described two raphi- venation (Engel, 2002). Four autapomorphies have been suggested diopterans from the Lower Jurassic of Dorset: Metaraphidia confusa for the order, but only one of which can be applied to fossil wings: (Mesoraphidiidae) and Priscaenigma obtusa (Priscaenigmatidae). the forewing subcosta running into the anterior margin (Aspo¨ck Priscaenigmatidae, however, has subsequently been excluded and Aspo¨ck, 2004). The habitat preference of extant snakeflies is from the order (Aspo¨ck and Aspo¨ck, 2004). Clifford et al. (1994) arboreal, in areas of cool weather with at least some periods of near figured a snakefly (Mesoraphidia sp.) from the Purbeck Limestone freezing temperatures. This restricts them to cold temperate Group of Dorset. This specimen, described herein along with the regions and excludes them from areas with hotter climates (Gri- other Purbeck specimens, brings the total number of described maldi and Engel, 2005). Mesozoic raphidiopterans, however, have species from the Mesozoic of Britain to five, all in the family been discovered in much warmer conditions such as inferred Mesoraphidiidae. tropical climates (Engel, 2002). In Britain today snakeflies are rare, with four species in four genera in the family Raphidiidae; these are 2. Geological setting restricted to arboreal environments (Plant, 1997). The Purbeck Limestone Group of southern England is a predom- inantly lagoonal succession currently considered to be basal Creta- ceous (Berriasian) in age, although the precise position of the * Corresponding author. Jurassic/Cretaceous boundary is still in dispute, and may in fact lie E-mail address: [email protected] (J.E. Jepson). within the Purbeck (Allen and Wimbledon, 1991; Ensom, 2002). The

0195-6671/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.cretres.2008.09.006 528 J.E. Jepson et al. / Cretaceous Research 30 (2009) 527–532

Group is divided into a lower Lulworth Formation and upper Durl- ston Formation; both of these have yielded fossils of Raphidioptera. The type section of the Purbeck Limestone Group, and the source of all currently recognised Purbeck raphidiopteran material, is Durlston Bay, near Swanage, Dorset (National Grid Reference SZ 035 780). Fossil insects from this site were first studied and named in the mid 19th Century, principally by Prof. J. O. Westwood (1854). It remains highly productive and has provided the holotypes of approximately 200 insect species to date.

3. Systematic palaeontology

The studied material is deposited in the following collections: Booth Museum of Natural History, Brighton (BMB) and Maidstone Museum and Bentlif Art Gallery, Kent (MNEMG). The material was studied using an Olympus SZH stereomicroscope with a camera lucida attachment for producing drawings. Some drawings are composite, merging the part and counterpart to estab- lish a more complete venation; this was achieved using a light box. 10% ethanol was used to view vein detail. Photographs were taken Fig. 1. A, B, Mesoraphidia durlstonensis sp. nov., holotype forewing BMB 0183737; using a Nikon D1X digital SLR camera mounted on a rostrum stand. Durlston Bay, Upper Berriasian. A, venation drawing. B, photograph. Scale bar repre- sents 1 mm. Wing vein abbreviations are as follows: C, costa; Sc, subcosta; R, radius; Rs, radial sector; Rs1, first branch of Rs; Rs2, second branch of Rs, M, media; MA, media anterior; MP, media posterior; MP1, first Diagnosis. Larger than other Purbeck Mesoraphidia. dc narrower branch of MP; MP2, second branch of MP; Cu, cubitus; CuA, cubitus than that of M. purbeckensis sp. nov. Costal area shorter and nar- anterior; CuP, cubitus posterior; A, analis; r, radial cell; dc, discal rower than in M. websteri sp. nov. CuA has fewer forks than cell; mc, medial cell; ac, anal cell; p, pterostigma. M. gobiensis Ponomarenko, 1988. Differs from all other Meso- raphidia in the following combination of characters: many costal Order: Raphidioptera Nava´s, 1918 veinlets; long pterostigma, almost half length of wing; Rs2 exhib- Family: Mesoraphidiidae Martynov, 1925 iting a kink in vein; three radial cells, 2r wide and 3r long; Rs and Genus Mesoraphidia Martynov, 1925 MA two branched, both branches simple; lack of a distinct fork for MA; MP two branched, both branches forking once. Type species. Mesoraphidia grandis Martynov, 1925 from Karatau, Description. Holotype forewing (Fig. 1): narrow and elongate; Kazakhstan (Upper Jurassic). 10.6 mm long and 2.9 mm wide. C complete; costal area Diagnosis. Wings short to long (5–18 mm long) and elongate. preserved with seven crossveins. Sc terminates approximately Narrow costal area. Costal veinlets not exceeding 10. Sc terminating 6 mm from wing base (just past mid-point of wing) on anterior at costal margin near mid-point of wing. Pterostigma long. R runs wing margin. R complete and long, terminating near wing apex; parallel to costal margin and curves with it towards apex. Rs one crossvein present between R and C, 9.0 mm from wing base. separates from R approximately one third of length of wing from Pterostigma present; over 3 mm long; beginning of pterostigma base; it is then connected to R by two crossveins forming the three difficult to determine as pigment fades towards wing base. Three radial cells. Elongated dc. Two or three medial cells posterior to MP radial cells present: 1r: 2.6 mm long and 0.5 mm wide; 2r: in forewing. Anal area similar to other raphidiopterans; eliptical ac. 2.4 mm long and 0.7 mm wide and 3r: 1.7 mm long and 0.4 mm wide. Rs arises from R approximately 3.5 mm from wing base. Rs Included species. In addition to the type species, 21 species from two branched; Rs1 and Rs2 simple, Rs2 exhibits a kink in vein; the Jurassic and Cretaceous of Europe, Asia and the USA (Engel, both Rs1 and Rs2 arise in distal part of wing near apex. dc formed 2002; Engel et al., 2006), and four species described below. by a crossvein between Rs and MA, 2.4 mm long and 0.5 mm wide (at widest point). MA arises 1.7 mm from R where M forks Mesoraphidia durlstonensis sp. nov. into MA and MP; it then runs into Rs 0.6 mm from the split with Figs. 1–3 M. MA coalesced with Rs for 1.9 mm. MA two branched with both branches simple. One crossvein present between MA and MP. MP 1994 Mesoraphidia sp. Clifford et al.: p.143, fig. 1 deeply forked with each branch forked near posterior wing Derivation of name. The species’ name refers to Durlston Bay. margin. Three cells present between MP and CuA: mc1: 1.8 mm long, 0.4 mm wide; mc2: 1.4 mm long, 0.8 mm wide and mc3: Holotype. Forewing: BMB 018371 and 0183737 [DB175/RAPH 1] 1.3 mm long, 0.6 mm wide. CuA coalesced with M for 0.6 mm. (part and counterpart). Clements (1993) Bed DB175, Corbula beds, CuA forks near posterior wing margin; distal branch forked, basal Durlston Formation, Purbeck Limestone Group; Durlston Bay, branch simple. Two crossveins present between CuA and CuP; Swanage, Dorset; Upper Berriasian; collected by R.A. Coram. CuP simple. Anal area is preserved with 1A, 2A and 3A present. ac oval shaped, 1.4 mm long, 0.6 mm wide. Paratype. MNEMG 2007.45 [DB175/RAPH 2] (part and counter- Paratype MNEMG 2007.46: hindwing lacking basal portion part). Fore- and overlapping hindwing. Horizon and locality as for (Fig. 3). Length as preserved 8.0 mm; estimated total length holotype; collected by R. A. Coram. 10.5 mm. Width 2.7 mm. C incomplete; costal area with three crossveins present but incomplete. Sc terminates approximately Paratype. Hindwing: MNEMG 2007.46 [DB175/RAPH 4] (part and 4.4 mm from wing apex (before mid-point of wing) on anterior counterpart). Horizon and locality as for holotype; collected by R. A. wing margin. R long, terminating near wing apex; one crossvein Coram. present between R and C, 1.0 mm from wing apex. Pterostigma J.E. Jepson et al. / Cretaceous Research 30 (2009) 527–532 529

Fig. 2. A, B, C, D, Mesoraphidia durlstonensis sp. nov. paratype MNEMG 2007.45; Durlston Bay, Upper Berriasian. A, venation drawing of overlapping fore- and hindwing. B, photograph under 10% ethanol. C, venation drawing of forewing. D, venation drawing of hindwing. Scale bar represents 1 mm.

(or pterostigma and dark membrane) long, from base of wing between MP and CuA; mc: 1.7 mm long, 0.5 mm wide. CuA forks (as preserved) to r-c crossvein. Three radial cells present: 1r: near posterior wing margin. Two crossveins present between MP 2.4 mm long and 0.5 mm wide, cut by crossvein; 2r: 2.3 mm long and CuA. One crossvein present between CuA and CuP; CuP forked. and 0.6 mm wide and 3r: 1.2 mm long and 0.3 mm wide. Rs arises One crossvein present between CuP and 1A. Anal area poorly 5.7 mm from wing apex; Rs two branched; Rs1 and Rs2 simple, Rs2 preserved; 1A present, but incomplete. has a kink in vein; both Rs1 and Rs2 arise in distal part of wing near Remarks. Three specimens have been assigned to this species apex. MA origin not preserved, MA and Rs coalesced for 1.1 mm comprising two forewings and two hindwings. The paratype before separating. dc: 2.7 mm long and 0.6 mm wide (at widest MNEMG 2007.45 shows slight differences in wing shape and vein point). MA two branched, both branches simple. Two crossveins proportions from the holotype. However, it has been assigned to M. present between MA and MP. MP origin 7.5 mm from wing apex; durlstonensis sp. nov. on the basis of its similar size, overall venation MP deeply forked with MP1 and MP2 branched. One cell present pattern and comparable variation observed within extant species of snakeflies (e.g. species described in Aspo¨ck et al., 1983; Rausch and Aspo¨ck, 1993; pers. obs. JEJ). The overlapping fore- and hindwing of MNEMG 2007.45 indicates that the paratype MNEMG 2007.46 is of the same species due to similarities in venation and size. This is the only species from the Purbeck that has a pigmented pterostigma, but this may simply be fortuitous preservation. M. durlstonensis sp. nov. is the largest species of snakefly from the Purbeck, being almost twice the size of the other Purbeck raphidiopterans described herein.

Mesoraphidia purbeckensis sp. nov. Fig. 4

Derivation of name. After the Isle of Purbeck, Dorset.

Holotype. Forewing: MNEMG 2007.47 [DB175/RAPH 6]. Clem- ents’ Bed DB175, Corbula beds, Durlston Formation, Purbeck Limestone Group; Durlston Bay, Swanage, Dorset; Upper Berria- sian; collected by R. A. Coram.

Diagnosis. Forewing small compared to most species of Meso- raphidia. Few costal veinlets. Differs from M. durlstonensis sp. nov. in possessing narrower mc3, CuA forking more proximally, and most distal r-c crossvein being closer to termination of R. dc wider than in M. durlstonensis sp. nov., M. websteri sp. nov. and M. mitchelli sp. nov. 1A, 2A and 3A long compared to other Purbeck Meso- Fig. 3. A, B, Mesoraphidia durlstonensis sp. nov., paratype hindwing MNEMG 2007.46; Durlston Bay, Upper Berriasian. A, composite venation drawing from part and coun- raphidia. Differences from other Mesoraphidia species as for M. terpart. B, photograph. Scale bar represents 1 mm. durlstonensis sp. nov. 530 J.E. Jepson et al. / Cretaceous Research 30 (2009) 527–532

Fig. 4. A, B, Mesoraphidia purbeckensis sp. nov., holotype forewing MNEMG 2007.47; Durlston Bay, Upper Berriasian. A, composite venation drawing from part and coun- terpart. B, under 10% ethanol. Scale bar represents 1 mm. Fig. 5. A, B, Mesoraphidia websteri sp. nov., holotype hindwing MNEMG 2007.48; Durlston Bay, Upper Berriasian. A. composite venation drawing from part and coun- Description. Forewing: 4.9 mm preserved length, 5.1 mm esti- terpart. B. photograph. Scale bar represents 1 mm. mated total length. Width 1.4 mm. C incomplete; costal area with four crossveins present but incomplete. Sc terminates approxi- mately 2.3 mm from wing apex (just past mid-point of wing) on the Limestone Group; Durlston Bay, Swanage, Dorset; Upper Berria- anterior wing margin. R complete and long, terminating near wing sian; collected by R. A. Coram. apex; one crossvein present between R and Sc, 3.8 mm from wing apex; two crossveins present between R and C, 0.4 mm and 1.9 mm Diagnosis. Wing shorter than M. durlstonensis sp. nov. but longer from wing apex. Pterostigma not preserved. Three radial cells than M. purbeckensis sp. nov. and M. mitchelli sp. nov. Costal area present: 1r: 1.2 mm long, 0.3 mm wide; 2r: 1.0 mm long, 0.4 mm longer than other Purbeck Mesoraphidia. R terminates further wide and 3r: 0.7 mm long, 0.2 mm wide. Rs arises approximately before apex than M. durlstonensis sp. nov. Differences from other 3.1 mm from wing apex. Rs two branched; Rs1 and Rs2 simple; both Mesoraphidia species as for M. durlstonensis sp. nov. Rs1 and Rs2 arise in distal part of wing near apex. dc: 1.2 mm long Description. Hindwing: incomplete, 5.2 mm preserved length, and 0.4 mm wide (at widest point). MA arises 0.7 mm from origin of 5.7 mm estimated total length. Width 2.1 mm. C incomplete; M þ CuA. MA coalesced with Rs for 0.7 mm. MA two branched with costal area poorly preserved and incomplete with two crossveins both branches simple. MP coalesced with CuA for short distance present. Sc terminates 2.1 mm from wing apex at two-thirds of (w0.2 mm). MP and CuA arise 3.9 mm from wing apex. MP deeply wing length. R long, terminating 0.4 mm from wing apex. Pter- forked with both branches forked. Three cells present between MP ostigma not preserved. Three radial cells present: 1r: 1.1 mm long and CuA: mc1: 0.7 mm long, 0.2 mm wide; mc2: 0.6 mm long, and 0.4 mm wide; 2r: 1.6 mm long and 0.4 mm wide and 3r: 0.3 mm wide and mc3: 0.3 mm long, 0.2 mm wide. CuA forks near 0.6 mm long and 0.3 mm wide. Rs and MA arise at different posterior wing margin; first branch forks again, second branch is points, Rs 3.4 mm from wing apex, MA 4.3 mm from wing apex. simple. Two crossveins present between CuA and CuP; CuP simple. Rs and MA coalesced for 0.6 mm before forking. Rs two branched; One crossvein present between CuP and 1A. 1A, 2A, 3A preserved, Rs1 and Rs2 simple; both Rs1 and Rs2 arise in distal part of wing all simple; only part of ac present, 0.3 mm wide. near apex. dc 1.6 mm long and 0.5 mm wide (at widest point). Remarks. M. purbeckensis sp. nov. has the smallest wing length of MA two branched, both branches simple. MP origin not the Purbeck Mesoraphidia and the second smallest wing length preserved; MP deeply forked with MP1 and MP2 two branched. within Mesoraphidia. It is more typical of the size of Purbeck One cell present between MP and CuA: mc: 1.3 mm long, 0.4 mm snakeflies, with the exception of M. durlstonensis sp. nov. The wide. CuA forks near the posterior wing margin. Two crossveins pterostigma is not present; this is most likely an artefact of pres- present between MP and CuA. One crossvein present between ervation where the pigmentation has not been preserved. CuA and CuP; CuP simple. Anal area poorly preserved; 1A, 2A and 3A present but incomplete. Mesoraphidia websteri sp. nov. Remarks. M. websteri sp. nov. has been tentatively assigned to Fig. 5 a new species as its small size and venation exclude it from M. durl- stonensis sp. nov. and its larger size excludes it from M. purbeckensis Derivation of name. After Mr Andrew Webster, collector of sp. nov. The apparent lack of a pterostigma, as with M. purbeckensis Purbeck insects. sp. nov., is probably an artefact of preservation.

Holotype. Hindwing: MNEMG 2007.48 [DB175/RAPH 8]. Clem- Mesoraphidia mitchelli sp. nov. ents’ Bed DB175, Corbula beds, Durlston Formation, Purbeck Fig. 6 J.E. Jepson et al. / Cretaceous Research 30 (2009) 527–532 531

4. Discussion

Prior to this research, there have been only five fossil snakefly specimens recorded in Britain from two families: Mesoraphidiidae and Priscaenigmatidae (Jarzembowski, 1984; Whalley, 1985, 1988; Clifford et al., 1994). The position of Priscaenigmatidae in Raphi- dioptera, however, has been questioned (see Aspo¨ck and Aspo¨ck, 2004 for discussion); therefore, if it is removed from the order, fossil snakeflies in Britain would only be represented by the family Mesoraphidiidae. The most notable features shared by the Purbeck fauna are a more ‘simplified’ venation: veins Rs and MA are two branched with both branches forking close to the posterior wing margin and MP with a deep fork with both branches forking once near the posterior wing margin. M. purbeckensis sp. nov., M. websteri sp. nov. and M. mitchelli sp. nov. are very small, with estimated wing sizes ranging from 5.1–5.7 mm; they are a little larger than the smallest known snakefly, Nanoraphidia, from the Lower Cretaceous of Burma with a wing length of 4.5 mm (Engel, 2002). This small size sepa- rates them from the majority of other Mesoraphidia species (exceptions being M. parvula (5.5 mm) and M. gaoi (5 mm) (Mar- Fig. 6. A, B, Mesoraphidia mitchelli sp. nov., holotype MNEMG 2007.49; Durlston Bay, tynov, 1925; Ren et al., 1995)) which are almost double or more the Lower Berriasian. A, venation drawing. B, photograph under 10% ethanol. Scale bar size of these snakeflies. M. durlstonensis sp. nov., however, is of represents 1 mm. a more average size for the genus. There are significant differences between the snakefly found in the older Upper Soft Cockle beds and Derivation of name. After Mr A. A. Mitchell, collector of the the snakeflies from the younger Corbula beds. These differences are holotype. the distinct fork of MA and the two medial cells and the posterior fusion of CuA and MP. Holotype. MNEMG 2007.49 [USCDB/TM 924]. Soft Cockle beds The Purbeck climate is interpreted as having been seasonally (probably Clements’ Bed DB66), Lulworth Formation, Purbeck semi-arid (Allen, 1998), providing further evidence that Mesozoic Limestone Group; Durlston Bay, Swanage, Dorset, UK; Lower Ber- raphidiopterans were less climatically restricted than they are riasian; collected by A. A. Mitchell. today. There is evidence that the climate became rather more Diagnosis. Differs from all other species of Mesoraphidia in the humid midway through Purbeck times: for example, evaporites possession of two medial cells: mc1 and mc2; and in the fusion become much less frequent and there is a sustained increase in the of MP2 and CuA for a short distance near the posterior wing clay mineral kaolinite (Deconinck, 1987), indicative of more intense margin. A distinct fork of MA also distinguishes it from all weathering of the massifs supplying sediment to the depositional congeners with the exception of M. gaoi Ren et al., 1995 and M. basin. The preserved material suggests that there may have been parvula Martynov, 1925. a raphidiopteran faunal change associated with this climate change, Description. 5.0 mm preserved length, 5.4 mm estimated total but more specimens are required to confirm this. length. Width 1.5 mm. C incomplete; costal area with four cross- Extant raphidiopterans are predators of small arthropods both veins present but incomplete. Sc termination not seen. R incom- as larvae and adults and live exclusively on trees, the larvae very pletely preserved, long, terminating near wing apex; one crossvein often developing under bark (Grimaldi and Engel, 2005). Purbeck present between R and C, 0.4 mm from wing apex. Pterostigma not raphidiopteran fossils are found alongside those of other tree- preserved. Three radial cells present: 1r: 1.4 mm long and 0.4 mm associated taxa (such as cupedoid ) and occasional remains wide; 2r: 1.1 mm long and 0.5 mm wide and 3r: 0.7 mm long and of conifer foliage, all suggestive of a forested hinterland and 0.3 mm wide. Rs arises approximately 3.2 mm from wing apex. Rs a raphidiopteran habitat similar to that of today. two branched; Rs1 and Rs2 simple; both Rs1 and Rs2 arise in distal part of wing near apex. dc 1.2 mm long and 0.4 mm wide (at widest point). MA origin not preserved. MA coalesced with Rs for 0.8 mm. Acknowledgements MA with distinct fork closer to wing margin than fork of RS; both branches simple. MP deeply forked with both branches forked near We wish to thank John Nudds, Karl Bates, Peter Falkingham (The posterior wing margin. Two cells present between MP and CuA: University of Manchester) for useful discussions, and Dmitri mc1: 0.4 mm long, 0.2 mm wide and mc2: 0.7 mm long, 0.4 mm Logunov and Philip Rispin (The Manchester Museum) for access to wide. CuA forked, fused for a short distance with MP2. Part of CuP extant specimens of Raphidioptera. Also, thanks to two anonymous preserved. Anal area not preserved. reviewers for their comments and suggestions. Remarks. M. mitchelli sp. nov. differs significantly from all the other Purbeck Mesoraphidia. The two cells of mc is the most References significant difference; this could be due to either one less cell in the forewing or one extra in the hindwing. However, it is difficult to Allen, P., Wimbledon, W.A., 1991. Correlation of NW European Purbeck-Wealden ascertain whether M. mitchelli sp. nov. is a fore- or hindwing (nonmarine Lower Cretaceous) as seen from the English type-areas. Cretaceous because the base of MA has not been preserved nor CuP and the Research 12, 511–526. Allen, P., 1998. Purbeck-Wealden (early Cretaceous) climates. Proceedings of the anal veins. The distinct fork of MA is also different from the other Geologists’ Association 109, 197–236. Purbeck Mesoraphidia. The apparent lack of a pterostigma is most Aspo¨ck, U., 2002. The phylogeny of the Neuropterida (Insecta: Holometabola). probably an artefact of preservation, because pigment is rarely Zoologica Scripta 31, 51–55. Aspo¨ck, U., Aspo¨ck, H., 2004. Two significant new snakeflies from Baltic amber, with preserved in this bed. M. mitchelli sp. nov. is the oldest snakefly discussion on autapomorphies of the order and its included taxa (Raphi- from the Purbeck. dioptera). Systematic Entomology 29, 11–19. 532 J.E. Jepson et al. / Cretaceous Research 30 (2009) 527–532

Aspo¨ck, H., Aspo¨ck, U., Rausch, H., 1983. Mauroraphidia maghrebina n. gen. n. Grimaldi, D.A., Engel, M.S., 2005. Evolution of the Insects. Cambridge University sp. – eine neue Raphidiiden-Spezies aus dem Atlas (Neuropteroidea: Raphi- Press, New York, p. 755. dioptera). Zeitschrift der Arbeitsgemeinschaft O¨ sterreichischer Entomologen Jarzembowski, E.A., 1984. Early Cretaceous insects from southern England. Modern 35, 27–32 (In German). Geology 9, 71–93. Clements, R.G., 1993. Type-section of the Purbeck Limestone Group, Durlston Bay, Martynov, A.V., 1925. To the knowledge of fossil insects from Jurassic beds in Turke- Swanage, Dorset. Proceedings of the Dorset Natural History and Archaeological stan. 1. Raphidioptera. 6 seryia. Izvestiya Rossiiskoi akademii Nauk 19, 235–246. Society 114 (for 1992), 181–206. Plant, C.W., 1997. A key to the adults of British lacewings and their allies (Neuro- Clifford, E., Coram, R., Jarzembowski, E.A., Ross, A.J., 1994. A supplement to the ptera, Megaloptera, Raphidoptera and Mecoptera). Field Studies 9, 179–269. insect fauna from the of Dorset. Proceedings of the Dorset Rausch, H., Aspo¨ck, H., 1993. Phaeostigma holzingeri n. sp. – eine neue Kamelhals- Natural History and Archaeological Society 115 (for 1993), 143–146. fliege aus Ipiros (Griechenland) (Neuropteroidea: Raphidioptera: Raphidiidae). Deconinck, J.F., 1987. Mine´raux argileux des facie`s Purbeckiens: Jura suisse et fra- Zeitschrift der Arbeitsgemeinschaft O¨ sterreichischer Entomologen 45, 19–26 (In nçais, Dorset (Angleterre) et Boulonnais (France). Annales Societe´ Ge´ologique German, with English summary). du Nord 106, 285–297. Ren, D., Lu, L., Guo, Z., Ji, S., 1995. Faunae and Stratigraphy of the Jurassic–Cretaceous Engel, M.S., 2002. The smallest snakefly (Raphidioptera: Mesoraphidiidae): a new in Beijing and Adjacent Areas. Seismic Publishing House, Beijing (In Chinese, species in Cretaceous amber from Myanmar, with a catalog of fossil snakeflies. with English summary). American Museum Novitates 3363, 1–22. Westwood, J.O., 1854. Contribution to fossil entomology. Quarterly Journal of the Engel, M.S., Lim, J.-D., Baek, K.-S., 2006. Fossil snakeflies from the Early Cretaceous Geological Society of London 10, 378–396. pls 14–18. of southern Korea (Raphidioptera: Mesoraphidiidae). Neues Jahrbuch fu¨ r Geo- Whalley, P.E.S., 1985. The systematics and palaeogeography of the Lower Jurassic logie und Pala¨ontologie 2006, 246–256. insects of Dorset, England. Bulletin of the British Museum (Natural History) Ensom, P.C., 2002. The Purbeck Limestone Group of Dorset, southern England: Geology 39, 107–189. a guide to lithostratigraphic terms. Special Papers in Palaeontology 68, Whalley, P.E.S., 1988. Mesozoic Neuroptera and Raphidioptera (Insecta) in Britain. 7–12. Bulletin of the British Museum (Natural History) Geology 44, 45–63.

PART ONE B:

NEUROPTERIDA OF THE LOWER CRETACEOUS

OF SOUTHERN ENGLAND

WEALDEN SUPERGROUP

And dead wings carried like a paper kite

Robert Frost, from Design (1936)

112

CHAPTER 6

2. NEUROPTERAN (INSECTA) PALAEODIVERSITY WITH PREDICTIONS FOR

THE CRETACEOUS FAUNA OF THE WEALDEN

113

Palaeogeography, Palaeoclimatology, Palaeoecology 248 (2007) 109–118 www.elsevier.com/locate/palaeo

Neuropteran (Insecta) palaeodiversity with predictions for the Cretaceous fauna of the Wealden ⁎ J.E. Jepson , D. Penney

School of Earth, Atmospheric and Environmental Sciences, University of Manchester, Manchester M13 9PL, Received 8 May 2006; received in revised form 23 November 2006; accepted 29 November 2006

Abstract

The Wealden Group of southern England is a succession of non-marine sediments from the lower Cretaceous (c. 141–119 Ma), which also outcrop in Belgium and northwest Germany. Most insect orders are present as fossils, including Neuroptera, one of the most basal holometabolous orders. The fossil record of Neuroptera is diverse, spanning Permian–Recent, with all major family radiations having occurred prior to the Wealden. However, Wealden Neuroptera have received little attention to date, with only three families reported and one species described. The fossil record in general is notoriously incomplete, and palaeoenvironmental reconstructions can facilitate predictions of what other organisms were present in the habitat, but which were either not preserved or only occasionally preserved and yet to be found as fossils. We examine the Wealden palaeohabitat and global neuropteran palaeodiversity and consider these in relation to palaeo/biogeography of fossil and extant taxa (including habitat preferences). We predict that the following families will form components of the Cretaceous Wealden fauna: Nymphidae, Polystoechotidae, Chrysopidae, Hemerobiidae and Coniopterygidae. © 2006 Elsevier B.V. All rights reserved.

Keywords: England; Fossil; Lacewings; Mesozoic; Neuroptera; Palaeobiogeography

1. Introduction Beds have been dated as Berriasian–Valanginian and the Weald Clays as Hauterivian–Aptian? (Allen and Wim- The Wealden Group of southern England is a suc- bledon, 1991; Rasnitsyn et al., 1998). The Wessex sub- cession of non-marine sediments from the lower Creta- basin comprises the Wessex (Hauterivian) and Vectis ceous (c. 141–119 Ma) (Allen and Wimbledon, 1991), Formations (Barremian–Aptian?) (Twitchett, 1994). In subdivided into the Wealden and Wessex sub-basins. addition to the distal facies, there are two principle facies The Wealden sub-basin consists of the Hastings Beds types in the Wealden: sediments of medial to proximal (Ashdown Beds, Wadhurst Clay, Lower and Upper Tun- alluvial origin and sediments of lacustrine–lagoonal bridge Wells Sand, and the Grinstead Clay Formations) origin (Allen, 1981, 1989; Radley, 2005). The Wealden and the Weald Clay (Upper and Lower) (Allen and Group also outcrops in Western Europe: Belgium and Wimbledon, 1991; Rasnitsyn et al., 1998). The Hastings northwest Germany (Allen and Wimbledon, 1991; Jar- zembowski, 1995; Elstner and Mutterlose, 1996; Gri- maldi and Engel, 2005)(Fig. 1). ⁎ Corresponding author. Fax: +44 161 275 3947. Most insect orders are present in the Wealden Group, E-mail address: [email protected] including Odonata, Blattodea, Orthoptera, Hemiptera, (J.E. Jepson). Neuroptera, Coleoptera, Mecoptera, Diptera, Trichoptera,

0031-0182/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.palaeo.2006.11.015 110 J.E. Jepson, D. Penney / Palaeogeography, Palaeoclimatology, Palaeoecology 248 (2007) 109–118

Fig. 1. Map showing the geographical extent of the Wealden Formation.

Hymenoptera, Isoptera, Raphidioptera (Jarzembowski, cribed extant species in 17 families (Aspöck, 2002). 1984, 1995) and Plecoptera (Sinitshenkova, 1998). How- Their defining features include diaphanous wings with ever, the continental deposits have not been as well intricate, laced venation, larval sucking mouthparts for- studied as the English deposits in terms of the insect med by the maxillae and mandible, and the association fauna. The insects are preserved as uncrushed and unoxi- of the ninth gonocoxites with the gonarcus (Grimaldi dized (Jarzembowski, 1984), disarticulated skeletal mate- and Engel, 2005). They range in size from small (wing- rial (usually only wings, the reason for this sorting is span 5 mm) barklouse-like waxy lacewings, to large unclear) in the calcareous (or ferruginous) siltstone (wingspan exceeding 150 mm) -like , developments or phosphatic or sideritic concretions in and when at rest hold their wings in a roof-like position the clay formations. A subtropical/warm temperate (with over their body. Most are poor fliers, however there are seasonal rainfall) climatic reconstruction has been gene- some exceptions e.g., the family Ascalaphidae, which rated for the Wealden, based on the presence of certain represents the strongest and most agile fliers of the order insect orders (e.g. Isoptera) (Jarzembowski, 1995)and (New, 1991). Neuroptera are widespread globally, but sedimentological evidence (Allen, 1998; Haywood et al., tend to be most abundant in the tropics, with several 2005). The occurrence of wet and dry (commonly with families having restricted distributions (New, 1991). wild fires) seasons has been proposed, as has year round Sixty-six extant species from six families (Chrysopidae, precipitation with little evidence for prolonged drought Coniopterygidae, Hemerobiidae, Myrmeleontidae, Os- (Batten, 1998; Haywood et al., 2005). Evidence for wild mylidae and Sisyridae) occur in Britain (Plant, 1997), fires exists in the form of fusainized elytra and plant with six additional families occurring on mainland remains (Watson and Alvin, 1996; Batten, 1998; Jarzem- Western Europe (Nevrorthidae, Berothidae, Ascalaphi- bowski, 2003). The Wealden palaeohabitat consisted of dae, Nemopteridae, Dilaridae and Mantispidae (Aspöck woodland bordering Wealden water courses (Jarzembow- et al., 1980)). Presumably, a combination of both ski, 1995). climatic and Channel barrier factors influence these Neuroptera is one of the most basal orders of holo- distributions. metabolous insects (Ponomarenko, 2002; Grimaldi and The fossil record of Neuroptera is diverse and spans a Engel, 2005) and consists of approximately 6000 des- long period of geological time (Permian–Recent) with J.E. Jepson, D. Penney / Palaeogeography, Palaeoclimatology, Palaeoecology 248 (2007) 109–118 111 all major family radiations having occurred prior to the tifiable as neuropterans, whereas isolated wings from Wealden (see later, and Ponomarenko, 2002; Grimaldi many hemerobiiform families would be considerably more and Engel, 2005 for comprehensive reviews). However, difficult to attribute to Neuroptera by a non-specialist. the Neuroptera of the Wealden have received little atten- According to Jarzembowski (1995) the frequency of tion to date, with only three myrmeleontiform families occurrence for Neuroptera in the Wealden fossil fauna is (†Brongniartiellidae, †Kalligrammatidae and Psychopsi- low, accounting for only 0.1–0.9% of a sample of 13,000 dae) reported (Jarzembowski, 1984) and only one species specimens. This is consistent with other fossil (and extant) described (Jarzembowski, 2001). The absence of hemer- assemblages. At least for the modern world and for obiiform families is enigmatic. One reason may be that the Palaeogene lacustrine shales, less than 1% representation myrmeleontiform families recorded are very easily iden- of Neuroptera is normal: Tadushi Formation, Primorye,

Fig. 2. Evolutionary tree of Neuroptera (after Grimaldi and Engel, 2005, based on Aspöck, 2002; updated to include Jarzembowski, 1980; Whalley, 1980; Schlüter, 1984; Ponomarenko, 1984; Lambkin, 1988; Makarkin, 1990a,b; Kukalova-Peck, 1991; Ross and Jarzembowski, 1993; Nel and Henrortay, 1994; Ren and Guo, 1996; Makarkin, 1996, 1997b; Martins-Neto, 1997; Makarkin, 1998, 1999; Azar et al., 2000; Martins-Neto, 2000; Makarkin and Archibald, 2003). 112 J.E. Jepson, D. Penney / Palaeogeography, Palaeoclimatology, Palaeoecology 248 (2007) 109–118

Russia, about 0.3%; Bembridge Marls, about 0.1%; none 2002, but with fossil taxa incorporated) was used to are known from Eckfeld Maar, Germany; Messel, generate observed and predicted palaeodiversity spin- Germany; or the Bournemouth Group, England. Higher dle diagrams (Fig. 3) using the technique explained by representations of Neuroptera in Palaeogene fossil Penney (2004a,b). This technique involves super- assemblages are rare: about 1.5% at the Okanagan imposing a cladogram over geological time and Highlands locality of McAbee; and approximately 3.8– calibrating it using fossil data. Where a node, or new 5.0% of the Danish Fur Formation insects (depending on fossilrecordappears,thenumberofrangesofknown museum collection) (Archibald and Makarkin, 2006 and taxa, based on direct fossil evidence, is counted go give references cited therein). In modern forest communities an observed palaeodiversity value for that particular Neuroptera have been consistently found to be less than point in geological time. At the same point in time, a 1% of insects: less than 1% in tropical forest canopy separate count is made of predicted taxa, from hy- fogging samples in Borneo; 0.45% in window and Ma- pothesized range extensions, based on the presence of laise trap samples in Queensland, Australia; and 0.25% in their sister taxon (derived from the cladogram) in the light trap samples in Primorye, Russia (Archibald and fossil record. Summing these two values gives a pre- Makarkin, 2006 and references cited therein). dicted palaeodiversity value. For details of tree All forms of fossilization are subject to unique pre- construction methodology see Smith (1994);seealso servation biases. In sedimentary fossil deposits such as Lane et al. (2005). The observed palaeodiversity data the Wealden, terrestrial invertebrates (Neuroptera do not were plotted over time (Fig. 4) and used to generate have aquatic larvae, except for the families Nevrorthidae maximum and minimum predictions for global neu- and Sisyridae) are usually allochthonous, and thus ropteran family richness at the time of the Wealden. represent only a small fraction of the biota alive in the Most neuropteroid families were already widespread by palaeohabitat. Given the incompleteness of the fossil the Cretaceous and adults of most families may be record, it is equally important that we elucidate what present in many different terrestrial environments other organisms were present in the environment at the (New, 1989), but not all can be expected to have same time as those preserved as fossils, but which did existed in the Wealden environment. Thus, predictions not get fossilized, or did so only occasionally and have specifically for the Wealden fauna were further refined yet to be found as fossils (e.g., Penney, 2005). Accurate based on closer scrutiny of fossil neuropteran palaeo- palaeoenvironmental reconstructions based on a broad biogeography (Fig. 5), in terms of taxa that had oc- range of observed fossil taxa can facilitate this. Here, we curred in the region earlier in geological time and also examine the global neuropteran fossil record and con- sider it in relation to the palaeo/biogeography of fossil and extant taxa in order to make predictions for expected palaeodiversity of Wealden Neuroptera during the Cretaceous.

2. Methods

Palaeontological data are scarce by their very nature. New fossil Neuroptera species are often described based on singletons and only occasionally from a series of individuals, which are usually small in number. Thus, species numerical data are of limited value for quan- titative assessments of palaeodiversity. Traditionally, analysis of terrestrial palaeocommunities has been at the family level, primarily as a result of a paucity of data, and has been applied extensively as the ‘higher taxon approach’ by neontologists charting the distribu- tion, ecological correlates, and diversity patterns of extant tropical insect faunas (Labandeira, 2005). Here we use a similar approach with the fossil data. Fig. 3. Spindle diagrams of observed and predicted neuropteran family The most recent evolutionary tree of the Neuroptera palaeodiversities based on data derived from the evolutionary tree (see (Fig. 2)(Grimaldi and Engel, 2005; based on Aspöck, Methods). J.E. Jepson, D. Penney / Palaeogeography, Palaeoclimatology, Palaeoecology 248 (2007) 109–118 113

Fig. 4. Observed neuropteran family richness over geological time. Data points are derived from Fig. 2 and relate to the total family richness value following the origin or extinction of a family. Hence the lack of data points from the late Cretaceous onwards, where there appears to have been little change in this respect. contemporaneous taxa from the proximate region of random, and fits perfectly that expected for diversification continental Europe. of a group through geological time (Sepkoski, 1979).

3. Results 4. Discussion

It is evident from the evolutionary tree (Fig. 2) that Neuroptera are archaic insects and underwent their the major family radiations within Neuroptera had major family radiations during the Triassic and Jurassic, occurred by the time of the Wealden (see also Nel et al., with all 17 extant families known or predicted as fossils 2005a). This is true for both observed and predicted by the Lower Cretaceous (Fig. 2). They may have been palaeodiversity as demonstrated by the spindle diagrams more diverse in the past than they are at present (Laban- (Fig. 3), which show barely any change over the past deira and Sepkoski, 1993; Grimaldi and Engel, 2005), a 100 million years. The increase of a taxon's diversity hypothesis supported by the presence of nine pre-Ter- over time is essentially a Markovian process, i.e., the tiary extinct families (Fig. 5). A number of additional newly evolved taxa are dependent on the presence of the fossil families, e.g. from the Crato Formation, were not ancestral taxa. Thus, the number of new taxa at any time included in the tree of Grimaldi and Engel (2005); these is dependent, in part, upon the number of existing taxa. are poorly delimited taxonomically and systematically, Therefore, the null model for the diversification process and require revision (Ponomarenko, 2002; Makarkin of a radiating group is a sigmoid curve with exponential and Archibald, 2003; Archibald and Makarkin, 2006), increase during the incremental phase (Sepkoski, 1979). thus nine pre-Tertiary families represents a minimum This is clearly evident in Fig. 4 and thus the palaeo- number. The need for revision of these poorly delimited diversity plot permitted accurate predictions for global fossil families obviously creates difficulties with ana- neuropteran family richness values at the time of the lyses such as we present here and this problem cannot be deposition of the Wealden sediments (c.141–119 Ma), ignored. However, all the families we predict as undis- and produced a lower limit of approximately 15 and upper covered components of the Wealden fauna are extant limit of approximately 17. However, it should be noted and so this issue loses its importance. The only effect of that no current phylogeny includes all fossil taxa, which inclusion or exclusion of these poorly defined families is may cast doubt on the accuracy of the above predictions. likely to be a slight shift in the slope of the diversity plot Unfortunately, this problem cannot be resolved until a during the incremental phase. This would change thorough phylogenetic revision including all neuropteran slightly the predicted global neuropteran palaeodiversity fossil families is performed. Given the current state of at the time of the Wealden, but it would not alter the taxonomic ‘chaos’ with regard to fossil Neuroptera, this family predictions we present here. does not seem likely to happen in the near future. None- Neuroptera were widespread from the Permian on- theless, the palaeodiversity pattern in Fig. 4 is far from wards, constituting a significant element of the global 114 J.E. Jepson, D. Penney / Palaeogeography, Palaeoclimatology, Palaeoecology 248 (2007) 109–118 J.E. Jepson, D. Penney / Palaeogeography, Palaeoclimatology, Palaeoecology 248 (2007) 109–118 115 insect fauna, with Permian fossils recorded from South blages (†Permithonidae, †Osmylopsychopidae, †Nym- Africa, South America and Australia and Triassic fossils phitidae, Nymphidae and Polystoechotidae) should recorded from the former USSR and England (Fig. 5; occur there too, although our current knowledge sug- Ross and Jarzembowski, 1993; Schlüter, 2003: Table 2 gests the first two of these families may have become and references cited therein). There is unequivocal fossil extinct by this time and the third may have been in evidence that the majority of neuropteran families had decline (Fig. 5). Nonetheless, the remaining two fami- evolved by the Late Jurassic (c.155 Ma) when Pangea lies would almost double our predicted value from that split to form Laurasia and Gondwana (Fig. 2). By the already observed. Also, based on the extensive fossil time of the Lower Cretaceous Wealden (c.141–119 Ma), record (including wide distributions, particularly in pre- many neuropteran families had a widespread distribu- Wealden continental European localities, see Fig. 5), tion, particularly in Western Europe and Eurasia, e.g., and the known habitat preferences of extant species the families †Brongniartiellidae, Nymphidae, Myrme- from the families Chrysopidae, Hemerobiidae, Coniop- leontidae, †Babinskaiidae, †Kalligrammatidae, Psy- terygidae, and possibly also Mantispidae, it is difficult to chopsidae, Polystoechotidae, Chrysopidae, Osmylidae, exclude the probable occurrence of at least some of these Hemerobiidae and Coniopterygidae (Fig. 5). families in the forests of the Wealden during the Creta- It would be unrealistic to expect that all neuropteran ceous. However, we appreciate that our arguments here families present globally in the Lower Cretaceous would are not as strong as for the families mentioned previously occur in the Wealden palaeoenvironment. Indeed, nume- because there is no evidence of their existence in England rous families have restricted distributions today and prior to the Wealden. Thus, to err on the side of caution, distribution ranges will have varied throughout their we cannot expect all of these families to have been geological history. Therefore the predicted family pa- present, so we restrict our predictions to those families laeodiversity range of 15–17 for this period represents in common in southern England and adjacent warm tem- excess of our upper predictable limit. Given that there are perate climates today: Chrysopidae, Hemerobiidae and three families already recorded from the Wealden (which Coniopterygidae. We appreciate that other factors may has been poorly studied with regard to Neuroptera), there control distribution, such as the existence of water are five extant families in southern England today barriers, e.g. the English Channel at present, but this (Aspöck et al., 1980) (with more in continental western does not appear to have stopped these families reaching Europe), and that the order is considered to have been the UK, and the current absence of Wealden fossils is not more diverse in the past, we believe that the observed evidence that they had not already done so by the Creta- value of three families in the Wealden fossil record is not ceous. Thus, our minimum family palaeodiversity esti- a true reflection of the actual palaeodiversity of Wealden mate for Wealden Neuroptera has risen to eight (including Neuroptera. Most families of Neuroptera occur in a wide those already known), which based on our reasoning does variety of habitats but they are particularly common in not seem unreasonable. woodland and open forest habitats close to water (New, The nature of predictions is that they cannot be made 1991; Plant, 1997; Engel, 2002), similar to the palaeo- retrospectively and thus, they may be considered specu- habitat for the Wealden as described earlier. lative axiomatically. However, given the incomplete Two families present from the Jurassic of England nature of the fossil record, we feel such predictions as (†Brongniartiellidae and †Kalligrammatidae) occur in presented here, that are based on sound reasoning, may the Wealden, therefore, it would be reasonable to expect be important when considering biodiversity of palaeo- that the other neuropteran families from these assem- communities. This work forms the rationale behind a

Fig. 5. The fossil record of Neuroptera (based on the tree of Grimaldi and Engel, 2005); geological timescale not to scale. References: 1, Engel, 2004a; 2, Schlüter, 1984;3,Makarkin et al., 2003;4,Nel and Jarzembowski, 1997;5,Grimaldi and Engel, 2005;5⁎, Grimaldi and Engel, 2005 (range taken from evolutionary tree, no fossil locality data); 6, Pérez-Gelabert and Flint, 2001;7,Engel, 2004c;8,Ren and Guo, 1996;9,Makarkin, 1998; 10, Nel et al., 2003;11,Peñalver et al., 1995; 12, Makarkin, 1996; 13, Carpenter, 1959; 14, Makarkin and Archibald, 2003; 15, Willmann and Brooks, 1991; 16, Carpenter, 1943; 17, Cockerell, 1914; 18, Jarzembowski, 1980; 19, Archibald and Makarkin, 2004; 20, Weitschat and Wichard, 2002; 21, Schlüter and Sturmer, 1984; 22, Engel, 1999; 23, Engel, 2004b; 24, Nel et al., 2005b; 25, Kukalova-Peck, 1991; 26, Engel, 2002; 27, Nel et al., 2005a; 28, Makarkin, 1994; 29, Rasnitsyn and Ross, 2000; 30, Makarkin, 1990b; 31, Ponomarenko, 1992; 32, Makarkin, 1997b; 33, Makarkin, 1997a; 34, Makarkin, 1990c; 35, Makarkin, 1999; 36, Martins-Neto and Vulcano, 1989, 37, Martins-Neto and Vulcano, 1997; 38, Martins-Neto, 1997; 39, Martins-Neto, 1994; 40, Martins-Neto and Vulcano, 1990; 41, Whalley, 1980; 42, Jarzembowski, 1984; 43, Jarzembowski, 2001; 44, Jepson et al., in preparation; 45, Ross and Jarzembowski, 1993;46Willmann, 1993; 47, Jikun and Youchong, 1990; 48, Jarzembowski, 1993; 49, Whalley, 1988; 50, Lambkin, 1994; 51, Engel, 2005; 52, Ren et al., 1995; 53, Nel and Henrortay, 1994; 54, Ren and Oswald, 2002; 55, Ren and Yin, 2003; 56, Ponomarenko, 1984; 57, Lambkin, 1988; 58, Ponomarenko and Scherbakov, 2004; 59, Novokschonov and Novokschonov, 1997. 116 J.E. Jepson, D. Penney / Palaeogeography, Palaeoclimatology, Palaeoecology 248 (2007) 109–118 research project being undertaken by the first author on Engel, M.S., 2004b. Thorny lacewings (Neuroptera: Rhachiberothi- Wealden Neuroptera, during which the predictions pre- dae) in Cretaceous amber from Myanmar. Journal of Systematic Palaeontology 2, 137–140. sented here will be evaluated. Such predictive techni- Engel, M.S., 2004c. Arthropods in Mexican amber. Biodiversidad, ques may prove useful for other insect groups and/or taxonomía y biogeografía de artrópodos de México: hacia una fossil localities. síntesis de su conocimiento, vol. 4. Universidad Nacional Autónoma de México, México, pp. 175–186. Acknowledgements Engel, M.S., 2005. A remarkable kalligrammatid lacewing from the Upper Jurassic of Kazakhstan (Neuroptera: Kalligramma- tidae). Transactions of the Kansas Academy of Science 108, J.J. thanks Dr. J.R. Nudds (University of Manchester), 59–62. Dr. D.V. Logunov (Manchester Museum) and Karl Bates Grimaldi, D., Engel, M.S., 2005. Evolution of the insects. Cambridge (University of Manchester) for the discussion. We thank University Press, New York. three anonymous reviewers for their comments. Haywood, A.M., Valdes, P.J., Markwick, P.J., 2005. Cretaceous (Wealden) climates: a modelling perspective. Cretaceous Research 25, 303–311. References Jarzembowski, E.A., 1980. Fossil insects from the Bembridge Marls, of the Isle of Wight, southern England. Bulletin of the Allen, P., 1981. Pursuit of Wealden models. Journal of the Geological British Museum of Natural History (Geology) 33, 237–293. Society (London) 138, 375–405. Jarzembowski, E.A., 1984. Early Cretaceous insects from southern Allen, P., 1989. Wealden research-ways ahead. Proceedings of the England. Modern Geology 9, 71–93. Geologists' Association 100, 529–564. Jarzembowski, E.A., 1993. A provisional checklist of the fossil insects Allen, P., 1998. Purbeck – Wealden (early Cretaceous) climates. from the Purbeck Beds of Dorset. Proceedings of the Dorset Proceedings of the Geologists' Association 109, 197–236. Natural History and Archaelogical Society 114, 175–179. Allen, P., Wimbledon, W.A., 1991. Correlation of NW European Jarzembowski, E.A., 1995. Early Cretaceous insect faunas and Purbeck–Wealden (non-marine Lower Cretaceous) as seen from palaeoenvironment. Cretaceous Research 16, 681–693. the English type-areas. Cretaceous Research 12, 511–526. Jarzembowski, E.A., 2001. A new Wealden fossil Lacewing. In: Archibald, S.B., Makarkin, V.N., 2004. New genus of minute Rowlands, J.L.J. (Ed.), Tunbridge Wells and Rusthall Commons: A Berothidae (Neuroptera) from the early Eocene amber of British History and Natural History. Tunbridge Wells Museum and Art Columbia. The Canadian Entomologist 136, 61–76. Gallery, Tunbridge Wells, pp. 56–58. Archibald, S.B., Makarkin, V.N., 2006. Tertiary giant lacewings Jarzembowski, E.A., 2003. ‘Burnt’ beetles (Insecta: Coleoptera) from (Neuroptera: Polystoechotidae): revision and description of new the Wealden of southern England. Acta Zoologica 46, 139–145 taxa from western North American and Denmark. Journal of Syste- (suppl-Fossil Insects). matic Palaetontology 4, 119–155. Jepson, J.E., Heads, S.W., Ren, D., Makarkin, V.N., in preparation. Aspöck, U., 2002. Phylogeny of the Neuropterida (Insecta: Holome- First Mesozoic Mantispids Possessing Bodies (Insecta, Neurop- tabola). Zoologica Scripta 31, 51–55. tera, Mantispidae) from China. Aspöck, H., Aspöck, U., Hölzel, H., 1980. Die Neuropteren Europas I Jikun, Y., Youchong, H., 1990. Drakochrysa, an early Cretaceous new and II. Goecke and Evers, Krefeld. genus of Chrysopidae (Insecta: Neuroptera) from Laiyang Basin, Azar, D., Nel, A., Solignac, M., 2000. A new Coniopterygidae from Shandong Province. Geoscience 4, 15–26 (In Chinese, English Lebanese amber. Acta Geologica Hispanica 35, 31–36. Summary). Batten, D.J., 1998. Palaeoenvironmental implications of plant, insect Kukalova-Peck, J., 1991. Fossil History and the Evolution of Hexapod and other organic-walled microfossils in the Weald Clay Formation structures. In: Naumann, I.D., Carne, P.B., Lawrence, J.F., Nielson, (Lower Cretaceous) of southeast England. Cretaceous Research E.S., Spradberry, J.P., Taylor, R.W., Whitten, M.J., Littlejohn, M.J. 19, 279–315. (Eds.), Second Edition. The Insects of Australia: A Textbook for Carpenter, F.M., 1943. Osmylidae of the Florissant Shales, Colorado Students and Research Workers, vol. 1. Melbourne University (Insecta–Neuroptera). American Journal of Science 241, 753–760. Press, Carlton, Victoria, pp. 141–179. Carpenter, F.M., 1959. Fossil Nemopteridae (Neuroptera). Psyche 66, Labandeira, C.C., 2005. The fossil record of insect extinction: new 20–24. approaches and future directions. American Entomologist 51, 14–29. Cockerell, T.D.A., 1914. New and little known insects from the Labandeira, C.C., Sepkoski Jr., J.J., 1993. Insect diversity in the fossil Miocene of Florissant, Colorado. Journal of Geology 22, 714–724. record. Science 261, 310–315. Elstner, F., Mutterlose, J., 1996. The lower Cretaceous (Berriasian and Lambkin, K.J., 1988. A re-examination of the Lithosmylidia Riek from Valanginian) in NW Germany. Cretaceous Research 17, 119–133. the Triassic of Queensland with notes on Mesozoic ‘osmylid-like’ Engel, M.S., 1999. The first fossil of a pleasing lacewing (Neuroptera: fossil Neuroptera (Insecta: Neuroptera). Memoirs of the Queens- Dilaridae). Proceedings of the Entomological Society of Washing- land Museum 25, 445–458. ton 101, 822–826. Lambkin, K.J., 1994. Palparites deichmuelleri Handlirsch from the Engel, M.S., 2002. A new Dustywing (Neuroptera: Coniopterygidae) Solnhofen Plattenkalk belongs to the Kalligrammatidae in amber from New Jersey, with a reassessment of (Insecta: Neuroptera). Palaeontologische Zeitschrift 68, 163–166. Glaesoconis in Neocomian amber from Lebanon. Journal of the Lane, A., Janis, C.M., Sepkoski Jr., J.J., 2005. Estimating paleodiver- Kansas Entomological Society 75, 38–42. sities: a test of the taxic and phylogenetic methods. Paleobiology Engel, M.S., 2004a. The Dustywings in Cretaceous Burmese amber 31, 21–34. (Insecta: Neuroptera: Coniopterygidae). Journal of Systematic Makarkin, V.N., 1990a. Baissoleon cretaceus gen. and sp. nov. Fossil Palaeontology 2, 133–136. Neuroptera of the Lower Cretaceous of Baisa, East Siberia. Part 2. J.E. Jepson, D. Penney / Palaeogeography, Palaeoclimatology, Palaeoecology 248 (2007) 109–118 117

Nymphitidae. Annales de la Société Entomologique de France (N.S.) Nel, A., Jarzembowski, E., 1997. New fossil Sisyridae and 26, 125–126. Nevrorthidae (Insecta: Neuroptera) from Eocene Baltic amber Makarkin, V.N., 1990b. New names for the Jurassic Neuroptera. and Upper Miocene of France. European Journal of Entomology Paleontologicheskii Zhurnal 1, 120 (In Russian). 94, 287–294. Makarkin, V.N., 1990c. A new fossil genus and species of Osmylidae Nel, A., Menier, J.-J., Waller, A., Hodebert, G., De Ploeg, G., 2003. New from the Lower Cretaceous of East Siberia (Neuroptera). Deutsche fossil spongilla-flies from the lowermost Eocene amber of France Entomologische Zeitschrift, Berlin (N.F.) 37, 101–103. (Insecta, Neuroptera, Sisyridae). Geodiversitas 25, 109–117. Makarkin, V.N., 1994. Upper Cretaceous Neuroptera from Russia and Nel, A., Perrichot, V., Azar, D., Neuraudeau, D., 2005a. New Kazakhstan. Annales de la Société Entomologique de France (N.S.) Rhachiberothidae (Insecta: Neuroptera) in early cretaceous and 30, 283–292. early Eocene from France and Lebanon. Neues Jahrbuch Makarkin, V.N., 1996. Fossil Neuroptera of the Lower Cretaceous of für Geologie und Palaeontologie 235, 51–85. Baisa, East Siberia. Part 5. Mantispidae. Russian Entonological Nel, A., Perrichot, V., Azar, D., 2005b. New and poorly known fossil Journal 5, 91–93. Coniopterygidae in Cretaceous and Cenozoic ambers (Insecta: Makarkin, V.N., 1997a. Fossil Neuroptera of the Lower Cretaceous of Neuroptera). Annales Zoologici 55, 1–7. Baisa, East Siberia. Part 3. Chrysopidae (Insecta). Spixiana 20, New, T.R., 1989. Handbook of Zoology, A Natural History of the 107–118. Phyla of the Animal Kingdom. Volume IV Arthropoda: Insecta Makarkin, V.N., 1997b. Fossil Neuroptera of the Lower Cretaceous of Planipennia Lacewings. Walter de Gruyter, Berlin, New York. Baisa, East Siberia. Part 4. Psychopsidae. Beiträge zur Entomo- New, T.R., 1991. Neuroptera (lacewings). In: Naumann, I.D., Carne, P.B., logie 47, 489–492. Lawrence, J.F., Nielson, E.S., Spradberry, J.P., Taylor, R.W., Whitten, Makarkin, V.N., 1998. New Tertiary Neuroptera (Insecta) from the M.J., Littlejohn, M.J. (Eds.), Second Edition. The Insects of Australia: Russian Far East. Tertiary Research 18, 77–83. A textbook for students and research workers, vol. 1. Melbourne Makarkin, V.N., 1999. Fossil Neuroptera from the Lower Cretaceous University Press, Carlton, Victoria, pp. 525–542. of Baisa, East Siberia. Part 6. Mesithonidae (Insecta). Neues Novokschonov, V., Novokschonov, E., 1997. Okolopania favorablis Jahrbuch für Geologie und Paläontologie 12, 705–712. n.sp. (Planipennia; Neuroptera; Permithonidae) aus dem unteren Makarkin, V.N., Archibald, S.B., 2003. Family affinity of the genus Perm des Urals. Palaontologische Zeitschrift 71, 89–90. Palaeopsychops Anderson with description of a new species from Peñalver, E., Nel, A., Martínez-Delclõs, X., 1995. New Notochrysinae the early Eocene of British Columbia, Canada (Neuroptera: from the Spanish Miocene (Neuroptera, Chrysopidae). Bulletin de Polystoechotidae). Annals of the Entomological Society of America la Société entomologique de France 100, 481–487. 96, 171–180. Penney, D., 2004a. Estimations of global extant family richness Makarkin, V.N., Archibald, S.B., Oswald, J.D., 2003. New early Eocene using neontological and palaeontological data. In: Logunov, D.V., brown lacewings (Neuroptera: Hemerobiidae) from western North Penney, D. (Eds.), Euroean Arachnology 2003. Proceedings of the America. The Canadian Entomologist 135, 637–653. 21st European Colloquium of Arachnology (St. Petersburg, Russia, Martins-Neto, R.G., 1994. Neurópterous (Insecta: Planipennia) da August 2003). KMK Scientific Press, Moscow, pp. 225–234. formação Santana (Cretáceo Inferior) Bacia do Araripe, Nordeste do Penney, D., 2004b. Does the fossil record of spiders track that of their Brasil. IX - Primeiros resultados da composicao da fauna e ddescricao principal prey, the insects? Transactions of the Royal Society of de novos taxons. Acta Geologica Leopoldensia 17, 269–288. Edinburgh, Earth Sciences 94, 275–281. Martins-Neto, R.G., 1997. Neurópterous (Insecta: Planipennia) da Penney, D., 2005. Importance of Dominican Republic Amber for formação Santana (Cretáceo Inferior) Bacia do Araripe, Nordeste determining taxonomic bias of fossil resin preservation—a case do Brasil. X - Descricao de novos taxons (Chrysopidae, Babin- study of spiders. Palaeogeography, Palaeoclimatology, Palaeoe- skaiidae, Myrmeleontidae, Ascalaphidae e Psychopsidae). Cien- cology 223, 1–8. cias Exatas e Tecnologicas 2, 68–83. Pérez-Gelabert, D.E., Flint Jr., O.S., 2001. Annotated list of the Neu- Martins-Neto, R.G., 2000. Remarks on the neuropterofauna (Insecta, roptera of Hispaniola with new faunatic records of some species. Neuroptera) from the Brazilian Cretaceous, with keys for the Journal of Neuropterology 3, 9–23. identification of known taxa. Acta Hispanica 35, 97–118. Plant, C.W., 1997. A key to the adults of British lacewings and their Martins-Neto, R.G., Vulcano, M.A., 1989. Neurópterous (Insecta: allies (Neuroptera, Megaloptera, Raphidoptera and Mecoptera. Planipennia) da formação Santana (Cretáceo Inferior) Bacia do Field Studies 9, 179–269. Araripe, Nordeste do Brasil. IV Complemento partes I & II, com Ponomarenko, A.G., 1984. Neuroptera from the Jurassic in Eastern descrição de novos taxa. Anais da Academia Brasileira de Ciencias Asia. Paleontological Journal 18, 59–69. 61, 312–318. Ponomarenko, A.G., 1992. Neuroptera (Insecta) from the Lower Martins-Neto, R.G., Vulcano, M.A., 1990. Neurópterous (Insecta: Cretaceous of Transbaykalia. Paleontological Journal 26, 56–66. Planipennia) da formação Santana (Cretáceo Inferior) Bacia do Ponomarenko, A.G., 2002. Superorder Myrmeleontidea Latreille, 1802 Araripe, Nordeste do Brasil. III - Superfamilia Mantispoidea. (=Neuropteroidea Handlirsch, 1903). In: Rasnitsyn, A.P., Quicke, Revista Brasileira da Entomologia 34, 619–625. D.L.J. (Eds.), History of Insects. Kluwer Academic Publishers, Martins-Neto, R.G., Vulcano, M.A., 1997. Neurópterous (Insecta: Dordrecht, pp. 176–192. Planipennia) da formação Santana (Cretáceo Inferior) Bacia do Ponomarenko, A.G., Scherbakov, D.E., 2004. New lacewings Araripe, Nordeste do Brasil. VIII - Descricao de novos taxa de Neuroptera from the Terminal Permian and Basal Triassic of Myrmeleontidae, Ascalaphidae e Nemopteridae. Ciencias Biolo- Siberia. Paleontological Journal 38, 197–203. gicas e da Saude 2, 64–81. Radley, J.D., 2005. Derived fossils in the southern English Wealden Nel, A., Henrortay, M., 1994. Les Chrysopidae Mésozoiques. État actual (non-marine early Cretaceous): a review. Cretaceous Research 26, des conaissances. Description d’un nouveau genre et nouvelle espèce 657–664. dans le Jurassique Inférior (Lias) (Insecta: Neuroptera). Annales de la Rasnitsyn, A.P., Ross, A.J., 2000. A preliminary list of arthropod Société Entomologique de France (N.S.) 30, 293–318. families present in Burmese amber collection at the Natural 118 J.E. Jepson, D. Penney / Palaeogeography, Palaeoclimatology, Palaeoecology 248 (2007) 109–118

History Museum, London. Bulletin of the Natural History Museum genographischer Methoden. In: Gepp, J., Aspöck, H., Hölzel, H. London (Geology) 56, 21–24. (Eds.), Proceedings of the 1st International Symposium on Neuro- Rasnitsyn, A.P., Jarzembowski, E.A., Ross, A.J., 1998. Wasps pterology in Graz (), pp. 49–55. (Insecta: Vespida=Hymenoptera) from the Purbeck and Wealden Sepkoski Jr., J.J., 1979. A kinetic model of Phanerozoic taxonomic (Lower Cretaceous) of southern England and their biostratigraphi- diversity II. Early Phanerozoic families and multiple equilibria. cal and palaeoenvironmental significance. Cretaceous Research 19, Paleobiology 5, 222–251. 329–391. Sinitshenkova, N.D., 1998. The first European Cretaceous stonefly Ren, D., Guo, Z.G., 1996. On the new fossil genera and species of (Insecta, Perlida=Plecoptera). Cretaceous Research 19, 317–321. Neuroptera (Insecta) from the late Jurassic of northeast China. Acta Smith, A.B., 1994. Systematics and the Fossil Record: Documenting Zootaxonomica Sinica 21, 461–479. Evolutionary Patterns. Blackwell Science, Oxford. Ren, D., Oswald, J.D., 2002. A new genus of kalligrammatid lacewing Twitchett, R.J., 1994. A new Lower Cretaceous insect fauna from the from the middle Jurassic of China. Stuttgarter Beiträge zur Vectis Formation (Wealden Group) of the Isle of Wight. Naturkunde Serie B (Geologie und Paläontologie) 317, 1–8. Proceedings of the Geologists' Association 106, 47–51. Ren, D., Yin, J., 2003. New ‘osmylid-like’ fossil Neuroptera from the Watson, J., Alvin, K.L., 1996. An English Wealden floral list, with middle Jurassic of Inner Mongolia, China. Journal of the New York comments on possible environmental indicators. Cretaceous Re- Entomological Society 111, 1–11. search 17, 5–26. Ren, D., Lu, L.W., Guo, Z.G., Ji, S.A., 1995. Faunae and stratigraphy Weitschat, W., Wichard, W., 2002. Atlas of Plants and Animals in of Jurassic–Cretaceous in Beijing and the adjacent areas. Seismic Baltic Amber. Verlag Dr Friedrich Pfiel, Munich. Publishing House, Beijing. (in Chinese, English Summary). Whalley, P.E.S., 1980. Neuroptera (Insecta) in amber from the Lower Ross, A.J., Jarzembowski, E.A., 1993. Arthropoda (; Cretaceous of Lebanon. Bulletin of the British Museum of Natural Insecta). In: Benton, M.S. (Ed.), The Fossil Record, vol. 2. History (Geology) 33, 157–164. Chapman & Hall, London, pp. 363–426. Whalley, P.E.S., 1988. Mesozoic Neuroptera and Raphidoptera (Insecta) Schlüter, T., 1984. Phylogeny of Chrysopidae. In: Canard, M., in Britain. Bulletin of the British Museum of Natural History Semeiria, Y., New, T.R. (Eds.), Biology of Chrysopidae. Dr. W. (Geology) 44, 45–63. Junk Publishers, The Hague, pp. 1–8. Willmann, R., 1993. Insekten aus der Fur-Formation von Dänemark Schlüter, T., 2003. Fossil insects in Gondwana-localities and diversity (Moler, ob. Paleozän/unt. Eozän?). 8. Zwei neue Vertreter der Chry- trends. Acta Zoologica Cracoviensia 46, 345–371 (supplement- sopidae (Neuroptera). Neues Jahrbuch für Geologie und Paläonto- Fossil Insects). logie 4, 239–245. Schlüter, T., Sturmer, W., 1984. Die identifikationeiner fossilen Willmann, R., Brooks, S.J., 1991. Insekten aus der Fur-Formation von Rhachiberothinae-Art (Planipennia: Berothidae oder Mantispidae) Dänemark (Moler, ob. Paleozän/unt. Eozän?). 6. Chrysopidae aus mittelkretazischem Bernstein NW-Frankreichs mit Hilfe ront- (Neuroptera). Meyniana 43, 125–135. CHAPTER 7

3. NEW LACEWINGS (INSECTA: NEUROPTERA) FROM THE LOWER

CRETACEOUS WEALDEN SUPERGROUP OF SOUTHERN ENGLAND

114

Cretaceous Research 30 (2009) 1325–1338

Contents lists available at ScienceDirect

Cretaceous Research

journal homepage: www.elsevier.com/locate/CretRes

New lacewings (Insecta: Neuroptera) from the Lower Cretaceous Wealden supergroup of Southern England

James E. Jepson a,*, Vladimir N. Makarkin b, Edmund A. Jarzembowski c a School of Earth, Atmospheric and Environmental Sciences, University of Manchester, Williamson Building, Oxford Road, Manchester, M13 9PL, UK b Institute of Biology and Soil Sciences, Far Eastern Branch of the Russian Academy of Sciences, Vladivostok 690022, Russian Federation c Maidstone Museum and Bentlif Art Gallery, St Faith’s Street, Maidstone, Kent ME14 1LH, UK article info abstract

Article history: Eight new genera and thirteen new species of lacewings (Neuroptera) are described from the Lower Received 5 February 2009 Cretaceous Wealden Supergroup, Weald Sub-basin: Principiala rudgwickensis sp. nov. (Ithonidae), Sten- Accepted in revised form 29 July 2009 omylina medialis gen. et sp. nov., Protosmylina bifasciata gen. et sp. nov. (Osmylidae), Cretapsychops corami Available online 5 August 2009 gen. et sp. nov., Micropsychops parallelus gen. et sp. nov., Valdipsychops logunovi gen. et sp. nov., V. brigidae gen. et sp. nov., V. maculosus gen. et sp. nov., V. proudlovei gen. et sp. nov., V. minimus gen. et sp. nov., Keywords: Psychopsites rolandi gen. et sp. nov. (Psychopsidae), Actinophlebioides valdensis gen. et sp. nov. and Pro- Ithonidae tohemerobius perexiguus gen. et sp. nov. (Neuroptera familia incertae sedis). Psychopsidae are shown to Kalligrammatidae Neuropterida dominate the neuropteran fauna, while the presence of Brongniartiellidae has not been confirmed in the Osmylidae Wealden, due to previously thought representatives of this family being in fact psychopsids. This is the Psychopsidae first fossil record of Ithonidae and Osmylidae in Britain. The range of Kalligrammatidae has been Weald Sub-basin extended in the Cretaceous of Britain from the mid-Valanginian to early Barremian. Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction been found in the Triassic (Tillyard, 1933), Jurassic (Whalley, 1988), Cretaceous (e.g. Jarzembowski, 1984,1993) and Palaeogene Neuroptera (lacewings) is one of three orders that make up the (Jarzembowski, 1980). superorder Neuropterida (Insecta: Holometabola) (Grimaldi and Engel, 2005) and consists of approximately 6000 described extant species in 17 families (Aspo¨ck, 2002). They are widespread globally, 2. Geological setting but tend to be most abundant in the tropics, with several families having restricted distributions (New, 1991). Sixty-six extant species The Lower Cretaceous Wealden Supergroup (c. 140–125 Ma: Gradstein et al., 2005) of southern England is subdivided into the from six families: Chrysopidae (green lacewings), Coniopterygidae (wax flies), Hemerobiidae (brown lacewings), Myrmeleontidae (ant Weald and Wessex Sub-basins (Allen and Wimbledon, 1991). The non-marine Weald Sub-basin has yielded the most insect remains lions), Osmylidae and Sisyridae (spongillaflies) occur in Britain (Plant, 1997). Prior to this study, the presence of three psychopsoid and it is from this sub-basin that the specimens described herein have been collected (Fig. 1). The Weald Sub-basin consists of the neuropteran families were recorded in the Wealden: Psychopsidae (silky lacewings), and the extinct Brongniartiellidae and Kalli- Hastings Beds Group (Ashdown Beds, Wadhurst Clay, Lower and Upper Tunbridge Wells Sand, and the Grinstead Clay formations) grammatidae (Jarzembowski, 1984; Ross and Cook, 1995) and one species of kalligrammatid has been formally described: Kalli- and the Weald Clay Group (Upper and Lower Weald Clay forma- tions) (Allen and Wimbledon, 1991; Rasnitsyn et al., 1998). The gramma roycrowsoni (Jarzembowski, 2001). Neuroptera have a long and diverse fossil record from the majority of insects and the specimens described herein are from the Weald Clay Group, which are dated as Hauterivian–Aptian? (Allen Permian to Recent with all the major family radiations having occurred prior to the Lower Cretaceous (Ponomarenko, 2002; and Wimbledon, 1991; Rasnitsyn et al., 1998). The Weald Clays Grimaldi and Engel, 2005). In Britain, fossil Neuroptera have represent a lacustrine–lagoonal environment (Allen, 1981,1989; Radley, 2005). The insect orders that have so far been recorded in the Weald Sub-basin are Odonata, ‘‘Blattaria’’, Orthoptera, Hemi- ptera, Neuroptera, Coleoptera, Mecoptera, Diptera, Trichoptera, * Corresponding author. Hymenoptera, Isoptera, Raphidioptera (Jarzembowski, 1984,1995) E-mail address: [email protected] (J.E. Jepson). and Plecoptera (Sinitshenkova, 1998). The insects are preserved as

0195-6671/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.cretres.2009.07.012 1326 J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338

Fig. 1. Geological map of the Wealden Supergroup, southeast England, with locations (brickworks) where Neuroptera have been found marked on. (Redrawn and modified from Gallois, 1978; Ross and Cook, 1995; Rasnitsyn et al., 1998). uncrushed and unoxidized, disarticulated skeletal material, with Oswald (1993a) and Archibald & Makarkin (2006). Wing vein isolated wings being the most common (Jarzembowski, 1984). They abbreviations are as follows: C, costa; Sc, subcosta; R, radius; R1, are found in the calcareous siltstone developments or phosphatic first branch of R; Rs, radial sector; Rs1, most proximal branch of Rs; or sideritic concretions in the clay formations. The climate has been M, media; MA, media anterior; MP, media posterior; Cu, cubitus; interpreted as subtropical/warm temperate with seasonal rainfall CuA, cubitus anterior; CuP, cubitus posterior; 1A–3A, first to third (Allen, 1998; Haywood et al., 2005). The occurrence of wet and dry anal veins. (commonly with wild fires) seasons has been proposed, as has year- round precipitation with little evidence for prolonged drought but 4. Systematic palaeontology high evaporation (Batten, 1998; Haywood et al., 2005). Wildfire evidence exists in the form of fusainized coleopteran elytra and Order: Neuroptera Linnaeus, 1758 plant remains (Watson and Alvin, 1996; Batten, 1998; Jarzembow- Family: Ithonidae Newman, 1853 (¼ Rapismatidae Kru¨ ger, 1922) ski, 2003). Genus Principiala Makarkin et Menon, 2007 Principiala rudgwickensis sp. nov. 3. Material and methods Fig. 2 Derivation of name. After the Rudgwick Brickworks locality The neuropteran material described herein was collected from where the specimen was collected. the Weald Clay Group in various brickworks: Clockhouse (Lower Weald Clay Formation, Hauterivian), Auclaye, Rudgwick and Material. Holotype BMB 025005, -6 (part and counterpart), an Smokejacks (Upper Weald Clay Formation, Barremian) (locations incomplete forewing. Rudgwick Brickworks, Upper Weald Clay, labelled in fig. 1). The material is deposited in the following insti- Lower Cretaceous, Barremian. tutions: Booth Museum of Natural History, Brighton (BMB), Natural Diagnosis. May be distinguished from Principiala incerta by the History Museum, London (BMNH Pal.) and Maidstone Museum and following forewing character states [alternative character states are Bentlif Art Gallery, Kent (MNEMG). given in brackets]: most proximal branch of Rs terminated on MA The material was studied using an Olympus SZH stereomicro- long [very short in P. incerta]; numerous crossveins between scope with a camera lucida attachment for producing drawings. proximal 5 branches of Rs: some veins having a ‘zig-zag’ appear- Some drawings are composites, merging the part and counterpart ance [not detected in P. incerta]; proximal 6 fused branches of Rs are to establish a more complete venation; this was achieved using straight [curved in P. incerta]. a light box. 10% ethanol was used to enhance vein detail. Photo- graphs were taken using a Nikon D1X digital SLR camera mounted Description. Forewing shape indeterminate, preserved wing on a rostrum stand, except Psychopsites rolandi gen. et sp. nov. length 9.4 mm (estimated length about 11–12 mm). C not preserved. which was drawn by EAJ using a camera lucida attachment on Few subcostal veinlets preserved. Gradate series of crossveins in a M4A Wild microscope and photographed by Tony Mitchell costal space partially preserved. Sc partially preserved appearing to (volunteer Maidstone Museum) using a Nikon Coolpix 5000 digital fuse with R1 distally: these veins at least, closely approach each other camera. towards wing apex. 5 subcostal crossveins preserved. R incomplete, We follow the traditional (sensu Wootton, 2003) venational origin and termination not preserved. 2 r-rs crossveins preserved. Rs terminology of Comstock (1918) with the recent interpretation of with 17 pectinate branches. Rs1 fuses with MA at approximately half J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338 1327

was recently discovered in the Lower Eocene of North America (Makarkin and Archibald, 2009). There are no ithonids in Britain today (Plant, 1997); extant ithonids are distributed in Australia, south-western North America to Honduras, and in Oriental regions, some living in forests in high altitude (900–2595 m) and some in desert and sandy areas (Makarkin and Menon, 2007). Family: Osmylidae Leach, 1815 Genus Stenosmylina gen. nov. Type and only species. Stenosmylina medialis gen. et sp. nov. Derivation of name. From Greek stenos narrow, and the genus Osmylus, in reference to narrow forewing of the type species. Gender feminine. Diagnosis. Forewing separable from those of other osmylid genera by the following combination of characters: costal space narrow; cross venation rather sparse; M forks very close to poste- rior wing margin, simple for most of length; CuA few-branched; CuP multi-branched, with more than ten pectinate branches. Remarks. The character state M being simple occurs very rarely in osmylids. Two Australian species represent the monotypic subfamilies Porisminae and Eidoporisminae, they also possess an M which forks very close to posterior wing margin (see New, 1983: figs. 1–4), but their venations are very different in other characters. The forewing of Stenosmylina gen. nov. most resembles that of Mesosmylina Bode,1953, known from three species from the Lower to Middle Jurassic of Germany and Siberia (Ponomarenko, 1984,1996; Ansorge, 1996). These two genera share similar size and shape, narrow costal space, and sparse cross-venation, but they are easily distinguished from each other by the configuration of CuP (few- branched in Mesosmylina), besides the different structure of M. Stenosmylina medialis gen. et sp. nov. Fig. 2. Principiala rudgwickensis sp. nov., holotype BMB 025005, -6. A, composite drawing. B, photograph of the same specimen; scale bar represents 1 mm. Fig. 3 Derivation of name. From median vein, in reference to its pecu- liar configuration in this species. MA length; other proximal 6 Rs branches fuse to form 1 vein, numerous crossveins between them forming polygonal cells. Material. Holotype BMB 018382, -3 (part and counterpart), an Majority of other Rs branches are simple, with few forked. 6 incomplete forewing. Clockhouse Brickworks, Lower Weald Clay, preserved r-ma crossveins. Origin of MA not preserved. MA long, Lower Cretaceous, Hauterivian. reaching wing base distad of midpoint of wing, forked two-thirds along its length. 5 ma-mp crossveins at midpoint of MA detected. MP origin not preserved. MP with posterior pectination of anterior branch and anterior pectination of posterior branch, branches mainly with marginal forking, and with numerous crossveins between them. Cu poorly preserved. CuA incomplete, pectinately branched distally. CuP, anal veins not preserved. Remarks. Hitherto, only one fossil ithonid has been described, Principiala incerta Makarkin et Menon, 2007 from the Lower Cretaceous Crato Formation, Brazil (Makarkin and Menon, 2007). The general wing structure of Principiala rudgwickensis sp. nov. is very similar to that of P. incerta, so that the assignment of this new species to the genus Principiala seems to be undoubted despite its fragmentary condition. Both species share some important char- acters in the forewing, including the fusion of the proximal branches of Rs with MA and each other, the peculiar configuration of MP, narrow space between R1 and Rs, and the presence of numerous costal and subcostal crossveins. The ithonid affinity of the genus is based on a set of distinctive head and thoracic char- acters, as well as general ithonid features of the venation found in the type species. Principiala rudgwickensis sp. nov. is the oldest ithonid formally described, pre-dating P. incerta from the younger Crato Formation (Upper Aptian). A third fossil species of Ithonidae, Fig. 3. Stenosmylina medialis gen. et sp. nov., holotype BMB 018382, -3. A, composite having similar venation and belonging to the same species group, drawing. B, photograph of the same specimen; scale bar represents 1 mm. 1328 J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338

Diagnosis. As for genus. Description. Forewing 13 mm preserved length, 4 mm preserved width (estimated length about 15–16 mm). Costal space narrow, basal subcostal veinlets, all simple. Sc poorly preserved, appears very close to R1. Numerous (preserved), regularly spaced crossveins between R1 and Rs. 5 Rs branches preserved. Crossveins in radial space rather sparse, not forming distinct series. M not branched, simple for most of length, forking 2 mm from posterior wing margin. 11 crossveins between M and CuA. Origin of Cu not detected, close to wing base; CuA simple for most of length, branched at two-thirds of its length, with 6 pectinate branches. 9 crossveins preserved between CuA and CuP. CuP pectinately branched with 14 branches. 1 crossvein preserved between CuP and 1A. 1A pectinately branched with 3 branches. 2A with 1 branch. Trichosors present on posterior wing margin. Remarks. One species of osmylid (Osmylus fulvicephalus Scopoli) is present in Britain today (Plant, 1997). Genus Protosmylina gen. nov. Type and only species. Protosmylina bifasciata gen. et sp. nov. Derivation of name. From the Greek protou, before, and the genus Osmylus, in reference to probable protosmyline subfamilial affinity of the genus. Gender feminine. Diagnosis. Forewing venation is most similar to that of proto- smyline genera, but may be distinguished from these by branches of Rs more numerous, closer spaced, more distal position of fork of M. Remarks. The forewing venation of Protosmylina is particularly Fig. 4. Protosmylina bifasciata gen. et sp. nov., holotype BMB 018569, -70. A, composite similar to that of Protosmylus Kru¨ ger, 1913 from the upper Eocene drawing. B, photograph of the same specimen; scale bar represents 1 mm. Baltic amber (see Pictet-Baraban and Hagen, 1856: Pl. 8, fig. 16), the type genus of the subfamily Protosmylinae. They are united by into CuA and CuP. CuA incomplete, pectinately branched distally, 4 the following character states: Sc fused with R1 far distant from branches preserved. CuP incomplete. Anal veins poorly preserved. wing apex; subcostal crossveins (at least one) present; all cross- Wing pattern consists mainly of 2 transverse, interrupted fuscous veins in the radial space arranged in three gradate series; few, bands in middle and apical parts of wing. sparsely spaced crossveins between R1 and Rs; proximal crossvein between Rs1 and MA strongly oblique. It is probable that Proto- Family: Psychopsidae Handlirsch, 1906–1908 smylina belongs to Protosmylinae; if so, this is the oldest known Remarks. Psychopsidae is the only living family of the ‘psy- genus in the subfamily. chopsid-like neuropterans’ (psychopsoids) which include in Protosmylina bifasciata gen. et sp. nov. particular the extinct Osmylopsychopidae, Brongniartiellidae, and Fig. 4 Kalligrammatidae. Psychopsoids are particularly diverse and Derivation of name. From the Latin bi, double, and fascia, band, in abundant in the Mesozoic, but the of this group has been reference to the forewing colour pattern of the holotype. very poorly studied. The Mesozoic psychopsids are often hard to distinguish from members of Osmylopsychopidae and Brongniar- Material. Holotype BMB 018569, -70 (part and counterpart), an tiellidae, especially if the wings are incomplete or fragmentary. The incomplete forewing. Auclaye Brickworks, Upper Weald Clay, diagnoses and compositions of these families have not been well Lower Cretaceous, Barremian. understood until now. A current study of very well preserved Diagnosis. As for genus. Triassic osmylopsychopids from the Madygen Formation (Central Description. Forewing 9 mm preserved length (estimated length Asia) and the type species of Brongniartiellidae from the Jurassic about 11 mm), 3.5 mm preserved width. Costal space broad, with , show that Osmylopsychopidae and Brong- 16 simple, closely spaced subcostal veinlets preserved. Sc fused niartiellidae might be readily distinguished from the Mesozoic with R distally, far distant from wing apex. 1 sc-r crossvein Psychopsidae, and from this we can say that there is no definite preserved at mid-point of wing. 5 r1-rs crossveins. Rs originates evidence of these two families among the material examined here near base of wing, with 12 pectinate branches preserved. Majority (VM, pers. obs., see also Makarkin, in press). of Rs branches simple for most of length; Rs5 forks at midpoint of The forewings of extant Psychopsidae are characterized by the vein, a few veins forked distally. 3 gradate series of crossveins in following character states [based on the reviews and revisions of radio-medial space: 5 crossveins in most proximal series, 11 in Tillyard (1919), Kimmins (1939), Tjeder (1960), New (1989),and middle series, and 8 in most distal series. Basal crossvein between Oswald (1993b)]: wing broad with apex always broadly rounded; stems of Rs and M present; basal crossvein connecting Rs1 and MA trichosors always present; ‘‘vena triplica’’ present (i.e., a complex strongly oblique. M deeply forked far distal to origin of Rs1. MA with character formed by Sc, R1, and Rs running parallel to ‘‘pterostigmal 4 pectinate branches, some branches forked distally. MP with 2 region’’, then anastomized or possesing 2 distal crossveins between pectinate distal branches, of which proximal one deeply forked. 2 them); pterostigma absent; humeral veinlets usually slightly recur- crossveins within pectinate branches of MA and 3 crossveins rent, pectinately branched; costal space very broad throughout between MA and MP. Cu origin close to base of wing, deeply forked (width in ‘‘pterostigmal region’’ >2 times the combined width of the J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338 1329 adjacent subcostal and R1 spaces [1]); subcostal space with more than four crossveins (usually >10) [2]; Sc and R1 distally connected by a strong crossvein or anastomosing; R1 (or Sc þ R1) distally with few branches (if any), usually not pectinate [3]; branches of Rs mostly not forked before ‘‘pterostigmal region’’ and approximately parallel to the hind margin; crossveins in the radial space arranged in gradate series (mostly 2 or 3 series) [4]; M with two or more deep branches, none of them pectinately branched distally; CuA and CuP in most species pectinately branched [5]; 1A multi-branched. The character states [1] and [2] are considered by Oswald (1993b) as synapomorphies of the Psychopsidae. The characters [3]–[5] are also characteristic of all (or the majority of) extant species. The typical ‘‘vena triplica’’ as described above is often absent in the fossil taxa. Therefore, the forewing of Psychopsidae may be characterized by a combination of the character states [1]–[5]. All the species described below are represented by incomplete or fragmentary wings, unfortunately lacking their basal portion. Nonetheless, the majority of the wings possess the characters [1]–[4]. Therefore, it seems most probable that the genera described, despite their fragmentary nature, are of psychopsid affinity. The structure of vein CuP, although not entirely preserved in any specimen, appears to be non-pectinate, at least in some species. However, because of problems with the identification of veins MA, MP, CuA and CuP due to the lack of a basal portion to the wings, these veins are often determined provisionally.

Genus Cretapsychops gen. nov.

Type and only species. Cretapsychops corami gen. et sp. nov. Derivation of name. From Cretaceous and -psychops, a traditional ending of psychopsoid generic names. Gender masculine. Diagnosis. May be distinguished from other psychopsid genera by the combination of non-pectinate structure of both CuA and CuP in the forewing. Remarks. A similar structure of CuA is seen in Triassopsychops (Tillyard, 1922) and the non-pectinate CuP is observed in Baisopy- schops (Makarkin, 1997). Fig. 5. Cretapsychops corami gen. et sp. nov., holotype BMNH Pal. PI II 1555a; scale bar Cretapsychops corami gen. et sp. nov. represents 1 mm. Fig. 5 Genus Micropsychops gen. nov. Derivation of name. After Dr Robert A. Coram, British palaeoentomologist. Derivation of name. From Greek mikros, small, and -psychops, a traditional ending of the psychopsoid generic names. Gender Material. Holotype BMNH Pal. PI II 1555a, an incomplete fore- masculine. wing. Smokejacks Brickworks, Upper Weald Clay, Lower Creta- ceous, Barremian. Type and only species. Micropsychops parallelus gen. et sp. nov. Diagnosis. As for genus Diagnosis. May be distinguished from other psychopsid genera by having non-dichotomously branching (parallel for most of Description. Forewing 10.6 mm preserved length (estimated length) branches of Rs and presence of four (at least) transverse length about 15 mm), 8.2 mm preserved width. C not preserved. gradate series of crossveins in radial space. Costal space broad, over three times width of ‘‘vena triplica’’ (i.e., subcostal, R1 spaces together), with 29 subcostal veinlets mostly Remarks. The genus possesses the wing venation typical of forked; 1 complete gradate series of 21 crossveins in costal space, Psychopsidae (e.g., the broad costal space, the presence of ‘‘vena running from its base to mid-point. 2 subcostal crossveins triplica’’, the numerous subcostal crossveins, the gradate series of preserved. 4 r1-rs crossveins preserved. 20 pectinate branches of Rs crossveins in the radial space), and it has been confidently placed in preserved, mainly simple for most of their length, few forked at this family despite its small size. mid-point of vein. Few crossveins preserved forming 5 interrupted Micropsychops parallelus gen. et sp. nov. gradate series in radial to cubital spaces. M poorly preserved, with 2 Fig. 6 branches. CuA peculiarly (in general, dichotomously) branched. CuP Derivation of name. From Greek parallelos, parallel, in reference poorly preserved, probably dichotomously branched (possibly 4 to the branches of Rs parallel for most of length. branches forking terminally to form 9 branches). Anal veins poorly preserved, difficult to differentiate. Wing membrane mainly Material. Holotype BMNH Pal. PI II 1518a, b (part and counter- fuscous, paler basally and apically, with broad transverse pale patch part), an incomplete forewing. Smokejacks Brickworks, Upper at mid-point of wing. Weald Clay, Lower Cretaceous, Barremian. 1330 J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338

Description. Forewing 7.6 mm preserved length (estimated length Remarks. This genus differs from other psychopsids mainly in about 9–10 mm), 5 mm preserved width. C not preserved. 33 sub- having the combination of a strongly dichotomously branched Rs costal veinlets, some forked. ‘‘Vena triplica’’ formed by Sc, R1, Rs and the presence of several short gradate series of crossveins in the configured nearlyas in extant taxa being closed by 2 distal crossveins. radial space. The other probable psychopsid genera that possess 14 sc-r1 and 6 r1-rs crossveins preserved. R1 not branched, only with dichotomously branched branches of Rs are Triassopsychops Till- terminal fork. Rs pectinately branched, 17 branches, simple for most yard, 1922, and Angaropsychops Martynova, 1949; they differ from of lengths, forking distal to most distal gradate series. All crossveins Valdipsychops gen. nov. by having a pectinate CuP (Triassopsychops) arranged into 4 gradate series in radio-medial space. Origin of M not and having radial crossveins that are not arranged in series preserved; MA probably deeply forked, each branch forked distal to (Angaropsychops). Other fossil psychopsids, for example Prop- most distal gradate series of crossveins; MP pectinately branched sychopsis MacLeod, 1971 and Baisopsychops Makarkin, 1997, and distally, with 3 long branches. CuA pectinately branched with 3 long extant genera have mostly unbranched veins of Rs before end- branches preserved. Anal veins not preserved. twigging, and two or three long gradate series of crossveins in the radial space. Some other Mesozoic genera usually ascribed to Psy- Genus Valdipsychops gen. nov. chopsidae that have a dichotomously branched Rs are Grammap- Type species. Valdipsychops logunovi gen. et sp. nov. sychops Martynova, 1954, Embaneura Zalessky, 1953, Pulchroptionia Martins-Neto, 1997 and Kagapsychops Fujiyama, 1978; however, Derivation of name. From valdi-, the latinized name for the they may actually (partly confidently, partly very probably) belong Wealden, in reference to the Wealden Supergroup, and -psychops, to another psychopsoid family, Osmylopsychopidae. This is implied a traditional ending of the psychopsoid generic names. Gender from their forewing characters, e.g. the subcostal space is narrow masculine. and lacks subcostal crossveins, Sc and R1 are fused, Rs1 is multi- Species included. V. logunovi gen. et sp. nov., V. brigidae gen. et sp. branched; all these features are very similar to those of osmylop- nov., V. maculosus gen. et sp. nov., V. proudlovei gen. et sp. nov., V. sychopid genera (VM, ongoing research). The majority of the genera minimus gen. et sp. nov., Valdipsychops sp., all from the Lower ascribed to Brongniartiellidae also have a dichotomously branched Cretaceous, Hauterivian and Barremian, Wealden Supergroup. Rs. The examination of the type species of this family (Brongniar- tiella gigas Weyenbergh) shows that all forewing character states of Diagnosis. May be distinguished from other genera by the combi- the Valdipsychops gen. nov. diagnosis are present also in B. gigas. nation of the following character states: branches of Rs dichoto- However, other features found in B. gigas are not present in Val- mously branched, CuA pectinate, CuP non-pectinate, and several dipsychops gen. nov. (as well as in other psychopsids): the costal (often numerous)short gradate series of crossveins in the radial space. space is relatively narrow (width in the ‘‘pterostigmal region’’ is equal to the combined width of the adjacent subcostal and R1 spaces); both subcostal and R1 spaces are broad; the crossveins are arranged mainly in long gradate series running from the cubital to radial spaces nearly parallel to the wing margin, like the ‘outer’ gradate series of Hemerobiidae or Polystoechotidae (VM, pers. obs.). Therefore, the venation of Valdipsychops gen. nov. is superfi- cially most similar to that of osmylopsychopsid and brongniartiellid genera, but it is separable from them by the presence of many subcostal crossveins and a broad costal space. Valdipsychops logunovi gen. et sp. nov. Fig. 7 Material. Holotype BMNH Pal. PI II 1551a, b (part and counter- part), an incomplete forewing. Smokejacks Brickworks, Upper Weald Clay, Lower Cretaceous, Barremian. Derivation of name. After Dr Dmitri V. Logunov, curator of arthropods at The Manchester Museum.

Diagnosis. May be distinguished from other species of the genus by scarce gradate crossveins in the radial space, and characteristic colour pattern.

Description. Forewing 14 mm preserved length (estimated length about 17–19 mm), 11 mm preserved width. C poorly preserved. Costal space very broad, approximately four times width of ‘‘vena triplica’’, with 33 subcostal veinlets; costal crossveins not detected. Sc, R1, Rs parallel for most of length, forming ‘‘vena trip- lica’’. Sc terminating on R1 (or connecting with R1 by short cross- vein). R1 probably with only 1 distal branch. 10 subcostal crossveins, and 3 r1-rs crossveins preserved. Rs with 15 pectinate branches, dichotomously forked into 29 branches. Many Rs branches fork before mid-point of vein. Few preserved crossveins in radial space forming 3 interrupted gradate series. Origin of M not preserved. M probably forked at mid-point of vein (alternatively, forked near wing base, if apparent most proximal branch of Rs Fig. 6. Micropsychops parallelus gen. et sp. nov., holotype BMNH Pal. PI II 1518a. A, belongs actually to M). CuA with 6 pectinate branches, most drawing. B, photograph of the same specimen; scale bar represents 1 mm. proximal branches deeply forked. CuP poorly preserved, probably J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338 1331

Fig. 7. Valdipsychops logunovi gen. et sp. nov., holotype BMNH Pal. PI II 1551a. A, drawing. B, photograph of the same specimen; scale bar represents 1 mm.

Fig. 8. Valdipsychops brigidae gen. et sp. nov., holotype MNEMG 2009.36 [S. 1569]. A, not pectinate. Anal veins not preserved. Colour pattern mainly drawing. B, photograph of the same specimen; scale bar represents 1 mm. fuscous, paler horizontal bars at mid-point of wing. Valdipsychops brigidae gen. et sp. nov. of Rs actually belongs actually to M). Cu origin not preserved. CuA Fig. 8 curved, pectinately branched, with 7 branches preserved. CuP Material. Holotype MNEMG 2009.36 [S. 1569a, b] (part and apparently deeply forked, both branches curved. 1A partially counterpart), an incomplete ?forewing. Smokejacks Brickworks, preserved, pectinately branched with 6 long branches. Upper Weald Clay, Lower Cretaceous, Barremian. Valdipsychops maculosus gen. et sp. nov. Derivation of name. After Mrs J. Brigid E. Jarzembowski, Wealden Fig. 9 insect collector. Material. Holotype BMB 018575, -6 (part and counterpart), an Diagnosis. May be distinguished from other species of the genus incomplete forewing. Auclaye Brickworks, Upper Weald Clay, by having strongly curved branches in proximal parts of Rs, M, and Barremian. Cu. Derivation of name. From the Latin maculosus, spotted, in refer- Description. ?Forewing 12.9 mm preserved length (estimated ence to the forewing colour pattern. length about 20–22 mm), 10 mm preserved width. C partially Diagnosis. May be distinguished from the other species of the preserved. Costal space broad, approximately four times width of genus by the characteristic spotted forewing pattern. ‘‘vena triplica’’, with 30 subcostal veinlets preserved; 3 crossveins forming a gradate series are preserved in apical portion of wing Description. Forewing 12.9 mm preserved length (estimated connecting veinlets of R1. Sc terminating on R1. 10 subcostal length about 25 mm) and 8.4 mm preserved width. C partially crossveins preserved. R1 with several distal branches. 12 r1-rs preserved with trichosors present. Costal space broad, approxi- crossveins preserved. Rs pectinate with 17 branches which are mately four times the width of ‘‘vena triplica’’, and with 38 sub- strongly dichotomously branched (up to 35 terminal branches). costal veinlets preserved, many forked near anterior wing margin. Many crossveins preserved in radio-cubital space forming up to 8 Sc terminated on R1 (alternatively, these veins connected by rather gradate series. Proximal branches of Rs curved. Origin of M not long, oblique crossvein). 4 subcostal crossveins preserved. R1 preserved; M slightly curved, forked at mid-point of vein (alter- incomplete. Rs pectinately branched, with 13 branches, which are natively, forked near wing base, if apparent most proximal branch dichotomously branched. Few crossveins preserved in radial space, 1332 J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338 with 1 poorly developed potential gradate series. Origin and fork of with 15 branches preserved, dichotomously branched (at least M not preserved. Cu forked close to wing base. CuA pectinately apically). Many crossveins preserved in distal portion of costal to branched, with 4 branches preserved. CuP few-branched, probably radial spaces, arranged mainly into over 10 gradate series. Striped dichotomously branched. 1A few-branched. 2A well developed, colour pattern is preserved. M, Cu and A not preserved. pectinately branched. 3A not preserved. Colour pattern represented Remarks. This specimen has been tentatively placed in Valdip- by small circular spots (diameter of about 0.3 mm) in radial to sychops gen. nov. based on the dichotomously branched Rs and the cubital spaces (over ten in number). presence of the gradate series of crossveins in the radial space. It Remarks. The apparent absence of gradate series of crossveins in has been assigned to a new species based on the large size of the this species is possibly an artefact of preservation. specimen, which is more than twice the size of V. logunovi gen. et sp. nov. Valdipsychops proudlovei gen. et sp. nov. Fig. 10 Valdipsychops minimus gen. et sp. nov. 2003 Silky lacewing: Austen et al., p. 11, fig. 2. Fig. 11 1987 Psychopsoidea: Jarzembowski, fig. 9.5. Derivation of name. Named after the late Mr. William Proudlove, general manager, for facilitating fossil collecting at Smokejacks Derivation of name. From the Latin minimus, small, in reference Brickworks. to the small wing size of the holotype. Material. Holotype MNEMG 2009.37 [S.1923a, b] (part and Material. Holotype MNEMG 2009.38 [CH643a, b] (part and counterpart), a fragmentary distal portion of a wing. Smokejacks counterpart), an incomplete?forewing. Clockhouse Brickworks, Brickworks, Upper Weald Clay, Lower Cretaceous, Barremian. Lower Weald Clay, Lower Cretaceous, Hauterivian. Diagnosis. May be distinguished from the other species of the Diagnosis. May be distinguished from V. logunovi gen. et sp. nov., genus by the large size. V. brigidae gen. et sp. nov., and V. maculosus gen. et sp. nov. by having fewer branches of CuA; the smaller size excludes it from Description. 21.9 mm preserved length (estimated length about V. proudlovei gen. et sp. nov. 40 mm) and 14 mm preserved width. C not preserved. Subcostal veinlets in distal portion of costal space mainly branched. Sc not fused with R1, connected apically by long crossvein; other subcostal crossveins not detected. R1 with 4 distal veinlets. 7 crossveins between R and Rs preserved. Rs probably pectinately branched,

Fig. 9. Valdipsychops maculosus gen. et sp. nov., holotype BMB 018575. A, drawing. B, Fig. 10. Valdipsychops proudlovei gen. et sp. nov., holotype MNEMG 2009.37 [S.1923]. A, photograph of the same specimen; scale bar represents 1 mm. drawing. B, photograph of the same specimen; scale bar represents 1 mm. J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338 1333

Description. ?Forewing preserved length 12.5 mm (estimated V. minimus gen. et sp. nov. Unfortunately, in the latter species the length about 15 mm), preserved width 7 mm. C not preserved. 33 colour pattern is not preserved. subcostal veinlets preserved, some forked. 3 crossveins preserved Genus Psychopsites gen. nov. in costal space. Sc and R1 distally not fused, only slightly approaching each other towards wing apex. R1 with several ante- Derivation of name. Psychops- (from Psychopsis, a psychopsid riorly directed veinlets. 7 subcostal crossveins and 5 r1-rs cross- genus-group name), and -ites a traditional ending of fossil generic veins preserved. R1 space (between R1 and Rs) narrow. Rs names. In reference to the similar forewing colouration of the type pectinately branched with 14 branches, many dichotomously species of Psychopsis mimica Newman. Gender masculine. forked before mid-way point of vein. Numerous crossveins Type and only species. Psychopsites rolandi gen. et sp. nov. preserved in radial space, forming at least 5 gradate series. M forked near mid-point of vein to give 2 branches. CuA pectinately Diagnosis. Largest psychopsid (forewing of type species is branched with 3 branches preserved. CuP incompletely preserved, 45–50 mm long), which may be distinguished from other genera by with 2 branches. Anal veins not preserved. a combination of the forewing features: subcostal space very broad with densely spaced crossveins; Sc and R1 are distally fused; Valdipsychops sp. crossveins in radial space form >10 long gradate series. Fig. 12 Material. MNEMG 2009.39 [S.1387a, b] (part and counterpart), Remarks. Comparison with Recent Psychopsidae in BMNH shows a fragmentary wing. Smokejacks Brickworks, Upper Weald Clay, that the fossil cannot be assigned to a Recent genus. The fossil wing Lower Cretaceous, Barremian. is larger than any extant psychopsid. Of Recent species, Mega- psychops illedgei (Froggatt) from Australia has the largest forewing Description. Preserved length 9.9 mm (estimated length about which is only 30 mm. The fossil shows better developed gradate 20–22 mm), preserved width 9 mm. C, Sc and R not preserved. Rs series than any extant species. The colour pattern resembles that of with more than ten dichotomously forked branches. Numerous Psychopsis mimica Newman from Australia, but stripes are double in crossveins, arranged in approximately eight–ten gradate series. the latter. The affinities of Psychopsites gen. nov. are unclear Presumably M with two branches preserved. Presumably CuA pectinately branched with two branches preserved. Presumably CuP non-pectinately branched. Anal veins not preserved. Colour pattern preserved, mainly pale with light stripe thinning anteriorly. Remarks. Tentatively placed in the genus Valdipsychops gen. nov. on the basis of having dichotomously branched Rs and the presence of the several gradate series of crossveins in the radial space. The venation of this fragmentary wing is most similar to that of

Fig. 11. Valdipsychops minimus gen. et sp. nov., holotype MNEMG 2009.38 [CH643a]. A, Fig. 12. Valdipsychops sp. A, specimen MNEMG 2009.39 [S.1387a]. A, drawing. B, drawing. B, photograph of the same specimen; scale bar represents 1 mm. photograph of the same specimen; scale bar represents 1 mm. 1334 J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338 compared with other fossil psychopsids due to the incompleteness crossveins. Sc terminated on R1 distally; Sc þ R1 lacking (or at most of the wing. It is probably most closely related to two other with 1–2) veinlets. Rs and R1 apparently not fused distally, but Wealden genera, i.e. Valdipsychops gen. nov. and Cretapsychops gen. connected by oblique crossvein. Crossveins in R1 space poorly nov. developed. Proximal branches of Rs and M difficult to separate because wing base not preserved. Rs apparently with c. 15 Psychopsites rolandi gen. et sp. nov. branches, distal branches inclined to Rs at acute angle; Rs branches Fig. 13 dividing dichotomously. Crossveins in radial and medial spaces 1984 Psychopsidae: Jarzembowski, p. 82, fig. 32 form 12–14 long transverse gradate series. Colour pattern 1987 Psychopsidae: Jarzembowski, figs 9.9, 9.10 preserved consisting of three oblique, brown stripes; the middle 1988 Psychopsidae: Whalley, p. 61. one interrupted posteriorly. 2006 Psychopsidites rolandi Jarzembowski and Makarkin, p. 29 ?Psychopsites sp. (nomen nudum). Fig. 14 Derivation of name. After the late Prof. Roland Goldring, Material. BMB 016397, -8 (part and counterpart), wing fragment. University of Reading. Auclaye Brickworks, Upper Weald Clay, Barremian. Material. BMB 614916, -7 [CH34xa, b] (part and counterpart), an Description. Partially preserved wing fragment, 14 mm long incomplete ?right ?forewing. Clockhouse Brickworks, Lower Weald (estimated wing length over 40 mm). 4 parallel veins of Rs Clay, Lower Cretaceous, Hauterivian. preserved. Preserved part of MA simple, concave. MP pectinately forked distally with 2 branches preserved. CuA pectinately Diagnosis. As for genus. branched with 5 branches, first dichotomously branched. CuP Description. ?Forewing preserved length 33 mm (estimated pectinately branched, with at least 10 branches preserved. Cross- length about 45–50 mm), preserved width 24 mm. C partially veins cover wing throughout, mostly arranged in series. preserved along costal margin. Costal space relatively narrow (for Remarks. This specimen resembles the family Kalligrammatidae psychopsids), only two times wider than subcostal and R1 spaces due to the large size and numerous crossveins. However, the combined. Preserved subcostal veinlets mainly forked, not con- preserved venation and the crossveins arranged in numerous series nected by crossveins (except occasional one). Subcostal space very clearly show that it belongs to the Psychopsidae. This specimen broad, twice wider than R1 space, with numerous closely spaced may belong to Psychopsites rolandi gen. et sp. nov. The fragmentary

Fig. 13. Psychopsites rolandi gen. et sp. nov., holotype BMB 614916, -7. A, composite Fig. 14. ?Psychopsites sp., specimen BMB 016397. A, drawing. B, photograph of the drawing. B, photograph of the same specimen; scale bar represents 1 mm. same specimen; scale bar represents 1 mm. J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338 1335 nature of the specimen, however, does not allow us to associate Remarks. In general, Actinophlebioides gen. nov. might be them with certainty. assigned to some psychopsoid family, but its features do not fit confidently in any of them. The classification of psychopsoids is Family: Kalligrammatidae Handlirsch, 1906–1908 generally very poorly understood. Superficially, this genus most Kalligrammatidae gen. et sp. indet. closely resembles the genera usually treated as belonging to the Fig. 15 Brongniartiellidae (e.g. Actinophlebia Handlirsch, 1906–1908, Pter- Material. MNEMG 2009.40 [S.1635], a proximal wing fragment. inoblattina Scudder, 1885), particularly in having the similar general Smokejacks Brickworks, Upper Weald Clay, Lower Cretaceous, venation pattern and in lack of crossveins. However, the brong- Barremian. niartiellid affinities of these genera need confirmation, as the type genus Brongniartiella Meunier, 1897 is inadequately described. At Description. 21 mm preserved length (estimated length over present, we cannot place this genus in a family with certainty. 50 mm) and 9 mm preserved width. C not preserved. Costal space with 30 subcostal veinlets preserved of which 2 deeply forked. 7 Actinophlebioides valdensis gen. et sp. nov. crossveins preserved in costal space forming gradate series. Sc Fig. 16 approaches R at wing base. 9 subcostal crossveins preserved Derivation of name. From vald, the latinized name for the Weald, basally. Origin of Rs 6.2 mm from origin of R. 6 crossveins preserved in reference to the Weald Sub-basin. between R1 and Rs basally. Rs pectinately branched with 8 Material. Holotype BMB 025007, -8 (part and counterpart), branches preserved in proximal portion of wing, Rs1 and Rs3 a nearly complete forewing. Clockhouse Brickworks, Lower Weald deeply forked. 17 crossveins randomly distributed in radial space Clay, Lower Cretaceous, Hauterivian. basally. M fragment preserved. 7 randomly arranged crossveins preserved between R, Rs and M basally. Cu and A not preserved. Diagnosis. As for genus. Remarks. The specimen has tentatively been placed in the family Description. Forewing preserved length 7.4 mm (estimated Kalligrammatidae based on the size of the wing and the dense length about 9 mm), preserved width 4.3 mm. C partially preserved pattern of crossveins in the subcostal to medial spaces. This is with 24 subcostal veinlets, some forking towards anterior wing a small portion of a much larger wing indicated by the size of the margin. Costal space broad, narrowed distally. Sc terminating on R1. fragment; it is much bigger than the psychopsid wings found in the R1 with some forked veinlets distally. Rs with 11 branches, many of Wealden, thus excluding it from this family. them forked before mid-point of vein; Rs1 multi-branched, with 4 deep forks. M simple most of its length forking distally. CuA pec- Family: Neuroptera familia incertae sedis tinately branched with 10 branches. CuP probably simple, or few- Genus Actinophlebioides gen. nov. branched distally. 1A pectinately branched with 8 branches. 2A Derivation of name. From Actinophlebia, a genus-group name, in appears pectinately branched. No crossveins detected. Colour reference to a superficial resemblance of the venation of the type pattern faint and complex. species to that of Actinophlebia. Gender masculine. Type and only species. Actinophlebioides valdensis gen. et sp. nov. Diagnosis. May be distinguished from other neuropteran genera by the combination of the following character states: costal space broad, narrowed distally; Sc terminating on R1; Rs1 multi- branched, with four deep forks; M simple most of its length; CuA pectinately branched; CuP without deep forking; 1A pectinately branched; no crossveins present.

Fig. 15. Kalligrammatidae gen. et sp. indet., specimen MNEMG 2009.40 [S.1635]. A, Fig. 16. Actinophlebioides valdensis gen. et sp. nov., holotype BMB 025007, -8. A, drawing. B, photograph of the same specimen; scale bar represents 1 mm. drawing. B, photograph of the same specimen; scale bar represents 1 mm. 1336 J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338

Remarks. The forewing of this species is noteworthy for the absence of crossveins. However, this may possibly be an artefact of preservation. Genus Protohemerobius gen. nov. Derivation of name. From the Greek protou, before, and Hemer- obius, a genus-group name, in reference to the general similarity of the venation of the type species to that of Hemerobiidae. Gender masculine. Type and only species. Protohemerobius perexiguus gen. et sp. nov. Diagnosis. May be distinguished from other neuropteran genera by the combination of the following characters: very small wing size; subcostal veinlets forked once; Sc probably not terminating on R1; MA, MP and CuA pectinately branched distally; CuP probably dichotomously branched; 1A pectinately branched. Remarks. The venation of this genus is similar (possibly super- ficially) to that of Hemerobiidae (brown lacewings), e.g. the similar branching of the subcostal veinlets together with the small size, the presence of trichosors, and the similar general configuration of the venation. Unfortunately, Protohemerobius gen. nov. lacks critical areas of the wing, and the determination of its family placing is not possible. However, the occurrence in the Chinese Jurassic (Daohu- gou) of taxa with rather similar venation (Ren, 2003) suggests that this genus may represent an undescribed family. Protohemerobius perexiguus gen. et sp. nov. Fig. 17 Derivation of name. From the Latin perexiguus, tiny, in reference to the tiny wing size of the holotype. Fig. 17. Protohemerobius perexiguus gen. et sp. nov., holotype BMB 018581, -2. A, drawing. B, photograph of the same specimen; scale bar represents 1 mm. Diagnosis. As for genus. Material. Holotype BMB 018581, -2 (part and counterpart), an incomplete, basally crumpled forewing. Auclaye Brickworks, Upper other neuropteran families have been recorded: Ithonidae, Osmylidae Weald Clay, Barremian. and at least one undescribed family. The predictive method may not have been successful because the predicted families have not yet been Description. Forewing preserved length 4 mm (estimated length found, been preserved, or the analytical method that was successful 4.6 mm), preserved width 2.5 mm. C partially preserved; trichosors for spiders may not be transferable to fossil insects. present on anterior and posterior wing margin. 14 subcostal vein- The psychopsoids make up the majority of the neuropteran lets preserved, forking towards anterior wing margin. Sc curves fauna of the Wealden, and among them the family Psychopsidae towards R1 apically, apparently not terminating on it. No subcostal dominates. Genera from the family Psychopsidae are described crossveins preserved. 1 crossvein between Rs and R1 preserved. Rs herein (Cretapsychops gen. nov., Valdipsychops gen. nov., Micro- incomplete, pectinately branched, apparently with 8 preserved psychops gen. nov. and Psychopsites gen. nov.). The vast majority of branches forking distally; 1 branch appears to be dichotomously undescribed fragmentary specimens of Wealden Neuroptera from branched. Crossveins in radial space poorly preserved, apparently museum collections are also psychopsids (or have psychopsoid present. M forked before mid-point of vein (fork not preserved); affinities) (JEJ, pers. obs.). MA incomplete with 2 pectinate branches; MP pectinately Kalligrammatids are now known to have persisted in the Weal- branched with 3 branches. 1 crossvein between MA and MP, and 2 den from the Valanginian Wadhurst Clay (Hastings Beds Group) to crossveins between MP and CuA preserved. CuA pectinately the Barremian Upper Weald Clay (Weald Clay Group). Kalli- branched with 4 branches and 2 preserved crossveins between grammatidae are represented in the Wealden by small wing frag- them. CuP incomplete, crumpled, probably dichotomously forked. ments. Kalligramma roycrowsoni Jarzembowski, 2001 defined by an 1A incomplete, pectinately branched. impressive eyespot and fragmentary venation typical of the genus Kalligramma, was previously described from the Wadhurst Clay 5. Discussion (Hastings Beds Group) (Jarzembowski, 2001). The specimen described herein, Kalligrammatidae gen. et sp. indet., is from the Prior to this study only a few neuropterans of the psychopsoid younger Upper Weald Clay, and is placed in the family on the basis of families Psychopsidae, Brongniartiellidae and Kalligrammatidaewere subcostal and medial space crossveins and the large size of the wing. recorded from the Wealden (Jarzembowski, 1984; Ross and Cook, The fact that a kalligrammatid has been observed in the Hastings 1995). It was predicted by Jepson and Penney (2007) that the families Beds Group and Upper Weald Clay Formation suggests that the Nymphidae (split-footed lacewings), Polystoechotidae (giant lace- family Kalligrammatidae persisted from the mid-Valanginian to the wings), Chrysopidae (green lacewings), Coniopterygidae (wax flies) early Barremian in the Wealden and should be present in the Grin- and Hemerobiidae (brown lacewings) should be present in the stead Clay. Wealden fauna (as well as the already recorded Psychopsidae (silky The occurrence of the family Brongniartiellidae has not been lacewings), Brongniartiellidae and Kalligrammatidae). This work, confirmed in the Wealden. Prior to this study, brongniartiellids were however, has not shown any of those predicted to be present: instead thought to be a major constituent of the Wealden insect fauna, J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338 1337 however, what was previously thought to be members of this family Osmylopsychopidae from the Madygen Formation and providing are in fact from the family Psychopsidae. For example, a fragmentary VM with their photographs; S. Bruce Archibald (Simon Fraser wing from Smokejacks Brickworks identified by Ross and Cook University, Burnaby, Canada) for providing VM with photographs of (1995: fig. 5) as the brongniartiellid Pterinoblattina sp., is most Brongniartiella gigas. Also, thanks to three anonymous reviewers for probably a representative of the genus Valdipsychops gen. nov., or their comments and suggestions. JEJ acknowledges the Alkins Fund some other psychopsid genus. There is no evidence that this family (University of Manchester), Palaeontographical Society Fund and was present in the Wealden despite being recorded from the Lower Wingate Scholarship for financial support. Cretaceous of southern England by Martynova (1962). The discovery of the ithonid genus Principiala is most note- References worthy. Hitherto, only one genus of Neuroptera was known with certainty to occur in both the European and South American Lower Allen, P., 1981. Pursuit of Wealden models. Journal of the Geological Society, London Cretaceous, i.e., Triangulochrysopa Nel et al., 2005 (Menon and 138, 375–405. Allen, P., 1989. Wealden researchdways ahead. Proceedings of the Geologists’ Makarkin, 2008). Association 100, 529–564. Osmylidae prior to this study were unknown in the Wealden; Allen, P., 1998. Purbeck – Wealden (Early Cretaceous) climates. Proceedings of the two genera have been described herein, Stenosmylina gen. nov. and Geologists’ Association 109, 197–236. Allen, P., Wimbledon, W.A., 1991. Correlation of NW European Purbeck-Wealden Protosmylina gen. nov. The presence of Osmylidae in Britain has (non-marine Lower Cretaceous) as seen from the English type-areas. Creta- been extended back to the Lower Cretaceous. Both genera are ceous Research 12, 511–526. closely related to those of the subfamilies Mesosmylininae and Ansorge, J., 1996. Insekten aus dem oberen Lias von Grimmen (Vorpommern, Norddeutschland). Neue Pala¨ontologische Abhandlungen 2, 1–132. Protosmylinae, which are known to occur in the Jurassic to Eocene Archibald, S.B., Makarkin, V.N., 2006. Tertiary giant lacewings (Neuroptera: Poly- of continental Europe, and are very different to those of the stoechotidae): revision and description of new taxa from western North subfamily Osmylinae, present in Britain today. America and Denmark. Journal of Systematic Palaeontology 4, 119–155, p. 307 (errata). The two genera of uncertain placement might be representa- Aspo¨ck, U., 2002. Phylogeny of the Neuropterida (Insecta: Holometabola). Zoologica tives of undescribed families. Unfortunately, the poor state of Scripta 31, 51–55. preservation of these fossils prevents us from establishing new Austen, P.A., Agar, R., Jarzembowski, E.A., 2003. Smokejacks & Clockhouse Brick- families. works–7th June 2003. GA Magazine of the Geologists’ Association 2 (4), 11. Batten, D.J., 1998. Palaeoenvironmental implications of plant, insect and other The absence in the Wealden fauna of such groups as Chrys- organic-walled microfossils in the Weald Clay Formation (Lower Cretaceous) of opidae and Myrmeleontoidea is important. Chrysopids often occur southeast England. Cretaceous Research 19, 279–315. in Cretaceous deposits, and their absence in the Wealden is Batten, D.J., 2008. Spores and pollen from the Crato Formation: biostratigraphic and palaeoenvironmental implications. In: Martill, D.M., Bechly, G., Loveridge, R.F. somewhat surprising. The myrmeleontoids are the dominant (Eds.), The Crato Fossil Beds of Brazil. Window into an Ancient World. Cam- superfamily among Neuroptera in the Crato Formation of Brazil, bridge University Press, Cambridge, pp. 566–573. indicative of arid and semi-arid conditions of the environment in Bode, A., 1953. Die Insektenfauna des Ostniedersa¨chsischen Oberen Lias. Palae- ontographica (A) 103, 1–375. the surrounding areas of the Crato lake/lagoon (Martins Neto, Comstock, J.H., 1918. The Wings of Insects. Comstock Publishing Company, Ithaca, 2006; Batten, 2008). This group is also rarely recorded from other NY, 430 pp. Cretaceous localities. Fujiyama, I., 1978. Some fossil insects from the Tedori Group (Upper Jurassic - Lower Cretaceous), Japan. Bulletin of the National Science Museum, Tokyo (C) 4, The Wealden Neuroptera would most likely have been arboreal, 181–194. based on the habitat preferences of modern analogues. Modern psy- Gallois, R.W., 1978. British Regional Geology: The Wealden District, fourth ed. chopsids, ithonids and hemerobiids (potential modern analogues of HMSO, London. Gradstein, F.M., Ogg, J.G., Smith, A.G. (Eds.), 2005. A Geologic Time Scale 2004. Protohemerobius gen. nov.)are often (but not always)found living in or Cambridge University Press, Cambridge, 589 pp. around forested habitats, usually amongst the branches and trunks of Grimaldi, D., Engel, M.S., 2005. Evolution of the Insects. Cambridge University Press, trees (New, 1989; Grimaldi and Engel, 2005). This being the case, the Cambridge, xv þ 755 pp. neuropterans would probably have inhabited the woodland, which Handlirsch, A., 1906–1908. Die fossilen Insekten und die Phylogenie der rezenten Formen. Wilhelm Engelmann, Leipzig, 1430 pp. (in German). bordered the Wealden watercourses (Jarzembowski, 1995). The Haywood, A.M., Valdes, P.J., Markwick, P.J., 2005. Cretaceous (Wealden) climates: majority of modern osmylids have semi-aquatic larvae living in damp a modelling perspective. Cretaceous Research 25, 303–311. conditions (wet moss near water, under wet stones immediately Jarzembowski, E.A., 1980. Fossil insects from the Bembridge Marls, Palaeogene of the Isle of Wight, southern England. Bulletin of the British Museum (Natural adjacent to water); others occur in soil or under the tree bark (Riek, History) (Geology) 33, 237–293. 1970; New,1986, VM, pers. obs.). Their adults are often found close to Jarzembowski, E.A., 1984. Early Cretaceous insects from southern England. Modern water. Therefore if the Wealden osmylids had semi-aquatic larvae, the Geology 9, 71–93. Jarzembowski, E.A., 1987. Early Cretaceous Insects from Southern England. adults would be found near water, which compliments the suggestion Unpublished Ph.D thesis, University of Reading, 421 pp. that the neuropterans inhabited the woodland bordering water Jarzembowski, E.A., 1993. A provisional checklist of the fossil insects from the courses. The specimens of Neuroptera are all fragmentary and no Purbeck Beds of Dorset. Proceedings of the Dorset Natural History and Arch- aelogical Society 114, 175–179. body parts are present: this indicates that the insect bodies decayed, Jarzembowski, E.A., 1995. Early Cretaceous insect faunas and palaeoenvironment. disarticulated and were sorted, most probably by fluvial process, prior Cretaceous Research 16, 681–693. to their deposition (Jarzembowski, 1995). The fragmentary nature of Jarzembowski, E.A., 2001. A new Wealden fossil lacewing. In: Rowlands, J.L.J. (Ed.), Tunbridge Wells and Rusthall Commons: a History and Natural History. Tun- the wings may suggest a long transportation distance, a long decay bridge Wells Museum and Art Gallery, Tunbridge Wells, pp. 56–58. period on land or on the water surface, or predation. Jarzembowski, E.A., 2003. ‘Burnt’ beetles (Insecta: Coleoptera) from the Wealden of southern England. Acta Zoologica 46 (Suppl. Fossil Insects), 139–145. Acknowledgements Jarzembowski, E.A., Makarkin, V.N., 2006. A new species of silky lacewing (Insecta: Neuroptera: Psychopsidae) from the English Wealden (Early Cretaceous) with a critical review of other fossils referable to this family. In: Biotic-Sediment Many thanks to John Cooper and Gerald Legg (Booth Museum of Interactions. A Symposium in Memory of Roland Goldring (1928–2005), Natural History, Brighton) for help and loan of specimens, and University of Reading, 20–22 July 2006. Programme of Meeting. Curriculum Vitae. Abstracts. Ichron, p. 29. Andrew Ross (formerly of the Natural History Museum, London) Jepson, J.E., Penney, D., 2007. Neuropteran (Insecta) palaeodiversity with predic- and Maidstone Museum and Bentlif Art Gallery for loan of speci- tions for the Cretaceous fauna of the Wealden. Palaeogeography, Palae- mens; Tony Mitchell for photograph of Psychopsites rolandi gen. oclimatology, Palaeoecology 248, 109–118. Kimmins, D.E., 1939. A review of the genera of the Psychopsidae (Neuroptera), with et sp. nov.; Dmitri E. Shcherbakov (Palaeontological Institute of a description of a new species. Annals and Magazine of Natural History 11 (4), the Russian Adademy of Sciences) for loan of specimens of 144–153. 1338 J.E. Jepson et al. / Cretaceous Research 30 (2009) 1325–1338

Kru¨ ger, L., 1913. Osmylidae. Beitra¨ge zu einer Monographie der Neuropteren-Familie Oswald, J.D., 1993a. Revision and cladistic analysis of the world genera of the family der Osmyliden. II. Charakteristik der Familie, Unterfamilien und Gattungen auf Hemerobiidae (Insecta: Neuroptera). Journal of the New York Entomological Grund des Gea¨ders. Stettiner Entomologische Zeitung 74, 3–123. Society 101, 143–299. Kru¨ ger, L., 1922. Hemerobiidae. Beitra¨ge zu einer Monographie der Neuropteren- Oswald, J.D., 1993b. Phylogeny, taxonomy, and biogeography of extant silky lace- Familie der Hemerobiiden. Stettiner Entomologische Zeitung 83, 138–172. wings (Insecta: Neuroptera: Psychopsidae). Memoirs of the American Ento- Leach, W.E., 1815. Artikel Entomology. In: Brewster, D. (Ed.), Edinburgh Encyclo- mological Society 40, i–iii, pp. 1–65. paedia vol. 9, pp. 57–172. Pictet-Baraban, F.J., Hagen, H.A., 1856. Die im Bernstein befindlichen Neuropteren Linnaeus, C., 1758. Systema Naturae per Regna Tria Naturae, Secundum Classes, der Vorvelt. In: Berendt, G.C. (Ed.), Die im Bernstein befindlichen organischen Ordines, Genera, Species, cum Characteribus, Differentiis, Synonymis, Locis Reste der Vorwelt gesammelt, in verbindung mit mehreren bearbeitet und (tenth edn, revised). Salviae, Holmiae, Sweden, 824 pp. herausgegeben von Dr. Georg Carl Berendt. Bd. 2. Pt. 2. Nicholaischen Buch- MacLeod, E.G., 1971. The Neuroptera of the Baltic Amber. I. Ascalaphidae, Nym- handlung, Berlin, pp. 41–125. phidae, and Psychopsidae. Psyche 77 (for 1970), 147–180. Plant, C.W., 1997. A key to the adults of British lacewings and their allies (Neuro- Makarkin, V.N., 1997. Fossil Neuroptera of the Lower Cretaceous of Baisa, East ptera, Megaloptera, Raphidioptera and Mecoptera). Field Studies 9, 179–269. Siberia. Part 4. Psychopsidae. Beitra¨ge zur Entomologie 47, 489–492. Ponomarenko, A.G., 1984. Setchatokrylye iz yury Vostochnoi Azii. ([Neuroptera from Makarkin, V.N., New psychopsoid Neuroptera from the Lower Cretaceous of Baissa, the Jurassic of eastern Asia]). Paleontologicheskii Zhurnal 1984 (3), 64–73 (in Transbaikalia. Annales de la Socie´te´ Entomologique de France. (in press). Russian). Makarkin, V.N., Archibald, S.B., 2009. A new genus and first Cenozoic fossil record of Ponomarenko, A.G., 1996. Upper Liassic neuropterans (Insecta) from Lower Saxony, moth lacewings (Neuroptera: Ithonidae) from the Early Eocene of North Germany. Russian Entomological Journal 4 (for 1995), 73–89. America. Zootaxa 2063, 55–63. Ponomarenko, A.G., 2002. Superorder Myrmeleontidea Latreille, 1802 (¼ Neuro- Makarkin, V.N., Menon, F., 2007. First record of fossil ‘rapismatid-like’ Ithonidae pteroidea Handlirsch, 1903). In: Rasnitsyn, A.P., Quicke, D.L.J. (Eds.), History of (Insecta, Neuroptera) from the Lower Cretaceous Crato Formation of Brazil. Insects. Kluwer Academic Publishers, Dordrecht, pp. 176–192. Cretaceous Research 28, 743–753. Radley, J.D., 2005. Derived fossils in the southern English Wealden (non-marine Martins Neto, R.G., 2006. Insetos fo´ sseis como bioindicatores em depo´ sitos sed- early Cretaceous): a review. Cretaceous Research 26, 657–664. imentares: um estudo de casa para o Creta´ceo da Bacia do Araripe (Brasil). Rasnitsyn, A.P., Jarzembowski, E.A., Ross, A.J., 1998. Wasps (Insecta: Ves- Revista Brasileira de Zoocieˆncias 8, 155–183 (in Portuguese). pida ¼ Hymenoptera) from the Purbeck and Wealden (Lower Cretaceous) of Martins-Neto, R.G., 1997. Neuro´ pterous (Insecta: Planipennia) da formaça˜o Santana southern England and their biostratigraphical and palaeoenvironmental (Creta´ceo Inferior) Bacia do Araripe, Nordeste do Brasil. X – Descricao de novos significance. Cretaceous Research 19, 329–391. taxons (Chrysopidae, Babinskaiidae, Myrmeleontidae, Ascalaphidae e Psy- Ren, D., 2003. Study on the Jurassic fossil Raphidioptera and Neuroptera from chopsidae). Ciencias Exatas e Tecnologicas 2, 68–83 (in Portuguese). northeast China. Ph.D thesis, Beijing Forestry University, Beijing, China, v:202 Martynova, O.M., 1949. Mezozoiskie Setchatokrylye (Neuroptera) i ikh znachenie pp. (in Chinese, English abstract). dlya Ponimaniya filogenii i sistematiki otryada. ([Mesozoic Lacewings (Neuro- Riek, E.F., 1970. Neuroptera (lacewings). In: The Insects of Australia. Melbourne ptera) and Their Bearing on Concepts of Phylogeny and Systematics of the University Press, Melbourne, pp. 472–494. Order]), 20. Akademiya Nauk SSSR, Trudy Paleontologicheskogo Instituta (in Ross, A.J., Cook, E., 1995. The stratigraphy and palaeontology of the Upper Weald Russian)150–170. Clay (Barremian) at Smokejacks Brickworks, Ockley, Surrey, England. Creta- Martynova, O.M., 1954. Setchatokrylye nasekomye iz melovykh otlozhenii Sibiri. ceous Research 16, 705–716. ([Neuropterous insects from the Cretaceous deposits of Siberia]). Doklady Scudder, S.H., 1885. Notes on Mesozoic . Proceedings of the Academy of Akademii nauk SSSR (N.S.) 94, 1167–1169 (in Russian). Natural Sciences, Philadelphia 37, 105–115. Martynova, O.M., 1962. Otryad Neuroptera. Setchatokrylye. ([Order Neuroptera. Sinitshenkova, N.D., 1998. The first European Cretaceous stonefly (Insecta, Perli- Lacewings]). In: Rohdendorf, B.B. (Ed.), Osnovy Paleontologii: Chlnistonogie – da ¼ Plecoptera). Cretaceous Research 19, 317–321. Trakheinye i Khelitserovye. ([Fundamentals of Palaeontology. Arthropoda, Tra- Tillyard, R.J., 1919. Studies in Australian Neuroptera. No. 6. The family Psychopsidae, cheata and Chelicerata]). Akademiya Nauk SSSR, Moskva, pp. 272–282 (in with descriptions of new genera and species. Proceedings of the Linnean Russian). Society of New South Wales 43 (for 1918), 750–786. Menon, F., Makarkin, V.N., 2008. New fossil lacewings and antlions (Insecta, Neu- Tillyard, R.J., 1922. Mesozoic insects of Queensland. No. 9. Orthoptera, and additions roptera) from Cretaceous Crato Formation of Brazil. Palaeontology 51, 149–162. to the Protorthoptera, Odonata, Hemiptera and Planipennia. Proceedings of the Meunier, F., 1897. Revue critique de quelques insectes fossiles du Musee Teyler. Linnean Society of New South Wales 47, 447–470. Archives du Musee Teyler 5, 217–239 (in French). Tillyard, R.J., 1933. The panorpoid complex in the British Rhaetic and Lias. British Nel, A., Delclos, X., Hutin, A., 2005. Mesozoic chrysopid-like Planipennia: a phylo- Museum (Natural History). Fossil Insects 3, 1–79. genetic approach (Insecta: Neuroptera). Annales de la Socie´te´ Entomologique Tjeder, B., 1960. Neuroptera-Planipennia. The Lace-wings of Southern Africa. 3. de France 41, 29–69. Family Psychopsidae. In: Hanstro¨m, B., Brinck, P., Rudebeck, G. (Eds.), South New, T.R., 1983. Revision of the osmylid subfamilies Porisminae and Eidoporisminae African Animal Life. Results of the Lund University Expedition in 1950–1951, (Insecta: Neuroptera). Australian Journal of Zoology 31, 763–770. vol. VI. Swedish Natural Science Research Council, Stockholm, pp. 164–209. New, T.R., 1986. A review of the biology of Neuroptera Planipennia. Neuroptera Watson, J., Alvin, K.L., 1996. An English Wealden floral list, with comments on International, Supplemental Series 1, 1–57. possible environmental indicators. Cretaceous Research 17, 5–26. New, T.R., 1989. The Psychopsidae (Insecta: Neuroptera) of Australia and the Whalley, P.E.S., 1988. Mesozoic Neuroptera and Raphidoptera (Insecta) in Britain. Oriental Region. Invertebrate Taxonomy 2 (for 1988), 841–883. Bulletin of the British Museum (Natural History) (Geology) 44, 45–63. New, T.R., 1991. Neuroptera (lacewings). In: Naumann, I.D., Carne, P.B., Lawrence, J.F., Wootton, R.J., 2003. Wings. In: Resh, V.H., Carde, R.T. (Eds.), Encyclopedia of Insects. Nielson, E.S., Spradberry, J.P., Taylor, R.W., Whitten, M.J., Littlejohn, M.J. Academic Press, London, pp. 1186–1192. (Eds.)The Insects of Australia: a Textbook for Students and Research Workers, Zalessky, G.M., 1953. Novye mestonakhozhdeniya melovykh nasekomykh v second ed., vol. 1. Melbourne University Press, Carlton, Victoria, pp. 525–542. Povolzh’e, Kazakhstane i Zabaikal’e. ([New localities of Cretaceous insects from Newman, E., 1853. Proposed division of Neuroptera into two classes. Zoologist 11 the Volga region, Kazakhstan and Transbaikalia]). Doklady Akademii Nauk SSSR (Appendix) clxxxi–cciv. (N.S.) 89, 163–166 (in Russian). CHAPTER 8

4. TWO NEW SPECIES OF SNAKEFLY (INSECTA: RAPHIDIOPTERA) FROM THE

LOWER CRETACEOUS OF ENGLAND AND SPAIN WITH A REVIEW OF OTHER

FOSSIL RAPHIDIOPTERANS FROM THE JURASSIC/CRETACEOUS TRANSITION

115

ALAVESIA, 2: 193-201 (2008) ISSN 1887-7419

Two new species of snakefly (Insecta: Raphidioptera) from the Lower Cretaceous of England and Spain with a review of other fossil raphidiopterans from the Jurassic/Cretaceous transition

James E. JEPSON1 & Edmund A. JARZEMBOWSKI2

1School of Earth, Atmospheric and Environmental Sciences, University of Manchester, Williamson Building, Oxford Road, Manchester M13 9PL, UK. E-mail: [email protected] 2Maidstone Museum and Bentlif Art Gallery, St. Faith’s Street, Maidstone, Kent ME14 1LH, UK. E-mail: [email protected]

ABSTRACT Two new species of snakefly (Insecta: Raphidioptera, Mesoraphidiidae) are described from the Weald Clay of the Weal- den Supergroup, southern England and the Lithographic Limestone of Montsec, Spain. Proraphidia hopkinsi sp. nov. is the first species of snakefly to be described from the Wealden and P. gomezi sp. nov. is the first snakefly to be formally described from Montsec. These species extend the range of the genus Proraphidia across the Jurassic/Cretaceous bound- ary and its geographical range from Kazakhstan to England and Spain. A review of the snakefly families and genera involved in the Jurassic/Cretaceous transition is given.

KEY WORDS: Wealden Supergroup. Montsec. Early Cretaceous. southern England. Jurassic/Cretaceous transition. new species.

INTRODUCTION 2004). At species level, the Mesozoic raphidiopterans are Raphidioptera is the least diverse order of the holom- very diverse, supporting the idea that extant Raphidiopt- etabolous insects with approximately 210 extant species era are relict (Engel 2002, Grimaldi & Engel 2005). Few described in two extant families: Raphidiidae and Inocel- attempts have been made at a cladistic study of fossil and liidae (Aspöck 2002, Grimaldi & Engel 2005, Engel et al. extant raphidiopterans (e.g. Ren & Hong 1994, Willmann 2006). The order is distinctive with an elongated protho- 1994) and the relationships have still not been satisfacto- rax, a long ovipositor seen in females, and typical vena- rily resolved (see Engel 2002). Proraphidia hopkinsi sp. tion, although the true autapomorphies are not immediately nov. and P. gomezi sp. nov. (Mesoraphidiidae) are the first clear (Engel 2002). However, Aspöck and Aspöck (2004) species of Raphidioptera to be formally described from the discussed and listed four autapomorphies of the order, Wealden Supergroup of southern England and the Litho- but the only one that can be applied to fossil wings is the graphic Limestone of Montsec, Spain. forewing subcosta running into the anterior margin. Extant snakeflies are arboreal, in areas of cold weather with near freezing winter temperatures. These cold temperatures are GEOLOGICAL SETTINGS required for the insect to complete development before The Wealden Supergroup is Lower Cretaceous in age pupation or just before eclosion (Grimaldi & Engel 2005). (Berriasian-Aptian?); the supergroup consists of the Hast- Therefore they are found in cold temperate regions and ings Beds (Ashdown, Wadhurst Clay, Lower and Upper are absent from hotter regions such as in South America, Tunbridge Wells Sand, and the Grinstead Clay formations) Australia and sub-Saharan Africa (loc cit.). However, some and the Weald Clay (Upper and Lower) (Allen & Wimble- Mesozoic raphidiopterans are to be found in much warmer don, 1991; Rasnitsyn et al. 1998). The insect remains are conditions such as inferred tropical climates (Engel 2002). almost solely found in stone developments in the clays and Mesozoic raphidiopterans are divided into three definite most common in the Weald Clay (Jarzembowski 1984) in extinct families: Baissopteridae, Mesoraphidiidae and All- which the specimen below was found. Many of the insect oraphidiidae, and one debatable extinct family: Priscaenig- orders are present: Odonata, Blattodea, Orthoptera, Hemi- matidae. It is debated whether Priscaenigmatidae belongs ptera, Neuroptera, Coleoptera, Mecoptera, Diptera, Tri- in Raphidioptera because the subcosta runs into the wing choptera, Hymenoptera, Isoptera, Raphidioptera (Jarzem- apex and not the anterior margin (see Aspöck & Aspöck bowski 1984, 1995). The insects are usually represented

193 Tw o n e w s p e c i e s o f s n a k e f l y (In s e c t a : Ra p h i d i o p t e r a ) James E. JEPSON & Edmund A. JARZEMBOWSKI by wing fragments and disarticulated body material; they (Whalley & Jarzembowski 1985, Gómez-Pallerola 1986, are preserved as uncrushed and unoxidized elements in Martínez-Delclòs 1989). The rocks were deposited in a siltstones and early diagenetic concretions in the clay lacustrine environment in subtropical conditions with alter- (Jarzembowski 1984). The palaeoenvironment is thought nating wet and not very pronounced dry seasons (Soriano to have been woodland bordering watercourses (Jarzem- et al. 2006). bowski 1995). The climate has been suggested as subtropi- cal/warm temperate (with seasonal rainfall), based on the MATERIAL AND METHODS presence of certain insect orders (e.g. Isoptera) (Jarzem- bowski 1995) and sedimentological evidence (Allen 1998, The Wealden holotype is preserved in a piece of silt- Haywood et al. 2005). The occurrence of a distinct wet and stone concretion associated with Blattodea, Hemiptera, dry season has been proposed, as well as year round pre- Mecoptera, Diptera, etc. and a crustacean antenna. The hol- otype is deposited in the collections of the Booth Museum cipitation with little evidence for prolonged drought (Bat- of Natural History, Brighton (BMB). ten 1998, Haywood et al. 2004). Wildfires are thought to have occurred due to the presence of fusainized elytra and The Montsec holotype is preserved on a piece of lime- plant remains (Watson & Alvin 1996, Batten 1998, Jarzem- stone. The holotype is deposited in the Institut d’Estudis bowski 2003). Ilerdencs, Lérida, Spain [L. 29]. The age of the Lithographic Limestone at Montsec is Drawings were made using a Wild Camera Lucida Early Barremian (Lower Cretaceous) (Soriano et al. 2006). microscope attachment. Both halves were drawn separately The insects are preserved as impressions, sometimes and a composite drawing was produced through the use of showing traces of cuticle. They occur in well-cemented a light-table. Photographs were taken using a Wild Phot- fine-grained brown limestone (Whalley & Jarzembowski omicroscope using Ilford FP4 film. Wetting using a weak 1985). Numerous insect orders have been found in the solution of ethyl alcohol was used to increase contrast. deposit: Blattodea, Odonata, Hemiptera, Coleoptera, Dip- Wing vein abbreviations are as follows: C, costa; Sc, tera, Hymenoptera, Ephemeroptera and Grylloblatodea subcosta; R, radius; Rs, radial sector; M, media; MA, media

Figure 1. A. Photograph of the forewing of Proraphidia hopkinsi sp. nov. BMB 014915 [CH 879ia] B. Camera lucida drawing showing venation of Proraphidia hopkinsi sp. nov. Drawing convention as in Jarzembowski (1980). Scale bar = 1mm.

194 James E. JEPSON & Edmund A. JARZEMBOWSKI Tw o n e w s p e c i e s o f s n a k e f l y (In s e c t a : Ra p h i d i o p t e r a ) anterior; MP, media posterior; Cu, cubitus; CuA, cubitus second 0.5mm. Pterostigma present terminating on third anterior; CuP, cubitus posterior; A, analis; r, radial cell; dc, crossvein in radial space and terminating more distally than discal cell; mc, medial cell; ac, anal cell; p, pterostigma. the most distal r-rs crossvein; pterostigma small, less than 1mm long. Three radial cells are present: 1r: 4mm long SYSTEMATIC PALAEONTOLOGY and 0.5mm wide; 2r: 2.3mm long and 0.7mm wide and 3r: 2.2mm long and 0.4mm wide. Rs and MA arise approxi- mately 6.5mm from wing apex; they are coalesced for Order: Raphidioptera Navás, 1918 1.6mm before forking. Rs2 is two branched forking from Family: Mesoraphidiidae Martynov, 1925 the stem of Rs 1.2mm from wing apex; both Rs1 and Rs2 arise in the distal part of the wing near the apex. The discal GENUS: Proraphidia Martynova, 1947 cell (dc) is formed by a crossvein between Rs and MA and is 2.2mm long and 0.7mm wide (at its widest point). MA is Type species: Proraphidia turkestanica Martynova, two branched with both branches being forked. MP is coa- 1947. Upper Jurassic (Malm), Karatau, Chimkent Prov- lesced with CuA for a short distance (0.2mm). M and CuA ince, Kazakhstan, based on an isolated forewing. arise 8.4mm from the wing apex. MP is deeply forked with Revised Diagnosis. Forewing. R close to Sc in proxi- each branch forked. Between MP and CuA three cells are mal half of wing, dc short. Second branch of CuA forked, present mc1, mc2 and mc3. CuA forks near the posterior ac wide. Pterostigma short, basal border of pterostigma wing margin; the first branch is simple, the second branch more distal than r-rs crossvein. Hindwing. Costal field nar- forks again. Two crossveins are preserved between CuA row, dc long, MP deeply forked, Cu deeply forked, pter- and CuP; CuP is simple. 1 crossvein is preserved between ostigma probably as in forewing and strongly developed. CuP and A. Anal area poorly preserved, only part of ac being present. Included species. Proraphidia turkestanica, P. hop- kinsi sp. nov. and P. gomezi sp. nov. Remarks. The fossil has been placed in the genus Pro- raphidia on the basis of the small pterostigma (excluding it from Mesoraphidia with its long pterostigma), the pter- Proraphidia hopkinsi sp. nov. ostigma’s basal border being located more distally than Figure 1 the most distal r-rs crossvein, and the structure of M and Cu. There are three species in the genus Proraphidia: P. 1984 Mesoraphidiidae, Jarzembowski, pp. 82–83, fig. turkestanica, P. hopkinsi sp. nov. and P. gomezi sp. nov. 30 Martynova (1961) identified a species Proraphidia viti- 1987 Proraphidia sp. nov., Jarzembowski, pp. 174– mica; this, however, was transfered to Alloraphidia by 176, fig. 8.1 Ponomarenko (1993) and later to Pararaphidia by Engel (2002). Proraphidia hopkinsi sp.nov. is the first species of 1988 Mesoraphidia sp., Whalley, pp. 48, 61 snakefly to be described from the Early Cretaceous Weal- 1990 Mesoraphidiidae, Nel, Séméria and Martins-Neto, den Supergroup. Snakeflies are present in the Jurassic in p. 11 England, represented by the mesoraphidiid Metaraphidia confusa and the debatable raphidiopteran Priscaenigma Etymology. Named after the late Dick Hopkins in rec- obtusa (Whalley 1985, 1988, Willmann 1994, Engel 2002). ognition of his long voluntary service to museum geology A Cretaceous snakefly (Mesoraphida sp.) has been figured at the Booth Museum, Brighton. from the Purbeck Limestone Group of England (Jarzem- Holotype. BMB 014915, - 8 [CH 879ia,b] single left bowski 1993, Clifford et al. 1994). There are ­­­­­­four species forewing. Deposited in the Booth Museum of Natural His- of the family Raphidiidae in England today (Plant 1997). tory, Brighton. Type locality. Lower Weald Clay, Clockhouse Brick- Proraphidia gomezi sp. nov. works, Surrey, England. Lower Cretaceous (Late Hauteriv- Figure 2 ian). Diagnosis. Wing not as broad as in P. turkestanica; R 1986 Proraphidia sp., Gómez-Pallerola, p. 45, fig. 7, not as close to Sc in proximal half of wing. Three cross- photograph 46. veins between R and Sc; dc shorter. Both branches of MA forked; MP deeply forked with both branches forked. Sec- Etymology. After Mr Gómez-Pallerola, fossil collector ond branch of CuA forked; ac much wider. and contributor to the Institut d’Estudis Ilerdencs, Lérida. Description and measurements. Forewing: 11.5mm Holotype. [L. 29]. Overlapping fore- and hindwings long and 3.2mm wide. C poorly preserved and incomplete; and part of body. Deposited in Institut d’Estudis Ilerdencs, Sc incomplete, terminating approximately 4mm from the Lérida, Spain. wing apex (slightly past the mid-point of the wing) on the anterior wing margin. Half of the subcostal area is pre- Type locality. Limestone, Pedrera de Meià, Rúbies, served, with three crossveins present. R is complete and Spain, Lower Cretaceous, Early Barremian. long terminating at the wing apex; in the radial area there Diagnosis. Hindwing differs from P. turkestanica are two crossveins preserved one 3.5mm from apex and the in length of the wing, having fewer costal veinlets, the

195 Tw o n e w s p e c i e s o f s n a k e f l y (In s e c t a : Ra p h i d i o p t e r a ) James E. JEPSON & Edmund A. JARZEMBOWSKI branches of Rs and MA being simple and mc being split the second 0.9mm from the wing apex. Pterostigma present into two. terminating on the second crossvein in radial area and ter- Description and measurements. Forewing: 13mm minating more distally than the most distal r-rs crossvein; long and 5.2mm wide. C is preserved with three costal pterostigma small, less than 1mm long. Three radial cells veinlets; Sc terminating approximately 5.6mm from the are present: 1r: 3.7mm long and 1mm wide; 2r: 3.6mm wing apex (past the mid-point of the wing) on the ante- long and 1.2mm wide and 3r: 2.2mm long and 0.7mm rior wing margin. R is complete and long terminating wide. Rs and MA arise approximately 9mm from wing just before the wing apex; in the radial area there are two apex; they are coalesced for 2.5mm before forking. Rs is crossveins preserved one 5.6mm from apex of wing and two branched; Rs1 and Rs2 are simple; both Rs1 and Rs2

Figure 2. A. Photograph of Proraphidia gomezi sp. nov. [L. 29]. B Camera lucida drawing of overlapping wings. C. Camera lucida drawing of venation of forewing. D. Camera lucida drawing of venation of hindwing. Drawing convention as in Jarzembowski (1980). Scale bar = 1mm.

196 James E. JEPSON & Edmund A. JARZEMBOWSKI Tw o n e w s p e c i e s o f s n a k e f l y (In s e c t a : Ra p h i d i o p t e r a ) arise in the distal part of the wing near the apex. The discal dled the Jurassic/Cretaceous boundary are Mesoraphidia, cell (dc) is 3.5mm long and 1mm wide (at its widest point). Proraphidia and Sinoraphidia. Out of these four gen- MA is forked with both branches simple. Origin of MP not era, Sinoraphidia (one species) is known from the Upper preserved. MP is deeply forked with each branch forked. Jurassic-Lower Cretaceous of Gansu, China (Hong 1982); Between MP and CuA three cells are present mc1, mc2 Proraphidia crosses the boundary: P. turkestanica is and mc3. Cu incomplete. CuA partially preserved. CuP known from the Upper Jurassic of Karatau, Kazakhstan and Analis not preserved. (Martynova 1947), P. hopkinsi sp. nov. from the Lower Hindwing: 12.2mm long and 4.8mm wide. C is pre- Cretaceous Wealden Supergroup and P. gomezi sp. nov. served with four costal veinlets; Costal area narrow. Sc ter- from Montsec in the Lower Cretaceous; however, there minating approximately 5.5mm from the wing apex (near is no evidence from the fossil record showing one good the mid-point of the wing) on the anterior wing margin. species continuing through the transition. Mesoraphidia R is complete and long terminating just before the wing shows a similar scenario to Proraphidia; there are no spe- apex; in the radial area there is one crossvein preserved cies that straddle the boundary. In the Jurassic, six species 0.8mm from apex of wing. Three radial cells are present: are known from Kazakhstan (Martynov 1925; Martynova, 1r: 3.8mm long and 1.1mm wide; 2r: 4.5mm long and 1947) and one species from Mongolia (Ponomarenko, 1.1mm wide and 3r: 2.2mm long and 0.6mm wide. Cell 1988; Willmann, 1994) has been described. In the Lower 1r has a crossvein before mid-point. Rs arises approxi- Cretaceous, eleven species have been recorded in China mately 9mm from wing apex and MA origin not seen. MA (Hong, 1992a; Ren et al. 1995; Ren, 1997; Engel 2002), and Rs are coalesced for 3mm before forking. Rs is two one unnamed species from England (Jarzembowski, 1993; branched; Rs1 and Rs2 are simple; both Rs1 and Rs2 arise Clifford et al. 1994), two species from southern Korea in the distal part of the wing near the apex. The discal cell (Engel et al. 2006) and one species has been recorded in (dc) is 3.5mm long and 0.9mm wide (at its widest point). the Upper Cretaceous New Jersey amber, USA (Grimaldi, Metaraphidia MA forks with both branches simple. MP origin not seen. 2000). The genus , however, did not survive into the Cretaceous; the fossil record currently suggests MP is deeply forked with each branch forked. Between the that the genus became extinct in the Jurassic. It comprises branches of MP the medial cell (mc) is split in two by a two species from the Lower Jurassic of England and Ger- crossvein. CuA forks once marginally; the origin of CuA is many (Whalley, 1988). not preserved. CuP is partially preserved with a crossvein joining it to CuP at the vein’s mid-point. CuP is simple. Baissopteridae have not been found in the Jurassic; Analis not preserved. however, they are found in the Lower Cretaceous. Five spe- cies of Baissoptera and four species of Cretoraphidia have Remarks. This species has been placed in Proraphidia been recorded in Baissa, Buryat Russia (Martynova 1961, on the basis of the small pterostigma, the pterostigma’s Ponomarenko 1993, Engel 2002). The family has also been basal border being located more distally than the most dis- recorded in the Lower Cretaceous of Brazil (one species of tal r-rs crossvein, and the structure of M and Cu. This is the Austroraphidia and two species of Baissoptera) (Oswald first species of snakefly to be formally described from the 1990, Willmann 1994, Engel 2002), China (three species Spanish Cretaceous and Montsec. There are twelve species of Baissoptera) (Ren 1994, 1997, Ren et al. 1995), and of snakeflies in Spain today in eight genera (http://www. Mongolia (one species of Cretinocellia, Cretoraphidiopsis fauna-ibericamncn.csic.es/faunaib/arthropoda/index.php). and Lugala) (Ponomarenko 1988, Willmann 1994, Engel 2002). RAPHIDIOPTERA FROM THE JURASSIC/ The family Alloraphidiidae is the smallest of three CRETACEOUS TRANSITION raphidiopteran families with three genera Alloraphidia, Archeraphidia and Pararaphidia (Engel 2002). All- A problem arises when one talks about a clear-cut oraphidia asiatica, Archeraphidia baisensis and Para- Jurassic/Cretaceous boundary in continental deposits, raphidia vitimica are found in the Lower Cretaceous of because many are badly dated or the dates given are debat- Baissa, Buryat, Russia (Ponomarenko 1993, Engel 2002); able. Here we have used the dates that are widely accepted other Cretaceous species are from Labrador, Canada (Car- by most workers. penter 1967) and Mongolia (Ponomarenko 1988, Willmann There has only been one raphidiopteran family recorded, 1994). Mesoraphidiidae that ranges from the Jurassic into the In the Cretaceous, raphidiopterans appear to be more Cretaceous. The family Raphidiidae has not been found in widespread geographically than in the Jurassic; however, pre-Tertiary rocks (Engel 2002) and Inocelliidae has been this may not be a true reflection because more fossil insect recorded in Lower Cretaceous rocks (Wang 1987). Table deposits are known in the Cretaceous than the Jurassic. In 1 lists the families (with species) involved in the Jurassic/ the Jurassic, the fossil evidence suggests that they were Cretaceous transition. only present in Eurasia (England, Germany, Mongolia, Mesoraphidiidae has two genera that crossed the tran- Kazakhstan and China). In the Cretaceous, fossils have sition, one genus that straddled the transition and one also been found in the USA, Canada and Brazil showing genus Metaraphidia that appears to have become extinct that they colonized the New World. The Jurassic, however, before the boundary. The three genera that crossed/strad- is thought to be the time when the major family radiations

197 Tw o n e w s p e c i e s o f s n a k e f l y (In s e c t a : Ra p h i d i o p t e r a ) James E. JEPSON & Edmund A. JARZEMBOWSKI

Age Family Species Locality/Deposit Reference Upper Mesoraphidiidae Mesoraphidia luzzii Grimaldi, 2000 New Jersey amber, USA Grimaldi 2000 Cretaceous Nanoraphidia electroburmica Engel, 2002 Myanmar amber Engel 2002 Middle Alloraphidiidae Alloraphidia dorfi Carpenter, 1967 Labrador, Canada Carpenter 1967 Cretaceous Lower Arariperaphidia rochai Martins-Neto and Martins-Neto and Incertae sedis Crato Formation, Brazil Cretaceous Vulcano, 1990 Vulcano, 1990 Austroraphidia brasiliensis (Nel, Séméria and Baissopteridae Crato Formation, Brazil Willmann 1994 Martins-Neto, 1990) Baissoptera brasiliensis Oswald, 1990 Crato Formation, Brazil Oswald 1990 Baissoptera pulchra (Martins-Neto and Nel, Crato Formation, Brazil Engel 2002 1992) Alloraphidiidae Alloraphidia anomala Ren, 1997 Beipiao, Liaoning, China Ren 1997 Baissopteridae Baissoptera euneura Ren, 1997 Beipiao, Liaoning, China Ren 1997 Baissoptera grandis Ren in Ren et al., 1995 Beipiao, Liaoning, China Ren et al. 1995 Baissoptera liaoningensis Ren, 1994 Beipiao, Liaoning, China Ren 1994 Mesoraphidiidae Huaxiaraphidia sinensis Hong, 1992a Beipiao, Liaoning, China Hong 1992a Huaxiaraphidia shandongensis Hong, 1992a Beipiao, Liaoning, China Hong 1992a Mesoraphidia koreenensis Engel et al., 2006 Southern Korea Engel et al., 2006 Mesoraphidia phantosma Engel et al., 2006 Southern Korea Engel et al., 2006 Mesoraphidia amoena Ren, 1997 Beipiao, Liaoning, China Ren 1997 Mesoraphidia furcivenata Ren in Ren et al., Beipiao, Liaoning, China Ren et al. 1995 1995 Mesoraphidia gaoi (Ren, 1995) Beipiao, Liaoning, China Engel 2002 Mesoraphidia glossophylla (Ren, 1997) Beipiao, Liaoning, China Engel 2002 Mesoraphidia longistimosa (Ren, 1994) Beipiao, Liaoning, China Engel 2002 Mesoraphidia heteroneura Ren, 1997 Beipiao, Liaoning, China Ren 1997 Mesoraphidia obliquivenatica (Ren, 1994) Beipiao, Liaoning, China Engel 2002 Mesoraphidia myrioneura (Ren, 1997) Beipiao, Liaoning, China Engel 2002 Mesoraphidia polyphlebia (Ren, 1994) Beipiao, Liaoning, China Engel 2002 Mesoraphidia shangyuanensis (Ren, 1994) Beipiao, Liaoning, China Engel 2002 Mesoraphidia sinica Ren, 1997 Beipiao, Liaoning, China Ren 1997 Siboptera fornicata (Ren, 1994) Beipiao, Liaoning, China Ren 1997 Xuraphidia liaoxiensis Hong, 1992b Kezuo, Liaoning, China Hong 1992b Kezuoraphidia kezuoensis (Hong, 1992c) Kezuo, Liaoning, China Willmann 1994 Jilinoraphidia dalaziensis Hong and Chang, Zhixian, Jilin province, China Hong & Chang 1989 1989 Lithographic Gómez-Pallerola Proraphidia gomezi sp. nov. Limestone, Montsec, Spain 1986; herein Wealden Supergroup, herein Proraphidia hopkinsi sp. nov. Southern England Purbeck Limestone Group, Mesoraphidia sp. Clifford et al. 1994 England Inocelliidae Sinoinocellia liaoxiensis Wang, 1987 Xiaoershijiazi, China Wang, 1987 Alloraphidiidae Alloraphidia petrosa Ponomarenko, 1988 Mongolia Ponomarenko 1988 Archeraphidia hosbayari Ponomarenko, 1988 Mongolia Ponomarenko 1988 Archeraphidia yakowlewi Ponomarenko, 1988 Mongolia Ponomarenko 1988 Pararaphidia deserta (Ponomarenko, 1988) Mongolia Willmann 1994 Baissopteridae Cretinocellia cellulosa Ponomarenko, 1988 Mongolia Ponomarenko 1988 Cretaraphidiopsis bontsaganensis Mongolia Engel 2002 (Ponomarenko, 1988) Lugala longissima (Ponomarenko, 1988) Mongolia Willmann 1994 Cretoraphidia certa Ponomarenko, 1993 Romanovka, Russia Ponomarenko 1993 Alloraphidiidae Alloraphidia asiatica Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Archeraphidia baisensis Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Pararaphidia vitimica (Martynova, 1961) Baissa, Buryat, Russia Engel 2002 Baissopteridae Baissoptera martinsoni Martynova, 1961 Baissa, Buryat, Russia Martynova 1961 198 James E. JEPSON & Edmund A. JARZEMBOWSKI Tw o n e w s p e c i e s o f s n a k e f l y (In s e c t a : Ra p h i d i o p t e r a )

Age Family Species Locality/Deposit Reference Baissoptera cellulosa Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Baissoptera elongata Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Baissoptera kolosnitsynae Martynova, 1961 Baissa, Buryat, Russia Martynova 1961 Baissoptera minima Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Baissoptera sibirica Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Cretoraphidia macrocella Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Cretoraphidia magna Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Cretoraphidia reticulata Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Baissoraphidia glossopteron Ponomarenko, Mesoraphidiidae Baissa, Buryat, Russia Ponomarenko 1993 1993 Baissoraphidia orientalis Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Siboptera eurydictyon Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Siboptera medialis Ponomarenko, 1993 Baissa, Buryat, Russia Ponomarenko 1993 Upper Jurassic Mesoraphidiidae Sinoraphidia viridis Hong, 1982 Gansu, China Hong 1982 – Lower Cretaceous Upper Mesoraphidiidae Mesoraphidia gobiensis Ponomarenko, 1988 Mongolia Ponomarenko 1988 Jurassic Mesoraphidia grandis Martynov 1925 Karatau, Kazakhstan Martynov 1925 Mesoraphidia elongata Martynov, 1925 Karatau, Kazakhstan Martynov 1925 Mesoraphidia inaequalis Martynov, 1925 Karatau, Kazakhstan Martynov 1925 Mesoraphidia parvula Martynov, 1925 Karatau, Kazakhstan Martynov 1925 Mesoraphidia similis Martynov, 1925 Karatau, Kazakhstan Martynov 1925 Mesoraphidia pterostigmalis Martynova, 1947 Karatau, Kazakhstan Martynova 1947 Proraphidia turkestanica Martynova, 1947 Karatau, Kazakhstan Martynova 1947 Lower Mesoraphidiidae Metaraphidia confusa Whalley, 1985 England Whalley 1985 Jurassic Metaraphidia vahldieki Willmann, 1994 Germany Willmann, 1994

Table 1. Extinct taxa of raphidiopterans known from the Jurassic and Cretaceous. occurred (see phylogenetic diagram in Grimaldi & Engel Dmitri Logunov (The Manchester Museum) for discussion 2005, fig. 9.4, p.336). However, the fact that there are no and translation of the Martynova (1947) paper, Drs Jörg fossils of the families Alloraphidiidae and Baissopteridae in Ansorge and Vladimir Makarkin for helpful comments the Jurassic means that based on the current fossil evidence, when reviewing the paper, Dr Carmen Soriano for infor- the major radiation of raphidiopterans could not have been mation on Spain, and INTAS for supporting Fossilsx3. before the latest Jurassic. The Cretaceous saw the extinc- tion of the fossil families Baissopteridae, Alloraphidiidae and Mesoraphidiidae (Engel 2002) and it is thought that REFERENCES snakefly biodiversity decreased around the middle Creta- ALLEN, P. 1998. Purbeck – Wealden (early Cretaceous) ceous (Aspöck 1998). Martynov (1925) suggested that the climates. Proceedings of the Geologists’ Association, extinction of families such as Mesoraphidiidae was due to 109: 197–236. ‘disharmony’ in the degree of evolution of the wings and ALLEN, P. & WIMBLEDON, W.A. 1991. Correlation of body. However, recently, the extraterrestrial impact at the NW European Purbeck-Wealden (non-marine Lower end of the Cretaceous has been hypothesized as a reason Cretaceous) as seen from the English type-areas. Creta- for the extinction of the tropical snakeflies with only the ceous Research, 12: 511–526. cold-climate adapted snakeflies (Inocelliidae and Raphidii- ASPÖCK, H. 1998. Distribution and biogeography of the dae) surviving (Aspöck 1998). This would account for the order Raphidioptera: updated facts and a new hypoth- absence of the order in warm-climate Palaeogene deposits esis. Acta Zoologica Fennica, 209: 33–44. such as the Bembridge Marls, southern England. ASPÖCK, U. 2002 The phylogeny of the Neuropter- ida (Insecta: Holometabola). Zoologica Scripta, 31: ACKNOWLEDGEMENTS 51–55. Thanks to Dr John Nudds, Peter Falkingham and ASPÖCK, U. & ASPÖCK, H. 2004. Two significant Karl Bates (University of Manchester) for discussion, Dr new snakeflies from Baltic amber, with discussion

199 Tw o n e w s p e c i e s o f s n a k e f l y (In s e c t a : Ra p h i d i o p t e r a ) James E. JEPSON & Edmund A. JARZEMBOWSKI

on autapomorphies of the order and its included taxa JARZEMBOWSKI, E.A. 1984. Early Cretaceous insects (Raphidioptera). Systematic Entomology, 29: 11–19. from southern England. Modern Geology, 9: 71–93. BATTEN, D.J. 1998. Palaeoenvironmental implications JARZEMBOWSKI, E.A. 1987. Early Cretaceous insects of plant, insect and other organic-walled microfossils from southern England. Unpublished PhD thesis, Uni- in the Weald Clay Formation (Lower Cretaceous) of versity of Reading, 421pp. southeast England. Cretaceous Research, 19: 279–315. JARZEMBOWSKI, E.A. 1993. A provisional checklist of CARPENTER, F.M. 1967. Cretaceous insects from Lab- the fossil insects from the Purbeck Beds of Dorset. Pro- rador. 2. A new family of snake-flies (Neuroptera: All- ceedings of the Dorset Natural History and Archaelogi- oraphidiidae). Psyche, 74: 270–275. cal Society, 114: 175–179. CLIFFORD, E., CORAM, R.A., JARZEMBOWSKI, E.A. JARZEMBOWSKI, E.A. 1995. Early Cretaceous insect & ROSS, A.J. 1994. A supplement to the insect fauna faunas and palaeoenvironment. Cretaceous Research, from the Purbeck Group of Dorset. Proceedings of the 16: 681–693. Dorset Natural History and Archaeological Society, JARZEMBOWSKI, E.A. 2003. ‘Burnt’ beetles (Insecta: 116: 146–150. Coleoptera) from the Wealden of southern England. ENGEL, M. S. 2002. The smallest Snakefly (Raphidiopt- Acta Zoologica, 46 (suppl- Fossil Insects): 139–145. era: Mesoraphidiidae): a new species in Cretaceous MARTÍNEZ-DELCLÒS, X. 1989. aquatica n. amber from Myanmar, with a catalog of fossil Snake- sp. insecto del Cretácico inferior de la flies. American Museum Novitates, 3363: 1–22. Sierra del Montsec (Lleida, España). Revista Española ENGEL, M.S., LIM, J.D. & BAEK, K.S. 2006. Fossil de Paleontología, 4, 67–74. snakeflies from the Early Cretaceous of southern Korea MARTINS-NETO, R. G. & VULCANO, M. A. 1990. Pri- (Raphidioptera: Mesoraphidiidae). Neues Jahrbuch für meiro registro de Raphidioptera (Neuropteroidea) na Geologie und Paläontologie, Monatshefte, 2006, 4: Formaçao Santana (Cretáceo Inferior), Bacia do Arar- 249–256. ipe, nordeste do Brasil. Revista Brasileira de Entomo- logia. 34: 241–249. GÓMEZ-PALLEROLA, J.E. 1986. Nuevos insectos fósiles de les calizas litográficas del Cretácico Inferior del MARTYNOV, A.V. 1925. To the knowledge of fossil Montsec (Lérida). Boletín Geológico y Minero, 97(6), insects from Jurassic beds in Turkestan. 1. Raphidiopt- 717–736. era. Izvestiya Rossiiskoi Akademii Nauk, 6 seryia, 19: 235–246. GRIMALDI, D.A. 2000. A diverse fauna of Neuroptero- dea in amber from the Cretaceous of New Jersey. In: MARTYNOVA, O.M. 1947. Two new snakeflies from the Jurassic shales of Karatau. Doklady Akademii Nauk Grimaldi, D.A. (ed.) Studies on fossils in amber, with SSSR, 56: 635­–637. [In Russian]. particular reference to the Cretaceous of New Jersey. Leiden, Backhuys: 259­–303. MARTYNOVA, O.M. 1961. Recent and fossil snakeflies (Insecta: Raphidioptera). Paleontologicheskiy Zhurnal, GRIMALDI, D.A. & ENGEL, M.S. 2005. Evolution of the 1961: 73–83. [In Russian]. insects. Cambridge University Press, New York. NAVÁS, L. 1918. Monografia de l’ordre dels Rafidiòpters HAYWOOD, A.M., VALDES, P.J. & MARKWICK, P.J. (Ins.). Arxius [Arxivs] de l’Institute [d’Estudis Cata- 2005. Cretaceous (Wealden) climates: a modelling per- lans, Seccio] de Ciencias, Barcelona. 1–90. spective. Cretaceous Research, 25: 303–311. NEL, A., SÉMÉRIA, Y. & MARTINS-NETO, R.G. 1990. HONG, Y.-C. 1982. Mesozoic fossil insects of Jiuquan Un Raphidioptera fossile du Cretacé Inferieur du Bresil Basin in Gansu Province. Beijing: Geological Publish- (Neuropteroidea). Neuroptera International, 6: 27–37. ing House. [In Chinese]. OSWALD, J.D. 1990. Raphidioptera. In: Grimaldi, D.A. HONG, Y.-C. 1992a. A new family of Mesozoic snake- (ed.) Insects from the Santana Formation, Lower Creta- flies (Insecta, Raphidioptera) from the Laiyang Basin, ceous, of Brazil. Bulletin of the American Museum of China. Paleontologicheskiy Zhurnal, 101–105. [In Rus- Natural History, 195: 154–163. sian, with English summary]. PLANT, C.W. 1997. A key to the adults of British lacewings HONG, Y.-C. 1992b The study of early Cretaceous Cole- and their allies (Neuroptera, Megaloptera, Raphidop- optera, Raphidioptera, Diptera of Kezuo, West Liaon- tera and Mecoptera). Field Studies, 9: 179–269. ing Province. Acta Geologica Gansu, 1: 1–13.­ [In Chi- PONOMARENKO, A.G. 1988. New Mesozoic insects. nese, with English summary]. Trudy Sovmestnaya Sovestko-Mongolskaya Paleonto- HONG, Y.-C. & CHANG, J.-P. 1989. A new family ‘Jilino- logicheskogo Ekspeditsiya [Transactions of the Joint- raphidiidae’ of Raphidioptera (Insecta). Geoscience, Soviet-Mongolian Palaeontological Expedition], 33: 290–296. [In Chinese, with English summary]. 71–80, 94. [In Russian]. JARZEMBOWSKI, E.A. 1980. Fossil insects from the PONOMARENKO, A.G. 1993. Lower Cretaceous snake- Bembridge Marls (Palaeogene) of the Isle of Wight, flies from Transbaikalia. Trudy Paleontologicheskaya southern England. Bulletin of the British Museum (Nat- Instituta, Akademii Nauk, SSSR, 252: 60–80. [In Rus- ural History), London (Geology), 33 (4): 237–239. sian].

200 James E. JEPSON & Edmund A. JARZEMBOWSKI Tw o n e w s p e c i e s o f s n a k e f l y (In s e c t a : Ra p h i d i o p t e r a )

RASNITSYN, A.P., JARZEMBOWSKI, E.A. & ROSS, Curculionoidea) from El Montsec, Spain. Cretaceous A.J. 1998. Wasps (Insecta: Vespida = Hymenoptera) Research, 25: 555–564. from the Purbeck and Wealden (Lower Cretaceous) of WANG, W. 1987. Early Mesozoic insects from western southern England and their biostratigraphical and pal- Liaoning. Mesozoic stratigraphy and palaeontology of aeoenvironmental significance. Cretaceous Research, western Liaoning. 3: 202-222. Geological Publishing 19: 329–391. House, Beijing. [In Chinese, with English summary]. REN, D. 1994. A new discovery of snake-flies (Insecta) WATSON, J. & ALVIN, K.L. 1996. An English Wealden from the late Jurassic of Beipiao, Liaoning. Professional floral list, with comments on possible environmental Papers of Stratigraphy and Paleontology, 25: 131–140. indicators. Cretaceous Research, 17: 5–26. [In Chinese, with English summary]. WHALLEY, P.E.S. 1985. The systematics and palaeogeog- REN, D. 1997. Studies on the late Mesozoic snakeflies of raphy of the Lower Jurassic insects of Dorset, England. China (Raphidioptera: Baissopteridae, Mesoraphidii- Bulletin of the British Museum (Natural History) Geol- dae, Alloraphidiidae). Acta Zootaxonomica Sinica, 22: ogy, 39: 107–189. 172–188. [In Chinese, with English summary]. WHALLEY, P.E.S. 1988. Mesozoic Neuroptera and REN, D. & HONG, Y.-C. 1994. A cladistic study on the Raphidioptera (Insecta) in Britain. Bulletin of the Brit- familial phylogeny of fossil and living Raphidioptera ish Museum (Natural History) Geology, 44: 45–63. (Insecta). Bulletin of the Chinese Academy of Geological WHALLEY, P.E.S .& JARZEMBOWSKI, E.A. 1985. Sciences, 29: 103–117. [In Chinese, English summary]. Fossil insects from the Lithographic Limestone of REN, D., LU, L., GUO, Z. & JI, S. 1995. Faunae and Montsech (late Jurassic-early Cretaceous), Lérida stratigraphy of the Jurassic–Cretaceous in Beijing and Province, Spain. Bulletin of the British Museum (Natu- adjacent areas. Beijing: Seismic Publishing House. [In ral History) Geology, 38: 381–412. Chinese, with English summary]. WILLMANN, R. 1994. Raphidoidea aus dem Lias und SORIANO, C., GRATSHEV, V.G. & DELCLÒS, X. 2006. die Phylogenie der Kamelhalsfliegen (Insecta: Holom- New Early Cretaceous weevils (Insecta, Coleoptera, etabola). Paläontologische Zeitschrift, 68: 167–197.

Received: 5th June 2007 Accepted: 19th February 2008

201 CHAPTER 9

5. NEW SNAKEFLIES (INSECTA: RAPHIDIOPTERA) FROM THE

LOWER CRETACEOUS OF THE UK, SPAIN AND BRAZIL

116

Palaeontology

NEW SNAKEFLIES (INSECTA: RAPHIDIOPTERA) FROM THE LOWER CRETACEOUS OF THE UK, SPAIN AND BRAZIL

Journal: Palaeontology

Manuscript ID: PALA-11-09-2957-OA.R2

mstype: Original Article

Date Submitted by the n/a Author:

Complete List of Authors: Jepson, James; University of Manchester, School of Earth, Atmospheric and Environmental Sciences Ansorge, Jörg; Ernst-Moritz-Andt Universität Greifswald, Institut für Geographie und Geologie Jarzembowski, Ed; Maidstone Museum, Natural History

Baissopteridae, Mesoraphidiidae, Mesozoic, Crato Formation, Key words: Wealden, snakefly, Montsec

Palaeontology Page 1 of 33 Palaeontology

1 2 3 4 1 NEW SNAKEFLIES (INSECTA: RAPHIDIOPTERA) FROM 5 6 7 2 THE LOWER CRETACEOUS OF THE UK, SPAIN AND 8 9 10 3 BRAZIL 11 12 13 4 14 15 5 JAMES E. JEPSON*, JÖRG ANSORGE† and EDMUND A. JARZEMBOWSKI‡ 16 17 18 6 19 20 7 * School of Earth, Atmospheric and Environmental Sciences, University of 21 22 8 Manchester, Williamson Building, Oxford Road, Manchester, M13 9PL, UK. e-mail: 23 24 25 9 [email protected] 26 27 10 28 29 11 † Institut für Geographie und Geologie, Ernst-Moritz-Arndt Universität Greifswald, 30 31 32 12 Friedrich-Ludwig-Jahn Strasse 17a, D-17489 Greifswald, Germany. e-mail: 33 34 13 [email protected] 35 36 37 14 38 39 15 ‡ Maidstone Museum and Bentlif Art Gallery, St. Faith’s St., Maidstone, Kent ME14 40 41 16 1LH, UK. e-mail: [email protected] 42 43 44 17 45 46 18 Abstract: Several new taxa of snakeflies (Raphidioptera) are described from the 47 48 19 Lower Cretaceous deposits of the Wealden, UK (Barremian), Montsec, Spain 49 50 51 20 (Barremian) and the Crato Formation, Brazil (Aptian). Mesoraphidia ednae sp. nov. 52 53 21 and M. hilli sp. nov. are described from the Wealden; Nanoraphidia lithographica sp. 54 55 22 nov. and Iberoraphidia dividua gen. et sp. nov. are described from Montsec, and 56 57 58 23 Baissoptera lisae sp. nov. is described from the Crato Formation. The geographical 59 60 24 range of Nanoraphidia has potentially been extended.

1 Palaeontology Palaeontology Page 2 of 33

1 2 3 25 4 5 6 26 Key words : Baissopteridae, Crato Formation, Mesoraphidiidae, Mesozoic, Montsec, 7 8 27 Wealden, snakefly 9 10 11 28 12 13 29 RAPHIDIOPTERA are a member of the holometabolous superorder Neuropterida 14 15 30 along with the orders Neuroptera and Megaloptera (Aspöck, U. 2002). Snakeflies are 16 17 18 31 the least diverse order of the holometabolous insects with 225 extant species 19 20 32 described in two families Raphidiidae Latreille, 1810 and Inocelliidae Navás, 1913 21 22 33 (Aspöck, U. and Aspöck, H. 2003; Engel et al. 2006). Extant snakeflies are arboreal 23 24 25 34 and found in cold temperate regions; they require near-freezing temperatures to aid 26 27 35 development (Aspöck, H. et al. 1991; Aspöck, H. 2002; Grimaldi and Engel 2005). 28 29 36 Mesozoic snakeflies comprise three valid families: Baissopteridae Martynova, 1961, 30 31 32 37 Mesoraphidiidae Martynov, 1925, and Alloraphidiidae Carpenter, 1967, and a 33 34 38 doubtful family, Priscaenigmatidae Engel, 2002 (Aspöck, U. and Aspöck, H. 2004). 35 36 37 39 The fossil snakefly species, on the whole, are diagnosed on the basis of wing 38 39 40 venation, whereas extant species are described on body morphology, in particular 40 41 41 genitalic differences. In extant species wing venation is of little importance and it is 42 43 44 42 likely that the venation-based species of the Mesozoic represent a complex of species, 45 46 43 rather than a single species. This could mean that the Mesozoic diversity could be 47 48 44 much greater than previously thought, thus supporting the idea that extant snakeflies 49 50 51 45 are relict (Aspöck, H. et al. 1991; Engel 2002). This could also be true for other 52 53 46 neuropterid groups (e.g. Myrmeleontiformia). Unfortunately, this is difficult to prove 54 55 47 in the fossil record without preserved terminalia. 56 57 58 48 Only one other snakefly has been formally described from the Lower 59 60 49 Cretaceous Wealden Supergroup, Proraphidia hopkinsi Jepson and Jarzembowski,

2 Palaeontology Page 3 of 33 Palaeontology

1 2 3 50 2008. P. hopkinsi belongs to the family Mesoraphidiidae and four species of the 4 5 6 51 nominotypical genus Mesoraphidia occur in the underlying Lower Cretaceous 7 8 52 Purbeck Limestone Group (Jepson et al. 2009). In addition, Metaraphidia is known 9 10 11 53 from the Jurassic of Dorset (Lower Lias, Charmouth) (Whalley 1988). Montsec has 12 13 54 had one species of snakefly formally described ( Proraphidia gomezi Jepson and 14 15 55 Jarzembowski, 2008) in the family Mesoraphidiidae. The Crato Formation has four 16 17 18 56 species of snakefly described in two families Baissopteridae: Austroraphidia 19 20 57 brasiliensis Nel et al., 1990, Baissoptera brasiliensis Oswald, 1990, B. pulchra 21 22 58 (Martins-Neto and Nel, 1992) and familia incertae sedis: Arariperaphidia rochai 23 24 25 59 Martins-Neto and Vulcano, 1990. 26 27 60 28 29 61 GEOLOGICAL SETTINGS 30 31 32 62 33 34 63 Upper Weald Clay, UK 35 36 37 64 38 39 65 The Upper Weald Clay is the youngest formation of the Lower Cretaceous Wealden 40 41 66 Supergroup. The age of the Upper Weald Clay is Barremian, and the specimens 42 43 44 67 described herein were collected from two localities: Auclaye Brickworks, Surrey and 45 46 68 Rudgwick Brickworks, Sussex. The Upper Weald Clay represents a lacustrine- 47 48 69 lagoonal environment, with a warm-temperate to sub-tropical climate (Allen 1998; 49 50 51 70 Haywood et al . 2005). Many insects have been discovered in the Upper Weald Clay 52 53 71 including the orders Odonata, Grylloblattodea, Orthoptera, Neuroptera, 54 55 72 Raphidioptera, Hymenoptera, Diptera, and Trichoptera (Jarzembowski 1984, 1995). 56 57 58 73 59 60

3 Palaeontology Palaeontology Page 4 of 33

1 2 3 74 Montsec, Spain 4 5 6 75 7 8 76 The lithographic limestones of Sierra del Montsec outcrop at two sites in the 9 10 11 77 mountain range Montsec de Rubies, Lerida Province, northeast Spain. The La Cabrua 12 13 78 locality is situated at a steep local road near Santa Maria de Meia; the former quarry is 14 15 79 near the abandoned village Rubies. The carbonate mud which builds up the 16 17 18 80 lithographic limestones was deposited in the deepest parts of a fresh or brackish water 19 20 81 lagoon without direct connection to the epicontinental sea (Ansorge 1993; Fregenal- 21 22 82 Martinez and Melendez 1995). Deposition of charophytic limestones took place in 23 24 25 83 shallow water. The age of the lithographic limestones has been determined as Lower 26 27 84 Barremian based on the occurrence of the charophyte Atopochara trivolvis triquetra 28 29 85 Grambast, 1968 (morphotype ‘typique’) (Ansorge 1993; Martín-Closas and López- 30 31 32 86 Móron 1995). The rich and well-preserved insect fauna of Montsec consists of aquatic 33 34 87 larvae and imagines. Up to now representatives of the following insect orders have 35 36 37 88 been recorded: Ephemeroptera, Odonata, Blattodea, Isoptera, Orthoptera, Hemiptera, 38 39 89 Psocoptera, Trichoptera, Diptera, Hymenoptera, Coleoptera, Neuroptera and 40 41 90 Raphidioptera (Ansorge 1993; Martinez-Delclos and Nel 1995; Whalley and 42 43 44 91 Jarzembowski 1985). The most common fossil insect is the aquatic larva 45 46 92 Mesopalingea leridae Whalley and Jarzembowski, 1985. Cockoaches and beetles are 47 48 93 the most common flying insects. Together with the plant macrofossils (Barale 1995) 49 50 51 94 the insects are typical of a drier climate with seasonal rainfall. 52 53 95 54 55 96 Crato Formation, Brazil 56 57 58 97 59 60

4 Palaeontology Page 5 of 33 Palaeontology

1 2 3 98 The specimen described herein is from the Nova Olinda Member of the Crato 4 5 6 99 Formation, Brazil. This is the lowest unit of the formation, Aptian in age, consisting 7 8 100 of finely laminated limestones deposited in a hypersaline lagoonal environment 9 10 11 101 (Martill et al . 2007). The Crato Formation represents a Konservat Lagerstätte and is 12 13 102 famous for its exceptionally preserved and diverse insect assemblage, with 21 14 15 103 hexapod orders present, including, Diplura, , Ephemeroptera, Odonata, 16 17 18 104 Dermaptera, Mantodea, ‘Blattaria’, Isoptera, Chresmododea, , 19 20 105 Orthoptera, Hemiptera, Neuroptera, Raphidioptera, Megaloptera, Coleoptera, 21 22 106 Hymenoptera, Mecoptera, Diptera, Trichoptera and Lepidoptera (see chapters in 23 24 25 107 Martill et al . 2007). 26 27 108 28 29 109 MATERIAL AND METHODS 30 31 32 110 33 34 111 The studied material is deposited in the following collections: Booth Museum of 35 36 37 112 Natural History, Brighton (BMB), Maidstone Museum and Bentlif Art Gallery, Kent 38 39 113 (MNEMG), Sedgwick Museum of Earth Sciences, Cambridge (CAMSM), Institute of 40 41 114 Geography and Geology, University of Greifswald (FGWG) and 42 43 44 115 Geowissenschaftliches Zentrum der Universität Göttingen (GZG). 45 46 116 Mesoraphidia ednae sp. nov. was drawn by JEJ using an Olympus SZH 47 48 117 stereomicroscope with a camera lucida attachment. The drawing is a composite, 49 50 51 118 merging the part and counterpart to establish a more complete venation; this was 52 53 119 achieved using a light box. 10% ethanol was used to enhance vein detail. Photographs 54 55 120 were taken using a Canon EOS 450D digital SLR camera mounted on a rostrum 56 57 58 121 stand. 59 60

5 Palaeontology Palaeontology Page 6 of 33

1 2 3 122 The Mesoraphidia hilli sp. nov. holotype was drawn by Robert A. Coram 4 5 6 123 (British Fossils) and EAJ using a camera lucida attachment on a Zeiss Stemi SVII 7 8 124 microscope and photographed by JEJ using a Canon EOS 450D digital SLR camera 9 10 11 125 mounted on a rostrum stand. The M. hilli sp. nov. paratype was photographed by 12 13 126 Tony Mitchell (volunteer Maidstone Museum) using a Nikon Coolpix 5000 digital 14 15 127 camera. 16 17 18 128 The Montsec specimens were drawn by JA using a camera lucida attached to a SM 19 20 129 XX Zeiss stereomicroscope and photographed with a Nikon Coolpix 4500 mounted in 21 22 130 a macrophotographic bellows. 23 24 25 131 Wing vein abbreviations are as follows: C, costa; Sc, subcosta; R, radius; Rs, 26 27 132 radial sector; Rs1, first branch of Rs; Rs2, second branch of Rs; M, media; MA, 28 29 133 anterior media; MP, posterior media; MP1, first branch of MP; MP2, second branch 30 31 32 134 of MP; Cu, cubitus; CuA, anterior cubitus; CuP, posterior cubitus; A, analis; r, radial 33 34 135 cell; dc, discal cell; mc, medial cell; ac, anal cell; p, pterostigma. 35 36 37 136 38 39 137 SYSTEMATIC PALAEONTOLOGY 40 41 138 42 43 44 139 Order RAPHIDIOPTERA Navás, 1918 45 46 140 Family MESORAPHIDIIDAE Martynov, 1925 47 48 141 Genus MESORAPHIDIA Martynov, 1925 49 50 51 142 52 53 143 Type species: Mesoraphidia grandis Martynov, 1925 from the Late Jurassic of 54 55 144 Kazakhstan. 56 57 58 145 59 60 146 Mesoraphidia ednae sp. nov.

6 Palaeontology Page 7 of 33 Palaeontology

1 2 3 147 Text-fig. 1 4 5 6 148 7 8 149 Derivation of name . In honour of Edna Clifford who inspired the online EDNA fossil 9 10 11 150 insect database. 12 13 151 14 15 152 Holotype. MNEMG 2009.42 [AY36a, b], disarticulated forewing, Auclaye 16 17 18 153 Brickworks, Upper Weald Clay, Barremian. 19 20 154 21 22 155 Diagnosis. Separated from all other Mesoraphidia species by combination of MA 23 24 25 156 bifurcating at exact point of fork of MP, long basal branch of MA, small size of mc3, 26 27 157 wide distance between cua-cup crossveins, and 3A forking close to posterior wing 28 29 158 margin. 30 31 32 159 33 34 160 Description . Holotype forewing (Text-fig. 1): narrow and elongate; 8 mm long and 35 36 37 161 2.6 mm wide. C complete; costal area preserved with five crossveins. Sc terminates 38 39 162 approximately 3.4 mm from wing base (just past midpoint of wing) on anterior wing 40 41 163 margin. R complete and long, origin 7.5 mm from wing apex, terminating near wing 42 43 44 164 apex; three crossveins present between R and C, 0.8 mm, 1.1 mm and 2.8 mm from 45 46 165 wing apex. Pterostigma present; over 1.7 mm long; beginning of pterostigma difficult 47 48 166 to determine as pigment fades proximally. Three radial cells present: 1r: 1.6 mm long 49 50 51 167 and 0.5 mm wide; 2r: 1.9 mm long and 0.9 mm wide and 3r: 1.3 mm long and 0.3 mm 52 53 168 wide. Rs arises from R approximately 4.9 mm from wing base. Rs two branched; Rs1 54 55 169 and Rs2 simple; both Rs1 and Rs2 arise in distal part of wing near apex. dc formed by 56 57 58 170 a crossvein between Rs and MA, 1.9 mm long and 0.5 mm wide (at widest point). MA 59 60 171 arises 1.4 mm from R where M forks into MA and MP; it then runs into Rs 0.7 mm

7 Palaeontology Palaeontology Page 8 of 33

1 2 3 172 from fork with M. MA coalesced with Rs for 1.1 mm. MA two branched with both 4 5 6 173 branches simple. Two crossveins present between MA and MP. MP deeply forked 7 8 174 with each branch forked near posterior wing margin. Three cells present between MP 9 10 11 175 and CuA: mc1: 0.9 mm long, 0.5 mm wide; mc2: 1 mm long, 0.5 mm wide and mc3: 12 13 176 0.6 mm long, 0.3 mm wide. CuA coalesced with M for 0.1 mm. CuA forks near 14 15 177 posterior wing margin, distal branch forked, basal branch simple. Two crossveins 16 17 18 178 present between CuA and CuP; CuP simple. Anal area is preserved with 1A simple, 19 20 179 2A and 3A fused dividing 0.2 mm from posterior wing margin; 2A simple and 3A 21 22 180 forked near posterior wing margin. ac oval shaped, 1.1 mm long, 0.4 mm wide. 23 24 25 181 26 27 182 Remarks. Placed in Mesoraphidia Martynov, 1925 on the basis of typical venation, 28 29 183 having a long pterostigma and MA arising at the origin of MP. 30 31 32 184 33 34 185 Mesoraphidia hilli sp. nov. 35 36 37 186 Text-figs. 2–3 38 39 187 40 41 188 Derivation of name. In honour of the collector of the holotype, Prof. Chris R. Hill 42 43 44 189 (Nanjing Institute of Geology and Palaeontology). 45 46 190 47 48 191 Holotype. BMB 016381, 016382 (part and counterpart), disarticulated hind wing, 49 50 51 192 Rudgwick Brickworks, Upper Weald Clay, Barremian. 52 53 193 54 55 194 Paratype. BMB 025010, 025011 (part and counterpart), disarticulated wing, fragment 56 57 58 195 of apical half, Rudgwick Brickworks, Upper Weald Clay, Barremian. 59 60 196

8 Palaeontology Page 9 of 33 Palaeontology

1 2 3 197 Diagnosis. Separated from other Mesoraphidia species by the combination of Sc 4 5 6 198 terminating well past midpoint of wing above centre of 2r, a near horizontal (relative 7 8 199 to the anterior wing margin) cua-mp2 crossvein, 9 10 11 200 12 13 201 Description. Holotype hind wing (Text-fig. 2): narrow and elongate; 11.5 mm long 14 15 202 and 7.1 mm wide. C incomplete; costal area preserved with four crossveins. Sc 16 17 18 203 terminates approximately 4.3 mm from wing base (1.9 mm past mid-point of wing) on 19 20 204 anterior wing margin. R complete and long, origin 7.5 mm from wing apex, 21 22 205 terminating far from wing apex; two crossveins present between R and C, 2.7 mm and 23 24 25 206 3.6 mm from wing apex. Pterostigma present; over 0.9 mm long; beginning of 26 27 207 pterostigma difficult to determine as pigment fades towards wing apex. Three radial 28 29 208 cells present: 1r: 3.8 mm long and 0.9 mm wide with crossvein halfway; 2r: 3 mm 30 31 32 209 long and 0.7 mm wide and 3r: 1.9 mm long and 0.3 mm wide curving towards 33 34 210 anterior wing margin. Rs arises from R approximately 9.1 mm from wing apex. Rs 35 36 37 211 two branched; Rs1 branches pectinately to form four branches at antero-posterior 38 39 212 wing margin, Rs2 simple; both Rs1 and Rs2 arise in distal part of wing near apex. dc 40 41 213 3 mm long and 0.8 mm wide (at widest point). MA arises close to wing base, 10.9 42 43 44 214 mm from wing apex. MA runs into Rs, 2.2 mm from origin. MA coalesced with Rs 45 46 215 for 2.2 mm. MA is dichotomously branched, splitting into four branches at posterior 47 48 216 wing margin. Two crossveins present between MA and MP and MP1 and MP2. MP 49 50 51 217 origin 11.4 mm from wing apex. MP appears to be deeply forked; however, fork not 52 53 218 preserved. MP1 forks 3.6 mm from posterior wing margin, first branch forks 1.4 mm 54 55 219 from posterior wing margin; second branch incomplete, however, most likely forked, 56 57 58 220 based on vein fragment terminating on posterior wing margin next to first branch. 59 60 221 MP2 forks 1.4 mm from posterior wing margin; first branch forks 1.62 mm from

9 Palaeontology Palaeontology Page 10 of 33

1 2 3 222 posterior wing margin, with most distal branch possibly forking again but incomplete. 4 5 6 223 Second branch forks 1 mm from posterior wing margin. Medial cell: 1.3 mm long, 0.5 7 8 224 mm wide; mc2: 1 mm long, 0.8 mm wide. One crossvein between CuA and MP and 9 10 11 225 two crossveins between MP2 and CuA. CuA simple for most of its length forking 12 13 226 0.75 mm from posterior wing margin. CuP incomplete, termination on posterior wing 14 15 227 margin preserved. Anal area not preserved. 16 17 18 228 19 20 229 21 22 230 Description of paratype. Apical wing fragment (Text-fig. 3): 6.4 mm preserved 23 24 25 231 length, 4 mm preserved width. C incomplete, costal area incomplete. Sc incomplete, 26 27 232 terminates 4.5 mm from wing apex, on anterior wing margin. R incomplete, 28 29 233 terminating before wing apex. Three crossveins between R and C: 1.5 mm, 2.5 mm 30 31 32 234 and 3.3 mm away from wing apex. Pterostigma preserved 1.9 mm long, faded towards 33 34 235 beginning. Three radial cells, 1r: incomplete 1.5 mm preserved length, 0.8 mm width, 35 36 37 236 2r: 2.4 mm long, 0.8 mm wide and 3r: 2.2 mm long, 0.6 mm wide. Rs origin not 38 39 237 preserved, Rs separates from MA 5.7 mm from wing apex. Rs two-branched, with 40 41 238 Rs1 forking 1.8 mm from wing margin, first branch forking 0.8 mm from posterior 42 43 44 239 wing margin, second branch simple. dc 2.5 mm long and 0.7 mm wide (at widest 45 46 240 point). MA origin not preserved, dichotomously branched, first fork 3.2 mm from 47 48 241 posterior wing margin, MA1 forks 1 mm from wing margin, and MA2 forks 1.8 mm 49 50 51 242 from posterior wing margin. Two crossveins between MA and MP. MP incomplete, 52 53 243 origin not preserved. MP1 dichotomously branched 2.5 mm from posterior wing 54 55 244 margin, first branch forks 1.3 mm from posterior wing margin, second branch forks 56 57 58 245 0.8 mm from posterior wing margin. MP2 forks 1.7 mm from posterior wing margin, 59 60 246 first branch forks 1 mm from posterior wing margin, second branch simple. Part of

10 Palaeontology Page 11 of 33 Palaeontology

1 2 3 247 mc2? preserved, but incomplete. CuA partially preserved. CuP and anal area not 4 5 6 248 preserved. 7 8 249 9 10 11 250 Remarks. It is unsure whether the specimen represents a fore- or hind wing due to the 12 13 251 fragmentary nature of the wing. The paratype has been provisionally attributed to 14 15 252 Mesoraphidia hilli sp. nov. subject to more material being found. The provisional 16 17 18 253 placement is on the basis of similarity in the structure of Rs, a dichotomously 19 20 254 branched MA and the termination of Sc being well past the midpoint of the wing. The 21 22 255 placement is tentative due to the fragmentary nature of the wing. 23 24 25 256 26 27 257 Genus NANORAPHIDIA Engel, 2002 28 29 258 30 31 32 259 Type species: Nanoraphidia electroburmica Engel, 2002 from the Lower Cretaceous 33 34 260 Burmese amber. 35 36 37 261 38 39 262 Nanoraphidia lithographica sp. nov. 40 41 263 Text-Fig. 4 42 43 44 264 45 46 265 Derivation of name. The species epithet lithographica is derived from the rock source 47 48 266 – a fine-grained lithographic limestone. 49 50 51 267 52 53 268 Holotype. FGWG 147 (MA 20 a,b, part and counterpart) disarticulated body with four 54 55 269 overlapping wings, La Cabrua outcrop, Sierra del Montsec, Lower Barremian; for 56 57 58 270 locality reference see Ansorge (1993, pp. 21–23) and Fregenal- 59 60 271 Martinez and Melendez (1995).

11 Palaeontology Palaeontology Page 12 of 33

1 2 3 272 4 5 6 273 Diagnosis. Differs from N. electroburmica in having larger size ( N. lithographica, 5.3 7 8 274 mm, N. electroburmica , 4.26 mm), radial cells and dc narrower, CuA with one fork, 9 10 11 275 forking further away from posterior wing margin. 12 13 276 14 15 277 Description. Holotype forewing (Text-fig. 4): 5 mm preserved length, estimated 16 17 18 278 length 5.3 mm. C incomplete. Sc incomplete, termination not observed. R complete 19 20 279 and long, terminating 0.4 mm from wing apex; one crossvein between R and C 0.6 21 22 280 mm from wing apex. Pterostigma not preserved, seems to be absent. Three radial cells 23 24 25 281 present: 1r: incomplete, 0.4 mm wide; 2r: 1.1 mm long and 0.4 mm wide and 3r: 0.6 26 27 282 mm long and 0.2 mm wide. Rs two branched, both branches simple; both Rs1 and Rs2 28 29 283 arise in distal part of wing near apex. dc 1 mm long and 0.3 mm wide (at widest 30 31 32 284 point). MA arises after fork of MP and CuA. MA coalesced with Rs for 1.2 mm. MA 33 34 285 two branched with both branches simple. Two crossveins present between MA and 35 36 37 286 MP. MP deeply forked with MP1 forking 0.8 mm and MP2 forking 0.7 mm from 38 39 287 posterior wing margin. One medial cell posterior to MP between MP and CuA: mc1: 40 41 288 0.8 mm long, 0.4 mm wide and mc2: 0.8 mm long, 0.4 mm. CuA coalesced with M 42 43 44 289 distally. CuA forks 0.6 mm from posterior wing margin. One cua-cup crossvein 45 46 290 partially preserved before M-CuA separation. CuP, 1A, 2A and 3A incomplete. 47 48 291 49 50 51 292 Body parts. Poorly preserved; part of thorax and abdomen present. Thorax preserved 52 53 293 length 2.3 mm. Abdomen preserved: length 1.8 mm. 54 55 294 56 57 58 295 Remarks . All four wings of the specimen are overlapping obscuring the venation, 59 60 296 making it difficult to distinguish. The hind wing venation was indecipherable. The

12 Palaeontology Page 13 of 33 Palaeontology

1 2 3 297 specimen has been tentatively placed in the genus Nanoraphidia due to the 4 5 6 298 combination of its small size and venation: the one medial cell (separating it from 7 8 299 Mesoraphidia ) and the first cua-cup crossvein occurring before the separation of M 9 10 11 300 and CuA. Other diagnostic features are not preserved, e.g. the costal area and 12 13 301 pterostigma (possibly an artefact of preservation); however, the pterostigmal area has 14 15 302 no crossveins which is consistent with Nanoraphidia . Nanoraphidia lithographica sp. 16 17 18 303 nov. has a larger wing size (estimated 5.3 mm) than N. electroburmica (4.26 mm); the 19 20 304 size of the wing is similar to other snakeflies, e.g. Mesoraphidia parvula (wing length 21 22 305 5.5 mm (Martynov 1925)), M. gaoi (wing length 5 mm (Ren et al . 1995)) , M. 23 24 25 306 mitchelli (estimated wing length 5.4 mm) , M. purbeckensis (estimated wing length 5.1 26 27 307 mm) , M. websteri (estimated wing length 5.7 mm) (Jepson et al . 2009). However, the 28 29 308 venation is sufficiently different from these species and closer to Nanorapidia to 30 31 32 309 allow a tentative placement in the latter genus. The geographical disjunction of N. 33 34 310 lithographica sp. nov. in Spain and N. electroburmica in Burma, again only supports 35 36 37 311 a tentative and not a confident placement within Nanoraphidia . 38 39 312 40 41 313 Genus IBERORAPHIDIA gen . nov . 42 43 44 314 45 46 315 Type and only species . Iberoraphidia dividua gen. et sp. nov. 47 48 316 49 50 51 317 Derivation of name. After Iberia, the Iberian Peninsula comprising Spain and Portugal 52 53 318 and Raphidia , a raphidiopteran genus. 54 55 319 56 57 58 320 Diagnosis . Divided pterostigma separates it from all other known snakeflies. MA 59 60 321 separates distinctly prior to fork of MP, which separates it from almost all

13 Palaeontology Palaeontology Page 14 of 33

1 2 3 322 Mesoraphidiidae and Alloraphidiidae - only Ororaphidia Engel and Ren, 2008 shows 4 5 6 323 a similar shift of MA towards the wing base. 7 8 324 9 10 11 325 Iberoraphidia dividua sp. nov. 12 13 326 Text-fig. 5 14 15 327 16 17 18 328 Derivation of name. Meaning divided, in reference to the divided pterostigma. 19 20 329 21 22 330 Diagnosis . As for the genus. 23 24 25 331 26 27 332 Holotype. GZG.RF.7563 (part and counterpart), isolated forewing; La Cabrua 28 29 333 outcrop, Sierra del Montsec, Lower Barremian, for locality reference see Ansorge 30 31 32 334 (1993, pp. 21–23) and Fregenal-Martinez and Melendez (1995). 33 34 335 35 36 37 336 Description. Holotype forewing (Text-fig. 5): 12.5 mm long. C complete; costal area 38 39 337 preserved with six crossveins (originally seven - basal one is extremely faint). 40 41 338 Macrotrichiae on C and the distal part of R (Text-fig. 5C). Sc terminates 42 43 44 339 approximately 5.3 mm from wing apex (just past midpoint of wing) on anterior wing 45 46 340 margin. R complete and long, terminating near wing apex; one crossvein between R 47 48 341 and Sc 9.5 mm from wing apex, six crossveins present between R and C, 0.6 mm, 1.3 49 50 51 342 mm, 2.1 mm, 2.5 mm, 4.3 mm and 4.8 mm from wing apex. Pterostigma divided into 52 53 343 two, over 1.7 mm long; beginning of first part of pterostigma 4.8 mm from wing apex, 54 55 344 end of first part 4.3 mm, beginning of second part 2.5 mm and end 2.1 mm. Three 56 57 58 345 radial cells present: 1r: 3.2 mm long and 0.9 mm wide; 2r: 2.5 mm long and 1 mm 59 60 346 wide and 3r: 2.5 mm long and 0.6 mm wide. 3r narrows towards apex of wing. Rs

14 Palaeontology Page 15 of 33 Palaeontology

1 2 3 347 arises from R approximately 8 mm from wing apex. Rs two branched; Rs1 forks at its 4 5 6 348 midpoint and first branch forks again 0.7 mm from wing margin; Rs2 is simple; both 7 8 349 Rs1 and Rs2 arise in distal part of wing near apex. dc 2.5 mm long and 0.9 mm wide 9 10 11 350 (at widest point). MA arises 3 mm from R, where M forks into MA and MP; it then 12 13 351 runs into Rs 0.8 mm from split with M. MA coalesced with Rs for 2.1 mm. MA two 14 15 352 branched with MA1 forking 2 mm from posterior wing margin; MA2 simple. Two 16 17 18 353 crossveins present between MA and MP. MP deeply forked with MP1 forking 2 mm 19 20 354 and first branch forking 1.8 mm from posterior wing margin; MP2 forks 1.2 mm from 21 22 355 posterior wing margin. Three cells present between MP and CuA: mc1: 1.3 mm long, 23 24 25 356 0.5 mm wide; mc2: 1.6 mm long, 0.9 mm wide and mc3: 1.5 mm long, 0.8 mm wide. 26 27 357 CuA coalesced with M for 1.7 mm. CuA forks near posterior wing margin; distal 28 29 358 branch forked 0.7 mm from wing margin, basal branch simple. Two crossveins 30 31 32 359 present between CuA and CuP; CuP simple. One crossvein between CuP and 1A at 33 34 360 midpoint of these two veins. Anal area preserved with 1A simple, 2A and 3A present. 35 36 37 361 ac oval shaped, 1.9 mm long, 0.7 mm wide. 38 39 362 40 41 363 Remarks. The species is excluded from the family Baissopteridae due to its less 42 43 44 364 complex venation, and from Alloraphidiidae due to the basal shift of MA. It has been 45 46 365 placed in Mesoraphidiidae due to the typical venation and long pterostigmal area. The 47 48 366 divided pterostigma is a unique feature of Iberoraphidia dividua gen. et sp. nov. 49 50 51 367 distinguishing it from all other Mesoraphidiidae. 52 53 368 54 55 369 Family BAISSOPTERIDAE Martynova, 1961 56 57 58 370 Genus BAISSOPTERA Martynova, 1961 59 60 371

15 Palaeontology Palaeontology Page 16 of 33

1 2 3 372 Type species . Baissoptera martinsoni Martynova, 1961 from the Lower Cretaceous of 4 5 6 373 Baissa, Russia. 7 8 374 9 10 11 375 Baissoptera lisae sp. nov. 12 13 376 Text-fig. 6 14 15 377 16 17 18 378 Derivation of name . Named after Lisa Abbott, geology student at the University of 19 20 379 Manchester. 21 22 380 23 24 25 381 Holotype . CAMSM X.50155.2, female insect, some body parts preserved and fore- 26 27 382 and hind wings overlapping on left and right side, Crato Formation, Brazil, Lower 28 29 383 Cretaceous (Aptian). The provenance of the specimen has been inferred from the 30 31 32 384 characteristic nature of the sediment and preservation of the insect. No provenance 33 34 385 details were supplied with the specimen. 35 36 37 386 38 39 387 Diagnosis . Separated from other species of Baissoptera by a combination of Sc and C 40 41 388 anastomising very proximally to pterostigma; few crossveins and cells in medio- 42 43 44 389 radial area, but crossveins present forming two gradate series in posterior part of wing 45 46 390 R forking distally in hind wing. Small body size (excluding ovipositor); head and 47 48 391 pronotum, almost same size, head does not narrow posteriorly, separating it from B. 49 50 51 392 brasiliensis . 52 53 393 54 55 394 Description . Holotype forewing (Text-fig. 6A): 12.4 mm preserved length, 3.7 mm 56 57 58 395 preserved width. C complete; costal area poorly preserved with three crossveins. Sc 59 60 396 terminates approximately 5.5 mm from wing apex (past mid-point of wing) on

16 Palaeontology Page 17 of 33 Palaeontology

1 2 3 397 anterior wing margin. R complete and long but terminating before wing apex; two 4 5 6 398 crossveins between R and C 1.8 mm and 4 mm from wing apex, one crossvein present 7 8 399 between R and Sc 11 mm from wing apex. Pterostigma preserved; 2 mm long, 0.3 9 10 11 400 mm wide between two r-c crossveins. Three radial cells present: 1r: 2.3 mm long and 12 13 401 0.7 mm wide; 2r: 1.3 mm long and 0.6 mm wide and 3r: 1.2 mm long and 0.3 mm 14 15 402 wide. 3r narrows towards apex of wing due to fork of Rs. Rs arises from R 16 17 18 403 approximately 8 mm from wing apex. Rs four branched; Rs1 forks near wing apex, 19 20 404 other branches simple. MA origin not preserved. MA coalesced with Rs for 0.8 mm. 21 22 405 MA simple most of length, forking close to posterior wing margin. Two crossveins 23 24 25 406 present between MA and MP. MP appears deeply forked, although point of forking 26 27 407 not preserved; both branches forked; cubital and anal veins not preserved. 28 29 408 30 31 32 409 Hind wing : 12.5 mm preserved length and 3.5 mm preserved width. C incomplete; 33 34 410 costal area preserved with three crossveins. Sc terminates approximately 5.7 mm from 35 36 37 411 wing base (past mid-point of wing) on anterior wing margin. R long, origin not 38 39 412 preserved, forking before termination, terminates away from wing apex. Two 40 41 413 crossveins preserved between R and C, 1.5 mm and 3.6 mm from wing apex. 42 43 44 414 Pterostigma present, incomplete. Three radial cells present: 1r: 2.6 mm long and 0.7 45 46 415 mm wide with crossvein halfway; 2r: 1.3 mm long and 0.4 mm wide and 3r: 1 mm 47 48 416 long and 0.3 mm wide. Rs arises from R approximately 7.9 mm from wing apex. Rs 49 50 51 417 four branched, Rs1 forking near wing margin. MA arises close to wing base, 10 mm 52 53 418 from wing apex. MA runs into Rs, 2.3 mm from origin. MA coalesced with Rs for 1 54 55 419 mm. MA dichotomously branched, splitting into four branches at posterior wing 56 57 58 420 margin. Four crossveins present between MA and MP. MP origin 10 mm from wing 59 60 421 apex, very close to MA. MP deeply forked. MP1 forks at 1.1 mm and 1.4 mm from

17 Palaeontology Palaeontology Page 18 of 33

1 2 3 422 posterior wing margin. MP2 dichotomously forked, main fork 1.3 mm from posterior 4 5 6 423 wing margin, both branches forking 0.5 mm from posterior wing margin. Three 7 8 424 crossveins between CuA and MP. CuA incomplete. CuP and anal area not preserved. 9 10 11 425 12 13 426 Head . 2.9 mm long, 1.8 mm wide. Eyes partially preserved, left eye better preserved 14 15 427 than right, 1 mm long, 0.5 mm wide. Three ocelli present and preserved in a triangular 16 17 18 428 pattern close together. Antennae incomplete, partially preserved, scape 0.3 mm long, 19 20 429 0.2 mm wide. 21 22 430 23 24 25 431 Thorax. Pronotum elongate 3.1 mm long, 1.5 mm wide, enclosing prothorax. Meso- 26 27 432 and metathorax poorly preserved, combined length 4.4 mm. Legs preserved, obscured 28 29 433 by sediment (legs go down vertically into the rock). 30 31 32 434 33 34 435 Abdomen. Shorter than wings, 6 mm long (excluding ovipositor), 1.8 mm wide, 35 36 37 436 poorly preserved; elongate ovipositor incomplete; preserved length 2.3 mm. 38 39 437 40 41 438 Remarks : The specimen is preserved in three-dimensions with many parts of the fossil 42 43 44 439 (e.g. the legs) going down vertically into the rock concealing features of the body 45 46 440 morphology. It is tentatively placed in the genus Baissoptera because of the numerous 47 48 441 crossveins arranged in gradate series posteriorly, this being a diagnostic character of 49 50 51 442 the genus (Oswald 1990). The wing venation is similar to the Crato Formation genus 52 53 443 Austroraphidia Willmann, 1994; however, B. lisae sp. nov. differs from it in having 54 55 444 arranged crossveins, more branches of Rs and MA, and having a more elongate wing. 56 57 58 445 The absent/incomplete cubital and anal areas have meant only a tentative placement. 59 60 446 B. lisae sp. nov. differs from the Brazilian Baissoptera brasiliensis in having a smaller

18 Palaeontology Page 19 of 33 Palaeontology

1 2 3 447 body length (B. lisae sp. nov. 16.5 mm, B. brasiliensis 19.5 mm), smaller wing size 4 5 6 448 (B. brasiliensis larger), and the ratio of size of head to pronotum is greater in B. 7 8 449 brasiliensis sp. nov. Also the head of B. lisae sp. nov. does not taper posteriorly. 9 10 11 450 12 13 451 DISCUSSION 14 15 452 16 17 18 453 This study has increased the number of species of snakeflies in the Wealden 19 20 454 from one to three. The Cretaceous snakeflies in the UK are not diverse at family or 21 22 455 generic level, being represented by only one family, Mesoraphidiidae, and two 23 24 25 456 genera, Mesoraphidia and Proraphidia (Jepson and Jarzembowski 2008; Jepson et al . 26 27 457 2009). Mesoraphidiidae were present in the UK from the Jurassic (Lias) to the 28 29 458 Cretaceous (Weald Clay: Barremian), with specimens found in between in the 30 31 32 459 Purbeck Limestone Group (Berriasian) (Whalley 1988; Jepson et al . 2009). The genus 33 34 460 Mesoraphidia has also had a long history in the UK, being recorded first in the Lower 35 36 37 461 Berriasian of the Purbeck Limestone Group by Mesoraphidia mitchelli (Jepson et al . 38 39 462 2009) and last recorded in the Upper Weald Clay by Mesoraphidia ednae sp. nov. and 40 41 463 M. hilli sp. nov. The genus developed from being small, just greater than 5 mm in 42 43 44 464 wing length, e.g. M. mitchelli, M. purbeckensis , M. websteri , getting larger (over 10 45 46 465 mm wing length, a more normal size for the genus) in the later Berriasian M. 47 48 466 durlstonensis and the Barremian M. ednae sp. nov. and M. hilli sp. nov., possibly 49 50 51 467 linked to the changing climate from semi-arid to humid from the early Purbeck to 52 53 468 Wealden (Jepson et al . 2009). 54 55 469 Spain has a slightly more diverse Cretaceous fauna with three genera of 56 57 58 470 Mesoraphidiidae: Proraphidia (Jepson and Jarzembowski 2008), Nanoraphidia and 59 60 471 Iberoraphidia gen. nov. herein. The discovery of a possible Nanoraphidia in Montsec

19 Palaeontology Palaeontology Page 20 of 33

1 2 3 472 could potentially mean that the genus is older and more geographically widespread 4 5 6 473 than previously thought. Originally the genus was only known from the younger 7 8 474 Burmese amber (Engel 2002). It is interesting to note that as yet no Mesoraphidia 9 10 11 475 have been found in Montsec. 12 13 476 The Brazilian snakefly is younger than the European ones being of Aptian age. 14 15 477 Other than Baissoptera lisae sp. nov., the Brazilian snakefly fauna includes other 16 17 18 478 baissopterids: Austroraphidia brasiliensis Nel et al ., 1990, Baissoptera brasiliensis 19 20 479 Oswald, 1990, B. pulchra (Martins-Neto and Nel, 1992); also familia incertae sedis 21 22 480 Arariperaphidia rochai Martins-Neto and Vulcano, 1990 and a Raphidioptera sp. 23 24 25 481 (Oswald 1990). The Brazilian snakeflies are very important from a historical 26 27 482 biogeography point of view, because prior to their discovery, snakeflies were only 28 29 483 observed in the northern hemisphere. The southernmost regions where extant 30 31 32 484 Raphidioptera are found are Mexico (Central America), north of the (Africa), 33 34 485 northern India, Burma and northern Thailand (Asia) (Aspöck, U. et al . 1992; Aspöck, 35 36 37 486 U. and Aspöck, H. 2003). The discovery of a snakefly fauna in the Crato Formation 38 39 487 extends their range to the southern hemisphere (Aspöck, H. 1986; Oswald 1990). 40 41 488 Aspöck, U. and Aspöck, H. (2004) supposed that the rich Mesozoic 42 43 44 489 raphidiopteran fauna ended dramatically through climatic changes caused by the end- 45 46 490 Cretaceous bolide impact. Snakeflies of tropical and subtropical regions died out, and 47 48 491 only those branches adapted to colder climates like Raphidiidae and Inocellidae 49 50 51 492 survived. 52 53 493 54 55 494 Acknowledgements 56 57 58 495 59 60

20 Palaeontology Page 21 of 33 Palaeontology

1 2 3 496 Thanks to John Cooper and Gerald Legg at the Booth Museum of Natural History, 4 5 6 497 Brighton; also to Maidstone Museum and Bentlif Art Gallery and Dan Pemberton and 7 8 498 Matt Riley at the Sedgwick Museum of Earth Sciences for loan of specimens, Robert 9 10 11 499 Coram (British Fossils) for the drawing of M. hilli sp. nov., Tony Mitchell (Maidstone 12 13 500 Museum) for the photograph of M. hilli sp. nov. paratype. Thanks to Ulrike Aspöck 14 15 501 and Sam Heads for reviewing and improving the manuscript. JEJ acknowledges the 16 17 18 502 Alkins Fund (The University of Manchester) and a Wingate Scholarship for financial 19 20 503 support. 21 22 504 23 24 25 505 References 26 27 506 28 29 507 ALLEN, P. 1998. Purbeck-Wealden (early Cretaceous) climates. Proceedings of the 30 31 32 508 Geologists’ Association , 109 , 197–236. 33 34 509 ANSORGE, J. 1993. Bemerkenswerte Lebensspuren und ? Cretosphex catalunicus n. 35 36 37 510 sp. (Insecta; Hymenoptera) aus den unterkretazischen Plattenkalken der Sierra 38 39 511 del Montsec (Provinz Lerida, NE - Spanien). Neues Jahrbuch für Geologie und 40 41 512 Paläontologie Abhandlungen, 190 (1), 19-35. 42 43 44 513 ASPÖCK, H., 1986. The Raphidioptera of the world: a review of present knowledge. 45 46 514 15–29. In GEPP, J., ASPÖCK, H. and HÖLZEL, H. (eds), Recent research in 47 48 515 neuropterology. Proceedings of the 2 nd symposium of Neuropterology, Austria, 49 50 51 516 Graz, 176 pp. 52 53 517 — 2002. The biology of Raphidioptera: a review of present knowledge. 35–50. In 54 55 518 SZIRÁKI, G. (ed.), Neuropterology 2000. Proceedings of the Seventh 56 57 58 519 International Symposium on Neuropterology. Acta Zoologica Academiae 59 60 520 Scientiarum Hungaricae, 431 pp.

21 Palaeontology Palaeontology Page 22 of 33

1 2 3 521 — ASPÖCK, U. and RAUSCH, H. 1991. Die Raphidiopteran der Erde: Eine 4 5 6 522 monographische Darstellung der Systematik, Taxonomie, Biologie, Ökologie 7 8 523 und Chorologie der rezenten Raphidiopteran der Erde, mit einer 9 10 11 524 zusammenfassenden Übersicht der fossilen Raphidiopteran (Insecta: 12 13 525 Neuropteroidea) . Goecke and Evers, Krefeld, 2 Volumes, 730 & 550 pp. 14 15 526 ASPÖCK, U. 2002. The phylogeny of the Neuropterida (Insecta: Holometabola). 16 17 18 527 Zoologica Scripta , 31 , 51–55. 19 20 528 — and ASPÖCK, H. 2003. Raphidioptera (snakeflies). 973–975. In RESH, V. H. and 21 22 529 CARDÉ, R. T. (eds) Encyclopedia of Insects . Elsevier Academic press, 23 24 25 530 Amsterdam, 1266 pp. 26 27 531 — and ASPÖCK, H. 2004. Two significant new snakeflies from Baltic amber, with 28 29 532 discussion on autapomorphies of the order and its included taxa 30 31 32 533 (Raphidioptera). Systematic Entomology , 29 , 11–19. 33 34 534 — ASPÖCK, H. and RAUSCH, H. 1992. Rezente Südgrenzen der Ordnung 35 36 37 535 Raphidioptera in Amerika (Insecta: Neuropteroidea). Entomologia Generalis , 38 39 536 17 , 169–184. 40 41 537 BARALE, G. 1995. The fossil flora Megarests and Microrests. 31–38. In 42 43 44 538 MARTINEZ-DELCLOS, X. (ed.) Montsec & Montral-Alcover. Two 45 46 539 Konservat-Lagerstätten Catalonia, Spain . Institut d'Estudis Ilerdencs, Lleida, 47 48 540 Spain, 97 pp. 49 50 51 541 ENGEL, M. S., 2002. The smallest snakefly (Raphidioptera: Mesoraphidiidae): a new 52 53 542 species in Cretaceous amber from Myanmar, with a catalog of fossil 54 55 56 543 snakeflies. American Museum Novitates , 3363 , 1–22. 57 58 59 60

22 Palaeontology Page 23 of 33 Palaeontology

1 2 3 544 — and REN, D. 2008. New snakeflies from the Jiulongshan Formation of Inner 4 5 6 545 Mongolia, China (Raphidioptera). The Journal of the Kansas Entomological 7 8 546 Society, 81 (3), 188–193. 9 10 11 547 — LIM, J. -D. and BAEK, K. -S., 2006. Fossil snakeflies from the Early Cretaceous 12 13 548 of southern Korea (Raphidioptera: Mesoraphidiidae). Neues Jahrbuch für 14 15 549 Geologie und Paläontologie, 2006 , 246–256. 16 17 18 550 FREGENAL-MARTINEZ, M. And MELENDEZ, N. 1995. Geological setting. 12– 19 20 551 24. In MARTINEZ-DELCLOS, X. (ed.) Montsec & Montral-Alcover. Two 21 22 552 Konservat-Lagerstätten Catalonia, Spain . Institut d'Estudis Ilerdencs, Lleida, 23 24 25 553 Spain, 97 pp. 26 27 554 GRIMALDI, D. A. and ENGEL, M. S. 2005. Evolution of the insects . Cambridge 28 29 555 University Press, New York. 755 pp. 30 31 32 556 HAYWOOD, A. M., VALDES, P. J. and MARKWICK, P. J. 2005. Cretaceous 33 34 557 (Wealden) climates: a modelling perspective. Cretaceous Research , 25 , 303- 35 36 37 558 311. 38 39 559 JARZEMBOWSKI, E. A. 1984. Early Cretaceous insects from southern England. 40 41 560 Modern Geology , 9, 71–93. 42 43 44 561 — 1995. Early Cretaceous insect faunas and palaeoenvironment. Cretaceous 45 46 562 Research, 16 , 681–693. 47 48 563 JEPSON, J. E. and JARZEMBOWSKI, E. A. 2008. Two new species of snakefly 49 50 51 564 (Insecta: Raphidioptera) from the Lower Cretaceous of England and Spain 52 53 565 with a review of other fossil raphidiopterans from the Jurassic/Cretaceous. 54 55 566 Alavesia , 2, 193–201. 56 57 58 59 60

23 Palaeontology Palaeontology Page 24 of 33

1 2 3 567 — CORAM, R. A. and JARZEMBOWSKI, E. A. 2009. Raphidioptera (Insecta: 4 5 6 568 Neuropterida) from the Lower Cretaceous Purbeck Limestone Group, Dorset, 7 8 569 UK. Cretaceous Research , 30 (3), 527–532. 9 10 11 570 LATREILLE, P. A. 1810. Considérations générales sur l'ordre naturel des animaux 12 13 571 composant les classes des Crustacés, des Arachnides, et des Insectes; avec un 14 15 572 tableau méthodique de leurs genres, disposés en familles. Schoell, Paris, 444 16 17 18 573 pp. 19 20 574 MARTILL, D. M., BECHLY, G. and LOVERIDGE, R. F. 2007. The Crato fossil 21 22 575 beds of Brazil . Cambridge University Press, New York, 625 pp. 23 24 25 576 MARTÍN-CLOSAS, C. and LÓPEZ-MÓRON, N. 1995. The Charophyte flora. 29– 26 27 577 31. In MARTINEZ-DELCLOS, X. (ed.) Montsec & Montral-Alcover. Two 28 29 578 Konservat-Lagerstätten Catalonia, Spain . Institut d'Estudis Ilerdencs, Lleida, 30 31 32 579 Spain, 97 pp. 33 34 580 MARTINEZ-DELCLOS, X. and NEL, A, 1995. Arthropods. 39–46. In MARTINEZ- 35 36 37 581 DELCLOS, X. (ed.) Montsec & Montral-Alcover. Two Konservat- 38 39 582 Lagerstätten Catalonia, Spain . Institut d'Estudis Ilerdencs, Lleida, Spain, 97 40 41 583 pp. 42 43 44 584 MARTINS-NETO, R. G. and VULCANO, M. A. 1990. Primeiro registro de 45 46 585 Raphidioptera (Neuropteroidea) na Formação Santana (Cretáceo Inferior), 47 48 49 586 Bacia do Araripe, nordeste do Brasil. Revista Brasileira de Entomologia , 34 50 51 587 (1), 241–249. 52 53 588 — and NEL, A. 1992. Un nouveau fossile de Raphidoptère de la formation Santana, 54 55 56 589 Crétacé inférieur du Brésil (Neuropteroidea, Raphidioptera). Bulletin de la 57 58 590 Société Entomologique de France , 97 , 425–428. 59 60

24 Palaeontology Page 25 of 33 Palaeontology

1 2 3 591 MARTYNOV, A. V. 1925. To the knowledge of fossil insects from Jurassic beds in 4 5 6 592 Turkestan. 1. Raphidioptera. Izvestiya Rossiiskoi Akademii Nauk, 6 seryia , 19 , 7 8 593 235–246. 9 10 11 594 MARTYNOVA, O. M. 1961. Recent and fossil snakeflies (Insecta: Raphidioptera). 12 13 595 Paleontologicheskiy Zhurnal , 1961 , 73–83. [In Russian] 14 15 596 NAVÁS, L. 1913. Neurópteros del R. Museo Zoológico de Nápoles. Annuario del 16 17 18 597 [R.] Museo Zoologico della R. Università di Napoli (N.S.) , 4, 1–11. 19 20 598 — 1918. Monografia de l’ordre dels Rafidiòpters (Ins.). Arxius [Arxivus] de l’ 21 22 599 Institute [d’Estudis Catalans, Seccio] de Ciencias , Barcelona. 1–190. 23 24 25 600 NEL, A., SÉMÉRIA, Y. and MARTINS-NETO, R. G. 1990. Un Raphidioptera fossile 26 27 601 du Cretacé Inferieur du Bresil (Neuropteroidea). Neuroptera International , 6, 28 29 602 27–37. 30 31 32 603 OSWALD, J. D. 1990. Raphidioptera. 154–163. In GRIMALDI, D. A. (ed.) Insects 33 34 604 from the Santana Formation, Lower Cretaceous , of Brazil . Bulletin of the 35 36 37 605 American Museum of Natural History, 195, American Museum of Natural 38 39 606 History, New York, 191 pp. 40 41 607 REN, D., LU, L., GUO, Z. and JI, S. 1995 . Fauna and stratigraphy of the Jurassic– 42 43 44 608 Cretaceous in Beijing and adjacent areas . Seismic Publishing House, Beijing, 45 46 609 222 pp. [In Chinese, with English summary] 47 48 610 WHALLEY, P. E. S. 1988. Mesozoic Neuroptera and Raphidioptera (Insecta) in 49 50 51 611 Britain. Bulletin of the British Museum (Natural History) Geology , 44 , 45–63. 52 53 612 — and JARZEMBOWSKI, E. A. 1985. Fossil insects from the lithographic limestone 54 55 613 of Montsech (late Jurassic - early Cretaceous), Lerida Province, Spain. 56 57 58 614 Bulletin of the British Museum (Natural History) Geology, 38 (5), 381-412. 59 60

25 Palaeontology Palaeontology Page 26 of 33

1 2 3 615 WILLMANN, R. 1994. Raphidiodea aus dem Lias und die Phylogenie der 4 5 6 616 Kamelhalsfliegen (Insecta: Holometabola). Paläontologische Zeitschrift , 68 , 7 8 617 167–197. 9 10 11 618 12 13 619 FIGURE CAPTIONS 14 15 620 16 17 18 621 TEXT-FIG. 1. Mesoraphidia ednae sp. nov. holotype MNEMG 2009.42 [AY36a, b] 19 20 622 A, composite venation drawing of the forewing. B, photographs part and counterpart. 21 22 623 Scale bar represents 1 mm. 23 24 25 624 26 27 625 TEXT-FIG. 2. Mesoraphidia hilli sp. nov. holotype BMB 016381, -2. A, composite 28 29 626 venation drawing of the hind wing. B, photograph. Scale bar represents 1mm. 30 31 32 627 33 34 628 TEXT-FIG. 3. Mesoraphidia hilli sp. nov. paratype BMB 025010–12. A, composite 35 36 37 629 venation drawing. B, photograph. Scale bar represents 1 mm. 38 39 630 40 41 631 TEXT-FIG. 4. Nanoraphidia lithographica sp. nov. holotype FGWG 174 (MA 20 42 43 44 632 a,b). A, venation drawing of the forewing. B, photograph part and counterpart. 45 46 633 47 48 634 TEXT-FIG. 5. Iberoraphidia dividua gen. et sp. nov. holotype GZG.RF.7563 A, 49 50 51 635 venation drawing of the forewing. B, photograph of wing. C, photograph of wing tip 52 53 636 showing macrotrichiae. 54 55 637 56 57 58 59 60

26 Palaeontology Page 27 of 33 Palaeontology

1 2 3 638 TEXT-FIG. 6. Baissoptera lisae sp. nov. holotype CAMSM X.50155.2. A, venation 4 5 6 639 drawing of the forewing. B, venation drawing of the hind wings. C, drawing of body. 7 8 640 D. photograph. Scale bar represents 1 mm. 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

27 Palaeontology Palaeontology Page 28 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 TEXT-FIG. 1. Mesoraphidia ednae sp. nov. holotype MNEMG 2009.42 [AY36a, b] A. composite 29 venation drawing of the forewing. B. photographs part and counterpart, scale bar = 1 mm. 30 176x111mm (200 x 200 DPI) 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Palaeontology Page 29 of 33 Palaeontology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 TEXT-FIG. 2. Mesoraphidia hilli sp. nov. holotype BMB 016381, -2. A. composite venation drawing of 47 the hind wing. B. photograph, scale bar = 1mm. 48 188x222mm (300 x 300 DPI) 49 50 51 52 53 54 55 56 57 58 59 60 Palaeontology Palaeontology Page 30 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 TEXT-FIG. 3. Mesoraphidia hilli sp. nov. paratype BMB 025010–12 . A. composite venation drawing. 30 B. photograph, scale bar = 1 mm. 186x120mm (200 x 200 DPI) 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Palaeontology Page 31 of 33 Palaeontology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 TEXT-FIG. 4. Nanoraphidia lithographica sp. nov. holotype FGWG 174 (MA 20 a,b). A. venation 49 drawing of the forewing. B. photograph part and counterpart. 131x177mm (200 x 200 DPI) 50 51 52 53 54 55 56 57 58 59 60 Palaeontology Palaeontology Page 32 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 TEXT-FIG. 5. Iberoraphidia dividua gen. et sp. nov. holotype GZG.RF.7563 A. venation drawing of the forewing. B. photograph of wing. C. photograph of wing tip showing macrotrichiae. 48 189x263mm (300 x 300 DPI) 49 50 51 52 53 54 55 56 57 58 59 60 Palaeontology Page 33 of 33 Palaeontology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 TEXT-FIG. 6. Baissoptera lisae sp. nov. holotype CAMSM X.50155.2. A. venation drawing of the 39 forewing. B. venation drawing of the hind wings. C. drawing of body. D. photograph, scale bar = 1 40 mm. 180x167mm (200 x 200 DPI) 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Palaeontology CHAPTER 10. PALAEOECOLOGY OF THE LOWER CRETACEOUS PURBECK

AND WEALDEN NEUROPTERIDS (INSECTA: NEUROPTERA, RAPHIDIOPTERA)

(Jepson, J.E. Palaeoecology of the Lower Cretaceous Purbeck and Wealden neuropterids (Insecta: Neuroptera, Raphidioptera). To be submitted to Palaeogeography, Palaeoclimatology, Palaeoecology)

10.1. Abstract

The neuropterid fauna of the Lower Cretaceous Purbeck Limestone Group and Wealden Supergroup of southern England is discussed. The palaeoenvironment and palaeoecology are discussed and compared with other Lower Cretaceous deposits world-wide. The most dominant neuropterids in the Lower Cretaceous of England are the Psychopsoid neuropterans forming over 60% of the neuropterid fauna. Kalligrammatidae, Psychopsidae and Osmylidae persist from the Lower Purbeck to the Weald Clay. Some families differ between the deposits, some of which may be a function of the changing climate at this period in England‘s history. The Purbeck and Wealden are now seen to be major deposits for the study of fossil Neuropterida, filling gaps in the knowledge of Neuropterida in Europe during the Cretaceous.

Key words. England, Lacewing, Neuropterida, Palaeoenvironment, Snakefly, Wealden, Purbeck

10.2. Introduction

Neuropterida are a superorder of insects comprising snakeflies (Raphidioptera), and dobsonflies (Megaloptera) and lacewings (Neuroptera) (Aspöck, 2002). Until recently little was known of the neuropterid fauna of the Lower Cretaceous Purbeck and Wealden of southern England. The neuropteran families Ithonidae, Berothidae, ?Hemerobiidae, Chrysopidae, Nymphidae, Prohemerobiidae, Osmylidae, Kalligrammatidae, Psychopsidae, an undescribed family possibly related to Hemerobiidae and the raphidiopteran family Mesoraphidiidae are now known to have been present, however, Megaloptera have not been discovered (Jepson & Jarzembowski 2008 (chapter 7); chapter 4; chapter 9; Jepson et al. 2009a (chapter 5); Jepson et al. 2009b (chapter 8)). A review of the published literature on the

117 palaeontology, geology and modern analogues will be used to interpret the environments in which these Lower Cretaceous neuropterids once lived.

10.3. Materials and Methods

The fossil specimens represent extinct species and finding a direct modern analogue (i.e. species and genera) is therefore difficult. However, with regards to Neuroptera from these deposits many of the families are still living today (for example, Psychopsidae, Berothidae, Chrysopidae, Nymphidae, ?Hemerobiidae, Ithonidae and Osmylidae) and therefore, information can be taken from modern analogues and the mode of life of the extinct can be interpreted using the extant. Problems do occur, however, when dealing with the extinct and the incertae sedis families. With some, for example, Protohemerobius it may be possible to use Hemerobiidae as an analogue due to their potential relationship (Jepson et al. 2009b (chapter 7)) and the Mesoraphidiidae could be interpreted from general extant raphidiopteran habits. With regards to the other extinct taxa it may be more difficult to interpret their mode of life, but not impossible because the fossils themselves can give information such as the eyespot on Kalligrammatidae which could be interpreted as having a similar function to that of extant Lepidoptera. With others however, it may be more difficult and only general neuropteran modes of life can be applied.

10.4. Geological Settings

The fossil neuropterids in the Lower Cretaceous are deposited in the Wessex Sub- basin (Purbeck Limestone Group) and the Weald Sub-basin (Wealden Supergroup). These two sub-basins were surrounded by four massifs, the Welsh massif to the north, the Armorica massif to the south, the Londinia-Brabant massif to the east and the Cornubian massif to the west (Fig. 10.1). Two of these massifs, the Londinia and Cornubia, are the most important as these are the areas were sediment was supplied to the sub-basins (Fig. 10.1).

118

Figure 10.1. The palaeogeography of the Lower Cretaceous of southern England showing the massifs, sub-basins and drainage pattern (Constructed with data from Sladen & Batten 1984; Hancock & Rawson 1992; Jarzembowski 1995; Allen 1998; Radley 2005).

10.4.1. Purbeck Limestone Group

The Purbeck Limestone Group is Lower Cretaceous (Berriasian) in age, consisting mainly of lagoonal sediments (Allen & Wimbledon 1991). The group is divided into two formations, the older Lulworth Formation and the younger Durlston Formation. The type locality at Durlston Bay was extensively logged by Clements (1993), who divided the group up into numerous beds. Fossil insects are found in many horizons in the Purbeck, and in some places there can be hundreds found on a single bedding plane (Coram 2003).

In the Lulworth Formation the climate was dry, with little evidence of fluvial activity; it has been interpreted as being semi-arid, based on sedimentological evidence, with hypersaline deposits common, for example, evaporites (Clements 1993; Allen 1998; Batten 2002; Coram 2005), and clay mineralogy such as the presence of the mineral palygorskite, which indicates an arid/semi-arid environment (Sladen & Batten 1984; Hallam et al. 1991; Ruffell & Batten 1994; Coram 2005). In contrast the Durlston Formation climate has been interpreted as being more humid,

119 shown by the lack of hypersaline deposits and the dominance of the clay mineral kaolinite (Deconinck 1987).

The neuropterid specimens are from Durlston Bay, Swanage, Dorset and Teffont, Wiltshire.

10.4.2. Wealden Supergroup, Weald Sub-basin

The Wealden Supergroup of southern England is Berriasian–Aptian in age; it is subdivided into the Weald and Wessex Sub-basin (Allen & Wimbledon 1991). The non-marine Weald Sub-basin has yielded the most insect remains and it is from this basin that the neuropterid fossils derive. The Weald Sub-basin consists of the Hastings Group (Ashdown Beds, Wadhurst Clay, Lower and Upper Tunbridge Wells Sand, and Grinstead Clay formations) and the Weald Clay Group (Upper and Lower Weald Clay formations) (Allen & Wimbledon 1991; Rasnitsyn et al. 1998). The neuropterids are from the Wadhurst Clay and Weald Clay Group (Jarzembowski 2001; Jepson & Jarzembowski 2008 (chapter 8); Jepson et al. 2009b (chapter 7)). The Weald Clays represent a lacustrine–lagoonal environment (Allen 1981, 1989; Radley 2005). The climate has been interpretated as subtropical/warm temperate (with seasonal rainfall): the climatic reconstruction has been generated for the Wealden based on geological evidence, modelling (Allen 1998; Haywood et al. 2005) and the presence of certain insect Orders, for example, Isoptera (Jarzembowski 1995). The occurrence of wet and dry seasons (commonly with wild fires) has been proposed, as has year-round precipitation with little evidence for prolonged drought but high evaporation (Batten 1998; Haywood et al. 2005). Wildfire evidence exists in the form of fusainized elytra and plant remains (Watson & Alvin 1996; Batten 1998; Jarzembowski 2003).

The neuropterid specimens have been found in numerous localities: Hastings Beds, Wadhurst Clay, Quarry Hill, Tunbridge Wells and Weald Clay, Lower Weald Clay, Clockhouse Brickworks, and the Upper Weald Clay, Keymer Tileworks, Rudgwick, Auclaye, and Smokejacks Brickworks.

10.5. Taphonomy

The neuropterids from the Purbeck and Wealden are preserved as disarticulated (often fragmentary) wings; there have been no body parts of neuropterids found. The

120 proposed reason for the absence of complete specimens, or specimens that have some body parts preserved is transport, disarticulation and sorting (Jarzembowski 1995).

The taphonomy of Purbeck insects was described in detail by Coram (2003). Herein only the information that is relevant to Neuropterida has been added. In the Purbeck it has been suggested that the neuropterids most likely lived in a distant forested land and on death they underwent disarticulation and transport, with their wings being deposited into the Purbeck lagoon (Coram 2003 Figure 6 p. 316). The long transport also manifests itself in the poor state of the wings, many of which are fragmentary, and in some cases they cannot be described to lower taxonomic levels.

The neuropterids in the Wealden, as mentioned previously, most probably lived in a distant wooded area, for example, on the south side of the Londinian massif (Jarzembowski 1995). The Purbeck and Wealden wings are of a similar fragmentary state and have both been interpreted as being transported from a distant wooded habitat. A schematic figure of the palaeoenvironment within the Wealden is shown in figure 10.2, showing a similar scenario to the Purbeck (Coram 2003 Figure 6 p. 316).

121

Figure 10.2. A schematic reconstruction of the Wealden showing the distant forested land and area of deposition

Most of the neuropterids in the Wealden are found in the localities close to the Londinia-Brabant massif (Smokejacks, Clockhouse and Auclaye, see Fig. 10.3), whereas they are very rare and fragmented in the most distant locality of Keymer Tileworks (Fig. 10.3). Being a further distance away means the neuropterid remains would have a higher chance of being destroyed by the longer period of transport. This long transport is also the reason why few fossil insects and no neuropterids have been found in the Wealden in the Wessex Sub-basin on the Isle of Wight, and also why the insects which have been found are in a poor fragmentary state dominated by the hard elytra of beetles.

122

Figure 10.3. Showing the position of the Wealden localities with respect to the massif. The furthest locality away from the massif, Keymer tileworks, has only one fragmentary neuropterid recorded (Constructed with data from Sladen & Batten 1984; Jarzembowski 1995; Allen 1998; Jepson et al. 2009b (chapter 7)).

There are numerous possible scenarios to explain why the insects in the Purbeck and Wealden are only represented by their wings. The insects could have died on the land, disarticulated, their bodies decomposing or forming food for predators, who often discard wings. Their wings would then be transported in the wet season to one of the watercourses, which would transport them, causing damage in the process, until deposition (in the Purbeck Lagoon, or Weald Sub-basin).

Alternatively the insect could have died and fallen into the watercourse or died on land and washed in whole into the watercourse. When on the water the insect disarticulated and decomposed. The decomposing body parts on the surface of the water would be an easy target for predation, again leaving just the wings.

Another scenario could be that the insect was alive and became trapped in the watercourse, for example, during heavy rain, or blown in during a storm. When on the water surface, if the insect could not escape it would eventually die through asphyxiation, either on the surface, or if it broke through the surface tension, within the water column. The insect would then disarticulate, due to decomposition or predation, separating the wings from the body, which eventually get deposited in the basin.

123

The neuropterid fossils in the Wealden are more fragmented than those of the Purbeck, indicating that they have undergone more disarticulation. The (estimated) wingsize range of the described taxa shows that the majority are between 6–25 mm, with 1–5 mm being very rare and few 40+ mm. The rarity of <5 mm and 40+ mm wingsized neuropterids, may be due to the fact that insects with these wing sizes find it hard to break through the water surface tension (Martínez-Delclos & Martinell 1993), and therefore are an easy target for predation and total decomposition on the water surface. This may be a reason why the family Coniopterygidae (very small wing sizes) predicted by Jepson and Penney (2007 (chapter 6)) to be in the Wealden, has not been discovered, and also why there are few large neuropterans.

It is difficult to state confidently which of the above scenarios are true for the Purbeck and Wealden. It is probably most likely that a mixture of all three took place. The fate of the neuropteran body parts is unknown; they could have decomposed, been consumed by predators, or separated during transport and deposited in another location awaiting discovery.

10.6. Lower Cretaceous Neuropterida from Britain

The knowledge of the neuropterid fauna of the Purbeck and Wealden within the last few years has increased significantly. Table 10.1 and 10.2 list the taxa described and undescribed (fragments from museum collections) from the deposits.

124

Order Family Species Formation Bed Locality Age Reference Neuroptera Kalligrammatidae Sophogramma wimbledoni Lulworth ? Teffont, Wiltshire Lower Berriasian Jepson et al. submitted Nymphidae Sialium sipylus Lulworth ? Durlston Bay, Dorset Lower Berriasian Jepson et al. submitted Osmylidae Osmylopsis duplicata Lulworth ? Durlston Bay, Dorset Lower Berriasian Jepson et al. submitted Psychopsoid indet. (Port/Neur1) Lulworth Lower Cypris freestone Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Psychopsoid indet. (POX/DBartonPB11) Lulworth Upper Cypris freestone Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Incertae sedis Paraksneura alleni Lulworth Hard Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson et al. submitted Raphidioptera ? (DB53/Neur1) Lulworth Lower Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Coram pers comm. 2006 Neuroptera Psychopsidae indet. (USCDB/TM986) Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Psychopsidae indet. (DB64-8/Wing2) Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Psychopsidae indet. (USCDB/TM176) Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Psychopsidae indet. (DB64-8/Wing14) Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Psychopsidae indet. (USCDB/TM233) Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Indet. Indet. (DB59/Neur2) Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Indet. Indet. (USCDB/TM211) Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Indet. Indet. (USCDB/TM169) Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Berothidae Epimesoberothus osmylodes Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson et al. submitted Raphidioptera Mesoraphidiidae Mesoraphidia websteri Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson et al. 2009 Mesoraphidiidae ? Mesoraphidia websteri (DB64-8/Raph1) Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Mesoraphidiidae ? Mesoraphidia websteri (USCDB/TM857) Lulworth Upper Soft Cockle Bed Durlston Bay, Dorset Lower Berriasian Jepson pers. obs. Neuroptera Psychopsidae Pterinoblattina penna Durlston ? Durlston Bay, Dorset Lower Berriasian Jepson et al. submitted Psychopsidae Pterinoblattina pluma Durlston ? Durlston Bay, Dorset Lower Berriasian Jepson et al. submitted Psychopsidae indet. (DB107/Neur1) Durlston Cherty Freshwater Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Raphidioptera Mesoraphidiidae M. durlstonensis Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. 2009 Mesoraphidiidae M. durlstonensis Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. 2009 Mesoraphidiidae M. durlstonensis Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. 2009 Mesoraphidiidae M. purbeckensis Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. 2009 Mesoraphidiidae M. mitchelli Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. 2009 Indet. Indet. (DB175/Raph5) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Indet. Indet. (DB175/Raph3) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Neuroptera Osmylidae Mesosmylidus vulgaris Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. submitted Osmylidae Osmylochrysa anomola Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. submitted Osmylidae Osmylochrysa minima Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. submitted Osmylidae Stenochrysa gradata Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. submitted Prohemerobiidae Ovalorobius purbeckensis Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. submitted ?Hemerobiidae Purbemerobius medialis Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. submitted Chrysopidae Mesypochrysa minuta Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. submitted Psychopsidae Pterinoblattina purbeckensis Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. 2009 Psychopsidae Pterinoblattina fasciata Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson et al. 2009 Psychopsidae indet. (DB175/Neur3) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/Neur23) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/TM1183) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/Neur10) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/Neur22) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/TM1274) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/Neur26) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/Neur6) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/Neur21) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/Neur19) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/Neur15) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/Neur20) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs. Psychopsoid indet. (DB175/wing3) Durlston Corbula beds Durlston Bay, Dorset Upper Berriasian Jepson pers. obs.

Table 10.1. The specimens of Neuropterida from the Purbeck Limestone Group.

125

Order Family Genus and Species Group Formation Locality Age Reference Neuroptera Kalligrammatidae Kalligramma roycrowsoni Hastings Wadhurst Clay Quarry Hill, Kent Mid Valanginian Jarzembowski, 2001 Raphidioptera Mesoraphidiidae Proraphidia hopkinsi Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson and Jarzembowski, 2008 Neuroptera Osmylidae Stenosmylina medialis Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson et al. 2009 Psychopsidae Pterinoblattina minimus Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson et al. 2009 ?Pterinoblattina sp Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson et al. 2009 Psychopsites rolandi Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson et al. 2009 indet. (MNEMG CH25c) Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson pers. obs. indet. (MNEMG CH772) Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson pers. obs. indet. (MNEMG CH315) Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson pers. obs. indet. (MNEMG CH115b) Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson pers. obs. indet. (MNEMG 25F) Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson pers. obs. "Psychopsoid" indet. (BMB RC) Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson pers. obs. Incertae sedis Actinophlebioides valdensis Weald Clay Lower Weald Clay Clockhouse Brickworks, Surrey Late Hauterivian Jepson et al. 2009 Psychopsidae Cretopsychops corami Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson et al. 2009 ? Cretapsychops sp. (BMNH Pal. 1539) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Cretapsychops sp. (BMNH Pal. 1542) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. Micropsychops parallelus Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson et al. 2009 Pterinoblattina logunovi Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson et al. 2009 Pterinoblattina brigidae Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson et al. 2009 Pterinoblattina proudlovei Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson et al. 2009 Pterinoblattina sp. Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson et al. 2009 ? Pterinoblattina sp. (BMNH Pal. 1544) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Pterinoblattina sp. (BMNH Pal. 1540) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Pterinoblattina sp. (BMNH Pal. 1524) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Pterinoblattina sp. (BMNH Pal. 1543) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Pterinoblattina sp. (BMNH Pal. 1515) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Pterinoblattina sp. (BMNH Pal. 1553) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Pterinoblattina sp. (BMNH Pal. 1520) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Pterinoblattina sp. (BMNH Pal. 1528) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Pterinoblattina sp. (BMNH Pal. 1535) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Pterinoblattina sp. (BMNH Pal. 1519) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Pterinoblattina sp. (BMNH Pal. 1532) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. ? Pterinoblattina sp. (BMNH Pal.1554) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. indet. (MNEMG S1631) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. indet. (MNEMG S823b) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. indet. (MNEMG S87a) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. indet. (BMNH Pal. 1516) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. indet. (BMNH Pal. 1549) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. indet. (MNEMG S72) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. Kalligrammatidae indet. Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson et al. 2009 Indet. Indet. (BMNH Pal. 1522) Weald Clay Upper Weald Clay Smokejacks Brickworks, Surrey Barremian Jepson pers. obs. Raphidioptera Mesoraphidiidae Mesoraphidia ednae Weald Clay Upper Weald Clay Auclaye Brickworks, Surrey Barremian Jepson et al. submitted Neuroptera Psychopsidae Pterinoblattina maculosus Weald Clay Upper Weald Clay Auclaye Brickworks, Surrey Barremian Jepson et al. 2009 ? Pterinoblattina sp. (BMB 018571-2) Weald Clay Upper Weald Clay Auclaye Brickworks, Surrey Barremian Jepson pers. obs. ? Psychopsites (BMB 018579) Weald Clay Upper Weald Clay Auclaye Brickworks, Surrey Barremian Jepson pers. obs. indet. (BMB 01851-2) Weald Clay Upper Weald Clay Auclaye Brickworks, Surrey Barremian Jepson pers. obs. indet. (BMB 018577-8) Weald Clay Upper Weald Clay Auclaye Brickworks, Surrey Barremian Jepson pers. obs. Osmylidae Protosmylina bifasciata Weald Clay Upper Weald Clay Auclaye Brickworks, Surrey Barremian Jepson et al. 2009 Incertae sedis Protohemerobius perexiguus Weald Clay Upper Weald Clay Auclaye Brickworks, Surrey Barremian Jepson et al. 2009 Indet. indet. (BMB 0185773-4) Weald Clay Upper Weald Clay Auclaye Brickworks, Surrey Barremian Jepson pers. obs. Raphidioptera Mesoraphidiidae Mesoraphidia hilli Weald Clay Upper Weald Clay Rudgwick Brickworks, Sussex Barremian Jepson et al. submitted Neuroptera Ithonidae Principiala rudgwickensis Weald Clay Upper Weald Clay Rudgwick Brickworks, Sussex Barremian Jepson et al. 2009 Psychopsidae indet. (BMB 20409-10) Weald Clay Upper Weald Clay Keymer Brickworks, Surrey Barremian Jepson pers. obs.

Table 10.2. The specimens of Neuropterida from the Wealden Supergroup.

The neuropterid fauna of Britain comprises Raphidioptera and Neuroptera, with Megaloptera currently not having been found. Neuroptera are the most numerous representatives of the superorder, while Raphidioptera is only represented by one family and two genera. Figure 10.4 displays the percentages of the described species of each deposit and the percentages of families within the neuropterid fauna of the Purbeck and Wealden.

126

Figure 10.4. Pie charts showing the percentage of described Neuropterida compared to other described species of other insect Orders in the Purbeck (A) and Wealden (B). The percentage of families within the neuropterid fauna of the Lower Cretaceous southern England is shown in C.

Within the described insect fauna of the Purbeck and Wealden, Neuroptera is shown as quite a high percentage. This unfortunately does not represent true diversity; instead it represents a bias to those insect groups which have been worked on in detail, for example, Hymenoptera (Rasnitsyn et al. 1998) and Orthoptera (Gorochov et al. 2006). With regards to the neuropterid fauna (Fig 10.4C) it is shown that the psychopsoid neuropterans dominate comprising 66%, with Osmylidae forming quite a high percentage.

The number of specimens found at each locality in the Purbeck and Wealden is shown in the spindle diagrams of figure 10.5. Neuropterans show an increase in number from the Lower Purbeck to the Upper Weald Clay (Wealden) (dashed line, Fig. 10.5), whereas the raphidiopterans remain approximately constant. However, little is known from the Middle Purbeck to the top of the Lower Weald Clay due to

127 poor preservation conditions and little work done, especially in the clay formations of the Hastings Group.

Figure 10.5. Spindle diagram showing the number of specimens from the fossil localities within the Purbeck and Wealden. The dashed lines shows an extrapolation through the areas of little information.

The increase in numbers could be due to the taphonomy of the deposits; the Purbeck is drier than the Wealden, and therefore there is less water to transport neuropterans from their distant habitat to the basin of deposition. Whereas the increased fluvial activity within the Wealden could have increased the chances of transport, deposition and therefore preservation. Although the number of specimens is higher in the Wealden the diversity of neuropterans is lower, with more families known in the Purbeck.

The geological ranges of the families of Neuropterida from the Purbeck into the Wealden are shown in figure 10.6.

128

Figure 10.6. The geological ranges of the neuropterid families in the Lower Cretaceous of Britain.

The families Mesoraphidiidae, Kalligrammatidae, Psychopsidae, Osmylidae and the indeterminate psychopsoid neuropterans are shown to have been present in the Lower Cretaceous of England from the Lower Purbeck into the Weald Clay (Fig. 10.6). The differences between the two deposits are that in the Purbeck the families Berothidae, Nymphidae, Prohemerobiidae, ?Hemerobiidae and Chrysopidae are recorded and in the Wealden Ithonidae is present. The fauna is discussed in detail below.

10.6.1. Raphidioptera

Raphidiopterans are found in both the Purbeck and Wealden; the oldest Cretaceous specimen is undescribed from the Lower Soft Cockle Beds, Lulworth Formation (Table 10.1). The next occurrence of raphidiopterans is in the Upper Soft Cockle

129

Bed (Lulworth Formation) with the species Mesoraphidia mitchelli (Jepson et al. 2009a (chapter 5)). This species is found with two other fragmentary specimens most likely of the same species. Raphidiopterans are then not observed until the Corbula Beds (Durlston Formation), where they are more diverse (at species level), and represented by Mesoraphidia durlstonensis, M. purbeckenisis and M. websteri (Jepson et al. 2009a (chapter 5)). Snakeflies are next recorded in the top of the Lower Weald Clay, where they are represented by a larger species and a different genus Proraphidia hopkinsi (Jepson & Jarzembowski 2008 (chapter 8)), and their last occurrence is in the lower part of the Upper Weald Clay with Mesoraphidia hilli and M. ednae (chapter 9).

The Lower Cretaceous British raphidiopterans are not very diverse, and are represented by one family Mesoraphidiidae and only two genera, the dominant Mesoraphidia, and Proraphidia.

In the Lower Purbeck the snakeflies were much smaller than the subsequent fossils in the Purbeck and Wealden and their modern counterparts (Fig. 10.7). The oldest described snakefly from the Lulworth Formation of the Purbeck is M. mitchelli with an estimated wing length of 5.4 mm. Moving up through the Purbeck into the Durlston Formation there are still small raphidiopterans M. purbeckensis (5.1 mm) and M. websteri (5.2 mm), but there is a larger more dominant species M. durlstonensis (8–10.6 mm) represented by three specimens. The Wealden raphidiopterans are of a larger size; Proraphidia hopkinsi from the Lower Weald Clay has a wing length of 11.5 mm and in the Upper Weald Clay Mesoraphidia hilli and M. ednae have wing lengths 11.5 mm and 8 mm respectively.

130

Figure 10.7. The wing size of fossil raphidiopterans from the Purbeck and Wealden with the proposed climate chage highlighted (incorporating data from Sladen & Batten 1984; Ruffell & Batten 1994; Allen 1998; Coram 2005; Jepson & Jarzembowski 2008 (chapter 8); Jepson et al. 2009a (chapter 5); chapter 9).

A reason for the size differences could be due to the changing climate. The climate change has been interpreted by the presence of evaporites, lack of fluvial sediments in the Lower Purbeck, changing to a loss of evaporates and increase in fluvial sediments from the mid-Purbeck to the Wealden. The presence of certain clay minerals in the Purbeck and Wealden also indicate a change in climate, for example, palygorskite in the Lower Purbeck indicates arid/semi arid conditions (Deconinck 1987). The snakeflies in the Lower-Mid Purbeck may have been small to cope with the harsher environment, e.g., high temperatures and lack of prey.

10.6.2. Neuroptera

There are 12 neuropteran families present in the Purbeck and Wealden; the most numerous representatives are the psychopsoid neuropterans (Table 10.1 and 10.2, Fig. 10.4).

131

The psychopsoid neuropterans are represented by the families Psychopsidae and Kalligrammatidae which are easily recognized by their dense venation in both fore- and hind wings.

Psychopsoid neuropterans are first recorded in the British Lower Cretaceous in the Lower and Upper Cypris Freestones (Lulworth Formation) with 2 indeterminate psychopsoids. They are next seen in the Lulworth Formation in the Upper Soft Cockle Bed, with 5 indeterminate psychopsid specimens. The kalligrammatid Sophogramma wimbledoni is found in the Lulworth Formation in Teffont, Wiltshire. Psychopsoids are next recorded in the Cherty Freshwater Beds in the Durlston Formation again with an indeterminate specimen. Numerous specimens are known from the Corbula Beds; with the species Pterinoblattina fasciata and P. purbeckensis and 1 indeterminate specimen and 10 psychopsoid neuropteran fragments.

The earliest record of a psychopsoid in the Wealden is the kalligrammatid Kalligramma roycrowsoni from the Wadhurst Clay, Hastings Group. The next records are the psychopsids, Pterinoblattina minimus, ?P. sp. A and Psychopsites rolandi and 5 indeterminate psychopsids and 1 indeterminate psychopsoid in the Lower Weald Clay, Clockhouse Brickworks. In the Upper Weald Clay psychopsids are recorded at Auclaye Brickworks with Pterinoblattina maculosus, P. sp., Psychopsites sp. and 2 indeterminate psychopsids. At Smokejacks Brickworks (Upper Weald Clay) there are numerous psychopsoids known (Cretapsychops corami, 2 ?Cretapsychops sp., Micropsychops parallelus, Pterinoblattina logunovi, P. brigidae, P.proudlovei, P. sp., 12 ?Pterinoblattina sp., 6 indeterminate psychopsids and an indeterminate kalligrammatid). An indeterminate psychopsid is also known from Keymer Tileworks (Jarzembowski 2001; Jepson et al. 2009b (chapter 7); Jepson pers. obs. 2009).

Osmylidae are fairly common within the neuropterid fauna of Lower Cretaceous Britain. They are first recorded in the Corbula beds (Durlston Formation) in the Purbeck, with the species Osmylopsis duplicata, Osmylochrysa minima, O. anomola, Mesosmylidus vulgaris and Stenochrysa gradata. Osmylids are then observed within the Weald Clay, represented by Stenosmylina medialis and Protosmylina bifasciata.

132

One species of ithonid is known from the Lower Cretaceous of Britain, Principiala rudgwickensis, which is of the same genus of P. incerta from the younger Crato Formation of Brazil (Makarkin & Menon 2007; Jepson et al. 2009b (chapter 7)). Extant ithonids may be broadly divided into arid and forest dwelling genera (Makarkin & Archibald 2009). It is most likely that the genus Principiala would have been a forest dweller in the Wealden, based on the assemblages of insects and interpreted palaeoenvironment (Jarzembowski 1995).

Nymphidae is the only myrmeleontoid known from the Lower Cretaceous of England. It is represented by the species Sialium sipylus, which was one of the first neuropterids described from the deposit (Westwood 1854). Unfortunately the locality information is unknown, except that it is from the Lulworth Formation, Purbeck Limestone Group and was collected at Durlston Bay.

Prohemerobiidae is only represented by the genus Ovalorobius purbeckensis in the Corbula Beds.

One species of ?Hemerobiidae is known from the Corbula Beds (Purbeck); Purbemerobius medialis is known from a poorly preserved hind wing. This is the first record of a potential hemerobiid in the fossil record of Britain.

Chrysopidae is represented by one species in the Corbula Beds (Purbeck); Mesypochrysa minuta is the smallest chrysopid known in the fossil record with an estimated wing span of ~ 7 mm. No other chrysopids are known in the fossil record of Britain.

Two specimens of Berothidae represent the species Epimesoberothus osmylodes known from the Purbeck Limestone Group. Both of the specimens are from the Upper Soft Cockle Beds (Lulworth Formation). These represent the smallest neuropterans in the Purbeck.

Incertae sedis neuropterans are found in both the Purbeck and Wealden. The only one in the Purbeck is the species Paraksneura alleni which may be related to chrysopoids or even megalopterans; unfortunately with no body parts preserved the true affinity of this species cannot be ascertained. The Wealden has two species that cannot be placed within a described family; Actinophlebia valdenisis, is possibly from a psychopsoid family, for example Brongniartiellidae, but, due to the poor

133 taxonomy of these neuropterans it has been placed as incertae sedis. A. valdenisis is from the Lower Weald Clay. In the Upper Weald Clay Protohemerobius perexiguus is present and is superficially similar to Hemerobiidae; however, it is most likely a new family of Neuroptera as it is similar to an undescribed neuropteran family from China. The preservation of P. perexiguus is such that it cannot be described as a new family, because many of the diagnostic features required are not preserved.

Megalopterans have not yet been found in the Purbeck or Wealden.

10.7. Neuropterid palaeoenvironment and palaeoecology

Extant neuropterids of the families Psychopsidae, Ithonidae, Berothidae, Nymphidae, Hemerobiidae, Chrysopidae and the Order Raphidioptera are found in a variety of habitats; they are mainly arboreal, often (but not always) found living in or around forested habitats. They are quite often found living on the trunk, or amongst the branches. Many of the neuropterids are weak flyers and therefore do not travel far from their place of hatching. Some extant Osmylidae have a different habitat preference to the other neuropterids, tending to live close to water. It has been suggested that they may have a semi-aquatic larva, because of the habits the adult are similar to insects that have semi-/aquatic larva, for example, the megalopteran family Sialidae.

Neuropterans and raphdiopterans are on the whole entomophagous insects, but, some are known to feed on nectar and pollen, especially in their adult form (New 1989; Aspöck et al. 1991). The fact that they are predators means that they inhabit a higher trophic level in the food chain, and therefore will be fewer in number than phytophagous insects. This is reflected in their rarity in the insect fossil record of the Purbeck and Wealden and modern faunas. Extant neuropterids (both larva and adult) are predominantly generalist predators feeding on small invertebrates, for example, scale insects, aphids, bark lice, and mites. However, some families are more specialized feeding solely on termites, or ants, or sponges, or even spider egg cases (New 1989). They are generally active hunters, either seeking prey (for example, they can be observed walking through aphid colonies) or sit and wait predators, waiting for prey to come to them; or on the wing hunters, like dragonflies; scavenging has also been observed. Extant neuropterids have also been observed eating plant matter, such as pollen and honeydew, and in some cases the larva are

134 voracious predators, whereas the adult feeds only on pollen or honeydew (New 1989).

In the Purbeck it is likely that the neuropterans and raphidiopterans would have fed on the small invertebrate fauna, for example the smaller forest dwelling arthropods. In the distant forest, where the neuropterids have been proposed to have lived (Coram 2003), potential prey are also observed including fungus gnats, cupedoid beetles, aphids and other plant bugs. It is likely that these insects would have been the main food source for the majority of neuropterids. To add support to this the fossil raphidiopterans from the Purbeck have been found associated with these tree dwelling insects, for example cupedoid beetles (Jepson et al. 2009a (chapter 5)). With regards to the family Berothidae, the extant members of the family are specialized feeders on termites; however, no fossils of termites have been found within the Purbeck as yet, although they are present in the Wealden (Jarzembowski 1995).

The Wealden Neuroptera and Raphidioptera would have had similar prey such as barklice (producers of Paleoscolytus), aphids (e.g. the Wealden aphid: Penaphis woolardi), and other scale insects that are found in the deposit (Jarzembowski 1995).

The neuropterids from both localities would have been prey for other predatory insects, for example, the abundant dragonflies known from these deposits. There also would most likely have been predation by vertebrates, using the modern analogue of insects with cryptic colour patterns and in particularly eye-spots (e.g. on the wings of extant Lepidoptera). Kalligrammatids, which are observed in the Wealden, in particularly Kalligramma roycrowsoni has a well-preserved eyespot (a feature observed in the genus Kalligramma) (Jarzembowski 2001), and many of the psychopsoids from both the Purbeck and Wealden have cryptic wing patterns, which may indicate a defence against vertebrate prey (Jepson et al. 2009b (chapter 7)).

10.8. Faunal change

The major change from the neuropterid fauna of the Purbeck to the Wealden is the loss of some and gaining of other families. Only the families Psychopsidae, Kalligrammatidae, Osmylidae and Mesoraphidiidae persist from the Lower Purbeck up into the Weald Clay (Fig. 10.6). The change in fauna also represents the change in

135 climate to some extent. The change from small to larger snakeflies has already been mentioned, with regards to size in neuropterans there is little difference between the two deposits. Berothidae are the smallest neuropterans within the Purbeck and these happen to be from the same beds as the smallest raphidiopterans, but, little can be inferred from this because of the small sample size and the fact that other neuropterans are not of such a size. Also no berothids have been found in younger beds for comparison. The families themselves may indicate climate. Berothids and nymphids are usually found in arid/semi-arid environments in the present day, so their presence in the Lower Purbeck and not in the younger beds could be related to climatic change.

The Purbeck neuropterid fauna contains the families Chrysopidae, ?Hemerobiidae, Nymphidae, Berothidae and Prohemerobiidae, which are not present in the Wealden. Berothidae and Nymphidae has been excused on climatic grounds, but there is no reason why Chrysopidae and ?Hemrobiidae and even Prohemerobiidae should not be present in the Wealden and they may yet be discovered. Ithonidae is only present in the Wealden; again it may have been present in Purbeck times but not discovered because it has a wide habitat preference today from high altitudes to sandy desert environments (Menon & Makarkin 2008).

What has to be remembered is that the fossils from these distant woodland areas of the Purbeck and Wealden are just ‗snapshots‘ of a habitat and not all the insects will survive transportation to be preserved.

10.9. Lower Cretaceous neuropterid faunas world-wide

Many neuropterids have been described from numerous localities in the Lower Cretaceous world-wide. Below is a review of some of these localities of a similar age to the Purbeck and Wealden (Berriasian–Aptian) which have a neuropterid fauna.

Baissa, Transbaikalia (Valanginian–Barremian): This is a very rich insect deposit comprised of the neuropteran families: Osmylidae, Mesithonidae, Nymphidae, Mantispidae, Psychopsidae, Chrysopidae, Babanskaiidae, Brongniartiellidae and Myrmeleontoidea (Makarkin 1990a, b, 1996, 1997a, b, 1999, in press; Wedmann & Makarkin 2007). The Raphidiopterans are also diverse in this deposit, with all the extinct families represented: Mesoraphidiidae (genera: Siboptera, Baisoraphidia),

136

Baissopteridae (genera: Cretoraphidia, Baissoptera) and Alloraphidiidae (genera: Pararaphidia, Archeraphidia, Alloraphidia) (Martynova 1961; Ponomarenko 1993).

Spain: Montsec and Las Hoyas (Barremian): The neuropteran families include Mesochrysopidae, Incertae sedis (Chrysopoid), and Kalligrammatidae (Martínez- Delclos & Martinell 1993; Nel et al. 2005). The snakefly fauna consists only of the family Mesoraphidiidae (genera: Proraphidia, Iberoraphidia, Nanoraphidia) (Jepson & Jarzembowski 2008 (chapter 8); chapter 9).

China: Yixian Formation (Barremian–Aptian): Another formation, which has an abundant neuropterid fauna, it comprises the neuropteran families: Myrmeleontidae, Myrmeleontoidea, Aetheogrammatidae, Kalligrammatidae, Osmylidae, Chrysopidae and Berothidae (Ren & Guo 1996; Ren & Yin 2002; Ren & Engel 2008a, b). The snakefly fauna is also very diverse with all extinct families present: Mesoraphidiidae (genera: Siboptera, Mesoraphidia, Huaxiaraphidia), Baissopteridae (genus: Baissoptera) and Alloraphidiidae (genus: Alloraphidia) (Ren 1994, 1997; Ren et al. 1995).

Mongolia (Barremian–Aptian): The neuropteran families represented are Kalligrammatidae, Prohemereobiidae, Hemerobiidae, Grammostylidae, Mesochrysopidae, and Myrmeleontidae (Ponomarenko 1992b). The snakefly fauna has the families Baissopteridae (genera: Lugala, Cretaraphidiopsis, Cretinocellia), and Alloraphidiidae (genera: Pararaphidia, Archeraphidia, Alloraphidia) present (Ponomarenko 1988).

Lebanon: Lebanese amber (Barremian–Aptian) has few neuropterids. Neuropterans include Coniopterygidae, Berothidae and Rhachiberothidae, while snakeflies are only represented by a ?Mesoraphidiidae larva (Engel & Grimaldi 2007).

Brazil: Crato Formation (Aptian): the Crato Formation of Brazil has a very extensive and diverse neuropterid fauna. Neuropterans are represented by the families: Nemopteridae, Ascalaphidae, Nymphidae, Myrmeleontoidea, Berothidae, Palaeoleontidae, Araripeneuridae, Babinskaiidae, Chrysopidae, Mesochrysopidae, Hemerobiidae, Makarkiniidae, Brongniartiellidae, Kalligrammatidae, Ithonidae, Osmylidae, Sisyridae, and Limaiidae. The snakefly fauna consists of the family Baissopteridae (genus: Baissoptera, Austroraphidia), and one familia incertae sedis

137

(genus: Arariperaphidia) (Martins-Neto et al., 2007). Megalopterans are also known from this deposit (Martins-Neto et al. 2007; Jepson pers. obs.).

Lower Cretaceous localities with few neuropterids recorded:

China: Laiyang Formation, (Barremian): Chrysopidae and Mesochrysopidae (Yang and Hong, 1990; Zhang, 1991); Xiaoershijiazi: Inocelliidae (Sinoinocellia) (Wang 1987); Zhixian, Jilin province: Mesoraphidiidae (genus: Jilinoraphidia) (Hong & Chang 1989); Kezuo, Liaoning: Mesoraphidiidae (genus: Kezuoraphidia) (Hong 1992).

South Korea: Jinju Formation: Mesoraphidiidae (genus: Mesoraphidia) (Engel et al. 2006).

Japan: Osmylopsychopidae (Fujiyama 1978).

Romanovka, Russia: Baissopteridae (Cretoraphidia) (Ponomarenko 1993).

10.9.1. Palaeobiogeography

Neuropterids are now a relict group of insects, but during the Cretaceous they were globally widespread. Further evidence for this has come from the study of the Purbeck and Wealden fauna, which shares genera with more geographically distant insect localities. For example, the genus Sophogramma is also present in Russia and China (Ren & Guo 1996; Yang et al. 2009; Makarkin in press; chapter 4), Kalligramma is widespread in Kazakhstan, Germany and China (Walther 1904; Panfilov 1980; Ren & Guo 1996), Mesypochrysa is also present in Kazakhstan, Russia, China and Brazil (Makarkin 1997b; Nel et al. 2005) and the genus Principiala is present in Brazil (Makarkin & Menon 2007).

10.10. Discussion

The Lower Cretaceous Purbeck and Wealden are important deposits for Neuropterida; there are 12 families, 22 genera and 31 species of Neuroptera, and 1 family, 2 genera and 7 species of Raphidioptera. The neuropterids can be tracked from the Berriasian to the Barremian, but, there is a gap of 3 million years, where little is known of the insects. This gap is due mainly to the sediment not being conducive to insect preservation, for example, the Hastings Group is dominated by

138 sands. Crowson (1946) mentioned the presence of some insects from the Wadhurst Clay, and Jarzembowski (2001) described Kalligramma roycrowsoni from this unit. It is more than likely that further examination of the clays in the Hastings Group could yield more insects, and decrease this large gap within the Wealden.

The most common neuropterids in the Lower Cretaceous are the psychopsoid neuropterans (which include Psychopsidae and Kalligrammatidae). These neuropterans were widespread throughout the Cretaceous and Mesozoic, and have appeared in the majority of the fossil insect localities. Why they are so numerous is difficult to answer, and could be due to many possible reasons. It could be that they were the most numerous and successful group of neuropterans at that time, and this success has been recorded in the fossil record; alternatively their dominance could be due to their wings having many veins. Since the neuropterans are thought to have been transported some distance before deposition, the numerous veins of psychopsoid neuropterans could have made them more resilient to breaking down, and therefore giving them a better chance of preservation than insects with fewer veins. Human bias could also be a factor; the fact that psychopsoid neuropterans are readily recognized by non-specialists, whereas other families may be overlooked, or misidentified, could lead to them appearing more numerous. Osmylidae are also well represented in the Purbeck and Wealden, which may be because of a semi-aquatic larva. If adults are found near to water they have a better chance of being washed into the depositional area.

There are some absences in the fauna of the Purbeck and Wealden, such as the Order Megaloptera, and the families Chrysopidae (in the Wealden), Myrmeleontoidea (Myrmeleontidae, Palaeoleontidae, Ascalaphidae, Nymphidae) and Brongniartiellidae. The absence of Megaloptera may be due to an absence in England at that time. In the Jurassic they were present in Germany and in the Cretaceous they are known from Baissa, so it may be possible that they did not inhabit England. The environment may not have been suitable for their aquatic larvae, due to the water having varying salinities in both the Purbeck and Wealden. However, other insects with aquatic larva are known, and therefore there was most likely a suitable freshwater environment available. The other alternative is that they have not been preserved; fossil megalopterans are very rare in the fossil record, and the larvae would most likely have not survived the transport distances for deposition

139 in the Purbeck and Wealden. On the other hand it may be that they have yet to be found, and are awaiting discovery. The absence of Chrysopidae in the Wealden again is probably due to the fact that they are awaiting discovery, because chrysopids are common in many Cretaceous deposits, and a chrysopid is known from the Purbeck. Myrmeleontoidea are indicative of semi-arid conditions, and their absence in the more humid Wealden may be expected, but, the absence of all but Nymphidae in the Purbeck is more curious. Myrmeleontids are the major constituent in the neuropterid fauna of the Crato Formation of Brazil, but they are rare in the fossil record in the Cretaceous, with Palaeoleontidae being found in Israel and Brazil, Ascalaphidae in Brazil, Myrmeleontidae in China and Brazil. However, a species is known from the Jurassic of Germany, and therefore there may be a possibility of an English Cretaceous myrmeleontid being found. Brongniartiellidae was previously thought to have been present in the Lower Cretaceous of England, but the specimens assigned to this family are now known to be of the family Psychopsidae (Jepson et al. 2009b (chapter 7)) and therefore this family is not present.

Within the Purbeck and Wealden there is a very similar neuropterid fauna. Psychopsoids are dominant and osmylids and mesoraphidiids are present in each. They differ, however, with the families Berothidae, Chrysopidae, Prohemerobiidae, ?Hemerobiidae, Nymphidae (only found in the Purbeck) and Ithonidae (only found in the Wealden); the Incertae sedis families also differ between deposits. The Purbeck, with regards to Neuroptera, is more diverse than the Wealden, with more families present. Raphidioptera are more or less equal, the differences being that the Wealden has two genera (Mesoraphidia and Proraphidia), while the Purbeck is only represented by Mesoraphidia. The Purbeck raphidiopterans were also much smaller, than the Wealden. Neuropterids generally make up a small proportion of the insect fauna; extant neuropterids generally make up less than 1% of the insect fauna (for example, Borneo has less than 1%, and Queensland, Australia has ~0.45%) which is also consistent with Palaeogene fossil insect deposits (Archibald & Makarkin 2006). In the Purbeck the number of Neuropterids is ~1.5% of recognizable wings (Coram 2005) and in the Wealden they represent 0.1–0.9% from a sample of 13,000 specimens (Jarzembowski 1995) of the insect fauna. Therefore, their abundance in the Lower Cretaceous of Britain is consistent with the extant faunas.

140

The Purbeck and Wealden are amongst the better known neuropterid faunas, comparable to the diverse Asian faunas with regards to Neuroptera families. The snakefly fauna of the Lower Cretaceous of England, and indeed western Europe is not that diverse, compared to the Asian localities, such as Baissa, and China, for example. In western Europe only one family of snakefly has been recorded (Mesoraphidiidae) whereas the Asian localities have two or all three of the extinct families. The reason for this may be that more work is needed on these European deposits, and that more snakeflies could be found either in the field or in museum collections. It could also be a biogeographical reason; the families Alloraphidiidae and Baissopteridae may not have been present in Europe at that time, although this is unlikely because Baissopteridae and Alloraphidiidae at least must have been widespread because fossils of these families are recorded in the Lower Cretaceous of Brazil and the Middle Cretaceous of Canada respectively (Carpenter 1967; Oswald 1990).

The extant neuropterid fauna is very much a relict of their former diversity. Many families which had a global population are now restricted to isolated parts of the world. For example, the family Psychopsidae was widespread during the Cretaceous and it is now only found in isolated parts of the world (New 1988). Neuropterids were globally widespread into the ‗Tertiary‘, after which they started to reduce in numbers, maybe as a result of the end Cretaceous extinction, or the climate change during the ‗Tertiary‘.

141

CHAPTER 11. CONCLUSIONS

From the work on the fossil neuropterids from the Lower Cretaceous Purbeck and Wealden of southern England it has been found that:

1. There are 12 families, 22 genera, 33 species of Neuroptera and 1 family, 2 genera and 7 species of snakefly present in the Purbeck and Wealden.

2. In the Purbeck there are the families Kalligrammatidae, Psychopsidae, Osmylidae, Prohemerobiidae, Chrysopidae, ?Hemerobiidae, Berothidae, Nymphidae, familia incertae sedis and Mesoraphidiidae.

3. The Wealden has Kalligrammatidae, Psychopsidae, Osmylidae, Ithonidae, two familia incertae sedis and Mesoraphidiidae

4. The fauna of the Purbeck is much more diverse at family level than that of the Wealden. However, in both deposits the psychopsoid neuropterans dominate.

5. The first Mesozoic records in England of the families Chrysopidae, Hemerobiidae, Osmylidae, Berothidae and Ithonidae are documented.

6. The family Brongniartiellidae and the Order Megaloptera have been shown to be absent from the Purbeck and Wealden.

7. The oldest fossil record of the family Ithonidae is Principiala rudgwickenesis from the Wealden. The chrysopid Mesypochrysa minuta is the first record of Chrysopidae in the Mesozoic of Europe.

8. Biogeographical links have been shown with the Lower Cretaceous of England and Brazil, China, Russia, Kazakhstan and Mongolia. These links are interpreted from the presence of the genera Sophogramma, Mesypochrysa and Principiala.

9. Both palaeoenvironments for the neuropterids of the Purbeck and Wealden are distant woodlands away from the depositional basin. The insects were transported after undergoing periods of disarticulation before deposition in the basin.

10. Climate change from the Lower Purbeck to the Wealden may have affected the neuropterid fauna. This is shown by the small raphidiopterans within the Lower

142

Purbeck and the loss of Myrmeleontoids in the Wealden, which prefer arid/semi-arid environments.

143

CHAPTER 12. FURTHER WORK

There is much further work to be done with regards to the Purbeck and Wealden. With regards to the palaeoenvironment and palaeoecology, a detailed study of the sediments themselves, such as petrological and geochemical investigations would give more information on provenance of sediment, climate and the environment reconstruction. Also experimental taphonomy would give information on, for example, the transport dynamics of the wings and the decay rate of neuropterids and if the amount of damage of the wings can be correlated to distance travelled to pin- point a more accurate location of the distant habitat for these insects.

Other than Neuropterida, there are vast collections of Purbeck and Wealden insects which are awaiting description. Therefore further taxonomic studies on other insect groups would give a much fuller picture of the entomofauna of the Lower Cretaceous of England.

144

PART TWO:

A STUDY ON FOSSIL AND EXTANT

RAPHIDIOPTERA

The best-laid schemes o’ Mice an’ Men Gang aft a-gley

Robert Burns, from To a Mouse (1785)

145

CHAPTER 13. A STUDY ON FOSSIL AND EXTANT RAPHIDIOPTERA

13.1. Introduction

The phylogeny of raphidiopterans has rarely been investigated. Aspöck et al. (1991) with modifications from Aspöck (1998) constructed a hypothesis for the relationships of extant genera; Willmann (1994) and Ren and Hong (1994) tried to incorporate fossil material into a phylogenetic analysis. Ren and Hong‘s (1994) attempt was less successful than Willmann‘s (1994), whose work provides a good foundation to build upon.

This chapter will investigate the fossil raphidiopterans, producing a key for identification of all fossil families and genera, and a cladistic study to attempt to determine relationships between fossil taxa and extant genera.

13.2. Literature review

This section reviews the literature of relevance to the systematic and phylogenetic study of Raphidioptera. A review of raphidiopteran life-cycle, morphology and phylogeny will be given.

13.2.1. Raphidioptera

Raphidioptera (snakeflies) are a relict Order of insects, which along with Neuroptera and Megaloptera make up the superorder Neuropterida. Raphidiopterans are holometabolous insects, which means they have a distinct larval, adult and pupal stage (Aspöck 2002) (Fig 13.1).

146

Figure 13.1. Showing the eggs (A) of bractea, and the larva (B), pupa (C), adult male (D) and female (E) of ratzeburgi (photographs taken from Aspöck et al. 1991)

Raphidioptera are considered to be the sister-group of Megaloptera + Neuroptera (see Chapter 2.1 and below for detailed discussion). Extant raphidiopterans are split into two families Raphidiidae and Inocelliidae. Raphidiidae has been further subdivided into seven groups which are monophyletic (see below), while Inocellidae relationships still require much study. There are an estimated 260 species of extant Raphidioptera, with currently 185 species in Raphidiidae, and 21 species in Inocellidae (Aspöck 2002). Raphidioptera are considered relicts, or ‗living fossils‘, because they had a higher biodiversity in the Mesozoic, with many different types being found in the Lower Cretaceous (e.g. Aspöck et al. 1991; Aspöck 2002; Engel 2002a). In the Mesozoic they were divided up into three extinct families Baissopteridae, Alloraphidiidae and Mesoraphidiidae, with their first appearance in the fossil record in the Lower Jurassic of Europe. It was previously thought that the family Priscaenigmatidae belonged to the raphidiopterans, but, this family was removed from the Order on the basis of the subcosta not running into the costa (Aspöck & Aspöck 2004).

13.2.2. Raphidioptera morphology

The majority of systematic studies of raphidiopterans are based upon morphology of the body and in particular the terminalia and genitalia of the males. The body of raphidiopterans is easily identifiable from the combination of elongate pronotum and

147 long ovipositor in females. Figure 13.2 shows the morphology of an adult raphidiopteran.

Figure 13.2. A labelled photograph of an adult female raphidiopteran Phaeostigma notata (specimen in Manchester Museum collection).

The genitalia are of particular interest to raphidiopteran systamatics, and allow division of raphidiopterans into families, genera and species. Figure 13.3 shows the labelled genital morphology of adult raphidiopterans.

148

Figure 13.3. Genital morphology of Raphidioptera. A, lateral view of male; B, ventral view of male; C, ventral view of hypandium internum; D, lateral view of female; E, sternite 7 showing position of bursa copulatrix; F, ventral view of bursa copulatrix (redrawn from Pantaleoni et al. 2004).

Raphidioptera have four wings; which are held roof-like over the body when at rest. Figure 13.4 shows a labelled fore- and hind wing; the vein abbreviations are as follows: C, costa; Sc, subcosta; R, radius; Rs, radial sector; M, media; MA, media anterior; MP, media posterior; Cu, cubitus; CuA, cubitus anterior; CuP, cubitus

149 posterior; A, analis; 1-3A, anal veins 1-3; dc, discal cell; 1r, first radial cell; 2r, second radial cell; 3r, third radial cell; ac, anal cell.

Figure 13.4. The labelled fore- and hind wing venation of a raphidiopteran Phaeostigma notata (specimen from the Manchester Museum collection).

13.2.3. Potential problems of using wings for phylogenetic studies

As mentioned in chapter 3, there can be problems in using wing venation as a way of identifying species, because there is not the wealth of information in wing venation as there is in genital morphology, and the wing venation can vary within a species (sexual dimorphism) and even within an individual. Therefore, when dealing with an extant insect, the wing venation is of limited use. However, when dealing with an extinct insect, where the vast majority of fossils are wings, the venation becomes very important, as genitalia and body morphology are rarely preserved. Therefore care is needed to find informative characters in the venation and not ones which can be attributed to sexual dimorphism, or changes within an individual.

13.2.4. Phylogeny of the Raphidioptera

Much work has been done on the phylogeny of extant Raphidioptera at family level, which splits the raphidiopterans into two families; Raphidiidae and Inocellidae, with

150 a sister-group relationship. The apomorphies of the Order were discussed in detail by Aspöck and Aspöck (2004), who listed four: Sc running into the anterior wing margin (Fig. 13.5A), the ninth gonocoxites (ovipositor) of the female terminalia is elongate and tube shaped (Fig. 13.5B), imaginal tarsi with third tarsomere bilobed (Fig. 13.5C), and segment 9 in the male genital sclerites: the tergite and sternite form a ring (Fig. 13.5D). Other characters shared by extant raphidiopterans are the trichobothria in the male terminalia loosely forming a band (Fig. 13.5E) and the labial palp of the prementum in the larva is segment-like (Fig. 13.5F) (Aspöck et al. 2001; Aspöck & Aspöck 2004, 2008).

151

Figures 13.5. Apomorphies of Raphidioptera (Aspöck et al. 2001; Aspöck & Aspöck 2004, 2008). A. Sc running into C, B. Ovipositor (redrawn and modified from fig. 51 Aspöck et al. 1991). C. Bilobed 3rd tarsomere (redrawn and modified from fig. 17 Aspöck et al. 1991), D. Tergite 9 and Sternite 9 (redrawn and modified from fig. 3 Aspöck et al. 1991), E. Trichobothria (redrawn and modified from fig. 1036 Aspöck et al. 1991), F. Larval head capsule (redrawn and modified from figs. 55 and 60 Aspöck et al. 1991).

The apomorphies of the family Raphidiidae are: the lateral margins of the pronotum are bent down over to the ventral side, where they almost overlap in the middle (Fig. 13.6A), in Inocellidae the pronotum is rectangular and it does not appear ventrally, and the sternal sclerites are fused in the prothorax forming to large plates (Fig. 13.6B), in Inocellidae the sclerites are not fused medially.

152

The apomorphies of Inocellidae are an absence of ocelli (Fig. 13.6C), shortening and funnel-like shape of the second tarsi (Fig. 13.6D).

Figure 13.6. The apomorphic characters of Inocellidae and Raphidiidae. A. the Pronotum (redrawn and modified from figs. 9 and 33 Aspöck et al. 1991), B. The sternal sclerites (redrawn and modified from figs. 9 and 33 Aspöck et al. 1991), C. The ocelli (redrawn and modified from figs. 1 and 27 Aspöck et al. 1991), D. The 2nd tarsomere (redrawn and modified from figs. 17 and 42 Aspöck et al. 1991).

Other characters, which separate Raphidiidae and Inocellidae from each other are: sternite 8 being absent in Raphidiidae, is present in Inocellidae, the sclerites are fused in the paramere (Aspöck et al. 2001). Molecular data has corroborated the existence of the two raphidiopteran families (Haring & Aspöck 2004).

At generic level little work has been done on the relationships of extant Raphidioptera. Aspöck et al. (1991) proposed a hypothesis for a phylogeny based upon morphological characters (Fig 13.7 and 13.8). The extant raphidiopteran genera of Raphidiidae are split into 8 groups: Group I – Phaeostigma, ,

153

Tjederiraphidia, Turcoraphidia, Iranoraphidia, Tauroraphidia, Subilla, Ornatoraphidia, Xanthostigma, Parvoraphidia, Ulrike, Raphidia; Group II – Atlantoraphidia, Harrarphidia, Hispanoraphidia, Africoraphidia, Ohmella, Italoraphidia, Puncha, Group III – Venustoraphidia, Mauroraphidia; Group IV – Tadshikoraphidia; (Group V – Usbekoraphidia, synonymized with Mongoloraphidia (Aspöck et al. 1998), Group VI – Mongoloraphidia; Group VII – Agulla; Group VIII – Alena (Aspöck et al. 1991).

The characters used for Aspöck et al.‘s (1991) hypothesis are entirely based upon the morphology, in particularly the majority are based upon the genitalia and terminalia, especially of the male. Listed below are the characters used by Aspöck et al. (1991), which have been listed and numbered, the numbers corresponding to the character on the phylogenetic trees figures 13.7 and 13.8.

Figure 13.7. Aspöck et al.‘s (1991) hypothesis of relationships of Raphidiidae, with the characters numbered (key to the numbers in text) (modified and redrawn from Scheme 1, 2 and 3 Aspöck et al. 1991).

154

Raphidiidae characters (from Aspöck et al. 1991):

1. Median unpaired hypovalva 2. Massive sclerotization of parameres 3. Rod shaped parameres 4. Disc-shaped parameres 5. Lateral membranous hypovalva 6. Basally shortened ectoproct 7. Paramere-Hypovalva complex 8. Strengthening and dorsal shift of the stylus of the gonocoxites in the male 9. Narrow connection of the 7th sternite and 8th sternite in females 10. Large styli 11. Strong modification of stylus carrying part of the gonocoxites 12. Enlargement of the ectoproct 13. Differentiation of the paired contact places to the 9th sternite in males 14. Enlargement of the hypandium inturnum in the male 15. Shortening of MA in the hind wing 16. Pigmentation pattern in the larvae 17. Presence of intersegmental sternite 7/8 in females 18. Triangular, tapering head of the male 19. Reduction of hypovalva 20. Reduction of the gonocoxites base in males, breaking the connection with the hypovalva base. 21. Hypovalva with paired sclerites 22. Narrow coupling of the 8th sternite and the 9th sternite in males 23. Differentiation of the parameres 24. Anterior strongly convex in 8th tergite in the female 25. Paramere sclerite lying over hypovalva 26. Sclerite lying between styli in paramere 27. Enlarged ectoproct male 28. Strongly reduced paramere 29. Forked styli in the male 30. Poorly shaped apex of gonocoxites 31. Forked styli in the male

155

32. The form of the bursa copulatrix in the female 33. Terminal divergence of apices in hypovalva 34. Strongly sculptured ductus receptaculi in the female 35. Reduced gonocoxites 36. Large styli 37. Narrow coupling of the 8th sternite and 9th segment 38. Lack of parameres 39. Reduction of the ectoprocts 40. The obliteration of the 9th sternite 41. Trapezoid form of the 9th tergite 42. Lack of unambiguous trichobothria on the ectoproct of the male 43. Coupling of the 8th sternite at the 9th segment of the male 44. Paramere equipped with denticle sculpture 45. Possession of paramere 46. Paired differentiated hypovalva 47. Fusion of the basal sclerite with the gonocoxites in the male 48. Curved hypovalva 49. Tunnel-like merged parameres 50. The ‗hook and loop-fastener‘ connection 51. Bent styli 52. The apices of the gonocoxites arranged proximally, and black pronotum. 53. Fusion of the parameres arch with the hypovalva 54. Long sickle-shaped styli; unusually longer, positioned far forwards passing sacculus bursae 55. Hypovalva sunken base 56. Reduction of the parameres 57. 7th sternites with bristles in the female 58. Long, slender head 59. Long, slender pronotum 60. Construction of parameres 61. Strong extension of the 7th sternites of the female 62. Pigmentation pattern of the abdominal segments of larvae 63. The dorsal, plate-like extension of the stylus base 64. Construction of the parameres

156

65. The dentition of hypovalva 66. The highly developed intersegmental S 7 / 8 of the females. 67. Long yellow pterostigma 68. The shape of the parameres 69. Male: paired 8th Sternites 70. Ventral expansion of the ectoprocts 71. Glandular ducts of the seminal receptacle with spherical sections in the female 72. Modified Gonocoxites: the dorsal part with a ‗lever-like‘ stylus, while the ventral part is reduced almost completely 73. Reduction of the pterostigma 74. The unpaired, reduced parameres 75. The gonocoxites-apex ventrozephale 76. The bulge of the 8th tergite in the females 77. Dorsal transformation of the 9th tergites 78. U-shaped posterior border of gonocoxites 79. Tendency to a latero-caudal widening of the ectoprocts in males 80. A distinct, narrow ductus sacculi in females 81. The capsule-sclerotised bursa copulatrix of the female 82. Extension of the ductus receptaculi in females 83. The loss of the parameres in males 84. The widely divergent base of the hypovalva 85. The loss of the gonarcus 86. The loss of the parameres 87. The sclerotised, serrated hypovalva of the male 88. The huge schlid shaped parameres of male 89. The long sculpted receptaculi duct 90. The terminal cone on the spermatheca of the male 91. The angle of anterior apices of hypovalva and the associated median, large- sized and paired parameres in male 92. Paired sclerites in the bursa copulatrix 93. Stalked glandular ducts of the seminal receptacle in the female 94. Apex development on 9th tergite 95. Reduction of 9th sternites, linking the styli

157

96. Rod-shaped parameres in the male 97. A ring-shaped connection between the tergite and sternite of 8th segments among males 98. The long slender pronotum 101. The poor-shaped processus of the gonocixites of the male 102. Tendency to the obliteration of the crossveins vein in the pterostigma 103. The lightening of the abdominal sclerite 104. The form and dentition of parameres and gonarcus and their ‗hook-and- loopfastener-like‘ connections 105. The separation of gonocoxites of the 9th tergite and 9th sternite 106. Adhesion and fusion of atrium bursae and receptaculum seminis in females

Calabroraphidia was not included in Aspöck et al.‘s (1991) the study, because it was only described in 2004 (Rausch et al. 2004).

Figure 13.8. Aspöck et al.‘s (1991) hypothesis of inocelliid relationships, with the characters numbered (key to numbers in text) (modified and redrawn from Scheme 4 Aspöck et al. 1991).

Inocellidae characters (from Aspöck et al. 1991):

1. MA base in the hind wing not as a crossvein 2. MA base in hind wing cryptic 3. Serratulum absent

158

4. Strongly differentiated styli 5. 8th tergite divided, appearing lateral in females 6. Bristle platelets and groups in the endophallus base and inside of the gonocoxites 7. Hook-shaped pseudostyli 8. Monofiliform anntenae with bristles in male 9. Paired arcessus

Sinoinocellia: No male specimen therefore difficult to place within a phylogeny.

Aspöck et al., (1991) stress that these are only working hypotheses of the relationships within Raphidioptera, with many of the relationships not being strongly supported. For example, the inocelliid sister-group relationships are not based on very strong synapomorphies, and even the genera themselves are given tentative apomorphies (Aspöck et al. 1991). However, the hypotheses given do form an admirable starting point.

When it comes to incorporating fossils into the raphidiopteran phylogeny, very little has been done, with only two attempts made. One of these attempts is at the family level (Ren & Hong 1994) and the other at the generic level (Willmann 1994). Ren and Hong‘s (1994) attempt was strongly criticized by Engel (2002) for not being very rigorous, with only 6 characters used, and the characters themselves were dubious with no quantification given in many cases, for example, Prothorax: (0) shorter, (1) longer and Rs of forewing originates (0) just before midwing. (1) distad of midwing.

Two of Ren and Hong‘s (1994) taxa used have been synonymized; Jilionraphidiidae and Huaxiaraphidiidae are considered to be in the family Mesoraphidiidae (Engel 2002).

The characters used in this study were:

1. Oblique vein in the pterostigma 2. MA in hind wing arises from R or M 3. R and M regions with more crossveins or less 4. Prothorax (shorter or longer)

159

5. In forewing: MA arises from M before or after MP, or originates well beyond forking of MP 6. RS of forewing originates just before, or distad of midwing

Figure 13.9. Ren and Hong‘s (1994; Engel 2002) phylogeny of Raphidioptera families (redrawn from Ren & Hong 1994).

Willmann (1994) created a phylogeny based on selected extinct genera of Raphidioptera, as well as the two extant families Raphidiidae and Inocelliidae. Willmann‘s efforts are more robust than that of Ren and Hong (1994), using 36 characters to construct his phylogeny, all based on variations within the wing venation

Characters used:

1. Decrease in number of cells between R and RS. 2. Base of cell between Rs and M is shortened by the change in the base of MA. 3. The free base of M, between RS and M, and the proximal m-cu crossvein is shortened (later becoming reversed). 4. The 3rd m-cua crossvein (counted from the wing base) leads to the formation of a large cell behind M. 5. Sc runs into C. 6. Rs separates from R near the end of the first wing third (original state R-Rs fork near the wing base. Leads to a shortening of the radial cell. 7. The first cell between the Media branches, expands basalwards.

160

8. Basal stem of M becomes inverse 9. Lengthening of the basal m-cu crossvein. 10. Basal cell between M and Cu extended distally 11. Anal loop in hind wing 12. Length of prothorax with regard to head 13. Two Cu crossveins present in hind wing 14. Three rs-m crossveins 15. Lengthening and re-orientation of basal m-cu crossvein 16. Four crossveins between R and Rs in hind wing 17. Short basal m-cu crossvein slightly curved in hind wing 18. Formation of 2r and discal cell 19. Loss of one of three rs-m crossveins 20. Extension or shortening of r-rs crossvein 21. A) A3 merged with the rear curve of the anal loop. B) Comb-like arrangement of the rear branches of A2 and A3. 22. Hind wing: partial coalescence of CuA1 and M3+4, if divergent the cell gets lost 23. Narrow wing 24. Wing with narrow base. Intramedian cell lies basalwards of R/Rs fork and basal cell between Rs+M. 25. Hind wing: Cu2 partially fused with A1 26. Hind wing with only 3 r-rs crossveins 27. Broad wing 28. M and CuA touch 29. Basal cut-off of Rs is midwing 30. Hind wing: basal cut-off of Rs (basal branch of MA of some authors) arises from proximal part of M not R. 31. Disappearance of the basal cells behind the Intramedian cell 32. Disappearance of the Intramedian cell 33. Basal cell between Rs and M, misaligned in the direction of the wing apex 34. Crossvein present between R, M and Cu (basal segment of M). 35. Another crossvein between R and Rs 36. Hind wing: cell between M3+4 and CuA1 present

161

Figure 13.10. Willmann‘s (1994) raphidiopteran phylogenetic tree. A. Willmann‘s complete phylogenetic tree with characters (defined in text) (redrawn from fig. 7 Willmann 1994), B. Alternative arrangement for basal groups (redrawn from fig. 10 Willmann 1994), C. Alternative arrangement for younger groups (redrawn from fig. 11 Willmann 1994).

Aspöck et al. (1991), Ren and Hong (1994) and Willmann (1994) represent the only work as yet completed on trying to resolve the relationships of raphidiopterans. The work of Aspöck et al. (1991) is a preliminary working hypothesis, the work of Ren and Hong (1994) is unfortunately full of problems, and only based on family relationships. Willmann‘s (1994) work is a good foundation to build on, but, there may be some doubt as to his character choices, and he does not use all genera, just a select few. Hopefully the study herein will be able to shed some light upon the generic relationships within the Order Raphidioptera.

162

13.3. Keys to the Families and Genera of Fossil Snakeflies (Insecta: Raphidioptera)

(Jepson, J.E. Keys to the Families and Genera of Fossil Snakeflies (Insecta: Raphidioptera). To be submitted to Zootaxa)

13.3.1. Abstract

Identification keys have been constructed for the first time on all fossil raphidiopterans described to date. The keys offer an identification to family level, for all the five families (extant and extinct) and generic level within these families. All of the keys are primarily based upon the wing venation, which is predominantly used as the method of identification of the fossil taxa.

Keywords: Identification keys, Neuropterida, Raphidiidae, Inocelliidae, Alloraphidiidae, Mesoraphidiidae, Baissopteridae.

13.3.2. Introduction

The first record of Raphidioptera in the fossil record is from the Lower Jurassic with two species of Metaraphidia (Mesoraphidiidae) currently in the Lias of Britain and Germany (Whalley 1985; Willmann 1994). There are five families of Raphidioptera, both extant and extinct; these are the extinct Baissopteridae, Alloraphidiidae, Mesoraphidiidae and the extant Raphidiidae and Inocellidae (Engel 2002a). The family Priscaenigmatidae was previously thought to be raphidiopteran, however, it has now been removed from the Order (Aspöck & Aspöck 2004). There are approximately 32 genera and 96 species of fossil Raphidioptera described to date. For a review of fossil Raphidioptera see Engel (2002a) and Jepson and Jarzembowski (2008).

This study is the first attempt at a construction of an identification key for the families and genera of Raphidioptera. The keys are focused on the wings of fossil Raphidioptera, with some morphological data added. The keys to the families and genera are based on all the fossil raphidiopterans described to date.

The wings are the most diagnostic feature of fossil raphidiopterans; therefore they are the main focus of these keys. The vast amount of morphological data, especially the male genitalia used in the identification of extant species is very rarely preserved

163 in the fossil record to allow it to be of any use in identification.

Figure 13.11. The labelled wing venation of the fore- (top) and hind (bottom) wing of Mesoraphidia luzzi Grimaldi, 2000.

Figure 13.11 shows the venation of a fore- and hind wing of a fossil raphidiopteran (Mesoraphidia luzzii Grimaldi et al. 2000) with the terminology used in the keys below, other features are highlighted in figures referenced in the keys. The abbreviations are C, costa; Sc, subcosta; R1 radius; Rs, radial sector; M, media; MA, media anterior; MP, media posterior; Cu, cubitus; CuA, cubitus anterior; CuP, cubitus posterior; A, analis; 1A, first branch of analis; 2A, second branch of analis; 3A, third branch of analis; 1-3r, radial cells; dc, discal cell; mc1-3, medial cells; ac, anal cell.

13.3.3. Key to families

Included families: Baissopteridae, Alloraphidiidae, Mesoraphidiidae, Raphidiidae and Inocelliidae.

1. Sc terminating on the anterior wing margin, females with long ovipositor

164

Figure 13.12. Displaying the termination of Sc. A. is a raphdiopteran, B. is a neuropteran.

Yes...... 2 (Fig. 13.12A)

No...... Not Raphidioptera (Fig. 13.12B)

2. Wings long and slender

Figure 13.13. Displaying the different wing shapes within Raphidioptera. A. elongate, B. non-elongate.

Yes...... 3 (Fig 13.13A)

No...... 4 (Fig 13.13B)

3. Forewing: Many cells between Rs and MA (15+) in the forewing

165

Figure 13.14. Displaying the cells between Rs and MA. A many, B few.

Yes...... Baissopteridae (Fig. 13.14A)

No...... 5 (Fig. 13.14B)

4. Separation of CuA from M very close to the stem of R forming a trifurcation in forewing

Figure 13.15. Showing the trifurcation of R1, M and CuA.

Yes...... Alloraphidiidae (Fig. 13.15)

No...... 5

5. Free branch of MA fuses with R1 for short period in forewing (Fig 13.16A) and free analis in the hind wing (Fig 13.16B)

166

Figure 13.16. Displaying the fusion of MA with R1 (A) and the free analis in the hind wing (B).

Yes...... Inocelliidae

No...... 6

6. A combination of 2 medial cells in the forewing and a short pterostigma (pterostigma shorter or same size as 2r) (Fig 13.17)

Figure 13.17. Displaying the medial cells and short pterostigma.

Yes...... Raphidiidae

No...... Mesoraphidiidae

Any specimens that do not fit the above key = incertae sedis

13.3.4. Key to Genera

Baissopteridae Key:

Included genera: Baissoptera, Cretinocellia, Lugala, Cretoraphidia, Austroraphidia and Cretoraphidiopsis.

1. Number of medial cells (m, Fig. 13.11)

167

Few (3) ...... Cretinocellia

Many (4+)...... 2

2. Origin of the free base of MA with respect to the fork of MP in forewing

Figure 13.18. The origin of the free branch of MA with respect to the fork of MP. After (A), at (B) and before (C).

At or just before...... 3 (Fig. 13.18B, C)

After...... 4 (Fig. 13.18A)

3. Wings narrow and elongate, especially hind wing (see Fig. 13.13). Two cua-cup crossveins in hind wing (Fig. 13.19).

Figure 13.19. Showing the cua-cup crossveins in the hind wing.

Yes...... Lugala

No...... 5

168

4. Many branches of the radial sector (greater than 3) (Rs, Fig. 13.11), many medial cells (greater than 5) (m, Fig. 13.11), costal area broader than pterostigma in forewing.

Yes...... Cretoraphidia

No...... Austroraphidia

5. First cu-crossvein after fork of CuA and MP (Fig 13.20)

Figure 13.20. Showing the relationship between the first cu crossvein and the fork of CuA and MP.

Yes...... Cretoraphidiopsis

No...... Baissoptera

Any specimens that do not fit the above key = incertae sedis

Alloraphidiidae Key:

Included genera: Alloraphidia, Archeraphidia and Pararaphidia.

1. Both fore- and hindings very thin and slender (see Fig. 13.13)

Yes...... Alloraphidia

No...... 2

2. MA arising at or before the fork of MP in forewing (see Fig. 13.18)

Yes...... Archeraphidia

No...... 3

3. Distal part of fore- and hind wings broad and rounded at apex, 1r and 2r broad.

169

Yes...... Pararaphidia No...... Alloraphidia

Any specimens that do not fit the above key = incertae sedis

Mesoraphidiidae Key:

Included genera: Mesoraphidia, Proraphidia, Baisoraphidia, Siboptera, Huaxiaraphidia, Sinoraphidia, Kezuoraphidia, Xuraphidia, Metaraphidia, Nanoraphidia, Iberoraphidia and Cantabroraphidia.

Wing length below 6 mm:

Yes...... 1

No...... 3

1. Medial cells (m, Fig. 13.11):

Three...... Mesoraphidia

Two...... 2

One...... Nanoraphidia

2. 1r and 2r same width and thickness (Fig. 13.11):

Yes...... Cantabroraphidia

No...... Mesoraphidia

3. Pterostigma (pt, Fig. 13.11):

Short (same size or shorter than 2r, origin in the final wing third)...... Proraphidia

Divided...... Iberoraphidia

Long (longer than 2r)...... 4

4. M horizontal (~ parallel to R) before forking, after split with CuA (Fig. 13.21), Rs originates at mid-wing in the forewing, MA free branch originates very distal from M-CuA fork in the forewing

170

Figure 13.21. Showing the horizontal M ~parallel to R, origin of Rs, and the distal free branch of MA.

Yes...... Siboptera

No...... 5

5. Medial cells (m, Fig. 13.11) enlarged, and the discal cell ~one third of wing length (Fig. 13.22).

Figure 13.22. Showing the enlarged medial and discal cells

Yes...... Kezuoraphidia

No...... 6

6. Origin of MA very distal of fork of MP (see Fig. 13.18); MP forking close to R1

Yes...... 7

No...... 8

8. CuA pectinately branched (greater than 3)

Yes...... Sinoraphidia

No...... Xuraphidia

171

9. Numerous cup-a1 crossveins (greater than 3) and cua-cup crossveins, the presence of cua-m crossveins, and crossveins forming small cells between A2 and A3, ± pectinate A1 (Fig. 13.23):

Figure 13.23 showing the cup-a1, cua-cup and cua-m crossveins, small cells and pectinate A1.

Yes...... Huaxiaraphidia

No...... 10

10. A2 and A3 not fused in the forewing, 4 r1-c crossveins in final wing third (pterostigmal area) (Fig. 13.24):

Figure 13.24. Displaying the rs-c crossveins and not fused A3

Yes...... Metaraphidia

No...... 11

11. Pterostigma (pt, Fig. 13.11) starts before final wing third

172

Yes...... Mesoraphidia

No...... Baisoraphidia

Any specimens that do not fit the above key = incertae sedis

Incertae sedis key:

1. 2 r-sc (pterostigmal crossveins), extra rs-r1 crossvein, and lack of crossveins in radial and medial areas (this may be a preservational artefact) (Fig. 13.25)

Figure 13.25. Showing MA fused with R, the extra rs-r1 crossvein and the r-sc (pterostigmal) crossveins.

Yes...... Styporaphidia

No...... 2

2. 1 pterostigmal crossvein, cua-a1 crossvein in forewing and single medial cell in hind wing (Fig. 13.26).

Figure 13.26. Showing the r-sc and cup-a1 crossvein in forewing and medial cell in hind wing.

Yes...... Ororaphidia

173

No...... 3

3. Long pterostigma (longer than 2r), 3r small, MA and MP with gradate series of crossveins parallel to posterior wing margin. Head round and pronotum smaller than mesothorax.

Yes...... Arariperaphidia (based on fragmentary wing venation)

No...... 4

4. Less than 3 medial cells (m, Fig. 13.11) in forewing

Yes...... Archiinocellia

Raphidiidae Key:

Included genera: ‘Raphidia’, ‘Ohmella’, ‘Agulla’, Succinoraphidia

1. Short branch of M (before MA and MP split) in hind wing (Fig. 13.27)

Figure 13.27. Showing the stem of M in the hind wing. A, long; B, short.

Yes...... 2 (Fig. 13.27B)

No...... 3 (Fig. 13.27A)

2. Gonapophyses 9 of the male: rod-like, not being connected ventrally.

Yes...... Succinoraphidia

No...... ‘Raphidia’*

Note: Difficult to differentiate venationally.

174

3. Highly branched (over 2) apical terminus of R in hind wing, radial area not narrowed proximally (Fig. 13.28)

Figure 13.28 displaying the proximal radial area and apical terminus of R.

Yes...... ‗Agulla’*

No...... ‘Ohmella’*

*The placement of specimens represented by just wings in the extant genera Raphidia, Ohmella, and Agulla cannot be convincingly supported by wing venation (Aspöck & Aspöck 2004).

Inocelliidae Key:

Included genera: ‘Fibla’, Electrinocellia, Succinofibla

1. Narrow costal area in forewing; MP 3+4 not bifurcated in fore- and hind wing; MP with 2 medial cells in hind wing (Fig. 13.29)

Figure 13.29. Showing the features of Electrinocellia

175

Yes...... Electrinocellia

No...... 2

2. Wide costal area in forewing; second rs-sc before termination of Sc (Fig. 13.30)

Figure 13.30. Showing the second sc-r crossvein terminating before the termination of Sc.

Yes...... ‗Fibla’*

No...... Succinofibla

Note: Succinofibla is diagnosed on the basis of the Gonocoxites 9 of the male with an armlike processus in a terminal position and wartlike tubercles at the inner side.

*The placement of specimens represented by just wings in the extant genus Fibla cannot be convincingly supported by wing venation (Aspöck & Aspöck 2004).

176

13.4. A Cladistic Study on Fossil Snakeflies (Insecta, Neuropterida, Raphidioptera)

(Jepson, J.E. A Cladistic Study on Fossil Snakeflies (Insecta, Neuropterida, Raphidioptera). To be submitted to Systematic Entomology)

13.4.1. Abstract

A cladistic analysis of fossil snakeflies using 21 characters and 28 genera has been undertaken. The characters are based upon wing venation with some easily preserved body morphology. Little resolution has been obtained between the genera. The families Baissopteridae and Mesoraphidiidae have shown to be paraphyletic. Some poor to well supported clades have been resolved. The baissopterids are placed basally to the rest of the snakeflies, as expected and observed in previous cladistic studies. Fifteen synapomorphies have been established from the analysis. The problems of cladistic analyses upon fossil and extant snakeflies is also discussed.

Keywords. Cladistics, Snakeflies, Mesoraphidiidae, Baissopteridae, Alloraphidiidae, Raphidiidae, Inocelliidae

13.4.2. Introduction

Very few cladistic or phylogenetic studies have been completed on raphidiopterans. Aspöck et al. (1991), gave hypotheses on the relationships of the extant raphidiopterans, and also highlighted the problems associated with finding strong synapomorphies to support many of the relationships of the genera. They divided the extant Raphidiidae into eight groups; the largest group, group I contains the genera Phaeostigma, Dichrostigma, Tjederiraphidia, Turcoraphidia, Iranoraphidia, Tauroraphidia, Subilla, Ornatoraphidia, Xanthostigma, Parvoraphidia, Ulrike, and Raphidia, group II consists of Atlantoraphidia, Harraraphidia, Hispanoraphidia, Africoraphidia, Ohmella, Italoraphidia, and Puncha, group III has the genera Venustoraphidia and Mauroraphidia, group IV contains only one genus; Tadshikoraphidia, group V has Usbekoraphidia, group VII Agulla and group VIII Alena (Aspöck et al. 1991). Usbekoraphidia was synonymised with Mongoloraphida, therefore removing group V (Aspöck 1998). Aspöck et al. (1991) also did a tentative phylogenetic tree on the extant Inocelliidae. The Aspöcks‘

177 hypotheses, however, did not take into account the fossil raphidiopterans, the only two attempts of working out the relationships of fossil and extant are Ren and Hong (1994), who worked at the family level, and Willmann (1994), who used a selection of the genera. Ren and Hong‘s (1994) attempt at a phylogeny was not as rigorous as that of Willmann‘s (1994) whose phylogeny has created a good foundation to build upon.

This study is intended to give an updated starting point for the difficult task of incorporating fossil and extant raphidioptera into a phylogeny, however, the focus is on fossil raphidiopterans, with the relationships of extant genera being taken from the hypotheses put forward by Aspöck et al. (1991), Aspöck and Aspöck (1994), and Aspöck (1998).

13.4.3. Material and Methods

13.4.3.1. Body and wing morphology

The characters used in this study are a combination of both body and wing morphology; however the primary focus is on wing venation. This is because the fossil taxa are for the most part, described by their wing venation, because diagnostic body parts are rarely preserved. Genitalia characters are used to separate the extant taxa into their hypothesized relationships (Aspöck et al. 1991; Aspöck 1998); these however, are of little use for fossil raphidiopterans, which are usually preserved as isolated wings.

13.4.3.2. Taxa

The taxa used in this study are all the genera of Raphidioptera described, both extinct and extant. The extant taxa have been studied through the use of Aspöck et al.‘s (1991) monograph on the world fauna of Raphidioptera, with updates of revised taxonomic position and new genera (Aspöck & Aspöck 1994; Aspöck 1998; Rausch et al. 2004). The fossil material is studied from the numerous published work on Raphidioptera. The majority of the measurements have been taken from drawings, therefore, inaccuracies in drafting may give errors however, photographs have been used when they can. The taxa used with geological ranges are listed in table 13.1.

178

Family Genus Extinct/Extant Geological Range

Raphidiidae Phaeostigma Extant Recent

Raphidiidae Dichrostigma Extant Recent

Raphidiidae Tjederiraphidia Extant Recent

Raphidiidae Turcoraphidia Extant Recent

Raphidiidae Iranoraphidia Extant Recent

Raphidiidae Tauroraphidia Extant Recent

Raphidiidae Subilla Extant Recent

Raphidiidae Ornatoraphidia Extant Recent

Raphidiidae Xanthostigma Extant Recent

Raphidiidae Parvoraphidia Extant Recent

Raphidiidae Ulrike Extant Recent

Raphidiidae Raphidia Extant Eocene–Recent

Raphidiidae Atlantoraphidia Extant Recent

Raphidiidae Harraphidia Extant Recent

Raphidiidae Hispanoraphidia Extant Recent

Raphidiidae Africoraphidia Extant Recent

Raphidiidae Ohmella Extant Miocene–Recent

Raphidiidae Italoraphidia Extant Recent

Raphidiidae Puncha Extant Recent

Raphidiidae Venustoraphidia Extant Recent

Raphidiidae Mauroraphidia Extant Recent

Raphidiidae Tadshikoraphidia Extant Recent

Raphidiidae Mongoloraphidia Extant Recent

Raphidiidae Agulla Extant Miocene–Recent

Raphidiidae Alena Extant Recent

Raphidiidae Usbekoraphidia Extant Recent

Raphidiidae Calabroraphidia Extant Recent

Raphidiidae Succinoraphidia Extinct Eocene

179

Inocelliidae Fibla Extant Eocene–Recent

Inocellidae Parainocellia Extant Recent

Inocelliidae Inocellia Extant Recent

Inocellidae Indianocellia Extant Recent

Inocelliidae Extant Recent

Inocelliidae Succinofibla Extinct Eocene

Inocelliidae Electrinocellia Extinct Eocene

Inocelliidae Sinoinocellia Extinct Lower Cretaceous

Baissopteridae Austroraphidia Extinct Lower Cretaceous

Baissopteridae Baissoptera Extinct Lower Cretaceous

Baissopteridae Cretinocellia Extinct Lower Cretaceous

Baissopteridae Cretoraphidia Extinct Lower Cretaceous

Baissopteridae Cretoraphidiopsis Extinct Lower Cretaceous

Baissopteridae Lugala Extinct Lower Cretaceous

Alloraphidiidae Alloraphidia Extinct Lower–Middle Cretaceous

Alloraphidiidae Archeraphidia Extinct Lower Cretaceous

Alloraphidiidae Pararaphidia Extinct Lower Cretaceous

Mesoraphidiidae Iberoraphidia Extinct Lower Cretaceous

Mesoraphidiidae Baisoraphidia Extinct Lower Cretaceous

Mesoraphidiidae Huaxiaraphidia Extinct Lower Cretaceous

Mesoraphidiidae Jilinoraphidia Extinct Lower Cretaceous

Mesoraphidiidae Kezuoraphidia Extinct Lower Cretaceous

Mesoraphidiidae Mesoraphidia Extinct Upper Jurassic–Upper Cretaceous

Mesoraphidiidae Metaraphidia Extinct Lower Jurassic

Mesoraphidiidae Nanoraphidia Extinct Upper Cretaceous

Mesoraphidiidae Proraphidia Extinct Upper Jurassic– Lower Cretaceous

Mesoraphidiidae Siboptera Extinct Lower Cretaceous

Mesoraphidiidae Sinoraphidia Extinct Upper Jurassic– Lower

180

Cretaceous

Mesoraphidiidae Xuraphidia Extinct Lower Cretaceous

Mesoraphidiidae Cantabroraphidia Extinct Lower Cretaceous

Incertae sedis Ororaphidia Extinct Middle Jurassic

Incertae sedis Styporaphidia Extinct Middle Jurassic

Incertae sedis Arariperaphidia Extinct Lower Cretaceous

Incertae sedis Archiinocellia Extinct Oligocene

Table 13.1. The extant and extinct genera of snakeflies with geological ranges.

The outgroup taxa is the genus Permoberotha (: Permoberothidae) chosen because of its basal position to Neuropterida. (Carpenter 1943a; Grimaldi & Engel 2005). The genus Permoberotha (Fig. 13.31) is from the Permian of Kansas. It is likely that the Permoberothidae are stem-group Neuropterida (Grimaldi & Engel 2005).

Figure 13.31. The fore- and hind wing venation of the Permoberotha, the outgroup taxa of the study (after Carpenter 1943a).

13.4.3.3. Potential problems

The major problem with the incorporation of fossils into a phylogeny with extant raphidiopterans is that the vast amount of diagnostic characters are not preserved within the fossils. Most of the extant raphidiopteran species are diagnosed by the form and structure of the genitalia, or other body morphology, which very rarely get preserved in the fossil record, the rare occurrences are most often than not in amber fossils. For example, Aspöck and Aspöck (2004) were able to get many diagnostic characters from Succinofibla and Succinoraphidia in Baltic amber, than are normally

181 found in fossils. The majority of snakefly fossils are described based on wing venation, which in extant snakeflies can be very variable within genera, species and in some cases the individual. Therefore, how reliable the taxonomy of fossil snakeflies is, is debatable, many taxa in recent years have been synonymised (Engel 2002a).

Another problem which includes the extant taxa; is that apomorphies are very hard to decipher for genera, and when constructing relationships, synapomorphies are difficult to determine, and many of the resulting relationships are tentative. With fossil taxa the taxonomy of many groups is in need of revision, for example the genus Mesoraphidia appears to be a bucket group for many species, and some genera are in need of revision, and are possible synonyms of previously described genera. This therefore can cause erroneous results of any phylogenetic study, and care must be taken in interpreting the results.

13.4.4. Cladistic study

13.4.4.1. Characters

The charcters used in the analysis are listed below with their coding (Table 13.2). They are mainly based upon wing venation, however some commonly preserved morphological characters are used. A labelled figure of raphidiopteran wing venation is given (Fig. 13.32).

182

Figure 13.32. Showing the venation of a fore- and hind wing of a raphidiopteran (Mesoraphidia luzzi after Grimaldi 2000)

1: Number of cells of Rs and M in the forewing

[0] Many (>17), [1] Few (<17)

2: Number of medial cells in the forewing

[0] Many (>5), [1] Few (<5)

3: Trifurcation of R, M and Cu in the forewing

[0] absent, [1] present

4: Presence of a gradate series of crossveins in the forewing

[0] present, [1] absent

5: Pterostigma

[0] absent, [1] longer than 2r, [2] shorter than 2r

6: Head

[0] quadrate, [1] triangular

183

7: Number of cells of R and Rs in the forewing

[0] >6 [1] 3–6, [2] <3

8: The 2r and discal cell in forewing

[0] present, [1] absent

9: Number of r-rs crossveins in forewing

[0] many (>3), [1] few (<3)

10: Termination of Sc on C in both fore- and hind wings

[0] absent, [1] present

11: The presence of ocelli on the head capsule

[0] absent, [1] present

12: Basal cut off of Rs at the mid wing

[0] no, [1] yes

13: M and Cu touch

[0] no, [1] yes

14: ma-mp crossveins in the forewing

[0] >2, [1] <2

15: 2 cu-m crossveins in the hind wing

[0] no, [1] yes

16: r-rs crossveins in hind wing

[0] >4, [1] <4

17: Pterostigma

[0] not divided, [1] divided

18: Wing shape

184

[0] non-elongate, [1] elongate

19: Cubital area crossveins

[0] >2, [1] <2

20: M base

[0] at an angle, [1] horizontal

21: Wing size

[0] > 6 mm, [1] < 6 mm

185

TAXA CHARACTERS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Permoberotha 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Austroraphidia 0 0 0 0 1 0 0 0 0 1 0 0 0 1 0 1 0 0 0 0 0 Baissoptera 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Cretinocellia 0 0 0 0 1 0 0 0 0 1 0 0 0 1 1 1 0 0 0 0 0 Cretoraphidia 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 Cretoraphidiopsis 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 Lugala 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 1 0 1 0 Alloraphidia 1 1 0 1 2 1 1 1 1 1 0 0 0 1 1 1 0 1 1 0 0 Archeraphida 1 1 0 1 2 1 1 1 1 1 0 0 0 1 1 1 0 0 1 0 0 Pararaphidia 1 1 0 1 2 1 1 1 1 1 0 0 0 1 1 1 0 0 1 0 0 Iberoraphidia 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 1 0 1 0 0 Baisoraphidia 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 1 0 0 Huaxiaraphidia 1 1 1 1 1 0 1 1 1 1 0 0 0 1 ? 1 0 0 0 0 0 Kezuoraphidia 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 1 0 0 Mesoraphidia 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 1 0 0&1 Metaraphidia 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 1 0 0 Nanoraphidia 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 1 0 1 Proraphidia 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 1 0 0 Siboptera 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 1 1 0 Sinoraphidia 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 0 0 0 Xuraphidia 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 1 0 0 Cantabroraphidia 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 1 0 1 Ororaphidia 1 1 0 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 1 0 0 Styporaphidia 1 1 0 1 1 0 1 1 1 1 0 0 0 1 1 1 0 0 1 0 0 Arariperaphidia 1 1 0 1 ? 0 1 1 1 1 0 0 0 1 1 1 0 0 1 0 0 Archiinocellia 1 1 0 1 2 1 ? 1 1 1 ? 0 0 1 1 1 0 0 1 ? 0 RAPHIDIIDAE 1 1 0 1 1 0&1 1 1 1 1 0 1 1 1 1 1 0 0 1 0 0 INOCELLIIDAE 1 1 0 1 1 0 1 1 1 1 1 1 1 1 1 1 0 0 1 0 0 Table 13.2. Data matrix showing the coding for the genera of fossil snakeflies and extant.

13.4.4.2. Analysis

The cladistic analysis was undertaken on the 28 genera of snakefly using 21 characters (Table 13.2) using PAUP 4.0 beta 10 (Swofford 2002). A heuristic search resulted in 891 parsimonious topologies with little resolution. The 50% majority rule consensus tree showed some resolution with some well supported clades (Fig. 13.33).

186

Figure 13.33. Strict consensus Tree. The percentage support for each node is given in an oval (Consistency Index: 0.815, Retention Index: 0.934, Homoplasy Index: 0.185.). Synapomorphies are indicated by a black bar and number (numbers refer to characters in data matrix (Table 16.2) and are explained below).

13.4.4.3. Synapomorphies

Very few synapomorphies were found for raphidiopterans through this analysis giving very little resolution within the cladogram (Fig. 13.33). The synapomorphies found are listed below:

187

1: Termination of Sc on C in both fore- and hind wings

2: 3-6 cells of Rs in forewing

3: less than 4 r-rs crossveins in hind wing

4: less than 2 rs-m crossveins in the forewing

5: less than 5 medial cells in the forewing

6: less than 17 cells of Rs and M in the forewing

7: No gradate series of crossveins in forewing

8: less than 3 r-rs crossveins in forewing

9: 3 or less cells of Rs in forewing (loss of Rs cells)

10: less than 2 cubital area crossveins

11: less than 6 mm wing size

12: pterostigma shorter than 2r

13: triangular head

14: Basal cut off of Rs at the mid wing

15: M and Cu touch

Very few sister-group relationships were resolved with many polytomies existing. However, there are some well supported clades (Fig. 13.33). Figure 13.34 shows a composite of the analysis presented herein and the work on extant raphidiopterans by Aspöck et al. (1991), Aspöck and Aspöck (1994), Aspöck (1998), Rausch et al. (2004) with the addition of the fossil genera of Raphidiidae and Inocelliidae.

Figure 13.34. (Overleaf). Showing all raphidiopteran (fossil and extant) with the results of the analysis herein and the works of Aspöck et al. (1991), Aspöck and Aspöck (1994), Aspöck (1998), Rausch et al. (2004).

188

189

13.5. Discussion

The construction of the fossil raphidiopteran key and the cladistic study on fossil raphidiopterans has highlighted further the problems of finding defining and diagnostic characters of fossil raphidiopterans. The key is a good preliminary starting point which can aid the worker in identifying any fossil raphidiopteran to family and generic level. The keys are based upon few defining characters and if any of these characters are not preserved within the fossil the identification cannot be made. However, if the wing is well preserved then identification should be possible. At generic level the characters used are sometimes quite weak and it may be that they are the products of natural variation which could occur within different species in the same genus, therefore weakening the key. This however can only be tested when more material is described, because at the present time many of the genera are monospecific.

The cladistic study on raphidiopterans further highlighted the problems associated with raphidiopteran characters. Numerous characters were investigated on all the species to try to find phylogenetic informative characters. Many characters were rejected, most of them on the basis of variation between and within species of the same genus. All manner of characters were attempted from the position of the crossveins within the wings, angles between the veins, termination of veins and number of branches, to name a few. From over 40 potential characters, only 21 were used in the analysis. Another problem was that the photographs of the fossil specimens were often of such poor quality that the venation drawings were used to get character information. This is problematic because it relies on the authors creating an accurate representation of the fossil, especially when investigating characters such as angles between veins. Another subtle factor which could cause erroneous characters is the fact that slight post mortem deformation could take place on the wing causing slight changes in venation. There is also the fact that many of the wings are partially preserved and only represented by an isolated fore- or hind wing, which may lead to potential characters not being preserved in fossils, leading to unresolved relationships.

The first attempt of the cladistic study had every genus of snakefly both fossil and extant incorporated into the analysis. The characters used were based on wing

190 venation and easily preserved body morphology and the genitalia and body morphology from Aspöck et al., (1991) to resolve the relationships of the extant taxa. When running this analysis no resolution was found; the wing venation and easily preserved body morphology varied very little between the extant taxa and this caused an almost infinite number of trees to be generated, ultimately causing PAUP‘s memory to be exhausted, even after over 3 hours of running time. Therefore the extant taxa were condensed into the families Raphidiidae and Inocellliidae, which reduced dramatically the number of parsimonious topologies (to 897). When the 50% majority rule strict consensus tree was created, the work of Aspöck et al. (1991) with the additions of fossil raphidiid and inocelliid genera were incorporated to give a full view of raphidiopteran relationships at generic level. The analysis on the fossil raphidiopterans created little resolution of the relationships and only 15 potential synapomorphies were recovered. The families Baissopteridae and Mesoraphidiidae were shown to be paraphyletic and Alloraphidiidae was shown to be monophyletic, but forming a clade with incertae sedis genera and the mesoraphidiid Proraphidia. Baissopteridae was shown to be the basal family of Raphidioptera, as expected due to the dense venation pattern similar to that of the outgroup Permoberotha.The majority of genera within all families were resolved as polytomies, with few resolved clades. Some interesting relationships were shown such as the two Chinese taxa Huaxiaraphidia and Sinoraphidia forming a basal sister-group to the other mesoraphidiids, maybe reopening the possiblility of a separate family placement for the genera. However this was only based upon a symplesiomorphy. There is also the dubious accuracy of the drawings as highlighted by Engel (2002a); therefore redescription is needed before anything can be interpreted about their taxonomic position.

On the whole the cladistic analysis agreed to some extent with Willmann (1994), with regards to the basal placement of Baissopteridae, the paraphyletic arrangement of Baissopteridae and Mesoroaphidiidae, and the sister-group relationship of Raphidiidae and Inocelliidae. Alloraphidiid genera grouped into a monophyletic clade in the analysis herein, however, was shown to be paraphyletic in Willmann (1994). This is most probably due to the fact that Willmann used fewer taxa in his analysis, and many of the characters he used are variable within taxa of the same family, such as in Alloraphidiidae.

191

After completing extensive work looking at all the species of fossil raphidioptera and in compiling a key and analysing relationships the major problem observed time and again is the lack of defining/diagnostic characters. One problem is that fossil snakeflies are generally only represented by isolated wings with only rare occurrences of body parts. With extant genera the male genitalia holds a wealth of information that allows identification of species, genera and their potential relationships (Aspöck et al. 1991). This information is only very rarely found in fossil material, for example the Baltic amber taxa Succinoraphidia and Succinofibla had visible terminalia, which allowed a more detailed description and diagnosis. When it comes to wings very little information can be ascertained. Another problem is the lack of synapomorphies that have been found within extant taxa; therefore even the hypotheses of extant relationships are only tentative. The comparing of morphospecies (such as the fossil raphidiopterans) and biological species also causes problems, because it is likely that fossil species represent a complex of biological species, as opposed to just one. Therefore comparing these two different species concepts may always give unresolved relationships.

To find useful defining/diagnostic characters within fossil raphidiopterans a large scale taxonomic review is needed. Many genera and species are defined on very poor weak characters and the only way to sort out this confusion is for the redescription of type specimens. Some genera and families are used as ‗bucket‘ groups where many species are placed, based on dubious characters, such as the genus Mesoraphidia and the family Mesoraphidiidae. Engel (2002a) in his catalogue made a good start at sorting out the taxonomic confusion synonymizing many genera, families and bringing a more conservative systematic account of the Order. Until a major review is undertaken there is little hope in creating an accurate phylogenetic history of fossil Raphidioptera.

192

13.6. Conclusions

The work on raphidiopterans has shown that:

1. It is possible to complete a preliminary key at both family and generic level, which will help in avoiding errors when describing new taxa.

2. It may be premature to consider reconstructing the phylogenetic relationships of extant and fossil raphidiopterans. The characters that are useful for identifying fossil species are of little use at present to allow extant taxa at generic level to be placed into the same analysis. The problem of lack of variability within extant genera compared to extinct means that there are an infinite possibility of tree topologies, which cannot be resolved.

3. The phylogenetic relationships of fossil genera of raphidiopterans can give a preliminary hypothesis, but there is little resolution. There are only a few well supported clades.

4. Only 15 synapomorphies were found in the cladistic study of fossil raphidiopterans, out of the 21 characters used in the analysis.

5. Families of fossil raphidiopterans are on the whole paraphyletic, for example Baissopteridae and especially Mesoraphidiidae. Alloraphidiidae was shown to be monophyletic, but, only weakly supported. Baissopteridae however, are again shown to be basal with respect to all other raphidiopterans.

6. Much more work is needed on fossil raphidiopterans before their relationships and taxonomy can be fully understood.

193

13.6.1. Further Work

The first step for any further work on the fossils of Raphidioptera is a full world- wide taxonomic revison of all families, genera and species. Clear diagnoses at all taxonomic levels are needed, which will mean a complete redescription of all fossil material. Some taxa are very poorly described and figured, and they are diagnosed on either weak or dubious characters. To create a complete and accurate key to all raphidiopterans this revision needs to take place. With regards to phylogenetic studies, again the taxonomic revision will be required to find useful characters and synapomorphies, which will elucidate relationships between groups. It may also be worthwhile considering new techniques in identifying characters, due to the problems of wing venation. A possibility could be geometric morphometrics, which could help find more elusive characters within the wing venation.

194

REFERENCES

Adams, P. A. 1967. A review of the Mesochrysinae and Notochrysinae (Neuroptera: Chrysopidae). Bulletin of the Museum of Comparative Zoology, 135(4), 215– 238. Allen, P. 1975. Wealden of the Weald: a new model. Proceedings of the Geologists’ Association, 86, 389–437. Allen, P. 1981. Pursuit of Wealden models. Journal of the Geological Society of London, 138, 375–405. Allen, P. 1989. Wealden research—ways ahead. Proceedings of the Geologists' Association, 100, 529–564. Allen, P. 1998. Purbeck-Wealden (early Cretaceous) climates. Proceedings of the Geologists' Association, 109(3), 197–236. Allen, P. & Wimbledon, W. A. 1991. Correlation of NW European Purbeck- Wealden (non-marine Lower Cretaceous) as seen from the English type-areas. Cretaceous Research, 12, 511–526. Andersen, S. 2001. Silky lacewings (Neuroptera: Psychopsidae) from the Eocene- transition of Denmark with a review of the fossil record and comments on phylogeny and zoogeography. Insect Systematics & Evolution, 32, 419–438. Ansorge, J. 1993. Bemerkenswerte Lebensspuren und ? Cretosphex catalunicus n. sp. (Insecta; Hymenoptera) aus den unterkretazischen Plattenkalken der Sierra del Montsec (Provinz Lerida, NE - Spanien) [Remarkable ―Lebensspuren‖ and ?Cretosphex catalunicus n. sp. (Insecta: Hymenoptera) from the Lower Cretaceous lithographic limestones of the Sierra del Montsec (Province Lerida, Spain]. Neues Jahrbuch für Geologie und Palaeontologie, 190(1), 19–35. Ansorge, J. 1996. Insekten aus dem oberen Lias von Grimmen (Vorpommern, Norddeutschland). Neue Paläontologische Abhandlungen, 2, 1–132. Ansorge, J. & Schlüter, T. 1990. The earliest chrysopid: Liassochrysa stigmatica n. g., n. sp. from the Lower Jurassic of Dobbertin, Germany. Neuroptera International, 6, 87–93. Archibald, S. B. & Makarkin, V. N. 2004. New genus of minute Berothidae (Neuroptera) from Early Eocene amber of British Columbia. Canadian Entomologist, 136(1), 61–76.

195

Archibald, S. B. & Makarkin, V. N. 2006. Tertiary giant lacewings (Neuroptera: Polystoechotidae): revision and description of new taxa from western North American and Denmark. Journal of Systematic Palaeontology, 4, 119–155. Archibald, S. B., Makarkin, V. N. & Ansorge, J. 2009. New fossil species of Nymphidae (Neuroptera) from the Eocene of North America and Europe. Zootaxa, 2157, 59–68. Aspöck, H. 1986. Raphidioptera of the world: a review of present knowledge. In: Gepp, J., Aspöck, H. & Hölzel, H. (eds), Recent research in neuropterology. Proceedings of the 2nd symposium of Neuropterology, Austria, Graz: 15–29. Aspöck, H. 1998. Distribution and biogeography of the Order Raphidioptera: updated facts and a new hypothesis. Acta Zoologica Fennica, 209, 33–44. Aspöck, H. 2002. The biology of Raphidioptera: a review of present knowledge. In: Sziráki, G. (ed), Neuropterology 2000. Proceedings of the Seventh International Symposium on Neuropterology. Acta Zoologica Academiae Scientiarum Hungaricae: 35–50. Aspöck, H. & Aspöck, U. 1972. Raphidia (Magnoraphidia) flammi Asp. et Asp. und Raphidia (Magnoraphidia) horticola Asp. et Asp. – taxonomie und verbreitung (Neuropt., Raphidioptera, Raphidiidae). Zeitschrift der Arbeitsgemeinschaft Österr. Entomologen, 24(4), 140–146. Aspöck, H. & Aspöck, U. 1994. Mongoloraphidia (Hissaroraphidia) karatauica n. sp.: A new species from the Karatau mountains in the south of Kasachstan (Neuropteroidea, Raphidioptera). Entomologische Nachrichten und Berichte, 38(4), 217–220. Aspöck, H., Aspöck, U. & Hölzel, H. 1980. Neuropteren Europas I and II. Goecke and Evers. Aspöck, H., Aspöck, U. & Rausch, H. 1983. Mauroraphidia maghrebina n. gen. n. sp. – eine neue Raphidiiden-Spezies aus dem Atlas (Neuropteroidea: Raphidioptera). Zeitschrift der Arbeitsgemeinschaft Österr. Entomologen, 35(1), 27–32. Aspöck, H., Aspöck, U. & Rausch, H. 1991. Die Raphidiopteran der Erde: Eine monographische Darstellung der Systematik, Taxonomie, Biologie, Ökologie und Chorologie der rezenten Raphidiopteran der Erde, mit einer zusammenfassenden Übersicht der fossilen Raphidiopteran (Insecta: Neuropteroidea). Goecke and Evers.

196

Aspöck, H., Aspöck, U. & Rausch, H. 1998. Mongoloraphidia gulnara n.sp. – eine neue Kamalhalsfliege aus Kirgisistan (Neuropteroidea: Raphidioptera: Raphidiidae). Zeitschrift der Arbeitsgemeinschaft Österr. Entomologen, 50, 65–72. Aspöck, U. 1995. New hypotheses to the system of the Neuropterida. Mitteilungen Der Deutschen Gesellschaft Fur Allgemeine Und Angewandte Entomologie, Band 10, Heft 1-6, Dezember 1995, 10(1-6), 633–636. Aspöck, U. 2002a. Phylogeny of the Neuropterida (Insecta: Holometabola). Zoologica Scripta, 31, 51–55. Aspöck, U. 2002b. Homology of male genital sclerites in Neuropterida - An adventure. Acta Zoologica Academiae Scientiarum Hungaricae, 48, 51–51. Aspöck, U. 2002c. Male genital sclerites of Neuropterida: an attempt at homologisation (Insecta : Holometabola). Zoologischer Anzeiger, 241(2), 161– 171. Aspöck, U. & Mansell, M. W. 1994. A revision of the family Rhachiberothidae Tjeder, 1959, stat n (Neuroptera). Systematic Entomology, 19(3), 181–206. Aspöck, U. & Aspöck, H. 2004. Two significant new snakeflies from Baltic amber, with discussion on apomorphies of the Order and its included taxa (Raphidioptera). Systematic Entomology, 29(1), 11–19. Aspöck, U. & Aspöck, H. 2008. Phylogenetic relevance of the genital sclerites of Neuropterida (Insecta : Holometabola). Systematic Entomology, 33(1), 97–127. Aspöck, U., Aspöck, H. & Rausch, H. 1992. Extant southern boundaries of the distribution of the Order Raphidioptera in America (Insecta, Neuropteroidea). Entomologia Generalis, 17(3), 169–184. Aspöck, U., Plant, J. D. & Nemeschkal, H. L. 2001. Cladistic analysis of Neuroptera and their systematic position within Neuropterida (Insecta : Holometabola : Neuropterida : Neuroptera). Systematic Entomology, 26(1), 73–86. Aspöck, U., Aspöck, H. & Haring, E. 2003. Phylogeny of the Neuropterida -- morphological evidence and the molecular advocatus diaboli. Entomologische Abhandlungen, Staatliches Museum für Tierkunde in Dresden, 61(157–158). Azar, D., Nel, A. & Solignac, M. 2000. A new Coniopterygidae from Lebanese amber. Acta Geologica Hispanica, 35(1-2), 31–36. Babrikova, T. 1979. Studies on the biology of Chrysopa perla L. (Chrysopidae, Neuroptera). Rasteniev'dni Nauki, 16, 95–100.

197

Barker, D., Brown, C. E., Bugg, S. C. & Costin, J. 1975. Ostracods, land plants, and charales from the basal Purbeck Beds of Porteshem Quarry, Dorset. Palaeontology, 18(2), 419–436. Barnard, P. C., Brooks, S. J. & Stork, N. E. 1986. The seasonality and distribution of Neuroptera, Raphidioptera and Mecoptera on oaks in Richmond Park, Surrey as revealed by knock-down sampling. Journal of Natural History, 20(6), 1321–1331. Batten, D. J. 1982. Palynofacies and salinity in the Purbeck and Wealden of southern England. In: Banner, F. T. & Lord, A. R. (eds), Aspects of Micropalaeontology. George Allen & Unwin, London, 278–308. Batten, D. J. 1998. Palaeoenvironmental implications of plant, insect and other organic-walled microfossils in the Weald Clay Formation (Lower Cretaceous) of southeast England. Cretaceous Research, 19, 279–315. Batten, D. J. 2002. Palaeoenvironmental setting of the Purbeck Limestone Group of Dorset, southern England. In: Milner, A. & Batten, D. J. (eds), Life and environments in Purbeck times. Special Papers in Palaeontology. Palaeontological Association, London, 68, 13–20. Batten, D. J. 2008. Spores and pollen from the Crato Formation: biostratigraphic and palaeoenvironmental implications. In: Martill, D. M., Bechly, G. & Loveridge, R. F. (eds), The Crato Fossil Beds of Brazil. Window into an Ancient World. Cambridge University Press, Cambridge, 566–573. Bode, A. 1953. Die insektenfauna, des Ostinidersächsischen Oberen lias. Paleontographica: Beiträge zur Naturgeschichte der Vowelt, 103(1), 1–375. Brauckmann, C. & Schlüter, T. 1993. Insekten aus der Trias von Unter-Franken. Geologica et Palaeontologica, 27, 181–199. Brodie, P. B. 1845. A history of fossil insects in the secondary rocks of England. Van Voorst, 130 pp. Brooker, M. P. 1979. The life cycle and growth of Sialis lutaria L. (Megalotpera) in a drainage channel under different methods of plant management. Ecological Entomology, 4, 111–117. Cameron, S. L., Sullivan, J., Song, H., Miller, K. B. & Whiting, M. F. 2009. A mitochondrial genome phylogeny of the Neuropterida (lace-wings, alderflies and snakeflies) and their relationship to the other holometabolous insect Orders. Zoologica Scripta, 38(6), 575–590.

198

Canard, M. 1970. Incidences de la valeur alimentaire de divers pucerons (Homoptera, ) sur le potentiel de multiplication de Chrysopa perla (L.) (Neuroptera, Chrysopidae). Annales de Zoologie, Écologie Animale, 2, 345–355. Canard, M. 1973. Voltinisme, diapause et sex-ratio de Chrysopa perla (L.) (Neuroptera, Chrysopidae) dans de sud-ouest. Annales de Zoologie, Écologie Animale, 5, 29–37. Carpenter, F. M. 1929. A Jurassic Neuropteran from the Lithographic Limestones of Bavaria. Psyche, 36, 190–194. Carpenter, F. M. 1935. Tertiary Insects of the Family Chrysopidae. Journal of Paleontology, 9(3), 259–271. Carpenter, F. M. 1936. Revision of the Nearctic Raphidiodea (Recent and Fossil). Proceedings of the American Academy of Arts and Sciences, 71(2), 89–158. Carpenter, F. M. 1943a. The Lower Permian Insects of Kansas. Part 9. The Orders Neuroptera, Raphidiodea, Caloneurodea and Protorthoptera (Probnisidae), with Additional Protodonata and . Proceedings of the American Academy of Arts and Sciences, 75(2), 55–84. Carpenter, F. M. 1943b. Osmylidae, Florissant shales, Colorado (Insecta- Neuroptera). American Journal of Science, 241(12), 753–760. Carpenter, F. M. 1956. The Baltic amber snakeflies. Psyche, 63, 77–81. Carpenter, F. M. 1959. Fossil Nemopteridae (Neuroptera). Psyche, 66, 20–24. Carpenter, F. M. 1962. A Permian megasocopteran from Texas. Psyche, 69, 145– 153. Carpenter, F. M. 1967. Cretaceous insects from Labrador. 2. A new family of snake- flies (Neuroptera: Alloraphidiidae). Psyche, 74, 270–275. Carpenter, F. M. 1976. The Lower Permian Insects of Kansas. Part 12. Protorthoptera (Continued), Neuroptera, additional , and families of uncertain position. Psyche, 83, 336–376. Carpenter, F. M. 1992. Hexapoda. Treatise of invertebrate paleontology. Part R. Arthropoda 4 (3–4), Geological Society of America, Boulder, the University of Kansas Press, 677 pp. Clements, R. G. 1973. A study of certain non-marine Gastropoda from the Purbeck Beds of England, University of Hull.

199

Clements, R. G. 1993. Type-section of the Purbeck Limestone Group, Durlston Bay, Swanage, Dorset. Proceedings of the Dorset Natural History and Archaeological Society, 114, 181–206. Clifford, E., Coram, R., Jarzembowski, E. A. & Ross, A. J. 1994. A supplement to the insect fauna from the Purbeck Group of Dorset. Proceedings of the Dorset Natural History and Archaeological Society, 155(1993), 143–146. Cockerell, T. D. A. 1907. Some fossil arthropods from Florissant, Colorado. Bulletin of the American Museum of Natural History, 23, 605–616. Cockerell, T. D. A. 1908. Fossil insects from Florissant, Colorado. Bulletin of the American Museum of Natural History, 24, 59–69. Cockerell, T. D. A. 1909a. New fossil insects from Florissant, Colorado. Annals of the Entomological Society of America, 2, 251-258. Cockerell, T. D. A. 1909b. Fossil insects from Florissant, Colorado. Bulletin of the American Museum of Natural History, 26, 67–76. Cockerell, T. D. A. 1914. New and little known insects from the Miocene of Florissant, Colorado. Journal of Geology, 22, 714–724. Cockerell, T. D. A. 1921. Fossil arthropods in the British Museum. VI. Oligocene insects from Gurnet Bay, Isle of Wight. Annals and Magazine of Natural History, 9, 453–480. Cockerell, T. D. A. & Custer, C. 1925. A new fossil Inocellia (Neuroptera) from Florissant. Entomologist, 58, 295–297. Comstock, J. H. 1918. The wings of insects. Comstock Publishing Company. Coram, R. A. 2003. Taphonomy and ecology of Purbeck fossil insects. Acta Zoologica Cracoviensia, 46(October), 311-318. Coram, R. A. 2005. Palaeoecology and palaeoenvironmental significance of Early Creataceous fossil insects of the Purbeck Limestone Group, Dorset, UK, University of Reading. Coram, R. A. & Jarzembowski, E. A. 2002. Diversity and ecology of fossil insects in the Dorset Purbeck succession, southern England. In: Milner, A. R. & Batten, D. J. (eds), Life and Environments in Purbeck Times. Special Papers in Palaeontology, 68, 257–268. Crowson, R. A. 1946. The fossil insects of the Weald. In: Given, J. C. M. (ed), Royal Tunbridge Wells – Past and Present. Courier, Tunbridge Wells, 24–26.

200

Davis, C. 1943. A new species of Permithone (Neuroptera, Planipennia) from the Upper Permian of New South Wales. Proceedings of the Linnean Society of New South Wales, 68, 11–12. Deconinck, J.-F. 1987. Minéraux argileux des faciès Purbeckiens: Jura suisse et français, Dorset (Angleterre) et Boulonnais (France). Annales Societé Géologique du Nord, 106, 285–297. Dobruskina, I. A., Ponomarenko, A. G. & Rasnitsyn, A. P. 1997. Fossil insects found in Israel. Paleontologicheskii Zhurnal(5), 91–95. Dubois, A. M. & Geigy, R. 1935. Beiträge zur Oekologie, Fortpflanzungsbiologie und Metamorphose von Sialis lutaria L. (Studien am Sempachersee). Revue Suisse de Zoologie, 42, 169–248. Duelli, P. 1981. Is larval cannibalism in lacewings adaptive? (Neuroptera: Chrysopidae). Researches on Population Ecology, 23, 193–209. Duelli, P., Obrist, M. K. & Fluckiger, P. F. 2002. Forest edges are biodiversity hotspots - Also for Neuroptera. Acta Zoologica Academiae Scientiarum Hungaricae, 48, 75-87. Eisma, D. 1998. Intertidal deposits. CRC Press, 195 pp. Elliott, J. M. 1977. A key to the larvae and adults of British freshwater Megaloptera and Neuroptera with notes on their life cycles and ecology. Freshwater Biological Association Scientific Publication, 35, 1–52. El-Shahat, A. & West, I. M. 1983. Early and late lithification of aragonatic bivalve beds in the Purbeck Formation (Upper Jurassic – Lower Cretaceous) of southern England. Sedimentary Geology, 35, 15–41. Engel, M. S. 1995. A new fossil snake- species from Baltic amber (Raphidioptera: Inocellidae). Psyche, 102, 187–193. Engel, M. S. 1999. The first fossil of a pleasing lacewing (Neuroptera: Dilaridae). Proceedings of the Entomological Society of Washington, 101(4), 822–826. Engel, M. S. 2002a. The smallest Snakefly (Raphidioptera: Mesoraphidiidae): a new species in Cretaceous amber from Myanmar, with a catalog of fossil Snakeflies. American Museum Novitates, 3363, 1–22. Engel, M. S. 2002b. A new Dustywing (Neuroptera: Coniopterygidae) in Turonian amber from New Jersey, with a reassessment of Glaesoconis in Neocomian amber from Lebanon. Journal of the Kansas Entomological Society, 75, 38–42.

201

Engel, M. S. 2004a. The dustywings in Cretaceous Burmese amber (Insecta: Neuroptera: Coniopterygidae). Journal of Systematic Palaeontology, 2(2), 133–136. Engel, M. S. 2004b. The Alderflies of Kansas (Megalopter: Sialidae). Transactions of the Kansas Academy of Sciences, 107, 119–125. Engel, M. S. 2004c. Arthropods in Mexican amber, Biodiversidad, taxonomía y biogeografía de artrópodos de México: hacia una síntesis de su conocimiento volumen 4. Universidad Nacional Autónoma de México, México, 175–186. Engel, M. S. 2004d. Thorny lacewings (Neuroptera; Rhachiberothidae) in Cretaceous amber from Myanmar. Journal of Systematic Palaeontology, 2, 137–140. Engel, M. S. 2005. A remarkable kalligrammatid lacewing from the upper jurassic of Kazakhstan (Neuroptera: Kalligrammatidae). Transactions of the Kansas Academy of Science, 108(1-2), 59-62. Engel, M. S. & Grimaldi, D. A. 2007. The Neuropterid Fauna of Dominican and Mexican Amber (Neuropterida: Megaloptera, Neuroptera). American Museum Novitates, 1–58. Engel, M. S. & Grimaldi, D. A. 2008. Diverse Neuropterida in Cretaceous amber, with particular reference to the paleofauna of Myanmar (Insecta). Nova Supplementa Entomologica, Keltern (Deutsches Entomologisches Institut), 20, 1–86. Engel, M. S. & Ren, D. 2008. New snakeflies from the Jiulongshan Formation of Inner Mongolia, China (Raphidioptera). Journal of the Kansas Entomological Society, 81(3), 188–193. Engel, M. S., Lim, J. D. & Baek, K. S. 2006. Fossil snakeflies from the Early Cretaceous of southern Korea (Raphidioptera : Mesoraphidiidae). Neues Jahrbuch Fur Geologie Und Palaontologie-Monatshefte, 2006(4), 249–256. Ensom, P. C. 2002a. Vertebrate trace fossils in the Purbeck Limestone Group of Southern England. In: Milner, A. & Batten, D. J. (eds), Life and environments in Purbeck times. Special Papers in Palaeontology. Palaeontological Association, London, 68, 203–220. Ensom, P. C. 2002b. The Purbeck Limsetone Group of Dorset, southern England: a guide to lithostratigraphic terms. In: Milner, A. R. & Batten, D. J. (eds), Life and Environments in Purbeck Times. Special Papers in Palaeontology. The

202

Palaeontological Association, Oxford, Special Papers in Palaeontology, 68, 7– 11. Evans, E. D. 1972. A study of the Megaloptera of the Pacific coastal region of the United States, Oregon State University, 210 pp. Evans, H. E. 1978. A solitary wasp that preys upon lacewings (Hymenoptera, Sphecidae; Neuroptera, Chrysopidae). Psyche, 85, 81–84. Evans, H. E., Evans, M. A. & Hook, A. W. 1982. Observations on the nests and prey of Australian Bembix sand wasps (Hymenoptera: Sphecidae). Australian Journal of Zoology, 30(1), 71–80. Fisher, O. 1856. On the Purbeck strata of Dorsetshire. Transactions of the Cambridge Philosophical Society, 9, 555–581. Francis, J. E. 1983. The dominant Conifer of the Jurassic Purbeck Formation, England. Palaeontology, 26(2), 277–294. Francis, J. E. 1984. The seasonal environment of the Purbeck (Upper Jurassic) fossil forests. Palaeogeography, Palaeoclimatology, Palaeoecology, 48, 285–307. Fujiyama, I. 1978. Some fossil insects from the Tedori Group (Upper Jurassic – Lower Cretaceous). Bulletin of the National Science Museum Series C (Geology and Paleontology), 4(4), 181-192. Gallois, R. W. 1978. British regional Geology: The Wealden District. HMSO. Geinitz, F. E. 1880. Der Jura von Dobbertin in Mecklenburg und seine Versteinerungen. Zeitschrift der Deutschen Geologischen Gesellschaft, 32(3), 510–534. Geinitz, F. E. 1884. Ueber die Fauna des Dobbertiner Lias. Zeitschrift der Deutschen Geologischen Gesellschaft, 36(3), 566–583. Geinitz, F. E. 1887. Beitrag zur Geologie Mecklenburgs. Archiv des Vereins der Freunde der Naturgeschichte in Mecklenburg, 41, 143–216. Giani, N. & Laville, H. 1973. Cycle biologique et production de Sialis lutaria L. (Megaloptera) dans le lac de Port-Beilh (Pyrenees Centrales). Annales de Limnologie, 9, 45–61. Giebel, G. C. A. 1856. Vorwelt, Bd II. Die Insecten und Spinnen der Vorwelt. Vorwelt. F.U. Brodhaus, 511 pp. Gorochov, A. V., Jarzembowski, E. A. & Coram, R. A. 2006. Grasshoppers and crickets (Insecta: Orthoptera) from the Lower Cretaceous of southern England. Cretaceous Research, 27, 641–662.

203

Gradstein, F. M., Ogg, J. G. & Smith, A. G. (eds), 2005. A geologic time scale 2004. Cambridge University Press, 589 pp. Grimaldi, D. A. 2000. A diverse fauna of Neuropterodea in amber from the Cretaceous of New Jersey. In: Grimaldi, D. A. (ed), Studies on fossils in amber, with particular reference to the Cretaceous of New Jersey. Backhuys, Leiden, 259–303. Grimaldi, D. A. & Engel, M. S. 2005. Evolution of the insects. Cambridge University Press, 772 pp. Grimaldi, D. A., Shedrinsky, A. & Wampler, T. P. 2000. A remarkable deposit of fossiliferous amber from the Upper Cretaceous (Turonian) of New Jersey. In: Grimaldi, D. A. (ed), Studies on fossils in amber, with particular reference to the Cretaceous of New Jersey. Buckhuys Leiden, 1–76. Hagen, H. A. 1862a. Ueber die Neuropteren aus dem lithographischen Schiefer in Bayern. Palaeontographica, 10(2), 96–145. Hagen, H. A. 1862b. A comparison of the fossil insects of England and Bavaria. Entomologist's Annual, 1862, 1–10. Hallam, A. 1984. Continental humid and arid zones during the Jurassic and Cretaceous. Palaeogeography, Palaeoclimatology, Palaeoecology, 47, 195– 223. Hallam, A., Grose, J. A. & Ruffell, A. H. 1991. Palaeoclimatic significance of changes in clay mineralogy across the Jurassic – Cretaceous boundary in England and France. Palaeogeography, Palaeoclimatology, Palaeoecology, 81, 173–187. Hancock, J. M. & Rawson, P. F. 1992. Cretaceous. In: Cope, J. C. W., Ingham, J. K. & Rawson, P. F. (eds), Atlas of Palaeogeography and lithofacies. Geological Society, London, Memoir, London, 13, 131–139. Handlirsch, A. 1906–1908. Die fossilen Insekten und die Phylogenie der rezenten Formen. Wilhelm Engelmann, 1430 pp. Handlirsch, A. 1919. Eine neue Kalligrammidae (Neuroptera) aus dem Solnhofener Plattenkalke. Senckenbergiana, 1(3), 61–63. Handlirsch, A. 1920. Kapitel 7. Palaeontologie. In: Schroeder, C. (ed), Handbuch der Entomologie. G. Fischer, Jena, 3, 117–304. Handlirsch, A. 1939. Neue Untersuchungen über die fossilen Insekten mit Ergänzungen und Nachträgen sowie Ausblicken auf phylogenetische,

204

palaeogeographische und allgemein biologische Probleme. II. Theil. Annalen des Naturhistorischen Museums in Wien, 49, 1–240. Haring, E. & Aspöck, U. 2004. Phylogeny of the Neuropterida: a first molecular approach. Systematic Entomology, 29(3), 415–430. Haywood, A. M., Valdes, P. J. & Markwick, P. J. 2005. Cretaceous (Wealden) climates: a modelling perspective. Cretaceous Research, 25, 303–311. Heer, O. 1864. Ueber die fossilen Kakerlaken. Vierteljahrsschrift der Naturforschenden Gesellschaft in Zürich, 9(4), 273–30. Hennig, W. 1981. Insect phylogeny. John Wiley & Sons, 500 pp. Hill, C. R. 1996. A plant with flower-like organs from the Wealden of the Weald (Lower Cretaceous), southern England. Cretaceous Research, 17, 27–38. Hong, Y. C. 1982. Mesozoic fossil insects of Jiuquan Basin in Gansu Province. Geological Publishing House. Hong, Y.-C. 1983. Middle Jurassic fossil insects in North China. Geological Publishing House, 223 pp. Hong, Y. C. 1992a. The study of early Cretaceous Coleoptera, Raphidioptera, Diptera of Kezuo, West Liaoning Province. Acta Geologica Gansu, 1, 1–13. Hong, Y. C. 1992b. A new family of Mesozoic snake-flies (Insecta, Raphidioptera) from the Laiyang Basin, China. Paleontologicheskii Zhurnal, 1992(3), 101– 105. Hong, Y. C. & Chang, J. P. 1989. A new family ‗Jilinoraphidiidae‘ of Raphidioptera (Insecta). Geoscience, 290–296. Horne, D. J. 2002. Ostracod biostratigraphy and palaeoecology of the Purbeck Limestone Group in southern England. In: Milner, A. & Batten, D. J. (eds), Life and environments in Purbeck times. Special Papers in Palaeontology. Palaeontological Association, London, 68, 53–70. Hörnschmeyer, T. 1998. Morphologie und Evolution des Flügelgelenks der Coleoptera und Neuropterida. Bonner Zoologische Monographien, 43, 1–126. Hunt, C. O. 1985. Miospores from the Portland Stone Formation and the lower part of the Purbeck Formation (Upper Jurassic/Lower Cretaceous) from Dorset, England. Pollen et Spores, 27, 419–451. Hydorn, S. B. & Whitcomb, W. H. 1979. Effects of larval diet on Chrysopa rufilabris (Neuroptera, Chrysopidae). Florida Entomologist, 62(4), 293–298.

205

Illes, J. 1967. Megaloptera and Plecoptera (Ins.) aus den jungpliozänen Süßwassermergeln von Willershausen. Bericht der Naturhistorischen Gesselscahft zu Hannover, 111, 47–55. Iversen, T. M. & Thorup, J. 1987. Population dynamics and production of Sialis lutaria, L. (Megaloptera) in the Danish River Suså. Freshwater Biology, 17(461–469). Jarzembowski, E. A. 1980. Fossil insects from the Bembridge Marls, Paleogene of the Isle of Wight, southern England. Bulletin of the British Museum of Natural History (Geology), 33, 237–293. Jarzembowski, E. A. 1984. Early Cretaceous insects from southern England. Modern Geology, 9, 71–93. Jarzembowski, E. A. 1987. Early Cretaceous insects from southern England, University of Reading, 421 pp. Jarzembowski, E. A. 1991. New insects from the Weald Clay of the Weald. Proceedings of the Geologists' Association, 102(2), 93–108. Jarzembowski, E. A. 1993. A provisional checklist of the fossil insects from the Purbeck Beds of Dorset. Proceedings of the Dorset Natural History and Archaelogical Society, 114, 175–179. Jarzembowski, E. A. 1995. Early Cretaceous insect faunas and palaeoenvironment. Cretaceous Research, 16(6), 681–693. Jarzembowski, E. A. 2001. A new Wealden fossil Lacewing. In: Rowlands, J. L. J. (ed), Tunbridge Wells and Rusthall Commons: A History and Natural History. Tunbridge Wells Museum and Art Gallery, Tunbridge Wells, 56–58. Jarzembowski, E. A. 2003. ‗Burnt‘ beetles (Insecta: Coleoptera) from the Wealden of southern England. Acta Zoologica Cracoviensia, 46, 139–145. Jarzembowski, E. A. & Makarkin, V. N. 2006. A new species of silky lacewing (Insecta: Neuroptera: Psychopsidae) from the English Wealden (Early Cretaceous) with a critical review of other fossils referable to this family, Biotic-Sediment Interactions. A symposium in memory of Roland Goldring (1928-2005), University of Reading, 20–22 July 2006: 29. Jepson, J. E. & Penney, D. 2007. Neuropteran (Insecta) palaeodiversity with predictions for the Cretaceous fauna of the Wealden. Palaeogeography, Palaeoclimatology, Palaeoecology, 248(1-2), 109–118.

206

Jepson, J. E. & Jarzembowski, E. A. 2008. Two new species of snakefly (Insecta: Raphidioptera) from the Lower Cretaceous of England and Spain with a review of other fossil raphidiopterans from the Jurassic/Cretaceous transition. Alavesia, 2, 193–201. Jepson, J. E., Coram, R. A. & Jarzembowski, E. A. 2009a. Raphidioptera (Insecta: Neuropterida) from the Lower Cretaceous Purbeck Limestone Group, Dorset, UK. Cretaceous Research, 30(3), 527–532. Jepson, J. E., Makarkin, V. N. & Jarzembowski, E. A. 2009b. New lacewings (Insecta: Neuroptera) from the Lower Cretaceous Wealden supergroup of Southern England. Cretaceous Research, 30(5), 1325–1338. Jepson, J. E., Heads, S. W., Makarkin, V. N. & Ren, D. in prep. Mantispids (Insecta, Neuroptera) from the Lower Cretaceous of China. Killington, F. J. 1925. Notes on the prey of dragonflies. Entomologist, 58, 181–183. Kristensen, N. P. 1999. Phylogeny of endopterygote insects, the most successful lineage of living organisms. European Journal of Entomology, 96, 237–253. Kupryjanowicz, J. & Makarkin, V. N. 2008. Archiconiocompsa prisca Enderlein (Neuroptera : Coniopterygidae): The first neuropteran fossil in Rovno amber (Ukraine). Entomologica Fennica, 19(1), 25–31. Lambkin, K. J. 1987. A re-examination of Europrismites balli Tillyard from the Palaeocene of Queensland (Neuroptera: Osmylidae: Kempyninae). Neuroptera International, 4(4), 295–300. Lambkin, K. J. 1988. A re-examination of the Lithosmylidia Riek from the Triassic of Queensland with notes on Mesozoic ‗osmylid-like‘ fossil Neuroptera (Insecta: Neuroptera). Memoirs of the Queensland Museum, 25(2), 445–458. Lambkin, K. J. 1992. A record of Austrosialis Tillyard from the Queensland palaeocene (Insecta: Megaloptera: Sialidae). Queensland Naturalist, 31, 84– 86. Lambkin, K. J. 1994. Palparites deichmuelleri Handlirsch from the Tithonian Solnhofen Plattenkalk belongs to the Kalligrammatidae (Insecta: Neuroptera). Palaeontologische Zeitschrift, 68(1), 163–166. Leach, W. E. 1815. Artikel Entomology. In: Brewster, D. (ed), Edinburgh Encyclopaedia 9, 57–172.

207

Linnaeus, C. 1758. Systema Naturae per Regna Tria Naturae, Secundum Classes, Ordines, Genera, Species, cum Characteribus, Differentiis, Synonymis, Locis (tenth edition, revised). Salviae, Holmiae. MacLeod, E. G. 1964. A comparative morphological study of the head capsule and cervix of larval Neuroptera (Insecta), Harvard, 528 pp. MacLeod, E. G. 1970. The Neuroptera of the Baltic amber. I. Ascalaphidae, Nymphidae, and Psychopsidae. Psyche, 77, 147–180. Makarkin, V. N. 1990a. A new fossil genus and species of Osmylidae from the Lower Cretaceous of East Siberia (Neuroptera). Deutsche Entomologische Zeitschrift, 37(1-3), 101–103. Makarkin, V. N. 1990b. Baissoleon cretaceus gen. and sp. nov. Fossil Neuroptera of the Lower Cretaceous of Baisa, East Siberia. Part 2. Nymphitidae. Annales de la Société Entomologique de France (N.S.), 26, 125–126. Makarkin, V. N. 1990c. New names for Jurassic Neuroptera Paleontologicheskii Zhurnal, 1(1), 120. Makarkin, V. N. 1994. Upper Cretaceous Neuroptera from Russia and Kazakhstan. Annales de la Société Entomologique de France (N.S), 30(3), 283–292. Makarkin, V. N. 1996. Fossil Neuroptera of the Lower Cretaceous of Baisa, East Siberia. Part 5. Mantispidae. Russian Entomological Journal, 5, 91–93. Makarkin, V. N. 1997a. Fossil Neuroptera of the Lower Cretaceous of Baisa, East Siberia. Part 4. Psychopsidae. Beiträge zur Entomologie, 47, 489–492. Makarkin, V. N. 1997b. Fossil Neuroptera of the Lower Cretaceous of Baisa, East Siberia. Part 3. Chrysopidae (Insecta). Spixiana, 20, 107–118. Makarkin, V. N. 1998. New Tertiary Neuroptera (Insecta) from the Russian Far East. Tertiary Research, 18, 77–83. Makarkin, V. N. 1999. Fossil Neuroptera from the Lower Cretaceous of Baisa, East Siberia. Part 6. Mesithonidae (Insecta). Neues Jahrbuch für Geologie und Paläontologie, 12, 705–712. Makarkin, V. N. In Press. New psychopsoid Neuroptera from the Early Cretaceous of Baissa, Transbaikalia. Annales De La Societe Entomologique De France, 46(1), xxx. Makarkin, V. N. & Archibald, S. B. 2003. Family affinity of the genus Palaeopsychops Andersen with description of a new species from the Early

208

Eocene of British Columbia, Canada (Neuroptera : Polystoechotidae). Annals of the Entomological Society of America, 96(3), 171–180. Makarkin, V. N. & Archibald, S. B. 2005. Substitute names for three genera of fossil Neuroptera, with taxonomic notes. Zootaxa, 1054, 15–23. Makarkin, V. N. & Menon, F. 2005. New species of the Mesochrysopidae (Insecta, Neuroptera) from the Crato Formation of Brazil (Lower Cretaceous), with taxonomic treatment of the family. Cretaceous Research, 26(5), 801–812. Makarkin, V. N. & Menon, F. 2007. First record of fossil `rapismatid-like' Ithonidae (Insecta, Neuroptera) from the Lower Cretaceous Crato Formation of Brazil. Cretaceous Research, 28(5), 743–753. Makarkin, V. N. & Archibald, S. B. 2009. A new genus and first Cenozoic fossil record of moth lacewings (Neuroptera: Ithonidae) from the Early Eocene of North America. Zootaxa, 2063, 55–63. Makarkin, V. N. & Wedmann, S. 2009. First record of the genus Sympherobius (Neuroptera: Hemerobiidae) from Baltic amber. Zootaxa, 2078, 55–62. Makarkin, V. N., Archibald, S. B. & Oswald, J. D. 2003. New Early Eocene brown lacewings (Neuroptera : Hemerobiidae) from western north America. Canadian Entomologist, 135(5), 637–653. Makarkin, V. N., Ren, D. & Yang, Q. 2009. Two New Species of Kalligrammatidae (Neuroptera) From the Jurassic of China, With Comments on Venational Homologies. Annals of the Entomological Society of America, 102(6), 964– 969. Martín-Closas, C. & López-Móron, N. 1995. The Charophyte flora. In: Martínez- Delclòs, X. (ed), Montsec & Montral-Alcover. Two Konservat-Lagerstätten Catalonia, Spain., Lérida, 29–31. Martínez-Delclòs, X. 1989. Chresmoda aquatica n. sp. insecto Chresmodidae del Cretácico inferior de la Sierra del Montsec (Lleida, España). . Revista Española de Paleontología, 4, 67–74. Martínez-Delclos, X. & Martinell, J. 1993. Insect taphonomy experiments. Their application to the Cretaceous outcrops of the Lithographic Limestones from Spain. Darmstadter Beiträge zur Naturgeschichte, 2, 133–144. Martínez-Delclòs, X. & Nel, A. 1995. Arthropods. In: Martínez-Delclòs, X. (ed), Montsec & Montral-Alcover. Two Konservat-Lagerstätten Catalonia, Spain, Lérida, 39–46.

209

Martins-Neto, R. G. 1990. Neurópterous (Insecta: Planipennia) da formação Santana (Cretáceo Inferior) Bacia do Araripe, Nordeste do Brasil. VI – Ensaio flogenético das espécies do gênero Blittersdorffia Martins-Neto and Vulcano, com descrição de nova espécie. Acta Geologica Leopoldensia, 13, 3–12. Martins-Neto, R. G. 1994. Neurópterous (Insecta: Planipennia) da formação Santana (Cretáceo Inferior) Bacia do Araripe, Nordeste do Brasil. IX – Primeiros resultados da composicao da fauna e descricao de novos taxons. Geologica Leopoldensia, 17, 269–288. Martins-Neto, R. G. 1997. Neurópterous (Insecta: Planipennia) da formação Santana (Cretáceo Inferior) Bacia do Araripe, Nordeste do Brasil. X – Descricao de novos taxons (Chrysopidae, Babinskaiidae, Myrmeleontidae, Ascalaphidae e Psychopsidae). Ciencias Exatas e Tecnologicas, 2, 68–83. Martins-Neto, R. G. 2000. Remarks on the neuropterofauna (Insecta, Neuroptera) from the Brazilian Cretaceous, with keys for the identification of known taxa. Acta Hispanica, 35, 97–118. Martins-Neto, R. G. 2002. The Santana Formation palaeoentomofauna reviewed part 1 – Neuropteroida (Neuroptera and Raphidioptera): systematic and phylogeny, with description of new taxa. Acta Geologica Leopoldensia, 25, 36–66. Martins-Neto, R. G. 2005. Estágio atual da palaeoarthropodologia Brasilieira: Hexápodes, Miriápodes, Crustáceos (Isopoda, Decapoda, Eucrustacea e Copepoda) e quelicerados. Arquivos do Museu Nacional, Rio de Janeiro, 63(3), 471–494. Martins-Neto, R. G. 2006. Insetos fósseis como bioindicatores em depósitos sedimentares: um estudo de casa para o Cretáceo da Bacia do Araripe (Brasil). Brasileira de Zoociências, 8, 155–183. Martins-Neto, R. G. & Vulcano, M. A. 1989a. Neurópterous (Insecta: Planipennia) da formação Santana (Cretáceo Inferior) Bacia do Araripe, Nordeste do Brasil. II. Superfamilia Myrmeleontoidea. Revista Brasileira da Entomologia, 33(2), 367–402. Martins-Neto, R. G. & Vulcano, M. A. 1989b. Neurópterous (Insecta: Planipennia) da formação Santana (Cretáceo Inferior) Bacia do Araripe, Nordeste do Brasil. IV Complemento partes I & II, com descrição de novos taxa. Anais da Academia Brasileira de Ciencias, 61, 312–318.

210

Martins-Neto, R. G. & Vulcano, M. A. 1992. Neurópterous (Insecta: Planipennia) da formação Santana (Cretáceo Inferior) Bacia do Araripe, Nordeste do Brasil. VII – Palaeolentinae, nova subfamília de Myrmeleontidae e descrição de novos táxons. Revista Brasileira da Entomologia, 36(4), 803–815. Martins-Neto, R. G. & Nel, A. 1992. Un nouveau fossile de Raphidioptère de la formation santana, Crétacé infériur du Brésil (Neuropteroidea, Raphidioptera). Bulletin de la Société entomologique de France, 97, 425–428. Martins-Neto, R. G. & Vulcano, M. A. 1997. Neurópterous (Insecta: Planipennia) da formação Santana (Cretáceo Inferior) Bacia do Araripe, Nordeste do Brasil. VIII – Descricao de novos taxa de Myrmeleontidae, Ascalaphidae e Nemopteridae. Ciencias Biologicas e da Saude, 2(2), 64–81. Martins-Neto, R. G. & Rodrigues, V. Z. 2009. New Neuroptera (Insecta, Osmylidae and Mesochrysopidae) from the Santana Formation, Lower Cretaceous of northeast Brazil. Gaea, 5(1), 15–20. Martins-Neto, R. G., Heads, S. W. & Bechly, G. 2007. Neuropterida. In: Martill, D. M., Bechly, G. & Loveridge, R. F. (eds), he Crato Fossil Beds of Brazil: Window into an Ancient World. Cambridge University Press, 328–340. Martynov, A. V. 1925a. To the knowledge of fossil insects from Jurassic beds in Turkestan 1. Raphidioptera. Izvestiya Rossiiskoi akademii Nauk, 6 seryia, 19, 233–246. Martynov, A. V. 1925b. To the knowledge of fossil insects from Jurassic beds in Turkestan. 2. Raphidioptera (continued), Orthoptera (s.l.), Odonata, Neuroptera. Izvestiya Rossiiskoi akademii Nauk, 6 seryia, 19, 569–598. Martynov, A. V. 1927. Jurassic fossil insects from Turkestan. 7. some Odonata, Neuroptera, Thysanoptera. Bulletin de l’Académie des Sciences de l’URSS, 21(6), 757–768. Martynova, O. M. 1947. Two new snakeflies from the Jurassic shales of Karatau. Doklady Akademii Nauk SSSR, 56, 635–637. Martynova, O. M. 1949. Mezozoiskie setchatokrylye (Neuroptera) i ikh znachenie dlya ponimaniya filogenii i sistematiki otryada [Mesozoic Lacewings (Neuroptera) and their bearings on concepts of Phylogeny and Systematics]. Doklady Akademii Nauk SSSR, 20, 150–170. Martynova, O. M. 1952. Permskie setchatokrylye SSSR [=Permian Neuroptera of the USSR]. In: Orlov, Y. A. (ed), Materialy po faune paleozoya [=Materials

211

for the fauna]. Akademiia nauk SSSR, Trudy Paleontologicheskogo Instituta, 40, 197–237. Martynova, O. M. 1954. Setchatokrylye nasekomye iz melovykh otlozhenii Sibiri [=Neuropterous insects from the Cretaceous deposits of Siberia]. Doklady Akademii Nauk SSSR, 94, 1167–1169. Martynova, O. M. 1961. Recent and extinct snakeflies (Insecta: Raphidioptera). Paleontologicheskii Zhurnal, 1961, 73–83. Meinander, M. 1975. Fossil Coniopterygidae (Neuroptera). Natulae Entomologicae, 55, 53–57. Menge, A. 1856. Lebenszeichen vorweltlicher, im Bernstein eingeschlossener Thiere. Kafemann. Menon, F. & Makarkin, V. N. 2008. New fossil lacewings and antlions (Insecta, Neuroptera) from Cretaceous Crato Formation of Brazil. Palaeontology, 51, 149–162. Menon, F., Martins-Neto, R. G. & Martill, D. M. 2005. A new lower Cretaceous Nymphid (Insecta, Neuroptera, Nymphidae) from the lower Crato Formation of Brazil. Gaea, 1(1), 11–15. Meunier, F. 1897. Revue critique de quelques insectes fossiles du Museé Teyler. Archives du Museé Teyley, 2, 85–149. Mickoleit, G. 1973. Über den Ovipositor der Neuropteroidea und Coleoptera und seine phylogenetische Bedeutung (Insecta, Holometabola). Zeitschrift für Morphologie der Tiere, 74, 37–64. Miller, G. L., Oswald, J. D. & Miller, D. R. 2004. Lacewings and scale insects: A review of predator/prey associations between the Neuropterida and Coccoidea (Insecta : Neuroptera, Raphidioptera, Hemiptera). Annals of the Entomological Society of America, 97(6), 1103–1125. Milner, A. & Batten, D. J. (eds), 2002. Palaeoenvironmental setting of the Purbeck Limestone Group of Dorset, southern England, 68. Special Papers in Palaeontology, Palaeontological Associtaion, 268 pp. Navás, L. 1912. Monografia de l‘ordre dels Rafidiòpters (Ins.). Arxius [Arxivus] de l’ Institute [d’Estudis Catalans, Seccio] de Ciencias, Barcelona, 1–190. Nel, A. 1988a. Les Sialdiae (Megaloptera) fossiles des diatomites de Murat (Cantal, France) et des Bes-Konak (Anaolie, Turque). Neuroptera International, 5(1), 39–44.

212

Nel, A. 1988b. Deux nouveaux Mantispidae (Planipennia) fossilesde l‘Oligocene du sud-ouestde la France. Neuroptera International, 5, 103–109. Nel, A. 1990. Nouveaux Insectes Neuroptéroïdes fossiles de l‘Oligocène de France (Neuroptera et Megaloptera). Bulletin du Muséum National d’Histoire Naturelle, Série 4, Section C (Sciences de la Terre Paléontologie, Géologie, Minéralogie), 12, 327–349. Nel, A. 1993. Nouveaux Raphidioptères fossiles du Cénozoïque de France et d‘Espagne (Raphidioptera, Raphidiidae, Inocelliidae). Biologie et Évolution des Insectes, 6, 99–108. Nel, A. & Séméria, Y. 1988. Une nouvelle espece de chrysopide fossile du Stampien Superieur (Oligocene) d‘Aix-en-Provence, Notochrysa stampieni n. sp. (Neuroptera, Chrysopidae, Notochrysinae). Neuroptera International, 4(4), 23–30. Nel, A. & Henrortay, M. 1994. Les Chrysopidae Mésozoiques. État actual des conaissances. Description d‘un nouveau genre et nouvelle espèce dans le Jurassique Inférior (Lias) (Insecta: Neuroptera). Annales de la Société Entomologique de France (N.S.), 30, 293–318. Nel, A. & Jarzembowski, E. A. 1997. New fossil Sisyridae and Nevrorthidae (Insecta: Neuroptera) from Eocene Baltic amber and Upper Miocene of France. European Journal of Entomology, 94, 287–294. Nel, A., Séméria, Y. & Martins-Neto, R. G. 1990. Un Raphidioptera fossile du Cretace Inferieur du Bresil (Neuropteroidea). Neuroptera International, 6, 27– 37. Nel, A., Menier, J.-J., De Ploëg, G., Hodebert, G. & Danvin, L. 2002. Eosialis, a new genus in French Lowermost Eocene amber (Insecta, Megaloptera, Sialidae). Geobios, 35, 313–319. Nel, A., Menier, J.-J., Waller, A., Hodebert, G. & de Ploeg, G. 2003. New fossil spongilla-flies from the lowermost Eocene amber of France (Insecta, Neuroptera, Sisyridae). Geodiversitas, 25, 109–117. Nel, A., Martínez-Delclos, X. & Hutin, A. 2005a. Mesozoic chrysopid-like Planipennia a phylogenetic approach (Insecta: Neuroptera). Annales de la Société Entomologique de France (N.S.), 41, 26–68 Nel, A., Perrichot, V., Azar, D. & Neuraudeau, D. 2005b. New Rhachiberothidae (Insecta: Neuroptera) in early Cretaceous and early Eocene ambers from

213

France and Lebanon. Neues Jahrbuch für Geologie und Palaeontologie, 235, 51–85. Nel, A., Perrichot, V. & Azar, D. 2005c. New and poorly known fossil Coniopterygidae in Cretaceous and Cenozoic ambers (Insecta: Neuroptera). Annales Zoologici, 55(1), 1–7. New, T. R. 1983. Revision of the Osmylid Subfamilies Porsiminae and Eidoporsiminae (Insecta: Neuroptera). Australian Journal of Zoology, 31, 763– 770. New, T. R. 1986. A review of the biology of Neuroptera Planipennia. Neuroptera International, 1, 1–57. New, T. R. 1988. The Psychopsidae (Insecta: Neuroptera) of Australia and the Oriental Region. Invertebrate Taxonomy, 2, 841–883. New, T. R. 1989. Handbook of Zoology, A Natural History of the Phyla of the Animal Kingdom. Volume IV Arthropoda: Insecta Planipennia Lacewings. Walter de Gruyter, Berlin. New, T. R. 1991. Neuroptera (lacewings). In: Naumann, I. D., Carne, P. B. et al. (eds), The Insects of Australia: A textbook for students and research workers, Volume 1. Melbourne University Press, Victoria, 525–542. New, T. R. & Theischinger, G. 1993. Megaloptera, Alderflies and Dobsonflies. Handbuch der Zoologie, Vol. 4 (Arthropoda: Insecta), Part 33. Walter de Gruyter, 97 pp. Norris, G. 1969. Miospores from the Purbeck Beds and marine Upper Jurassic of southern England. Palaeontology, 12, 574–620. Novokshonov, V. G. 1998. New fossil insects (Insecta: Grylloblattida, Caloneurida, Hypoperlida?, ordinis incertis) from the Kungen beds of the middle Urals. Paleontological Journal, 32, 41–46. Ogg, J. G., Hasenyager, R. W. & Wimbledon, W. A. 1994. Jurassic–Cretaceous boundary: Portland–Purbeck magnetostratigraphy and possible corellation to the Tethyan faunal realm. Geobias, 17, 519–527. Oswald, J. D. 1990. Raphidioptera. In: Grimaldi, D. A. (ed), Insects from the Santana Formation, Lower Cretaceous, of Brazil. Bulletin of the American Museum of Natural History, New York, 195, 154–163.

214

Oswald, J. D. 1993a. Revision and cladistic analysis of the world genera of the family Hemerobiidae (Insecta: Neuroptera). Journal of the New York Entomological Society, 101, 143–299. Oswald, J. D. 1993b. Phylogeny, taxonomy, and biogeography of extant silky lacewings (Insecta: Neuroptera: Psychopsidae). Memoirs of the American Entomological Society, 40, 1 – 65. Panfilov, D. V. 1980. Novye predstaviteli setcharokrylykh (Neuroptera) iz yury Karatau [New representatives of lacewings (Neuroptera) from the Jurassic of Karatau]. In: Dolin, V. G., Panfilov, D. V. et al. (eds), Iskopaemye nasekomye mezozoya (Fossil insects of the Mesozoic). Naukova Dumka, Kiev, 82–111. Pantaleoni, R. A., Aspöck, U., Cao, O. V. & Aspöck, H. 2004. Subilla principiae n.sp., a new spectacular snakefly from Sardinia (Raphidioptera Raphidiidae). Redia, 87, 3–6. Parfin, S. I. 1952. The Megaloptera and Neuroptera of Minnesota. American Midland Naturalist, 47, 421–434. Peñalver, E., Nel, A. & Martínez-Delclõs, X. 1995. New Notochrysinae from the Spanish Miocene (Neuroptera, Chrysopidae). Bulletin de la Société entomologique de lá France, 100, 481–487. Penney, D. 2005. Importance of Dominican Republic Amber for determining taxonomic bias of fossil resin preservation—A case study of spiders. Palaeogeography, Palaeoclimatology, Palaeoecology, 223, 1–8. Plant, C. W. 1997. A key to the adults of British lacewings and their allies (Neuroptera, Megaloptera, Raphidioptera and Mecoptera). Field Studies, 9, 179–269. Poinar, G. 2006. Ferosta priscus (Neuroptera: Mantispidae), a new genus and species of mantidflies in Dominican Amber. Proceedings of the Entomological Society of Washington, 108(2), 411–417. Ponomarenko, A. G. 1976. Koridalidy (Megaloptera, Corydalidae) iz melovykh otlozhenii severnoi Azii [Corydalidae (Megaloptera) from the Cretaceous deposits of northern Asia]. Entomologicheskoe Obozrenie, 55, 425–433. Ponomarenko, A. G. 1977. Paleozoiskie predstaviteli vislokrylok (Insecta, Megaloptera) [Paleozoic members of the Megaloptera (Insecta)]. Paleontologicheskii Zhurnal, 1977(1), 78–86.

215

Ponomarenko, A. G. 1984. Neuroptera from the Jurassic in Eastern Asia. Paleontological Journal, 18, 59–69. Ponomarenko, A. G. 1986. Neuroptera from the Jurassic of Siberia and western Mongolia. 91–95. In: Rasnitsyn, A. P. (ed.). The Jurassic insects of Siberia and Mongolia. Trudy Paleontologicheskogo Instituta, 211, 192 pp. Ponomarenko, A. G. 1988. New Mesozoic insects. Trudy Sovmestnaya Sovestko- Mongolskaya Paleontologicheskogo Ekspeditsiya [Transactions of the Joint- Soviet-Mongolian Palaeontological Expedition], 33, 71–80. Ponomarenko, A. G. 1992a. Neuroptera (Insecta, Neuroptera) of the Mesozoic of Mongolia. Akademiia nauk SSSR, 41, 101–111. Ponomarenko, A. G. 1992b. Neuroptera (Insecta) from the Lower Cretaceous of Transbaykalia. Paleontological Journal, 26, 56–66. Ponomarenko, A. G. 1993. Lower Cretaceous snakeflies from Transbaikalia. Trudy Paleontologicheskaya Instituta, Akademii Nauk, SSSR, 252, 60–80. Ponomarenko, A. G. 1995. Upper Liassic neuropterans (Insecta) from Lower Saxony, Germany. Russian Entomological Journal, 4(1), 73–89. Ponomarenko, A. G. 1996. Upper Liassic neuropterans (Insecta) from Lower Saxony, Germany. Russian Entomological Journal, 4, 73–89. Ponomarenko, A. G. 2000. New alderflies (Megaloptera: Parasialidae) and Glosselytrodeans (Glosselytrodea: Glosselytridae) from the Permian of Mongolia. Paleontological Journal, 34, 309–311. Ponomarenko, A. G. 2002. Superorder Myrmeleontidea Latreille, 1802 (=Neuropteroidea Handlirsch, 1903). In: Rasnitsyn, A. P. & Quicke, D. L. J. (eds), History of Insects. Kluwer Academic Publishers, Dordrecht, 176–192. Ponomarenko, A. G. 2003. On some Neuroptera (Insecta) from Upper Jurrasic Solnhofen Limestone. Annals of the Upper Silesian Museum (Entomology), 12, 87–100. Ponomarenko, A. G. & Shcherbakov, D. E. 2004. New lacewings (Neuroptera) from the terminal Permian and basal Triassic of Siberia. Paleontological Journal, 38, 197-203. Pupedis, R. J. 1987. Foraging behavior and food of adult Spongila-flies [sic] (Neuroptera: Sisyridae). Annals of the Entomological Society of America, 80, 758–760.

216

Radley, J. D. 2002. Distribution and palaeoenvironmental significance of molluscs in the Late Jurassic–Early Cretaceous Purbeck Formation of Dorset, southern England: a review. In: Milner, A. & Batten, D. J. (eds), Life and environments in Purbeck times. Special Papers in Palaeontology. Palaeontological Association, London, 68, 41–51. Radley, J. D. 2005. Derived fossils in the southern English Wealden (non-marine early Cretaceous): a review. Cretaceous Research, 26, 657–664. Rambur, J. P. 1842. Histoire Naturelle des Insectes, Névroptères. Librairie encyclopédique de Roret. Fain et Thunot, 534 pp. Rasnitsyn, A. P. & Ross, A. J. 2000. A preliminary list of arthropod families present in Burmese amber collection at the Natural History Museum, London. Bulletin of the Natural History Museum London (Geology), 56, 21–24. Rasnitsyn, A. P., Jarzembowski, E. A. & Ross, A. J. 1998. Wasps (Insecta: Vespida = Hymenoptera) from the Purbeck and Wealden (Lower Cretaceous) of southern England and their biostratigraphical and palaeoenvironmental significance. Cretaceous Research, 19, 329–391. Rausch, H. & Aspöck, H. 1993. Phaeostigma holzingeri n. sp. – eine neue Kamelhalsfliege aus Ipiros (Greichenland) (Neuropteroidea, Raphidioptera: Raphidiidae). Zeitschrift der Arbeitsgemeinschaft Oesterreichischer Entomologen, 45(1), 19–26. Rausch, H., Aspöck, H. & Aspöck , U. 2004. Calabroraphidia renate n. gen. sp. - eine neue Spezies und ein neues Genus der Familie Raphidiidae aus Süditalien (Neuropterida, Raphidioptera). Entomologische Nachrichten und Berichte, 48(3), 159–165. Ren, D. 1994. A new discovery of snake-flies (Insecta) from late Jurassic of Beipiao, Liaoning. Professional Papers of Stratigraphy and Paleontology, 25, 131–140. Ren, D. 1997. Studies on the late Mesozoic snakeflies of China (Raphidioptera: Baissopteridae, Mesoraphidiidae, Alloraphidiidae). Acta Zootaxonomica Sinica, 22, 172–188. Ren, D. 2002. A new lacewing family (Neuroptera) from the Middle Jurassic of Inner Mongolia, China. Entomologica Sinica, 9(4), 53–67. Ren, D. 2003a. Two new Late Jurassic genera of Kalligrammatids from Beiapiao, Liaoning (Neuroptera: Kalligrammatidae). Acta Zootaxonomica Sinica, 28(1), 105–109.

217

Ren, D. 2003b. Study on the Jurassic fossil Raphidioptera and Neuroptera from northeast China, Beijing Forestry University, 202 pp. Ren, D. & Hong, Y.-C. 1994. A cladistic study on the familial phylogeny of fossil and living Raphidioptera (Insecta). Bulletin of the Chinese Academy of Geological Sciences, 29, 103–117. Ren, D. & Guo, Z. G. 1996. On the new fossil genera and species of Neuroptera (Insecta) from the late Jurassic of northeast China. Acta Zootaxonomica Sinica, 21, 461–479. Ren, D. & Yin, J.-C. 2002. A new genus and new species of lacewings from the Jurassic of China (Insecta: Myrmeleontoidea). Acta Zootaxonomica Sinica, 27(2), 269–273. Ren, D. & Oswald, J. D. 2002. A new genus of kalligrammatid lacewing from the middle Jurassic of China. Stuttgarter Beiträge zur Naturkunde Serie B (Geologie und Paläontologie), 317, 1–8. Ren, D. & Yin, J.-C. 2003. New ‗osmylid-like‘ fossil Neuroptera from the middle Jurassic of Inner Mongolia, China. Journal of the New York Entomological Society, 111(1), 1–11. Ren, D. & Engel, M. S. 2007. A split-footed lacewing and two Epiosmylines from the Jurassic of China (Neuroptera). Annales Zoologici, 57(2), 211–219. Ren, D. & Engel, M. S. 2008a. Aetheogrammatidae, a new family of lacewing from the Mesozoic of China (Neuroptera: Myrmeleontiformia). Journal of the Kansas Entomological Society, 81(3), 161–167. Ren, D. & Engel, M. S. 2008b. A second from the Mesozoic of northeastern China (Neuroptera: Myrmeleontidae). Alavesia, 2, 183–186. Ren, D., Lu, L. W., Guo, Z. G. & Ji, S. A. 1995. Faunae and stratigraphy of Jurassic-Cretaceous in Beijing and the adjacent areas. Seismic Publishing House. Ren, D., Engel, M. S. & Lü, W. 2002. New giant lacewings from the Middle Jurassic of Inner Mongolia, China (Neuroptera: Polystoechotidae). Journal of the Kansas Entomological Society, 75(3), 188–193. Rohwer, S. A. 1909. Three new fossil insects from Florissant, Colorado. American Journal of Science, 28, 533–536.

218

Ross, A. J. & Cook, E. 1995. The stratigraphy and palaeontology of the Upper Weald Clay (Barremian) at Smokejacks Brickworks, Ockley, Surrey, England. Cretaceous Research, 16, 705–716. Ross, A. J. & Jarzembowski, E. A. 1996. A provisional checklist of fossil insects from the Purbeck Group of Wiltshire. Wiltshire Archaeological and Natural history Magazine, 89, 106–115.. Ru, N., Whitcomb, W. H., Murphey, M. & Carlysle, T. C. 1975. Biology of Chrysopa lanata (Neuroptera: Chrysopidae). Annals of the Entomological Society of America, 68, 187–190. Ruffell, A. H. & Batten, D. J. 1994. Uppermost Wealden facies and Lower Greensand Group (Lower Cretaceous) in Dorset, southern England: correlation and palaeoenvironment. Proceedings of the Geologists’ Association, 105, 53– 69. Scherbakov, D. E. 1995. A new genus of the Palaeozoic Order Hypoperlida. Russian Entomological Journal, 3, 33–36. Schlüter, T. 1982. Cimbochrysa molerensis n. g. n. sp. und Hypochrysa hercyniensis n. sp., zwei fossile Chrysopidae-Arten (Insecta: Planipennia) aus dem europäischen Tertiär. Neues Jahrbuch für Geologie und Paläontologie, 1982, 257–264. Schlüter, T. 1983. A fossiliferous resin from the of the Paris and Aquitanian Basins of Northwestern France. Cretaceous Research, 4(3), 265– 269. Schlüter, T. 1984. The phylogeny of Chrysopidae. In: Canard, M., Séméria, Y. & New, T. R. (eds), Biology of Chrysopidae. Dr. W. Junk Publishers, , 1–8. Schlüter, T. 1986. Fossil Planipennia – a review. In: Gepp, J., Aspöck, H. & Hölzel, H. (eds), Proceedings of the 2nd International Symposium on Neuropterology in Graz (Austria): 103–111. Schlüter, T. & Sturmer, W. 1984. Die identifikationeiner fossilen Rhachiberothinae- Art (Planipennia: Berothidae oder Mantispidae) aus mittelkretazischem Bernstein NW-Frankreichs mit Hilfe rontgenographischer Methoden. Proceedings of the 1st International Symposium on Neuropterology in Graz (Austria), 49–55.

219

Schneider, W. G. 1851. Symbolae ad monographiam generis Chrysopae, Leach. . Apud Ferdinandum Hirt, Vratislaviae, 178 pp. Schreiber, B. C. 1986. Arid shorelines and evaporites. In: Reading, H. G. (ed), Sedimentary environments and facies. Blackwell Science, Oxford, 189–228. Scudder, S. H. 1885a. Systematische Übersicht der fossilen Myriopoden [sic], Arachnoideen, und Insekten. In: Zittel, K. A. (ed), Handbuch der Palaeontologie, Abteilung I, Zweiter Band, Palaeozoologie. Oldenbuch, Munich, 721–831. Scudder, S. H. 1885b. Notes on Mesozoic cockroaches. Proceedings of the Academy of Natural Sciences of Philadelphia, 37, 105–115. Scudder, S. H. 1886. Systematic review of our present knowledge of fossil insects including myriapods and arachnids. Bulletin of the United States Geological Survey, 5, 1–128. Scudder, S. H. 1890. The Tertiary insects of North America. Report of the United States Geological Survey of the Territories, 13, 1–663. Scudder, S. H. 1891. Index to the known fossil insects of the world including myriapods and arachnids. Bulletin of the United States Geological Survey, 71, 1–744. Selden, P.A. 2003. A new tool for fossil preparation. The Geological Curator, 7(9), 337–339. Sharov, A. G. 1968. Filogeniya ortopteroidnykh nasekomykh [=The phylogeny of orthopteroid insects]. Akademiia nauk SSSR, Trudy Paleontologicheskogo Instituta, 118, 1–217. Sinitshenkova, N. D. 1998. The first European Cretaceous stonefly (Insecta, Perlida = Plecoptera). Cretaceous Research, 19, 317–321. Sladen, C. P. & Batten, D. J. 1984. Source-area environments of Late Jurassic and Early Cretaceous sediments in southeast England. Proceedings of the Geologists’ Association, 95, 149–163. Storozhenko, S. Y. 1998. Sistematika, filogenia i evolyutsiya grilloblattidovskh nasekomykh (Insecta: Grylloblattida). [Systematics, phylogeny and evolution of Grylloblattid insects.]. Dal'nauka. Storozhenko, S. Y. & Novokshonov, V. G. 1995. Revision of the Permian family Sojanoraphidiidae (Grylloblattida). Russian Entomological Journal, 3(3), 37– 39.

220

Swofford, D. L. 2002. PAUP: phylogenetic analysis using parsimony (and other methods) 4.0 beta. Sunderland, MA: Sinauer Associates. Szentkirályi, F. & Kristín, A. 2002. Lacewings and snakeflies (Neuroptera, Raphidioptera) as prey for bird netslings in Slovakian forest habitats. Acta Zoologica Academie Scientiarum Hungaricae, 48, 329–340. Thomas, B. 1981. The evolution of plants and flowers. St. Martin‘s Press, 116 pp. Thorne, A. L. 1972. The Neuroptera suborder Planipennia of Wisconsin, part III – Mantispidae, Ascalaphidae, Myrmeleontidae and Coniopterygidae. Great Lakes Entomologist, 5, 119–128. Tillyard, R. J. 1917. Mesozoic insects of Queensland. No.1. Planipennia, trichoptera, and the new Order Protomecoptera. Proceedings of the Linnean Society of New South Wales, 42, 175–200. Tillyard, R. J. 1919. Mesozoic insects of Queensland. No.5. Mecoptera, the new Order Paratrichoptera, and additions to Planipennia. Proceedings of the Linnean Society of New South Wales, 44, 194–212. Tillyard, R. J. 1922. Mesozoic insects of Queensland. No. 9. Orthoptera, and additions to the Protorthoptera, Odonata, Hemiptera and Planipennia. Proceedings of the Linnean Society of New South Wales, 47, 447–470. Tillyard, R. J. 1923. Mesozoic insects of Queensland. No. 10. Summary of the upper Triassic insect fauna of Ipswich, Queensland (with an appedix describing new Hemiptera and Planipennia). Proceedings of the Linnean Society of New South Wales, 48, 481–498. Tillyard, R. J. 1926. Upper Permian insects of New South Wales. Part ii. The Orders Mecoptera, Paramecoptera and Neuroptera. Proceedings of the Linnean Society of New South Wales, 51, 265–268. Tillyard, R. J. 1932. Kansas Permian insects. Part 14. The Order Neuroptera. American Journal of Science, 23, 1–30. Tillyard, R. J. 1933. The panorpoid complex in the British Rhaetic and Lias. British Museum (Natural History), Fossil Insects, 3, 1–79. Tillyard, R. J. 1937. Kansas Permian Insects. Part 17. The Order Megasecoptera and additions to the Palaeodictyoptera, Odonata, Protoperlaria, Copeognatha and Neuroptera. American Journal of Science, 33, 81–110. Walther, J. 1904. Fauna der Solnhofener Plattenkalke. Denkschriften der Medizinisch-Naturnissenschaftlichen Gesellschaft, 11, 133–214.

221

Wang, W. 1987. Early Mesozoic insects from western Liaoning. Mesozoic stratigraphy and palaeontology of western Liaoning. Geological Publishing House, Beijing, 3, 202–222. Watson, J. & Alvin, K. L. 1996. An English Wealden floral list, with comments on possible environmental indicators. Cretaceous Research, 17, 5–26. Wedmann, S. & Makarkin, V. N. 2007. A new genus of Mantispidae (Insecta : Neuroptera) from the Eocene of Germany, with a review of the fossil record and palaeobiogeography of the family. Zoological Journal of the Linnean Society, 149(4), 701-716. West, I. M. 1975. Evaporites and associated sediments of the basal Purbeck Formation (Upper Jurassic) of Dorset. Proceedings of the Geologists’ Associtaion, 86, 205–225. West, I. M. 1979. Review of evaporite diagenesis in the Purbeck Formation of southern England, Symposium: Sedimentation Jurassique W. Europeen. A.S.F. Publication Speciale: 407–416. Westwood, J. O. 1854. Contributions to fossil entomology. Quarterly Journal of the Geological Society of London, 10, 378–396. Weyenbergh, H. 1869. Sur les insectes fossiles du calcaire lithographique de la Bavière, qui se trouvent au Museé Teyler. Archives du Museé Teyler, 2(1), 247–294. Whalley, P. E. S. 1980. Neuroptera (Insecta) in amber from the Lower Cretaceous of Lebanon. Bulletin of the British Museum of Natural History (Geology), 33, 157–164. Whalley, P. E. S. 1985. The systematics and palaeogeography of the Lower Jurassic insects of Dorset, England. Bulletin of the British Museum of Natural History (Geology), 39, 107–189. Whalley, P. E. S. 1988. Mesozoic Neuroptera and Raphidioptera (Insecta) in Britain Bulletin of the British Museum of Natural History (Geology), 44(1), 45–63. Whalley, P. E. S. & Jarzembowski, E. A. 1985. Fossil insects from the Lithographic Limestone of Montsech (late Jurassic-early Cretaceous), Lérida Bulletin of the British Museum of Natural History (Geology), 38, 381–412. Whiting, M. F., Carpenter, J. C., Wheeler, Q. D. & Wheeler, W. C. 1997. The Strepsiptera problem: phylogeny of the holometabolous insect Orders inferred

222

from 18S and 28S ribosomal DNA sequences and morphology. Systematic Biology, 46, 1–68. Whiting, M. F. 2002a. Phylogeney of Holometabola: state of the art. Zoologica Scripta, 31(1), 1–2. Whiting, M. F. 2002b. Phylogeny of the holometabolous insect Orders: molecular evidence. Zoologica Scripta, 31, 3–15. Wichard, W. 1997. Schammelfliegen aus baltischen Bernstein (Megaloptera, Sialidae). Mitteilungen aus dem Geologisch-Paläontologiscen Institut der Universität Hamburg, 80, 197–211. Wichard, W. 2002. Eine neue Schlammfliege aus dem baltischen Bernstein (Megaloptera, Silaidae). Mitteilungen aus dem Geologisch-Paläontologiscen Institut der Universität Hamburg, 87, 147–158. Wichard, W. 2003. Chauliodes, ein Großflügler im baltischen Bernstein (Megaloptera, Corydalidiae). Mitteilungen aus dem Geologisch- Paläontologiscen Institut der Universität Hamburg, 87, 147–158. Wichard, W. & Engel, M. S. 2006. A new Aldefly in Baltic amber (Megaloptera: Sialidae). American Museum Novitates, 3515, 1–9. Wichard, W., Chatterton, C. & Ross, A. J. 2005. Corydasialidae fam. n. (Megaloptera) from Baltic Amber. Insect Systematics and Evolution, 36, 279– 283. Willmann, R. 1990. The phylogenetic position of the Rhachiberothinae and the basal sister-group relationships within the Mantispidae (Neuroptera). Systematic Entomology, 15, 253–265. Willmann, R. 1993. Insekten aus der Fur-Formation von Dänemark (Moler, ob. ic) Neues Jahrbuch für Geologie und Palaeontologie, 4, 239–245. Willmann, R. 1994. Raphidiodea aus dem Lias und die Phylogenie der Kamelhalsfliegen (Insecta: Holometabola) Palaontologische Zeitschrift, 68, 167–197. Willmann, R. & Brooks, S. J. 1991. Insekten aus der Fur-Formation von Dänemark (Moler, ob. Paleozän/unt. Eozän?). 6. Chrysopidae (Neuroptera). Meyniana, 43, 125–135. Wise, K. A. J. 1995. Records concerning biological control of insect pests by Neuropteroidea (Insecta) in New Zealand. Records of the Auckland Institute and Museum, 32, 101–117.

223

Woodward, H. 1892. On a Neuropterous Insect from the Lower Lias, Barrow-on- Soar, Leicestershire. Geological Magazine, 9, 193–198. Wootton, R. J. 2002. Design, function and evolution in the wings of holometabolous insects. Zoologica Scripta, 31–40. Wootton, R. J. 2003. Wings. In: Resh, V. H. & Carde, R. T. (eds), Encyclopedia of Insects. Academic Press, London, 1186–1192. Yang, J. & Hong, Y. C. 1990. Drakochrysa, an early Cretaceous new genus of Chrysopidae (Insecta: Neuroptera) from Laiyang Basin, Shandong Province. Geoscience, 4, 15–26. Yang, Q., Zhao, Y. & Ren, D. 2009. An exceptionally well-preserved fossil kalligramatid from the Jehol Biota. Chinese Science Bulletin, 54(10), 1732– 1737. Yoshida, T., Sugiomoto, T. & Hayashi, F. 1985. Notes on the life history of the , Protohermes grandis Thunberg (Megaloptera, Corydalidae). Kontyû, 53, 734–742. Zalessky, G. M. 1953. New localities of Cretaceous insects from the Volga region, Kazakhstan and Transbaikalia. Doklady Akademii Nauk SSSR, 89, 163–166. Zhang, Z. F. 1991. A new family of Neuroptera (Insecta) from the Late Mesozoic of Shandong, China. Science in China (Series B), 34(9), 1105–1112.

224