A Molecular Phylogenetic Assessment of North America Lichen Moths (Lepidoptera; Erebidae: Arctiinae; Lithosiini) with Life Histo

A Molecular Phylogenetic Assessment of the North American Lichen Tiger Moths (Lepidoptera; Erebidae; Arctiinae; Lithosiini) with Life History Observations and a Description of a New Species from Central Arizona

Item Type text; Electronic Dissertation

Authors Palting, John Douglas

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 24/09/2021 12:38:11

Link to Item http://hdl.handle.net/10150/642057

A MOLECULAR PHYLOGENETIC ASSESSMENT OF NORTH AMERICA LICHEN MOTHS (LEPIDOPTERA; EREBIDAE: ARCTIINAE; LITHOSIINI) WITH LIFE HISTORY OBSERVATIONS AND A DESCRIPTION OF A NEW SPECIES FROM ARIZONA

A Dissertation Submitted to the Faculty of the

GRADUATE INTERDISCIPLINARY PROGRAM IN ENTOMOLOGY AND INSECT SCIENCE

In Partial Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2020

TABLE OF CONTENTS

ABSTRACT...... 4 CHAPTER 1 ...... 5 CHAPTER 2 ...... 9 REFERENCES ...... 11 APPENDIX A ...... 16 APPENDIX B ...... 46 APPENDIX C ...... 47

ABSTRACT

The tribe Lithosiini (Lepidoptera: Erebidae: Arctiinae) includes some 3600 described species worldwide, with probably another thousand or more awaiting description. As such, they represent one of the larger radiations among the ditrysian Lepidoptera. Yet the taxonomy of the group remains uncertain, their life histories largely unknown and at least 25% of their biodiversity undescribed. We inferred the phylogeny of North American lithosiines using molecular sequence data from three genes that have been show to phylogenetically informative in Lepidoptera: cytochrome oxidase subunit I (COI), ribosomal protein S5 (RPS5) and the large subunit 28S ribosomal DNA (28S). Results provide new insights as to how representatives of the 16 genera and 63 species that comprise the lithosiine fauna of the US are related. We also investigated a unique relationship that has evolved between ants and the caterpillars of lithosiines, the latter of which feed on lichens growing on rocks and tree trunks, habitats where ants are primary predators. Finally, we describe a new species of Hypoprepia, H. lampyroides, from the mountains of central Arizona. This is the fifth member of the genus Hypoprepia, all of which occur in the US.

CHAPTER 1. INTRODUCTION

1.1 Lichen Tiger Moths (tribe Lithosiini)

Tiger Moths (Erebidae, Arctiinae--with about 11000 species worldwide) are recognizable to most people because of their relatively large size and gaudy coloration. Their larval stages, popularly referred to as woolly bears, are also familiar to most. While nearly two-thirds of the subfamily is comprised of these familiar forms (tribe Arctiini contains about 7000 species worldwide), the much more diminutive and secretive Lichen Tiger Moths, tribe Lithosiini, make up about one third of the group (about 3600 species).

The Lichen Tiger Moths are so-named because those few with known life histories specialize in feeding on lichens (Comstock and Henne 1967, McCabe 1981, Wagner 2005, Wagner et al. 2008), symbiotic organisms comprised of algae, fungi and a cyanobacterium that are found throughout most of the world, as are the moths. As with other types of Tiger Moths, the diversity of the Lichen Tiger Moths is greatest in tropical regions, with diversity gradually dropping as one moves north and south from the equator (Holloway 2002, Wagner et al. 2008). The lithosiine fauna of the US consists of at least 63 species in 16 genera (Covell 2005, Schmidt and Opler 2008, Powell and Opler 2009), with nearly half of these occurring in the Southwest. This is likely due, in part, to the high diversity of lichens found here, representing some 1000 species, about 20% of the lichen fauna of North America (CNALH website 2020). The Southwest is also known as a biological blend zone, where typically northern or Rocky Mountain taxa meet more southern neotropical elements (Warshall 1995). Lichen moths may be small, but they are often brilliantly colored, a situation in nature which usually indicates them to be poisonous or harmful in some way. The fascinating chemical ecology of these aposematically-colored moths is just beginning to be studied, and preliminary data demonstrate that the larvae and adults of some sequester toxic polyphenolic compounds derived from the lichen on which they feed (Hesbacher et al. 1995, Scott et al. 2014, Scott-Chialvo et al. 2018). This and a few other tantalizing details known of their life histories suggests the group possesses a wealth of biological secrets, including potential myrmecophily (Ayre 1958, Komatsu and Itino 2014), mimicry rings and acoustic communication (Connor 2009, Nakano et al. 2003, Corcoran et al. 2013).

1.1.1 Taxonomy and Systematics

The taxonomic status of the Lichen Tiger Moths has been reconsidered many times, from family to subfamily and finally tribe, this latest taxonomic placement being determined using molecular phylogenetics (Zahiri et al. 2011, 2012, 2013). While the monophyly of the group as a whole is well supported, the major lineages within this large group are still being investigated, and it was only twenty years ago that systematists tried to place them into subtribal groups based on morphology (Bendib and Minet 1998, 2000, Jacobson and Weller 2001). More recently, molecular systematics was used to evaluate if these proposed subtribes represent natural groups (Scott et al. 2014, Zenker at al. 2017, Scott-Chialvo et al. 2018).

Morphology has been the cornerstone of taxonomy for more than two centuries, organisms bearing characteristics inferred as being both shared and derived being grouped together. The hierarchical ordering of kingdom, phylum, class, order, family, genus and species being based on the work of Linnaeus and others. Under consideration here is the order Lepidoptera, family Erebidae. Between family and genus are the subcategories of subfamily, tribe and subtribe. This is the area of our focus on Lichen Tiger Moth taxonomy.

The field of molecular phylogenetics has been evolving rapidly in the last few decades, and involves using DNA sequence data to infer organismal relationships. By using phylogenetic inference methods that incorporate models of nucleotide evolution, evolutionary relationships can be inferred independent of the morphological features of the organisms (Bybee et al. 2010).

The order Lepidoptera as a whole contains well over 100,000 species. By far the largest group of lepidopterans are the Noctuoidea, a huge radiation of moths containing nearly 40,000 species. About half of these are placed in the family Erebidae, and more than half of the Erebidae are the subfamily Arctiinae, with about 11,000 species. For many decades the Noctuiodea has been reorganized by taxonomists focusing on different suites of morphological characters. It was not until very recently that the use of molecular phylogenetics has allowed us to see through the dizzying array of morphological character loss and gain amongst this group, and allowed for a more stable taxonomic framework (Zahiri et al. 2011, 2012, 2013). Using this new framework, we are now able to refine our assessment of how genera and species are related, with an ultimate goal of more fully understanding how their magnificent diversity evolved.

1.1.2 Myrmecophily and other ant associations

Ants evolved rather early on during the Cretaceous period and have successfully radiated as a group across the planet (Moreau et al. 2006). The resource-rich nests of hard-working ants offer an attractive target for other organisms to exploit. Those which have evolved obligate dependency on ants are called myrmecophiles. Insect examples of myrmecophiles include ground beetles of the tribe Paussini (Maurizi et al. 2012; Moore and Robertson 2014) and certain butterflies belonging to the family Lycaenidae. In the latter family, there exists a spectrum of myrmecophily, the larvae of most lycaenids having special glands that the produce secretions that compel the ants to protect them like cattle (Devries 1992, Pierce et al. 2002). Others, such as the European genus Maculinia, chemically compel the ants to collect the larvae and bring them into their nests as if they are their own, at which point the caterpillars feed upon the ant larvae to complete their development into a butterfly (Thomas and Settele 2004, Thomas et al 2010).

Published observations suggest the caterpillars of lithosiines have some sort of relationship with ants, with some even calling them myrmecophiles (Nakano and Itino 2014). These observations makes some sense given the fact lichens grow on surfaces where ants are the most likely predators. Perhaps because of their secretive nature, there has yet to be any publication which actually proves the caterpillars of lithosiines are actually mymecophiles, just scattered accounts of behaviors that suggest they might be.

1.1.3 Identifying new species and protecting biodiversity

We as humans are having an enormous impact on other species on the planet (Sahney et al. 2010, Pimm et al. 2014, Vignieri 2014, Ceballos et al. 2015). It is therefore critical that we know what those species are and where they occur since we cannot protect what we do not know exists. Identifying and publishing new species is one way we can be both scientists and better stewards of our planet, by calling attention to the diversity of life around us.

Describing a new species involves comparison of the morphology to other known species, and recently can include molecular characters as well. In the case of closely related species that are morphologically very similar, analysis of molecular sequence data can often detect differences that morphology alone cannot. Such taxa are often referred to as cryptic or hidden species (Bickford et al. 2007). In this way using molecular data is a powerful tool to that allows us to better understand and protect genetic biodiversity as well as diversity among species that are more obviously different.

Lichen Tiger Moths are suspected of having many undescribed taxa among their taxonomic ranks (Common 1990, Wagner et al. 2008, Lewis 2014). Because of their intimate relationship with lichens, some of which have shown sensitivity to pollution and climate change, there is also interest in using lithosiines as models for studying these phenomena (Lawrey 1993, Simonson 1996, Teskey 2001, Government of Canada 2008). This not only requires knowing what lithosiines are present in an ecosystem, but also what species of lichens they depend upon in their larval stages.

1.2 Dissertation Format

The format of this dissertation follows the guidelines found in article IV of the University of Arizona Dissertation Formatting Guide, which outlines details for manuscript-based dissertations. The following chapter contains brief descriptions of the manuscripts prepared during the current research. Each manuscript can be found in the appendix of this dissertation. One manuscript (2.3) has already been published, while the remaining manuscripts will be submitted in the near future for publication in peer-reviewed journals.

CHAPTER 2. THE CURRENT STUDY

2.1 Title: A molecular phylogenetic assessment of North American members of the tribe Lithosiini (Lepidoptera: Erebidae: Arctiinae) yields some surprising subtribal relationships

Authors: John Palting and Wendy Moore

The few published studies evaluating the subtribal alliances in the Lithosiini have produced inconsistent results, suggesting further work was warranted. Here we used three genes (COI. 28S and RPS5) that have been found to be phylogenetically informative in Lepidoptera (Wahlberg et al., 2003, 2009, 2010, Wahlberg and Wheat 2008, Mutanen et al. 2010, Zahiri et al. 2011, 2013) to examine the relationships between the US lithosiine fauna. The 63 US species include representative members of three of the seven subtribe proposed for the group: Cistheniina, Eudesmiina and Lithosiina. While the Lithosiina are Holarctic in their distribution, the other two subtribes are neotropical, with the US representing the northern limits of their distribution. Our analysis suggests there exists an unnamed subtribe of lithosiines that includes the genus Clemensia (3 US species, previously placed in the Cistheniina) and some South American relatives, and that the remaining US taxa all fall within the Cistheniina and Lithosiina. We also find the proposed subtribe Eudesmiina may not be a natural group, instead being part of the Cistheniina. We place several genera not previously placed into subtribes and report the genus Gardinia, previously considered a member of the Lithosiina, is actually a member of the Cistheniina.

2.2 Title: Observations of the larval stages of lichen moths (Lepidoptera: Erebidae: Arctiinae: Lithosiini) and their interaction with ants (Formicidae)

Author: John Palting

With so little known about the biology of lithosiines, there are a few publications which suggest the larvae of some might be myrmecophilous, an intriguing possibility that would be a first for Lepidoptera outside the butterfly family Lycaenidae, many of which have evolved intimate relationships with ants. We investigated the potential for myrmecophily in the genus Crambida as well as any potential ant association of other genera where larval stages could be obtained. We find no evidence that Crambidia or any other US lithosiines are myrmecophilous, but rather that the larvae of all species evaluated possess some sort of chemical protection that repulses the ants and prevents the larvae from

being preyed upon by them. This unique ability allows the larvae to move about on the ground and on tree trunks feeding on the lichens with impunity from ant attack. As several authors have demonstrated the larvae of lithosiines sequester toxic lichen compounds in their tissues, we used histology to investigate potential cuticular structures which may be correlated with dispersal of these compounds.

2.3 Title: A new species of Hypoprepia from the mountains of central Arizona (Lepidoptera: Erebidae: Arctiinae: Lithosiini)

Authors: John Palting, Douglas Ferguson, Wendy Moore

Zookeys. 2018; (788): 19–38. Published online 2018 Oct 8. doi: 10.3897/zookeys.788.26885 During the course of this study we identified a new species in the lithosiine genus Hypoprepia that occurs in the mountains of central Arizona, as well as in two localities of the Sierra Madre Occidental of Mexico. We described and published this new taxa as Hypoprepia lampyroides Palting and Ferguson 2018, the species name reflecting the remarkable mimicry of this moth with toxic beetles of the family Lampyridae that co-occur with it. Many lithosiines seem to be part of mimicry rings with other moths and beetles, but this is the first US taxa found to be a firefly mimic, and the fifth member of the genus Hypoprepia, all of which occur in the US.

REFERENCES.

Ayre, G.L. (1958) Notes on insects found in or near nests of Formica subnitens Creighton (Hymenoptera: Formicidae) in British Columbia . Insectes Sociaux 5(1): 1- 7.

Bendib, A. and J. Minet (1999) Lithosiinae main lineages. Annales de Societe Entomologique de France 35(3-4): 241-263. Bendib, A. and J. Minet (1998) Female pheromone glands in Arctiidae (Lepidoptera): evolution and phylogenetic significance. C.R.Acad.Sci. Paris 321:1007-1014.

Bickford, D., D.J. Lohman, N.S. Sodhi, P.K.L. Ng, R. Meier, K. Winker, K.K. Ingram, I. Das. 2007. Cryptic species as a window on diversity and conservation. Trends in Ecology & Evolution 22(3): 148–155.

Bybee, S.M., J.M. Zaspel, K.A. Beuke, C.H. Scott, B. W. Smith and M.A. Branham (2010) Are molecular data supplanting morphological data in modern phylogenetic studies? Systematic Entomology 35: 2-5.

Ceballos, G., P.R. Ehrlich, A.D. Barnosky, A. García, R.M. Pringle and T.M. Palmer (2015). Accelerated modern human–induced species losses: Entering the sixth mass extinction. Science Advances. 1 (5): e1400253.

Comstock, J.A. and C. Henne (1967) Early stages of Lycomorpha regulus Grinnel with notes on the imago (Lepidoptera: Amatidae). Journal of Research on the Lepidoptera 6(4): 275-280.

Common, I.B.F. (1990) Moths of Australia, Melbourne University Press, Carlton.

Conner, W.L. ed. (2009) Tiger Moths and Woolly Bears: Behavior, Ecology and Evolution of the Arctiidae. Oxford University Press.

Consortium of North American Lichen Herbaria (2020) https://lichenportal.org/cnalh/index.php

Corcoran, A.J., R.D. Wagner and W.E. Connor (2013) Optimal predator risk assessment by the sonar-jamming arctiine moth Bertholdia trigona. PlosOne:1-13.

Covell, C.V.Jr. (2005) A Field Guide to the Moths of Eastern North America, 2nd ed. , Virginia Museum of Natural History, Martinsville, VA.

DeVries, P. J. (1992) Singing Caterpillars, Ants and Symbiosis. Scientific American 267: 76

Government of Canada (2008) Biobasics: the science and the issues. Bioindicators. http://biobasics.gc.ca/english/View.asp?x+740.

Hesbacher, S., I Giez, G. Embacher, K Fiedler, W. Max, A. Trawoger, R. Turk, O. L. Lange and P. Proksch (1995) Sequestration of lichen compounds by lichen-feeding members of the Arctiidae (Lepidoptera). Journal of Chemical Ecology 21(12): 2079- 2089.

Holloway, J.D. (2002) The moths of Borneo: family Arctiidae, subfamily Lithosiinae. The Malayan Journal 55:279-469.

Komatsu, T. and T. Itino (2014) Moth caterpillar solicits homopteran honeydew. Scientific Reports 4: 1-3.

Lawrey, J.D. (1993) Lichens as monitors of pollution elements at permanent sites in Maryland and Virginia. The Bryologist 96: 339-341.

Lewis, J.A. (2014) A new species of Crambidia (Lepidoptera: Noctuidae: Arctiinae: Lithosiini) from the Eastern United States. Proc. Entomol. Soc. Wash. 116(4):394- 401.

Maurizi, E., S. Fattorini, W. Moore and A. Di Giulio (2012) Behavior of Paussus favieri (Coleoptera, Carabidae, Paussini): a myrmecophilous beetle associated with Pheidole pallidula (Hymenoptera, Formicidae). Psyche: 1-9.

McCabe, T.L. (1981) Clemensia albata, an algal feeding arctiid. Journal of the Lepidopterists Society 35(1): 34-40.

Moore, W. and J. Robertson (2014) Explosive adaptive radiation and extreme phenotypic diversity within the ant-nest beetles. Current Biology 24: 2435-2439.

Moreau, C.S., C.D. Bell, R. Vila, S.B. Archibald and N. E. Pierce (2006). Phylogeny of the ants: diversification in the age of angiosperms. Science 312 (5770): 101–4.

Mutanen, M., N. Wahlberg and L. Kaila (2010) Comprehensive gene and taxon coverage elucidates radiation patterns in moths and butterflies. Proceedings of the Royal Society B, 277, 2839-2848.

Nakano, R. T. Takanashi, A. Surlykke, N. Skals and Y. Ishikawa (2003) Evolution of deceptive and true courtship songs in moths. Scientific Reports (3): 1-8.

Pierce, Naomi E., M. F. Braby A. Heath, D.J. Lohman, J. Mathew, D. B. Rand and M.A. Travassos (2002) Ecology and Evolution of Ant Association in the Lycaenidae (Lepidoptera). Annual Review of Entomology 47 (1): 259-267.

Pimm, S.L., C. N. Jenkins, R. Abell, T. M. Brooks, J. L. Gittleman, L. N. Joppa, P. H. Raven, C.M. Roberts and J.O. Sexton (2014). The biodiversity of species and their rates of extinction, distribution, and protection. Science. 344 (6187)

Powell, J. A. and P.A. Opler (2009) Moths of Western North America. University of California Press.

Sahney, S., M. J. Benton and P.A. Ferry (2010) Links between global taxonomic diversity, ecological diversity and the expansion of vertebrates on land. Biology Letters. 6(4): 544–547.

Schmidt, B.C. and P.A. Opler (2008) Revised checklist of the tiger moths of the Continental United States and Canada. Zootaxa 1677: 1-23.

Scott, C. H., J.M. Zaspel, P. Chialvo and S. J. Weller (2014) A preliminary molecular phylogenetic assessment of the lichen moths (Lepidoptera: Erebidae: Arctiinae: Lithosiini) with comments on palatability and chemical sequestration. Systematic Entomology 39: 286-303.

Scott Chialvo, C.H., P. Chailvo, J.D. Holland, T. J. Anderson. J. W. Breinholt, A.Y. Kawahara, X. Zhou, S. Liu and J.M. Zaspel (2018) A phylogenomic analysis of lichen-feeding tiger moths uncovers evolutionary origins of host chemical sequestration. Molecular Phylogenetics and Evolution 121:23-34.

Simonson, S. (1996) Lichens and Lichen-feeding moths as bioindicators of air pollution in the Rocky Mountain front range. www.interlibrary.narod.ru/GenCat/GenCat.Scient/400000011.htm.

Tesky, D. (2001) Biological Indicators. http://www.suite101.com/article.cfm/ecology/57858. Thomas, J.A. and J. Settele (2004) Evolutionary Biology: butterfly mimics of ants. Nature 432 (7015): 283-4.

Thomas, J., K. Schönrogge, S. Bonelli, F. Barbero and E. Balletto (2010). Corruption of ant acoustical signals by mimetic social parasites. Communicative and Integrative Biology. 3 (2): 169–171.

Vignieri, Sacha (2014). Vanishing fauna. Science 345 (6195): 392–395.

Wagner, D.L (2005) Caterpillars of eastern North America: a guide to identification and natural history. Princeton University Press, Princeton, N.J.

Wagner, D. L., J. Rota and T. L. McCabe (2008) Larva of Abablemma (Noctuidae) with notes on algivory and lichenivory in macrolepidoptera. Annals of the Entomological Society of America 101(1): 1-13.

Wahlberg, N. and C.W. Wheat (2008) Genomic outposts serve the phylogenetic pioneers: designing novel nuclear markers for genomic DNA extractions of Lepidoptera. Systematic Biology, 57, 231-242.

Wahlberg, N. E. Weingartner and S. Nylin (2003) Towards a better understanding of the higher systematics of Nymphalidae. Molecular Phylogenetics and Evolution, 28, 473-484.

Wahlberg, N., J. Leneveu, U. Kodandaramajab, C. Pena, S. Nylin, A.V.L. Frietas and A.V.Z. Brower (2009) Nymphalid butterflies diversify following near demise at the Cretaceous/Tertiary boundary. Proceedings of the Royal Society B, 276, 4295-4302.

Wahlberg, N., N. Snall, J. Vidalepp, K. Ruohomaki and T. Tammaru (2010) The evolution of female flightlessness among Enominae of the Holarctic forest zone (Lepidoptera, Geometridae). Molecular Phylogenetics and Evolution, 55, 929-938.

Warshall P. (1995) The Madrean Sky Island Archipelago: A planetary overview. In: DeBano, LH, Ffolliot PF, Ortega-Rubio A, Gottfried G, Hamre RH, Edmister CB (Eds) Biodiversity and management of the Madrean Archipelago: The Sky Islands of Southwestern United States and Northwestern Mexico. 1994 Sept. 19–23; Tucson, AZ. Tech. Rep. RM-GTR-264. Fort Collins, CO: US Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experimental Station.

Zahiri, R., I.J. Kitching, J. D. Lafontaine, L.K. Mutanene, J.D. Holloway and N. Wahlberg (2011) A new molecular phylogeny offers hope for a stable family level classification of the Noctuoidea (Lepidoptera). Zoologica Scripta, 40, 158-173.

Zahiri, R., J.D. Holloway, I.J. Kitching, D. Lafontaine, M. Mutanen and N. Walberg (2012) Molecular Phylogenetics of the Erebidae (Lepidoptera, Noctuoidea). Systematic Entomology 37:102-124.

Zahiri, R., D. Lafontaine, C. Schmidt, J.D. Holloway, I. J. Kitching, M. Mutanen and N. Wahlberg (2013) Relationships among the basal lineages of Noctuidae (Lepidoptera: Noctuoidea) based on eight gene regions. Zoologica Scripta 42(5): 488- 507.

Zenker, M.M., N. Wahlberg, G. Brehm, J.A. Teston, L. Przybylowicz, M. R. Pie and A.V.L. Freitas (2016) Systematics and origin of moths in the subfamily Arctiidae (Lepidoptera, Erebidae) in the Neotropical region. Zoologica Scripta 46(3): 348-362.

APPENDIX A.

Molecular phylogeny of the Lichen Tiger Moths (Lepidoptera: Erebidae: Arctiinae: Lithosiini) of the Western Hemisphere

John D. Palting1 and Wendy Moore2

Graduate Interdisciplinary Program in Entomology and Insect Science, University of Arizona, Forbes 410, Tucson, AZ, 85721-0036, USA

Email: [email protected]

Formatted for Zookeys.

Formatted for Zookeys

Molecular phylogeny of the Lichen Tiger Moths (Lepidoptera: Erebidae: Arctiinae: Lithosiini) of the Western Hemisphere

John D. Palting x 1,2 and Wendy Moore 2

1. Graduate Interdisciplinary Program in Entomology and Insect Science, University of Arizona, Tucson, Arizona, 85721-0036, USA, 2. Department of Entomology, University of Arizona, Forbes 410, Tucson, Arizona, 85721-0036, USA,

X Corresponding author: John Douglas Palting ([email protected])

Abstract

Few molecular-based studies have tested the monophyly of the Lithosiini subtribes proposed by Bendib and Minet (2000), and those have been limited by meager taxon sampling. Several studies have suggested some subtribes are not monophyletic as they are currently defined. We conducted a molecular phylogenetic study of representatives of the North American lithosiine fauna which are currently classified within the subtribes Acsalina, Lithosiina, Cisthenina and Eudesmiina, the latter having never been included in a molecular-based analysis before. Based on analyses of cytochrome oxidase subunit I (COI), ribosomal protein S5 (RPS5) and the large subunit 28S ribosomal DNA (28S), we assign some of the North American genera to subtribe for the first time and re-assign others contrary to previous placements. Then, we discuss the morphological characters that Bendib and Minet (2000) proposed to define subtribes and re-consider them in the context of our inferred phylogeny. We report high support for a monophyletic Lithosiina+ Agylla + Inopsis + Gnamptonychia, three genera being unplaced or of uncertain placement (Agylla) by Bendib and Minet (2000). We remove Gardinia from the subtribe Lithosiina and place it in the Cisthenina, along with Eudesmia, formerly placed in its own subtribe, the Eudesmiina. Two other genera, Bruceia and Ptychoglene, not previously assigned to a subtribe are found to be members of the Cisthenina. We remove Clemensia from the Cisthenina and report it, along with the neotropical Pronola, as part of undefined clade. After these changes, our phylogeny shows strong support for the monophyly of Cisthenina + Gardinia + Eudesmia + Bruceia + Ptychoglene. We find Acsala anomala occurs on a long branch by itself, confirming the uniqueness of this species and its placement in a monotypic subtribe.

Keywords: Arctiinae, Lithosiini, Lichen Tiger Moths, Cisthenina, Lithosiina, Eudesmiina, Acsalina, 17

Introduction:

While a vast majority of caterpillars are phytophagous, the larvae of the tribe Lithosiini, commonly known as the Lichen Tiger Moths, feed on lichens, a tough and often toxic mutualism of algae, fungus and cyanobacterium. The ability to utilize these peculiar, but widely available organisms as food is likely behind the successful radiation of lithosiines across the globe. The tribe Lithosiini currently consists of some 3600 described species of small to medium, often brightly colored moths of nearly worldwide distribution. Not only do lichens offer a food source generally free from competition, they possess a chemical arsenal that can be exploited. It has been demonstrated by many researchers that the larvae of lithosiines possess metabolic pathways to deal with lichen toxins and actually sequester modifications of these for their own protection in all life stages (Hesbacher et al. 1995; Wagner et al. 2008; Conner 2009; Scott et al. 2014; Anderson et al. 2017; Scott-Chailvo et al. 2018). Adults of most lithosiines exhibit aposematic coloration, however the larvae are dull- colored, secretive nocturnal feeders (Wagner 2005, Conner 2009). Hiding by day and feeding at night likely is an adaptation to avoid bird predation, as lichens offer little in the way of protective cover from visual predators. Because they are nocturnally active, lithosiine larvae are seldom encountered and poorly known, in contrast to the often conspicuous feeding of their well-known woolly bear relatives, the Arctiini (Fig. 1).

While no adult apomorphy uniting lithosiines has been identified to date, the larval stages provide two: a presence of a mandibular molar and the chaetotaxy of the labrum, specifically the unique arrangement of the labral setae M1 and M2 (Bendib and Minet 2000). We confirm the presence of both these features across the larvae reared as part of this study, many of which were previously unknown. The modified mandibular molars of some of these are shown in Fig. 2. This peculiar modification of the mandible, presumably for grinding the tough lichen thallus for digestion, was noted rather early on. The apical ends of the forth and sometimes the third scissoral teeth of the mandible have been modified from the blade-like structure seen in arctiines and noctuids (Gilligan and Passoa 1985) to a flattened grinding structure in lithosiines (Gardner, 1943; Issiki et al., 1965; Garcia-Barros, 1985; McCabe, 1981; Lafontaine et. al., 1982; Rawlins, 1984; Habeck, 1987, Bendib and Minet 2000). Despite being one of the larger radiations among the ditrysian Lepidoptera, few attempts had been made to define relationships among the numerous genera of lithosiines until Bendib and Minet (2000) proposed seven lineages and placed many of the genera into them. Originally considered part of the family Arctiidae, changes in taxon ranks and names were required when this family was subordinated under the Erebidae (Zahiri et al. 2012). The original Bendib and Minet classification was published prior to this change, thus their proposed tribes are now subtribes of the tribe Lithosiini. Of these proposed seven subtribes, only four that contain New World species are considered in the present study; the Lithosiina, Cisthenina, Eudesmiina and Acsalina. The other subtribes are the Phryganopterygiina (Madagascar), Nudariina (Old World) and Eudrosiina (Palearctic). 18

While these putative clades are not well defined by morphological apomorphies, Bendib and Minet (2000) note that there is “obvious morphological diversity in the imagines,” including size, color and resting posture that help support their subtribal classifications. The structure of the second abdominal sternite in the adults (Fig. 2) is described in detail and proposed as one possible apomorphy, the Cistheniina and Eudesmiina possessing the long, forked apodemes, while the Lithosiina (and arctiines) have shortened apodemes.

Among the North American fauna, adults of the Lithosiina are often on the larger side with elongate wings and a distinct adult resting posture, with the forewing covering most of the distal half of the other forewing. Most assume an almost flat profile when resting, while some genera like Crambidia Packard, 1864, rest in a cigar-like fashion, with the wings curved around the body. Adults of the Lithosiina are usually more somberly colored than many of the other tribes, with many members that are mostly white, grey or black. Bendib and Minet intended to publish a second paper focusing just on the Lithosiina, stating the subtribe is “extremely comprehensive and deserves, in our opinion, to be split into several precisely defined (sub)tribes.” Larvae of the Lithosiina have prominent verrucae, with an unusual arrangement of meso and metathoracical verrucae D and SD2. Genera assigned to this subtribe that occur in the US are Crambidia, with eleven species and Eilema Hubner, 1819, with one species. Agylla Walker, 1854 was initially assigned to this subtribe by Bendib and Minet (1998), but later they reconsidered and gave it uncertain status (Bendib and Minet 2000).

The Cisthenina are generally small moths and are often black banded with contrasting colors such as red or orange. The adults tend to hold their wings “roof-like” over their back without the wings overlapping. A suggested apomorphy is the lack of verrucae in the larvae (Bendib and Minet 2000), with the setae generally being short and sparse. US genera included by Bendib and Minet (2000) are the type genus Cisthene Walker, 1854 (with 20 U.S. sp.), Hypoprepia Hubner, 1831 (5 U.S. sp.), Lycomorpha Harris, 1839 (5 U.S. sp.), Lycomorphodes Hampson, 1900 (1 U.S. sp.) and Rhabdatomis Dyar, 1907 (1 U.S. sp.). The distinctive genus Clemensia Packard, 1844 (3 U.S. sp.) was also tentatively placed under the Cistheniina, although the authors found them different enough from other cisthenines that they suggested they be placed in a separate subtribe, the Clemensiiti. They listed three apomorphies for this designation: presence of a pair of metascutal membranous areas, sternum A2 with curved, moveable anterolateral processes and the abdomen of the female having corethrogyne. The subtribe Eudesmiina are brightly colored like the Cisthenina, often with alternating black and orange bands, but they are larger and their wing surface area a bit broader. Like Cisthenina, the wings are held over their back like a tent, without overlapping. The larval stages, known at least for the genus Eudesmia Hubner, 1823, are rather shortened like larvae of Megalopygidae, with verrucae bearing tufts of long, soft setae. Two species in this subtribe, both of the genus Eudesmia, occur in the U.S.

The aptly named Acsala anomala Benjamin, 1935 is the only known member of the subtribe Acsalina. The life history of this enigmatic arctic lithosiine has been described by Lafontaine et al. (1982). Originally considered a lymatriid (tussock moth), it was later recognized as a lithosiine from the larvae, which feed on various lichens, taking several years to mature with the cold temperatures and short daylength of the high arctic. Bendib and Minet (2000) placed this species it in its own subtribe based on many peculiar autapomorphies, including the hindwing venation, the translucent scales on the wings and flightless females.

Only a few molecular-based phylogenetic studies have tested the monophyly of the subtribes proposed by Bendib and Minet (2000) and those few have been limited by meager taxon sampling (Scott et al. 2014, Scott-Chailvo et al. 2018, Zenker et al. 2017). Even so, results of these studies suggest that some subtribes do not represent monophyletic groups. Thus, the present subtribal classification within the Lithosiini is tentative and will likely continue to be refined as more research is performed. Here we add to our knowledge of lithosiine evolution by conducting a molecular phylogenetic study which include representatives of the North American lithosiine fauna, which are classified within the subtribes Acsalina, Lithosiina, CIsthenina and Eudesmiina (representatives of the latter have never been included in a molecular- based analysis before).

Materials and Methods:

Taxon sampling

A total of 181 lithosiine specimens were collected from across North America (or obtained from Europe as comparators) from 2010-2018, with 121 yielding useful DNA (66%). All taxa used in this study are listed in Supplementary Table 1. Sequences for additional taxa were acquired from GenBank and the Barcode of Life Database. GenBank accession numbers for all sequences used in this study are listed in Supplementary Table 1. DNA voucher specimens used for this study have all been deposited in the University of Arizona Insect Collection (accession numbers UAIC 3250-3299, 3500-3599, 3900-3914 and 4162-4182).

DNA extraction, amplification, and sequencing

Total genomic DNA was extracted from right middle leg or the abdomen of each specimen the Qiagen® DNeasy Blood and Tissue Kit (Qiagen, Valencia, CA), according to manufacturer suggested protocol. Total genomic DNA was stored in buffer at -80°C. All DNA extractions used in this study are listed in Supplementary Table 1.

Three genes that have been found to be phylogenetically informative in Lepidoptera were amplified: the barcoding region of the mitochondrial gene cytochrome oxidase subunit 1 (COI) using primers LCO1490 and HCO2198 (Hebert et al. 2003), nuclear marker RPS5 (Ribosomal Protein S5) and a ribosome marker 28S (nuclear large subunit rRNA 28S D2 loop), using primers and PCR protocols per Scott et al. (2014)

PCR products were cleaned, quantified, normalized and sequenced in both directions at the University of Arizona’s Genomic and Technology Core Facility using a 3730 or 3730XL Applied Biosystems automatic sequencer. Chromatograms were assembled and initial base calls were made for each gene with Phred (Green and Ewing 2002) and Phrap (Green 1999) as orchestrated by Mesquite Ver. 3.6 (Maddison and Maddison 2018) and Chromaseq vers. 1.5 (Maddison and Maddison 2018). Final base calls were made in Mesquite and ambiguous bases were designated by a standard ambiguity code. Resulting sequences were deposited in GenBank (Table 1).

Alignment and phylogenetic analyses

Single gene matrices were aligned using default settings in MAFTT v. 7 as implemented in Mesquite. Aligned matrices were concatenated in Mesquite.

Maximum likelihood analyses were conducted on each gene individually using IQ- TREE version 1.6.10 (Nguyen et al. 2015), as orchestrated by the CIPRES Science Gateway (Miller et al 2010). The ModelFinder feature within IQ-TREE (Kalyaanamoorthy et al. 2017) was used to find the optimal character evolution models. The MFP model option was used for 28S, and the TESTMERGE option for the protein-coding genes. The TESTMERGE option sought the optimal partition of sites, beginning with the codon positions in different parts. In addition, analyses with the concatenated data were conducted, with the TESTMERGE option also being used, beginning with each codon position for each gene as a separate part (thus, the analysis began allowing for up to 7 parts (three for each of the two protein-coding fragments, and one for 28S). One hundred searches were conducted for the maximum-likelihood tree for each matrix; for bootstrap analyses, 500 replicates were used.

Results:

The maximum likelihood tree and bootstrap support values resulting from analysis of the concatentated dataset is in Fig. 3. Results of single gene analyses are presented as Supplement Fig. 1-3. One of the surprises is that our study challenges the separation of the Eudesmiina, represented here by Eudesmia arida Skinner, 1906 and Eudesmia menea Drury, 1782 as separate subtribe of the lithosiines. Despite obvious differences in the larval morphology from other Cisthenina, we find strong support that Eudesmia is a clade of the Cisthenina and not a separate subtribe of lithosiines. This creates a problem in the definition of the Cisthenina, the proposed apomorphy being that the larval stages lack verrucae (Bendib and Minet 2000). During the course of our research we reared Eudesmia arida and found the larvae indeed possess verrucae (Fig. 5B). While many larvae of Cisthenina (Cisthene, Hypoprepia, for example) lack verrucae, the placement of Eudesmia in the Cisthenina would require a new larval synapomorphic definition. In addition to the presence of verrucae, the larval stages of E. arida possess exceedingly long, soft setae, unlike the short, stiff setae of Cisthene, Hypoprepia etc. As observed by Bendib and Minet (2000), the second abdominal sternite of the imago has long apodemes (Fig. 4C), a feature shared with Cisthenina (and the Old World Nudariina). The sternite of Cisthene barnesii Dyar, 1904 is shown for comparison (Fig. 4A).

The genus Gardinia, represented here by a single American species found in southeastern Arizona, Gardinia anopla Hering, 1925, offered another surprise in our analysis. This neotropical genus was placed among the Lithosiina by both Bendib and Minet (2000) and Scott and Branham (2012) in a phylogeny based on morphological characters. While it resembles many of them in both size and color, our analysis indicates it is not a member of this subtribe. G. anopla is a relatively large moth (with an average wingspan of 4.5 cm), in fact, the largest lithosiine among our American fauna. Our molecular data (along with additional sequences from GenBank) place this species in the Cisthenina, a true giant among the group (most other members of which have a wingspan 2 cm or less). In Scott et al. (2014) the genus was placed among a polyphyletic Cisthenina in what they called “clade A,” which included putative members of the Cistheniina as well as Nudariina. Zenker et al. (2017) included a different species of Gardinia, G. paradoxa Hering, 1925, in their molecular phylogeny of neotropical tiger moths, and it also placed among other Cisthenina rather than among the Lithosiina.

In support of this new placement, we report the second abdominal sternite of Gardinia adults (Fig. 2B) exhibits the long apodemes associated with Cisthenina, Eudesmiina and Nudariina, and not the reduced apodemes associated with Lithosiina and other arctines (Bendib and Minet 2000). In addition, the living adults of G. anopla hold their wings “tent-like” over their backs and not flat like most Lithosiina. Other aspects of the biology of Gardinia lend support to its placement among the Cisthenina. One is that the genus Gardinia is neotropical, with G. anopla being the only member making 22 it into the US. While the subtribe Lithosiina is primarily a northern group, occurring in NA, Europe and Asia, the Cisthenina is a strictly new world with the center of diversity in the Neotropics proper. Imagos of two other new world genera (from GenBank), Balbura Walker 1854 and Chrysochlorosia Hampson, 1900, bear a resemblance to Gardinia and also fall within the Cisthenina. Another aspect of this placement that correlates is the use of acoustic aposematism by Gardinia, which is common among the Cisthenina and not known among the Lithosiina. When captured alive, Gardinia anopla produces surprisingly loud clicks, audible to most human ears because of its unusual size. Among the genera Cisthene and Hypoprepia, researchers have found the adult moths produce clicks in response to bat echolocation (Acharya and Fenton 1992, Conner 2009), thus one can surmise that the loud clicks of Gardinia are used in the same manner, to warn bats of their distastefulness. The larvae of Gardinia (Fig. 5A) possess verrucae bearing sprays of short, stiff setae. While the overall appearance and coloring is reminiscent of Hypoprepia, the presence of verrucae presents another challenge to this being a synapomorphy of Cisthenina.

The uniquely American genus Bruceia posed some challenges for Scott et al. (2014), the two members of the genus being polyphyletic in their original molecular analysis. In a later paper (Scott-Chialvo et al. 2018) they place both in a group they call Cisthenina NW (finding the Cisthenina polyphyletic), while in an earlier paper (Scott and Branham 2012) they placed it among the Cisthenina based on morphological characters. In our analysis both B. hubbardi and B. pulverina place squarely among a monophyletic Cisthenina. The overall aspect of the imagos certainly agrees with this, although the somber coloring is quite different from most other Cisthenina. The larval stage of B. hubbardi, reared as part of this study is rather Cisthene-like, lacking verrucae, although more flattened, with a distinctive lateral bulge along the thoracic and extending into the first two abdominal segments, giving it the appearance of a miniature Tolype Hubner, 1820 (Lasiocampidae). While Bendib and Minet (2000) did not specifically assign Bruceia to a subtribe, we assign it here to the Cisthenina. Our analysis also suggests the existence of an additional undescribed Mexican Bruceia, much smaller than either B. pulverina and B. hubbardi occurring near the US border in the Sierra Del Tigre of Sonora, Mexico. Being a rather drab genus compared to other Cisthenina, additional undescribed taxa likely have been overlooked in Mexico.

Similarly, Bendib and Minet did not specifically assign the neotropical genus Ptychoglene to a subtribe. Our molecular data for P. concinnea H. Edwards, 1886 places it in the Cisthenina, closely allied to Hypoprepia. This agrees with Scott and Branham (2012), Scott et al. (2014) and Scott-Chialvo et al. (2018), which included the same species. Like most of the Cisthenina, the larval stages lack verrucae. A recent rearing by R. Nagle of P. phrada Druce, 1889 also demonstrated that, like Hypoprepia, the immature stages overwinter as fully-grown larvae. This presumably gives them some advantage come spring when they pupate and emerge as adults. Hypoprepia lampyroides (Palting and Ferguson, 2018) group together as a distinct taxa from other Hypoprepia in our concatenated analysis. In the published description, using C01 alone did not provide convincing evidence of this taxa being unique from

23 the more widely distributed H. inculta H. Edwards, 1882, despite obvious morphological differences in the size, color, genitalia and antennal flagellomere structure (Palting et al. 2018). While widely used as the “barcode” of life, mitochondrial C01 represents only a tiny fraction of the total organismal genome. It is known among certain other arctiids, notably the genus Grammia Rambur, 1866, that C01 alone is of little taxonomic utility, apparently being decoupled from the rest of the genome and lacking the necessary interspecific variation among taxa that are morphologically distinct (Schmidt and Sperling 2008). Even with three genes, however, the position of one H. inculta specimen from the type locality of H. lampyroides suggests some hybridization events might be occurring where these species are sympatric.

An enigmatic lithosiine group that Bendib and Minet gave insertae sedis status to is the genus Agylla, represented in the US by A. septentrionalis Barnes and McDunnough, 1911. This species is known in the US only from extreme SE Arizona and only from high elevations of the Huachuca and Chiricahua Mountains. Other members of this large neotropical genus occur from Mexico to northern S. America. While Bendib and Minet stated the genus is likely part of the subtribe Lithosiina, larval characteristics (arrangement of the thoracic verrucae) were inconsistent with other members where the larvae are known. In addition, we confirm here that the apodemes of the second abdominal sternite of A. septentrionalis (Fig. 4D) are long, like Cisthenina, in contrast to other Lithosiina which have short apodemes. The Neotropical distribution of the genus is also notable, as already mentioned the center of diversity of the Lithosiina is northern, but the phylogeny of Zenker et al. (2017) found other neotropical members of the Lithosiina (including Agylla) that suggest the subtribe colonized the neotropics one or more times in the past. The similarity between the European Lithosiina Atolmis rubricollis Linnaeus 1758 and the neotropical Apistosia judas Hubner, 1827 is striking and provides morphological evidence that corroborates this. In the case of Agylla septentrionalis, there is a striking resemblance to some European members of the genus Lithosia Fabricius 1798, such as male L. quadra Linnaeus, 1758. Like Zenker (2017), our molecular phylogeny places Agylla in the Lithosiina, which creates issue with the proposed Lithosiina apomorphies of the larvae and adult second abdominal sternite. The adults of Agylla are predominantly white, fairly large (for a lithosiine), with elongate wings that are held tent-like over the body, another feature they share with the Cisthenina, and not with one wing partially covering the other as in the other Lithosiina.

Two additional genera of uncertain placement that occur in the American southwest are Inopsis (Felder 1874) and Gnamtonychia Hampson, 1900. Gnamtonychia ventralis Barnes and Lindsey, 1921 is found in SE Arizona to New Mexico at higher elevations, while Inopsis modulata Edwards, 1884 is a Mexican taxa that makes occasional incursions into the mountains of SE Arizona. The two taxa look remarkably similar, but side-by-side I. modulata is a much smaller moth with shorter, more rounded wing shape than G. ventralis. Both are evidently part of a Mullerian mimicry complex that includes the arctiine Pygotenucha terminalis Walker, 1854, included here as an

24 outgroup, which is similarly colored and a toxic milkweed-feeder in the larval stages. Despite it being less common, the larval stages of I. modulata are known, while that of G. ventralis is not. Inopsis larvae (Fig. 5C) are conspicuous feeders on lichens growing on tree branches and bear distinctive orange to red verrucae against a dark body. The second abdominal sternite of the imagos (Fig. 4E and 4F) as well as their resting posture is typical of the Lithosiina in both taxa.

Members of the North American genus Crambidia are unmarked and mostly white, tan or grey. Because of this, imagos of some species that occur sympatrically are often misidentified. Included in our analysis were specimens identified as representatives of nearly all Crambidia found in North America. A recent publication added a new member to the genus east of the Mississippi that was long confused with the widespread Western C. cephalica Grote and Robinson, 1820. Our analysis included new and published sequences of this new eastern C. xanthocorpa Lewis, 2014 as well as samples of Western C. cephalica, and lends support to C. xanthocorpa being a unique taxa separate from C. cephalica. While there has been some speculation that current C. cephalica might be a species complex harboring more cryptic taxa (C. Schmidt, pers. comm), our molecular data suggests it is a single taxa that occurs at many different elevations in the West (from desert to boreal environments) often with great disparity in the size and color of the imagos, some being uniformly white, while other having more grey hindwings. There is also variation in the extent and brightness of the namesake yellow head in this species, with specimens from Texas having much brighter and more extensive yellow than examples from Arizona.

The poorly known Mexican taxa C. myrlosea Dyar, 1917 occurs along the southern US border, from Texas to Arizona, the imagos being the same size as C. cephalica, but tan rather than white, and lacking the yellow head. Worn specimens of C. cephalica lose their bright white scales as well as many of the yellow head scales and have been misidentified as C. myrlosea. In Arizona, C. myrlosea is very rare along the border, but it can be common in adjacent Sonora, Mexico. Like C. cephalica, there is widespread variation in the size of the imagos, and they occur in habitat ranging from desert to pine forests, raising the possibility that there were cryptic taxa included under this name. Multiple adults were included in our molecular analysis with origins from Arizona, several localities in Sonora and from Tamaulipan scrub habitat of south Texas (courtesy of David Wagner). All grouped together, suggesting they are a single, widespread taxa. Other than size, the uniform tan coloring of both the forewing and hindwing is consistent among them.

The taxa C. impura Barnes and McDunnough, 1913 and C. pura Barnes and McDunnough, 1913 are often confused with each other and with other Crambidia. Specimens of C. impura from Arizona and Sonora all group as a single taxa in our analysis. This species can occur sympatrically with C. cephalica in the Southwest, but lacks the yellow head and is more drab and larger, particularly the surface area of the hindwing. The flight time of C. impura is also restricted to late fall (Sept.-Nov.) while C. cephalica can be taken during most of the warmer months. C. impura is also

25 sometimes confused with C. myrlosea as both lack the yellow on the head, but the latter is always smaller and uniform tan rather than drab white. Specimens identified as C. pura from Colorado (ex B. Bartell) actually appear to be C. impura, while published sequences of C. pura from the eastern US appear as a separate taxa.

Two species of Crambidia stand out due to the unusual rounded shape of their wings and the fact, noted by Forbes (1960), that they lack a forewing accessory cell found in all other members of the genus. Both C. uniformis Dyar 1898 and C. pallida Packard, 1864 are Eastern U.S. species. While such a significant morphological difference such as the absence of the accessory cell may indicate they belong in a genus distinct from Crambidia, the molecular data here supports their current placement amongst the Crambidia, and suggests that they lost this cell at some point in their evolutionary trajectory as part of this genus. The larvae look very much like other Crambidia reared by the author (JDP).

While we were unable to get fresh DNA of the genus Clemensia, we included published sequences in our analysis and found that the genus falls outside of the Cisthenina, where they were tentatively placed by Bendib and Minet (2000). This is the same result first reported by Zenker et al. 2017 as “Lithosiini Group 1,” suggesting Clemensia + Pronola +Garudinia may require creation of a new subtribe. In our analysis, Garudinia, a S.E. Asian genus, is not part of this clade, which includes just the neotropical genera, Clemensia and Pronola. Potential apomorphies for this new subtribe have already been defined by Bendib and Minet (2000), based on Clemensia, but it remains to be determined if members of Pronola possess the same traits. If they do, then the subtribe Clemensiita (type genus: Clemensia), previously proposed by Bendib and Minet (2000) as Clemensiiti, may take nomenclatural precedence as a subtribe separate from the Cisthenina.

Acsala anomala occurs on a long branch by itself, confirming the placement of this species in a subtribe of its own, the Ascalina.

Discussion and Conclusion:

The enormous radiation of the noctuoid moths and their relatives has been a formidable target for morphological systematists for a very long time, the confounding loss and change in characters having led to numerous relational configurations. Jacobson and Weller (2002) provides an excellent comparative table of these proposed historical relationships among tiger moths as a group, while Scott and Branham (2012) provides one for the Lithosiini in particular. It was not until Zahiri et al. (2011) applied molecular phylogenetic techniques using mitochondrial C0I and seven nuclear genes that a potentially stable taxonomy for the Noctuoidea was realized, one that included moving a large portion of Noctuidae into the resurrected family Erebidae and subordinating Arctiinae as a subfamily under this. This was followed by a molecular phylogenetic analysis of the Erebidae (Zahiri et al. 2012) where they defined the Arctiinae as consisting of three tribes: Lithosiini, Syntomimi and Arctiini. Further work using molecular techniques to elucidate subtribal taxonomic relationships among the tribe Lithosiini has met with results that were sometimes inconsistent and confounding, with the first large molecular analysis of the subtribe being performed by Scott et al. 2014, followed by Zenker et al. 2017 and Scott-Chialvo et al. 2018. These papers added tremendously to our evolving understanding of how these moths are related using molecular data, however, there were some placements that suggested additional work was warranted. We took a more focused approach using molecular techniques to investigate the diverse lithosiine fauna of the US using fresh specimen DNA with the hope of clearing up some of the inconsistencies and contribute to our body of knowledge about these organisms and their relationships. As with the greater Noctuoidea, we find trying to correlate these results with morphological characters a challenge, although we feel that many of the subtribal placements elucidated here using molecular characters provide a convincing framework for future work. We find support for the monophyly of the lithosiines as well as some of the subtribal classification of Bendib and Minet, with the exception of the Eudesmiini. Excluding Acsala and Clemensia, we find all the taxa known to occur in the Western Hemisphere belong to two subtribes, the monophyletic, neotropical Cisthenina and the circumpolar Lithosiina. We find the previously proposed apomorphy for the Cisthenina, the lack of larval verrucae, to not hold with the inclusion here of Gardinia and Eudesmia, thus a new apomorphy is needed to unite them. All Cisthenina examined have long apodemes on the second abdominal sternite, a morphological trait they share with the old world Nudariina (Bendib and Minet 2000). In addition to Gardinia and Eudesmia, we propose that the genera Bruceia and Ptychoglene be included in the Cisthenina. We unfortunately were not able to obtain fresh specimen DNA of Haematomis Hampson, 1900 for inclusion in our phylogeny, but speculate based on placement of the closely related genus Rhabdatomis that it also belongs among the Cisthenina. The genus Clemensia falls outside the Cisthenina, something also seen in Zenker et al. (2017). In our analysis it forms a clade with high bootstrap support with the small neotropical genus Pronola (5 species), the adults of which are similarly sized and have a similar peculiar rounded wing shape. Additional taxon sampling is needed to 27 determine the extent of this clade and hopefully resolve the polytomy seen in this area of our tree.

Agylla septentrionalis is found to have high support as a member of the Lithosiina, despite concerns expressed by Bendib and Minet (2000) that their larvae have a different verrucal configuration from other Lithosiina. We add to this the observation that living adults of A. septentrionalis hold their wings “tent-like” over the body rather than flat against it and that the adult possess a Cisthenina-like second abdominal sternite, both features inconsistent with other Lithosiina. Thus, the placement of Agylla within the Lithosiina, means that these morphological characteristics are more labile than previously thought.

Less impactful but interesting results of this study confirm the recently named C. xanthocorpa as a separate species, and that both C. myrlosea and C. cephalica are widespread species with significant variation in size and coloring, and that there are not cryptic species among them (at least not among the samples of these taxa which we sequenced here). We also find C. uniformis and C. pallida to be correctly placed among the other Crambidia, despite the lack of an accessory cell in the forewing found in other members of the genus (Forbes 1960). This absence may be why these taxa appear to have shorter, more rounded forewings than other members of the genus. Data for the subtribal placement of the small Mexican genera Gnamptonychia and Inopsis among the Lithosiina are presented here.

With a tribe as large as the Lithosiini, it is surprising that a subtribal classification was neglected for so long, but understandable given their worldwide diversity and confounding variation in morphological characters. Beginning with Bendib and Minet (2000), we started to conceptualize how these moths might be related. Some of our placements here, such as Gardinia among the Cisthenina, certainly show that the appearance of the imagos does not necessarily belie their phylogenetic relatedness. With the apparent lack of morphological synapomorphies identified thus far that unite subtribal alliances, molecular techniques provide a useful tool for understanding how their diversity evolved. As additional molecular data is published and made available, their evolutionary relationships will become more apparent and hopefully lead to secondary identification of new morphological apomorphies in both larvae and imagos. Presently the whole life history is currently known for only a small percentage of species. Thus, we have barely scratched the surface in understanding these remarkable lepidopterans and their unique relationship to their lichen hosts and each other.

Acknowledgements:

We are deeply grateful to Ray Nagle and David Wagner for their generous help in procuring fresh specimens of numerous lithosiines and for their photographs used in this work, to Reilly Beth McManus for her able assistance in the lab, and Christopher Palting for his computer help during this study. We also wish to express thanks to James Adams, Tim Anderson, Barbara Bartell, Dave Marsden, Cliff Ferris, Ann Hendrickson, Chris Schmidt, Christi Yeager, Ana Lilia Reina and Tom VanDevender for their help in collecting specimens or with other aspects of this study. This work is in partial fulfillment of JDP’s Doctorate of Philosophy degree in the Graduate Interdisciplinary Program in Entomology and Insect Science at the University of Arizona and is the product of the Arizona Sky Island Arthropod Project (ASAP) based in WM’s laboratory. JDP is grateful to Molly Hunter and the Entomology and Insect Science Program for their support during this project. He would also like to extend special thanks to committee members WM, Yves Carriere, Ray Nagle, Carol Schwalbe and Bruce Walsh for their support and mentoring. Funding for this work was provided by WM, and JDP is deeply indebted to her for this and grateful for her patient mentoring in molecular systematics.

References:

Acharya L, Fenton MB (1992) Echolocation behavior of vespertilionoid bats (Lasiurus cinereus and Lasiurus borealis) attacking airborne targets including arctiid moths. Canadian Journal of Zoology 70: 1292-1298. Anderson T, Wagner DL, Cooper BR, McCarty ME, Zaspel J (2017) HPLC-MS Analysis of lichen-derived metabolites in the life stages of Crambidia cephalica (Grote and Robinson). Journal of Chemical Ecology 43: 66-74.

Bendib A, Minet J (1999) Lithosiinae main lineages. Annales de Societe Entomologique de France 35(3-4): 241-263.

Bendib A, Minet J (1998) Female pheromone glands in Arctiidae (Lepidoptera): evolution and phylogenetic significance. Comptes rendus de l’Academie des Sciences Paris 321:1007-1014.

Burnham, KP, Anderson DR (2000) Model selection and multimodal inference: a practical information-theoretic approach. 2nd ed. Springer-Verlag, New York.

Common IBF (1990) Moths of Australia, Melbourne University Press, Carlton.

Comstock JA, Henne C (1967) Early stages of Lycomorpha regulus Grinnel with notes on the imago (Lepidoptera: Amatidae). Journal of Research on the Lepidoptera 6(4): 275-280.

Conner WL ed. (2009) Tiger moths and woolly bears: behavior, ecology and evolution of the Arctiidae. Oxford University Press.

Covell, CVJr. (2005) A field guide to the moths of Eastern North America. 2nd ed. Virginia Museum of Natural History, Martinsville, VA.

Dowdy NJ, Conner WE (2016) Acoustic aposematism and evasive action in select chemically defended arctiine (Lepidoptera: Erebidae) species: nonchalant or not? PLoS One 11(4): e0152981.

Dubatolov VV (2015) Taxonomic review of Manulea subgenus Setema (Lepidoptera: Erebidae: Arctiinae: Lithosiini) The Canadian Entomologist 147(5): 541-552.

Dyar HG (1913) The species Afrida Moscher. Insecutor Inscitiae Menstruss 1(3): 26- 33.

Forbes WTM (1960) Lepidoptera of New York and neighboring states (Agaristidae through Nymphalidae). Memoir 371, Part IV, Cornell University Agricultural Experiment Station, 41-45.

Garcia-Barros E (1985) Identificacion de la larva y datos biologicos de Eilema uniola Rambur, 1858. Caracterizacion provisional de la larva del genero Eilema Hubner (Lepidoptera: Arctiidae). Boletin de la Asociacion Espanola de Entomologia 9: 223- 237. Gardiner, JCM (1943) Immature stages of Indian Lepidoptera. Indian Journal of Entomology 5:89-102.

Gilligan T, Passoa S (2014) LepIntercept, An identification resource for intercepted Lepidoptera larvae. Identification Technology Program (ITP), USDA-APHIS-PPQ- S&T, Fort Collins, CO. [accessed at www.lepintercept.org].

Green P (1999) Phrap. Version 0.990329 http://phrap.org

Green P, Ewing B (2002) Phred. Version 0.020425c Error! Hyperlink reference not valid.Habeck DH (1987) Arctiidae (Noctuoidea). In: Stehr F.W. (Ed), Immature Insects. Kendall Hunt, Dubuque Iowa, 1: 538-542.

Hebert PDN, Cywinska A, Ball SL, deWaard JR (2003) Biological identifications through DNA barcodes. Proceedings of the Royal Society B 270 (1512) DOI: 10.1098/rspb.2002.2218

Hesbacher S, Giez I, Embacher G, Fiedler K, Max W, Trawoger A, Turk R, Lange OL, Proksch P (1995) Sequestration of lichen compounds by lichen-feeding members of the Arctiidae (Lepidoptera). Journal of Chemical Ecology 21(12): 2079-2089.

Holloway JD (1998) The classification of the Sarrothripinae, Chloephorinae, Camptolominae and Nolidae as the Nolidae (Lepidoptera, Noctuoidea). Quadrifina 1: 247-276.

Holloway JD (2002) The moths of Borneo: family Arctiidae, subfamily Lithosiinae. The Malayan Journal 55:279-469.

Issiki, S., Mutuura A, Yamamoto Y, Hattori I (1965) Early stages of Japanese moth in color, vol 1. Hoikusha Publishing Co, Osaka.

Jacobson NL, Weller SJ (2001) A cladistics study of the Arctiidae (Lepidoptera) using characters of immature and adults. Thomas Say Monographs, Entomological Society of America.

Kalyaanamoorthy S, Minh B, Wong T et al. (2017) ModelFinder: fast model selection for accurate phylogenetic estimates. Nature Methods 14: 587–589. https://doi.org/10.1038/nmeth.4285

Katoh K, Standley DM (2013) MAFFT Multiple Sequence Alignment Software Version 7: Improvement in Performance and Usability. Molecular Biolology and Evolution 30(4): 772-780. 31

Kitching IJ, Rawlins JE (1998) The Noctuoidea. In N.P. Kristensen (Ed.) Lepidoptera, Moths and Butterflies. Vol 1. Evolution, Systematics and Biogeography. Handbook of Zoology, 355-401. Berlin/New York: Walter de Gruyter.

Knowlton CBJr. (1967) A revision of the species of Cisthene known to occur north of the Mexican Border (Lepidoptera: Arctiidae: Lithosiinae). Transactions of the American Entomological Society 93:41-100 (33 figures).

Lafontaine JD, Franclemont JG, Ferguson DC (1982) Classification and life history of Acsala anomala (Arctiidae: Lithosiinae). Journal of the Lepidopterists Society 36: 218-226.

Lafontaine JD, Schmidt BC (2010) Annotated checklist of the Noctuoidea (Insecta, Lepidoptera) of North America north of Mexico. Zookeys 40: 1-239.

Lanfear R, Fradsen PB, Wright AM, Senfeld T, Calcott B (2016) PartitionFinder 2: new methods for selecting partitioned models of evolution for molecular and morphological phylogenetic analyses. Molecular Biology and Evolution 34(3): 772- 773.

Lewis JA (2014) A new species of Crambidia (Lepidoptera: Noctuidae: Arctiinae: Lithosiini) from the Eastern United States. Proceedings of the Entomological Society of Washington 116(4): 394-401.

Maddison DR, Maddison WP (2017) Chromaseq: a Mesquite package for analyzing sequence chromatograms. Version 1.3. http://mesquiteproject.org/packages/chromaseq

Maddison WP, Maddison DR (2018) Mesquite: a modular system for evolutionary analysis. Version 3.40 http://mesquiteproject.org

McCabe TL (1981) Clemensia albata, an algal feeding arctiid. Journal of the Lepidopterists Society 35(1): 34-40.

Miller MA, Pfeiffer W, Schwartz T (2010) Creating the CIPRES Science Gateway for inference of large phylogenetic trees. In: Proceedings of the Gateway Computing Environments Workshop (GCE). New Orleans, Louisiana, 1-8.

Mutanen M, Wahlberg N, Kaila L (2010) Comprehensive gene and taxon coverage elucidates radiation patterns in moths and butterflies. Proceedings of the Royal Society B 277, 2839-2848.

Nakano RT, Takanashi A, Surlykke N, Skals N, Ishikawa Y (2003) Evolution of deceptive and true courtship songs in moths. Scientific Reports (3): 1-8.

Nguyen LT, Schmidt HA, von Haeseler A, Minh BQ (2015) IQ-TREE: A fast and effective stochastic algorithm for estimating maximum likelihood phylogenies. Molecular Biology and Evolution 32:268-274. https://doi.org/10.1093/molbev/msu300

Palting JD, Ferguson DC, Moore W (2018) A new species of Hypoprepia from the mountains of central Arizona (Lepidoptera: Erebidae: Arctiinae: Lithosiini) Zookeys 788: 19-38.

Powell JA, Opler PA (2009) Moths of Western North America. University of California Press.

Schmidt BC, Opler PA (2008) Revised checklist of the tiger moths of the Continental United States and Canada. Zootaxa 1677: 1-23.

Schmidt BC, Sperling FAH (2008) Widespread decoupling of mtDNA variation and species intergrity in Grammia tiger moths (Lepidoptera: Noctuidae). Systematic Entomology 33:613-634.

Scott CH, Branham MA (2012) A preliminary phylogeny of the lichen moth tribe Lithosiini (Lepidoptera: Erebidae: Arctiinae) based on morphological characters. Insect Systematics and Evolution 43: 321-369.

Scott CH, Zaspel JM, Chialvo P, Weller SJ (2014) A preliminary molecular phylogenetic assessment of the lichen moths (Lepidoptera: Erebidae: Arctiinae: Lithosiini) with comments on palatability and chemical sequestration. Systematic Entomology 39: 286-303.

Scott-Chialvo CH, Chailvo P, Holland JD, Anderson TJ, Breinholt JW, Kawahara AY, Zhou X, Liu S, Zaspel JM (2018) A phylogenomic analysis of lichen-feeding tiger moths uncovers evolutionary origins of host chemical sequestration. Molecular Phylogenetics and Evolution 121:23-34.

Stamatakis A, Blagojevic F, Nikolopoulos DS, Antonopoulos CD (2007) Exploring new search algorithms and hardware for phylogenetics: RAxML meets the IBM cell. Journal of VLSI Signal Processing 48(3): 271-286.

Stamatakis A, Hoover P, Rougemont J (2008) A rapid bootstrap algorithm for the RAxML Web Servers. Systematic Biology 57(5): 758-771.

Stamatakis A (2014) RAxML version 8: a tool for phylogenetic analysis and post analysis of large phylogenies. Bioinformatics 30:1312–1313.

Wagner DL (2005) Caterpillars of eastern North America: a guide to identification and natural history. Princeton University Press, Princeton, NJ. Wagner DL, Rota J, McCabe TL (2008) Larva of Abablemma (Noctuidae) with notes on algivory and lichenivory in macrolepidoptera. Annals of the Entomological Society of America 101(1): 1-13.

Wagner DL, Schweitzer D, Sullivan JB (2011) Owlet caterpillars of Eastern North America. Princeton University Press, Princeton, NJ.

Wahlberg N, Wheat CW (2008) Genomic outposts serve the phylogenetic pioneers: designing novel nuclear markers for genomic DNA extractions of Lepidoptera. Systematic Biology 57, 231-242.

Wahlberg N, Weingartner E, Nylin S (2003) Towards a better understanding of the higher systematics of Nymphalidae. Molecular Phylogenetics and Evolution 28: 473- 484.

Wahlberg N, Leneveu J, Kodandaramajab U, Pena C, Nylin S, Frietas AVL, Brower AVZ (2009) Nymphalid butterflies diversify following near demise at the Cretaceous/Tertiary boundary. Proceedings of the Royal Society B 276: 4295-4302.

Wahlberg N, Snall N, Vidalepp J, Ruohomaki K, Tammaru T (2010) The evolution of female flightlessness among Enominae of the Holarctic forest zone (Lepidoptera, Geometridae). Molecular Phylogenetics and Evolution 55: 929-938.

Zahiri R, Kitching IJ, Lafontaine JD, Mutanene LK, Holloway JD, Wahlberg N (2011) A new molecular phylogeny offers hope for a stable family level classification of the Noctuoidea (Lepidoptera). Zoologica Scripta 40: 158-173.

Zahiri R, Holloway JD, Kitching IJ, Lafontaine JD, Mutanen M, Wahlberg N (2012) Molecular phylogenetics of Erebidae (lepidoptera, Noctuoidea). Systematic Entomology 37: 102-124.

Zahiri R, Lafontaine JD, Schmidt BC, Holloway JD, Kitching IJ, Mutanen M, Wahlberg N (2013) Relationships among the basal lineages of Noctuidae (Lepidoptera: Noctuoidea) based on eight gene regions. Zoologica Scripta 42(5): 488- 507.

Zahiri R, Lafontaine JD, Holloway JD, Kitching IJ, Schmidt BC, Kaila L, Wahlberg N (2013) Major lineages of Nolidae (lepidoptera, Noctuoidea) elucidated by molecular phylogenetics. Cladistics 29:337-359.

Zenker MM, Wahlberg N, Brehm G, Teston JA, Przybylowicz L, Pie MR, Freitas AVL (2017) Systematics and origin of moths in the subfamily Arctiidae (Lepidoptera, Erebidae) in the Neotropical region. Zoologica Scripta 46(3): 348-362. 34

Figure Legends:

Figure 1: Larvae of Arctiini are typically conspicuous diurnal feeders on vegetation, while the larvae of Lithosiini are cryptic nocturnal feeders on lichens: A) Lerina incarnata last instar larva feeding on Asclepias linaria B) Lerina incarnata adult C) Estigmene albida last instar larva D) Estigmene albida adult E) Lycomorpha regulus adult F) Lycomorpha regulus last instar larva G) Crambidia cephalica adult H) Crambidia cephalica larva on lichen.

Figure 2: mandibles of various lithosiines compared to an arctiine, arrows point to the scissoral region modified in the lithosiines: A) Lerina incarnata mandible (Arctiini) B) Lycomorpha fulgens C) Crambidia myrlosea D) Crambidia cephalica E) Hypoprepia lampyroides F) Agylla septentrionalis G) Eudesmia arida H) Cisthene kentuckiensis (all Lithosiini).

Figure 3: Phylogeny of concatenated dataset generated with IQtree with critical bootstrap values highlighted. Pictured on the right are: A) Virbia costata B) Pygoctenucha terminalis C) Inopsis modulata D) Agylla septentrionalis E) Gnamptonychia ventralis F) Manulea (Eilema) bicolor G) Crambidia cephalica H) Crambidia pallida I) Crambidia myrlosea J) Crambidia impura K) Ptychoglene coccinea L) Hypoprepia lampyroides M) Hypoprepia inculta N) Cisthene angelus O) Cisthene tenuifascia P) Cisthene juanita Q) Cisthene barnesii R) Gardinia anopla S) Eudesmia arida T) Lycomorpha fulgens U) Bruceia pulverina V) Bruceia hubbardi

Figure 4: Second abdominal sternites from various lithosiines with an arctiine comparator, arrows pointing to the apodemes: A) Cisthene barnesi B) Gardinia anopla C) Eudesmia arida D) Agylla septentrionalis E) Inopsis modulata F) Gnamptonychia ventralis (all Lithosiini) G) Pygarctia roseicapitus (Arctiini).

Figure 5: Larvae of select lithosiines: A) Gardinia anopla last instar (A1 closeup of verrucae) B) Eudesmia arida last instar (B1 closeup of verrucae) C) Inopsis modulata last instar.

Supplemental Figure 1: IQtree single gene tree for COI

Supplemental Figure 2: IQtree single gene tree for 28S

Supplemental Figure 3: IQtree single gene tree for RPS5

Supplementary Table 1

APPENDIX B.

Observations of the larval stages of lichen moths (Lepidoptera, Erebidae, Arctiinae, Lithosiini) and their interaction with ants (Formicidae)

John D. Palting

Graduate Interdisciplinary Program in Entomology and Insect Science, University of Arizona, Forbes 410, Tucson, AZ, 85721-0036, USA

Email: [email protected]

Formatted for Annals of Entomological Society of America.

APPENDIX C.

Formatted for Annals of Entomological Society of America Observations of the larval stages of lichen moths (Lepidoptera, Erebidae, Arctiinae, Lithosiini) and their interaction with ants (Formicidae) John Palting 1

Abstract

Lithosiini is a tribe of the tiger moths (subfamily Arctiinae), the larval stages of which have evolved to feed on a diet of lichen. In contrast to many tiger moth larvae which are aposematic, the larvae of lithosiines are cryptic, secretive and primarily feed on lichens at night. Lichens themselves are not a single organism, but a symbiotic community that includes a fungus and one or more photosynthetic microbes. Lichens are ancient organisms that thrive in inhospitable climates and often contain toxic compounds that discourage herbivory. Growing on rocks and trees, the main predators in the realm of lichens are usually ants. Lithosiine larvae have evolved mechanisms to protect themselves from ants. Several authors have demonstrated the presence of lichen-derived polyphenolic compounds in both the larvae and adults. As a result, the larvae are generally ignored by foraging ants, which actively attack unprotected larvae of other herbivorous Lepidoptera. Life histories are known for only a few species and the potential relationship between the larvae and ants has remained largely unstudied, however some genera are suspected to be myrmecophiles. This could not be confirmed in our study. However, after rearing the larvae of several species and observing their interactions with foraging ants, our study suggests the relationship is parabiotic, with the lithosiines being chemically protected and unpalatable to the ants. Histological sections of the cuticle of lithosiine larvae reveal a series of dorsal glands, some associated with hollow setae, which are postulated here to be a method of dispersal of these lichen- derived compounds (and perhaps other compounds) through the larval cuticle. Taken together, the lichen polyphenolics, ant-interaction observations and cuticular glands provide compelling evidence that lithosiines utilize lichen toxins to move about unhindered by foraging ants, and in certain cases utilize areas of ant nests as dry refugia and pupation sites. The complex chemical ecology of lithosiines remains poorly studied and offers a fertile area for future research.

Key words: Lithosiini, Crambidia, myrmecophily, parabiosis, chemical ecology, species interactions

1 Graduate Interdisciplinary Program in Entomology and Insect Science, University of Arizona, Tucson, Arizona, 85721-0036, USA. Corresponding author: John Douglas Palting: [email protected] Corresponding author: John Douglas Palting ([email protected]) 47

Introduction

The evolution of life on earth seems to be a combinatorial affair, with the smallest organisms banding together to form larger organisms. Thanks to the molecular revolution, we now know nearly all eukaryotic species contain other species or parts of species in their makeup, from the chloroplasts and mitochondria of cells to obligate endosymbionts. Lichens were among the earliest examples of this composite evolution of life. Cooperation between two or more primitive organisms, limited in their abilities to survive by themselves, allows the lichen partners to not only survive, but thrive in the most inhospitable parts of the planet. This amazing evolutionary feat has allowed lichens to colonize nearly every corner of the globe, from the arctic to the equator, from the coldest to the most arid regions, sometimes in places where few other plants or animals can exist. Lichens are everywhere. Lichens are not without defenses; millions of years of this cooperative evolution has resulted in lichens being able to produce a repertoire of chemicals to deter being eaten by other organisms.

But life finds a way to exploit other life: one diverse lineage of moths evolved the ability not only to consume the ubiquitous lichen, but also usurp their defenses and use these to their advantage. Likely due to a diet of lichens, the tiger moth tribe Lithosiini represents one of the largest radiations of the ditrysian Lepidoptera, with between 3000-5000 species (Common 1990, Wagner et al. 2008). The autapomorphy of the group, an enlarged mandibular mola in the larvae, has been suggested to be an adaptation for grinding up lichens for digestion (Gardiner 1943, Rawlins 1984, Bendib and Minet 1999). Lithosiines generally have small caterpillars, in keeping with the diminutive stature of the lichen forests on which they feed. Their distribution is as widespread as their lichen hosts, the greatest diversity being in the tropics (Holloway 2002, Wagner et al. 2008), but with members also found in very cold and very hot environments. In the US, some of the highest diversity occurs in the arid Southwest, where the great variety of elevations and microclimates contains around 1000 species of lichen, representing about 20% of the known American species (CNALH website 2020).

Lithosiine caterpillars are known to sequester lichen polyphenolic compounds in their tissues, and in some cases these are retained through the pupal and adult stages, although the compounds are not universally present and their concentration is variable across individuals (Hesbacher et al. 1995, Wagner et al. 2008, Conner 2009, Scott et al. 2014, Anderson et al. 2017, Scott-Chailvo et al. 2018). Adult lithosiines are often brightly colored, warning potential predators of their unpalatability. In addition to color, many adults exhibit acoustic aposematism, creating warning clicks with their tymbal organs while they are in flight, specifically in response to certain bat echolocation frequencies (Acharya and Fenton 1992, Conner 2009). In contrast, the larval stages of lithosiines, where known, are very cryptic and nocturnal, usually shades of brown or black, indicating they try to blend into their environment (Wagner 2005, Conner 2009). This difference in coloration between the larvae and adults likely reflects the different types of predators challenging the two life stages. Some adult lithosiines are diurnal, thus they have to avoid predation by visual predators like birds, leading to brilliant red and

48 black coloration of genera like Lycomorpha Harris, 1839 and Ptychoglene Felder, 1874. Others, like Cisthene Walker, 1854 and Gardinia Kirby, 1892, are brightly colored and produce audible clicks, affording them protection from both birds during the day when they are resting and bats at night when they are active (Conner 2009, Dowdy and Conner 2016). The larvae of lithosiines live among the lichens growing on rocks and trees, a realm ruled by ants. Their cryptic coloration combined with their habit of hiding during the day and feeding at night allows them to largely avoid bird predation, but the ants remain a constant threat. Myrmecophily has been suspected as part of lithosiine larval biology (Komatso and Itino 2014, Wagner 2008), but the nature of the association between larvae and ants remains poorly known.

The relationship between the lycaenid European Large Blue Butterfly, Phengaris arion Linneaus, 1758 and the ant Myrmica sabuleti Meinert, 1861 provides an example of truly obligate myrmecophily. M. sabuleti collect the last instar larvae P. arion from underneath Thymus plants on which they feed in early instars, carrying them to their nest as if they are their own larvae. The caterpillars employ both semiochemical and acoustical mimicry to deceive the ants (Thomas et al. 2010). Once inside the nest, the caterpillars feed on the ant larvae and complete their development (Thomas and Settele 2004). The potential ant- lithosiine relationship was not thought to be as intimate as that of P. arion and M. sabuleti, however, tantalizing observations of ant-lithosiine interactions have been published which suggest it worthy of further investigation.

Two papers, published decades apart, suggest an intimate relationship between ants and lithosiine larvae. Komatsu and Itino (2014) documented interactions between a Japanese lithosiine, Nudina artaxia Butler, 1881, and black ants, Lasius sp. Fabricius,1804. Not only were N. artaxia larvae able to move with impunity among the ants, they actually followed ant pheromone trails and participated in homopteran “milking” along with the ants. The authors speculate that the larvae are obtaining amino acids lacking in their lichen diet from this homopteran exudate. Ayre (1958) observed the larvae of the North American species, Crambidia casta Packard, 1869 moving in and out of Formica subnitens Creighton, 1940 (Hymenoptera, Formicidae) nests unhindered, and utilizing the refuse area of the ant nests as a safe pupation site. Ayre (1958) noted that once the adult moths emerged from their cocoons, however, they were no longer immune to attack by the ants. Both observations suggest lithosiine larvae are able to avoid predation by the ants and the latter paper led to speculation that C. casta has some sort of obligate relationship with F. subnitens Stehr, 1987 a member of the widespread Formica rufa Linneaus, 1761 species group, although there are no further records of this being observed again since its publication. Ayre (1958) detailed the numbers of larvae and pupae (hundreds) found in Formica nests in British Colombia as part of a larger inventory he conducted of arthropod guests of Formica. But whether their association is obligate or facultative remains unstudied. We do not even know whether the C. casta larvae derive any benefit from the ant association (such as homopteran milking in Nudina) or whether they simply have a chemical “cloak of invisibility” that allows them to opportunistically utilize the elevated Formica nests as a safe environment in which to shelter and pupate. Either way, if C. casta and N. artaxia are able to avoid ant attacks,

49 perhaps other members of the North American lithosiine fauna, particularly members of the genus Crambida, might also. In general, larvae of US lithosiines are rarely encountered in nature, and the life histories of many taxa remain unknown. Those that are known are associated with lichens, with several being successfully reared in the laboratory to adulthood (Comstock and Henne 1967, McCabe 1981, Wagner et al. 2008, unpublished data, Anderson et al 2016). Females of many species can be attracted to lights, and like their tiger-moth kin, will readily oviposit when confined in containers. This behavior allowed us to rear the larvae of several North American species in the laboratory on lichens so that interaction studies with ants could be conducted.

An experimental limitation of any work with a lichen-feeding organism is the heterogeneity of the lichens themselves. Lichens cannot be artificially cultured, and thus must be harvested from the wild. They seldom occur as one “species,” but rather as aggregates of species on branches and rocks, thus it can be difficult to know exactly what lichen the lithosiine larvae are feeding on. Spribille (2016) found in Montana that two very different-looking lichens had the same fungal and cyanobacterial component, but one bright yellow form that produced vulpinic acid also contained a yeast that the green, non-acid-producing form lacked. Anderson et al. (2017) analyzed the common lichen Physicia ({Schreber} Michaux 1803), and found anthraquinone and prasanic acid in all their samples, but one also contained barbatic acid, picrolichenic acid, and chrysophanol not present in the others. Additionally, the composition and amount of lichen toxins in a single “species” can vary due to temperature, photoperiod and other conditions (Rankovic and Kosanic 2015). These complex aspects of lichen biology compound the difficulty in controlling what types of metabolites the lithosiine larvae are ingesting. Several authors have mentioned that lithosiines preferentially feed on the algal component of the lichen, and seem to strip off the photosynthetic regions leaving the white fungus behind (McCabe 1981, Wagner et al. 2008). While this may appear to be the case, the integrated morphology of the lichen symbionts makes it difficult for the larvae to entirely avoid eating the fungus or other microbial components present.

Hesbacher et al. 1995 found that Eilema complana Linnaeus, 1758 larvae preferentially fed on the algal and cortical layers of the thallus of the lichen Cladonia pyxidata ({L.} Hoffm., 1796) which do not contain phenolic compounds, and avoided the medulla, which contained the bulk of these toxins, but that some adults demonstrated the presence of the compounds. It may be that some lithosiine larvae deliberately dose themselves with certain levels of toxins, deriving most of their nourishment from the photosynthetic symbiont(s), but obtaining just enough of the fungal (or other) toxins to provide protection (Wagner et al. 2008). A similar strategy has been proposed for the arctiine Grammia incorrupta (Hy.Edwards 1881) a generalist which seems to increase feeding on toxic plants when it is parasitized, a process the authors term “self-medication” (Singer et al. 2009). The tradeoff between food quality and consuming protective secondary metabolites for protection has also been termed “adaptive foraging” (Singer et al. 2004), and some lithosiines seem to fit this model in how they selectively feed on different parts of lichens.

It is also interesting to note that not all lithosiine adults are aposematically colored. Among North American genera Bruceia and Clemensia adults are rather cryptic, yet both genera have been reared on a diet of lichen. Clemensia is a genus that can also be reared successfully on just algae (McCabe 1981), so perhaps these and other somberly-colored lithosiines do preferentially feed on just the algal portion of the lichen and do not sequester and advertise lichen toxins in the same way as a majority of the lithosiine taxa. There likely exists a spectrum of feeding behaviors within the Lithosiini, but the presence of lichen-derived polyphenolic compounds has now been well-established in the group by several authors. Anderson et al. (2017) analyzed larvae, pupae and adults of Crambidia cephalica Grote and Robinson, 1870 for lichen-derivatives using HPLC-MS. Their study demonstrated the presence of an anthraquinone and chrysophanol in larvae and pupae, as well as barbatic, picrolenic and prasinic acid metabolites in the adults, providing evidence of lichen polyphenolics sequestered during the larval stages for use by the adult moths. Early work by Hesbacher et al. (1995) found the anthraquinone derivative parietin as well as four other lichen- polyphenolic compounds in the lithosiine Eilema complana, and Scott et al. (2014) identified 14 different lichen polyphenolic compounds from a sample of lithosiines, including vulpinic acid in Crambidia cephalica. Scott-Chialvo et al. (2018) recently published a study using transcriptomic data for lichen-derived phenolic metabolic pathways, the resulting phylogeny traces the evolution of lichenivory and sequestration of the secondary metabolites derived in the adults of 22 species of lithosiines. Anthraquinones and their derivatives occur across these studies and may be especially important at deterring ant predation, as documented by Hilker et al. (1992), who found that anthraquinones in the eggs and larvae of a chrysomelid beetle were a deterrent to predation by the ant Myrmica sabuleti.

While the exact type and quantity of lichen-derived toxins varied considerably in the above studies, a presumption is made here that by feeding on lichens, some level of these toxins are going to be present in the lithosiine larvae, thus the chemical analysis for the presence and identity of lichen-derived compounds was not performed. Instead we conducted interaction studies between these larvae and ants to investigate the potential use of these toxins as deterrents of ant predation, as well as histology of morphological features of the larval cuticle through which these compounds may be expressed.

Materials and Methods

Excavating ant nests

Formica ant nests were located visually in several localities in SE Arizona (Pima Co., Pinal Co., Cochise Co.), NE Sonora (Municipio de Cananea, Municipio de Santa Cruz) and in Colorado (Gilpin Co.) and excavated as far as the substrate would allow to search for evidence of Crambidia, including larval molts, pupal cases etc. using a graded sieve. Colorado Formica, locally called “wood ants,” built above-ground mounds, which required minimal excavation. In one case simply moving a large stump the mound was built around provided access to the detritus area of the nest where the Crambidia had

51 been found by Ayre (1958). Nest material was removed using a shovel and sieved well away from the agitated ants to search for Crambidia.

Rearing lithosiine larvae

Larvae of five species of North American lithosiines were reared for this study. The larvae of Hypoprepia inculta Hy. Edwards, 1882 were reared from eggs laid by females attracted to mercury vapor lights in Upper Pinery Canyon, Chiricahua Mountains, Cochise Co. Arizona, June 14, 2017, leg JDP. Larvae of Eudesmia arida Skinner, 1906 were obtained through D. Wagner, with the original source being a female collected in Ramsey Canyon, Huachuca Mts., Cochise Co., Arizona, July 2017. Cisthene tenuifascia Harvey, 1875 and Crambidia cephalica larvae were reared (3 generations) ex ova from females collected in Peppersauce Canyon, Santa Catalina Mtns., Pinal Co., Arizona, July 11, 2016, leg JDP. Crambidia myrlosea Dyar, 1917 was reared (2 generations) ex ova from a female collected at California Gulch, Santa Cruz Co., Arizona, August 15, 2018, leg JDP.

During rearing it proved critical that the larvae (and the lichen) not be kept too wet or too dry and that adequate ventilation be provided. Fine screen-topped bottles (as are often used for Drosophila cultures) provided ventilation but enough enclosure that the lichens remained moist for some time. Upon hatching from the egg, the larvae were individually picked up using a fine paintbrush and placed on lichen moistened with distilled water. An important note: prior to use, the lichen should be thoroughly inspected for the presence of predators, in particular neuropteran larvae, which can hide under the lichen thallus and decimate the newly hatched larvae if undetected. As the larvae grew, they were transferred to larger lichen-bearing branches that were misted with distilled water. Larger larvae were transferred to fresh lichen branches simply by tapping the old branch on a white paper towel, whereupon the viable larvae curled up and fell on the towel where they were picked up with a fine brush or larval forceps. Larvae that did not exhibit this behavior were either sick or dead and were discarded.

Lichen species offered varied, but often included a type of Physcia sp. growing in association with orange and grey crustose lichens. Lichen branches were primarily collected from various species of oak (Quercus arizonica and others) at mid-elevations in SE Arizona and stored dry in Ziploc bags in a refrigerator until needed.

Ant-caterpillar interaction studies

Ant-caterpillar interaction studies were staggered in time based on availability of the lithosiine larvae. For controls we used a variety of non-lithosiine larvae which were reared at the same time and included Hypercompe permaculata Packard, 1872 (Fig. 2D), Arachnis zuni Neumoegen, 1890 (Fig. 2A), Estigmene albida Stretch, 1873 (Fig 2B) (all Erebidae, Arctiinae) and a noctuid Trichoplusia ni Hubner, 1800 (Noctuidae, Plusinae) (Fig. 2C). All of the selected controls are generalist feeders that might include toxic 52 plants (Singer and Stireman 2005, Pankoke et al. 2012, Singer et al. 2014), even fungi (Moskowitz and Haramaty 2012), among their diet in nature, but the individuals used were reared ex-ova on a diet of edible produce (Romaine lettuce for the erebids or cabbage for the Trichoplusia) to eliminate this variable. We attempted to size-match the larvae, particularly the control species, as not to overwhelm the ants with a prey item that was too large. Generally, this was not a problem for the lithosiines, which are small enough to be easily handled by an ant, but the control larvae (genera Estigmene, Arachnis and Hypercompe) were second or third instars when the interaction studies were conducted, thus approximately the same size as later instar lithosiines. Foraging ants were visually located and the larvae were introduced using larval forceps ahead of the foraging individuals to minimize disturbance. Sometimes the foraging ants would move in a different direction and the process was repeated until it was certain an ant would sense the larvae before the interactions were recorded. For each lithosiine species, this was repeated three times with different individual larvae offered to different individual ants.

Histology

Larvae were injected anally using a tuberculin syringe with a solution of 10% neutral buffered formalin until slightly distended, then immersed in the same solution for several days to ensure they were fixed inside and out. Fixed larvae were then placed in a histological cassette and processed into paraffin using a Leica Biosystems tissue processor. The paraffin-embedded larvae were cut into 3-4um slide sections using a standard rotary microtome and the sections were allowed to dry. Slides were rehydrated and stained with either a standard hematoxylin/eosin setup, or stained with Periodic Acid-Schiffs (PAS) or mucicarmine reagents using standard histological protocols on a Roche Special Stains instrument. Photos were taken using an Olympus TH4-100 microscope with a CellSens photography program.

Results

Excavating ant nests

Arizona has a high diversity of lithosiines, including several members of the genus Crambidia. There is also a high diversity of ants and many myrmecologists routinely dig up ant nests and carefully sieve the soil for inquilines, including crickets of the family Mymecophilidae, as well as ant-mimicking salticid spiders. This careful sieving would certainly reveal the presence of lithosiine larvae and pupae if they were indeed regular inhabitants of ant nests, yet none of the researchers queried recalled finding anything resembling a lepidopteran inside an ant nest.

Attempts over several years to excavate Formica nests in the Southwest looking for any evidence of association with Crambidia, particularly in the nest refuse area, have proven unsuccessful. One challenge excavating Formica nests in arid environments is that they do not build the above-ground mounds that are familiar in more mesic environments, and 53 instead nest entirely underground. The haphazard tunnels meandering in and under rocks make excavation difficult. One promising site was located near Golden, Colorado at 9000’ elevation in the pine forests. Here C. impura occurs abundantly, along with at least two species of Formica which build above-ground mounds. The property owner, Barbara Bartell, is a keen naturalist and reported occasionally observing white Crambidia males sitting on or near the mounds. An additional curious observation she made was that she had only taken male C. impura at light, never a female. Perhaps the females do not respond to light in the same way as males, or could it be that the females are somehow tied to the mounds, either unable or unwilling to leave their shelter?

Two mounds of Formica rufa group were excavated that Bartell had watched undisturbed for the many years she had owned the property. Refuse areas of the mounds could be readily identified, as well as brood areas, and despite significant and uncomfortable effort sieving through the nest, no evidence of Crambidia could be found. The time of year this was conducted should have produced either fully mature larvae or pupae if it followed Ayre’s observations from British Colombia. Per Bartell’s collection data, the C. impura adults begin to fly in late August-September at this site. One feature we noted of the Colorado locality that may be relevant is the abundance of lichen-covered rock outcrops throughout the area. The Crambidia larvae need dry sites to hide and feed while they are developing, and these outcrops provide such an environment. Perhaps in a more mesic environment without rock outcrops, a dry Formica mound serves this purpose. Despite hours spent searching and brushing these rocks at night looking for lithosiine larvae, none were found. This however, is not unusual, as wild larvae are seldom encountered. Also, many different lichens colonize the inaccessible branches of the mostly pine canopy that the larvae could be feeding on.

Rearing lithosiine larvae

The larvae of the genera Hypoprepia and Eudesmia proved to be very slow to develop, often not pupating after many months (some not even after 1 year), while larvae of Crambidia and Cisthene grew more rapidly and pupated in a few weeks from eggs. The slow development of some lithosiines ultimately meant more opportunities for mold, disease and other factors that cause a demise of the larvae. Crambidia cephalica proved to be among the easiest lithosiines to rear, and captive-born imagos would mate and lay eggs for at least 3 generations.

Ant-caterpillar interaction studies

Rearing Crambidia and other species of lithosiines has shown there is indeed some chemical ecology at play that either renders the larvae invisible or unpalatable to ants. Various lithosiines were reared as part of this study, including Hypoprepia inculta (Fig. 2H), Crambidia cephalica (Fig 5A), C. myrlosea (Fig. 2G), Eudesmia arida (Fig 2D)and Cisthene tenuifascia (Fig 2E)and ant-interaction studies were conducted with these using several species of ants (Table 1). In all cases, the ants avoided the lithosiine larvae, while viciously attacking the non-lithosiine controls (Fig. 1). 54

What is providing the lithosiine larvae the observed immunity from ant predation? Possibilities considered include cuticular hydrocarbons that mimic the ants, chemicals such as crematoenones that “appease” ants and reduce aggression (Menzel et al. 2014) or, more likely, polyphenolics or other compounds obtained from the lichen diet that render the larvae unpalatable, eg. chemically aposematic. This latter hypothesis fits best both with the published evidence of lichen polyphenolics being detected in the larval stages of Crambidia, as well as the observed interactions between the ants and the larvae, but the use of other compounds, like bioamines, not necessarily derived from their lichen diet, cannot be ruled out. Both myrmicine (Aphaenogaster, Mayr 1853) and formicine (Formica Linnaeus, 1758, Camponotus, Mayr 1861, Myrmecocystus, Wesmael 1838) predatory ants were included in these studies. Crambidia larvae are most often ignored by the foraging ants, but occasionally an encounter resulted in a display of spread mandibles, followed by retreat of the ant, with the lichen polyphenolics likely being the reason for deterrence. Hydrocarbon mimicry and “appeasement” chemicals seems unlikely as the ants demonstrate no desire to interact with the lithosiine larvae. In some observed interactions, the ant would touch the lithosiine larvae with its antennae (“brief interaction”), sometimes resulting in the larvae arching its back slightly, at which point the ant would be repulsed and show no further interest. No such behavior was observed with any of the mentioned controls, all of which were quickly attacked by the ants. In addition to the published analytical presence of lichen polyphenolics in lithosiine larvae, some structure must be present through which they or their derivatives are dispersed from the cuticle.

Histology

Histological sectioning of the larval stages of some lithosiines was performed in an attempt to visualize cuticular structures that might be involved in dispersing semiochemicals to the ants. Five species were examined (Crambidia cephalica, Crambidia myrlosea, Eudesmia arida, Bruceia hubbardi and Lycomorpha fulgens), along with a larvae of a palatable arctiine control (Estigmene albida). All the larvae except the L. fulgens and B. hubbardi have verrucae (raised, setae-bearing cuticular structures) in pairs on the dorsum that are associated with underlying clusters of epithelial cells. These include various cell types associated with the production of the seta itself (trichogen cells), sensory cells around the sheath (thecogen cells), cells that produce the socket (tormogen cells) (Snodgrass 1935). In addition, there are sometimes cell clusters whose morphology suggest a secretory function. In the lithosiines examined, these enlarged cells or cell clusters are intimately associated with the seta-producing trichogen cells along the dorsum of the larvae, and were present in the larvae both with and without verrucae (Fig. 3c-f). In many venomous caterpillars, the glandular cells that secrete poison into the setae are thought to be a “sister cell” of the trichogen cell, perhaps derived from division of that cell (Snodgrass 1935, Gilmer 1925). Examination of multiple sections suggests the glandular cells of lithosiine cuticle are secreting a granular substance that stains with eosin (thus called eosinophilic). In the case of Crambidia, this substance seems to be secreted into modified setae. Lycomorpha fulgens and Bruceia hubbardi lacking verrucae, exhibit these glandular cells associated with dorsal tormogen

55 pores. The cells also seemed to be secreting a similar-looking granular eosinophilic substance, but its association with the setae was not clear from the specimens examined. Eudesmia also exhibited pores rather than hollow setae.

The palatable control, Estigme albida (Fig 2b), also has twin dorsal verrucae on each segment, with the clusters of cells associated with the setae being substantially larger than those in the lithosiines, particularly the trichogen cells that produce the setae (Fig. 3a, 3b). These cell clusters also contain some simple glandular cells, but these structures are not as complicated as those seen in the lithosiines. Prominent verrucal cell clusters with simple glands are apparently also present in another hairy arctiine species that proved to be a palatable control in the present study; Arachnis zuni (R. Nagle, pers. comm) as well as in the milkweed-feeding Lerina incarnata (JP, unpublished data). Both A. zuni and E. albida, as well as L. incarnata, are conspicuous (as opposed to secretive) feeders on low-growing plants and shrubs, and are clothed with long setae (relative to most lithosiines). Long setae have been suggested as a deterrent to parasitoids such as wasps and flies, making it more difficult for the females to oviposit on the body of the larva (Kageyama and Sugiura 2016, Dyer 1997).

What function the glandular cells associated with these setae play in the biology of Estigmene is unknown, however they were interrogated using two histological stains, PAS and mucicarmine, and found to exhibit different reactivity from the glandular cells of the lithosiines sampled (Fig 4). PAS stain is used to detect structures high in carbohydrates (glycoproteins, proteoglycans, glycolipids, glycogen) and this stain reacted with the secretion produced by the cuticular glands of the lithosiine larvae (Figs. 4c and 4d), but not the glands of the Estigmene control (Fig 4a). Mucicarmine stain is used to detect certain mucin compounds (especially acidic mucopolysaccharides), and was negative both for the lithosiines (Figs. 4d and 4f) and the arctiine control (Fig 4b), suggesting the nature the glandular secretion is not a mucin. Sections of both types of larvae contained inherent positive controls for each stain, especially the gut, to demonstrate the stain worked properly. Thus, in addition to morphological differences in the cuticular glands of both types of larvae, there seems to be a chemical difference, with the lithosiine cuticular glands producing a PAS-reactive substance, likely a glycoprotein, that was absent in the Estigmene control glands.

The dorsal gland-associated setae of Crambidia cephalica (Figs. 3e and 5A-E)) and C. myrlosea (Fig. 5F) larvae are particularly interesting. The seta itself appears to have a bulbous, hollow structure near the base that contains the granular, eosinophilic substance. These setae appear to break easily, releasing their contents. Below these setae are secretory cells which contain a similar-looking granular substance and are presumably the source, as the material within the bulbous seta and the gland cell both react with PAS stain. These unusual bulbous-based setae were only found in the Crambidia (both species examined) and not in Eudesmia or Lycomorpha. They are somewhat reminiscent of the balloon setae reported from certain myrmecophilous (and nonmyrmecophilous) Riodinidae. Larvae of the riodinid genus Calydna Doubleday, 1847 (a nonmyrmecophilous species) examined externally by scanning EM were found to have a

56 spongey material secreted inside the hollow acanthae of the specialized balloon setae. When the seta was damaged, the material released seemed to irritate ants (Hall et al. 2004). Unlike the riodinids, where the specialized balloon setae occur only on the prothoracic shield, the modified setae of C. cephalica occur all along the dorsum in association with the verrucae.

The cuticular glandular cells appear to be a feature that distinguishes lithosiine larvae from the Estigmene control, thus making these structures the likely source of semiochemicals responsible for deterring ant attacks. Other glandular epithelia were noted by histology deeper within the body of the thoracic segments of both the lithosiines and the arctiine control, the function of which are also unknown. Stehr (1987) states that many groups of lepidopteran larvae have midventral cervical glands anterior to the prothoracic legs, including the Noctuidae, Lymantriidae and Notodontidae, all groups within the Noctuoidea to which the lithosiines belong. Since these thoracic glands were present in both the ingroup and the outgroup, they were not considered the likely source of ant-related compounds, although they arguably could still play a role in the lithosiines. Modified setae and glands likely play an important role in the biology of many larval lepidopterans, especially those species that feed at ground level. Wagner et al. (2011) mention the distinctiveness of the setae of other erebid and some geometrid larvae that feed on leaf litter, algae and fungi, and speculate these may have a secretory function. The overall morphological similarities between these unrelated larvae and the lithosiines is rather striking (cryptic brown, with short setae), the convergence apparently caused by their living in the same microhabitat. It would be interesting to do comparative histological studies of these larvae to see if they possess cuticular glands similar to lithosiines. There might also prove to be convergence in the chemical ecology among these ground-dwelling lepidopteran larvae based on the similarity in their diet and need for defense against ant predation. North American genera that feed on leaf litter and fungus as larvae include Abablemma Nye, 1875 (Erebidae, Hypenodinae), Metalectra Hubner, 1823 (Erebidae, Boletobinae), Idia Hubner, 1813 (Erebidae, Herminiinae) and Idaea Treitschke, 1825 (Geometridae, Sterriinae). Wagner, Rota and McCabe (2008) provide an excellent review of algivory and lichenivory among larvae of North American macrolepidoptera that includes these genera and others.

Discussion and Conclusions

Lichens themselves are enigmatic and fascinating organisms. Recent molecular research has shown many to be an association of three microorganisms; a fungus, an algae and a yeast (Spribille 2016). Other authors argue that there are often many other microorganisms present beside these (Grube and Berg 2009, Grube et al. 2015, Wilkinson 2018) and that all of these might somehow contribute to the success of the lichen. They certainly defy our standard definition of “species” in many ways, and are usually classified based on the fungal component without regard for the other symbionts present (Mitchell 2015). Lichens also defy lab culture, as no one has ever created a lichen simply by putting these microorganisms together in a petri dish. Clearly there is much

57 that remains to be discovered about the complex biology of these ancient composite organisms, particularly in the chemical repertoire they possess. The lichen moths are proving to be equally fascinating, likely owing their success to the success of the lichens and the caterpillars’ ability to feed on them and sequester both nutrients and defensive chemicals. Avoidance of these caterpillars by ants extends across two major ant subfamilies (Formicinae and Myrmicinae) and across the major subtribes of US lithosiines, encompassing members of the Cistheninii (Cisthene and Hypoprepia), Eudesminii (Eudesmia) and Lithosiinii (Crambidia) (Bendib and Minet, 1999), suggesting this ability evolved very early on in the lineage when they adapted to a lichen diet. This correlates with the findings of Scott et al. (2014) and Scott-Chialvo et al. (2018), who reported finding lichen polyphenolics across all the major clades of lithosiines recovered in their analysis. The larvae of several US lithosiines reared here exhibit a variety of glands and pores on their dorsal cuticle which may be involved in the presentation of compounds that repulse the ants. Morphology alone is not sufficient to definitively conclude this, since there are also some glandular cells on each dorsal segment even in the palatable arctiine controls, but in the lithosiines these are more prominent and, in the case of Crambidia, associated with modified setae that appear to have a specialized function. It is worth noting that predators other than ants seem to not be deterred by the lithosiine caterpillar’s defenses; small centipedes, spiders and neuropteran larvae have all been inadvertently introduced via the lichens to rearing containers and these readily fed on the lithosiine larvae. Since all the lithosiine larvae used in this study were reared in captivity, we cannot say what effect, if any, the caterpillar’s chemical defenses have against dipteran and hymenopteran parasitoids they encounter in the wild. With the exception of E. arida, the lithosiines reared here all have relatively short setae compared to other arctiids, long setae having been postulated to serve as a barrier to oviposition by parasitoids.

One of the goals of this paper is to call attention to the complex and understudied chemical ecology of the lichen moths. Their small size, widespread distribution and the relative ease they can be reared in the lab using lichens make them attractive models. Lichens are an unusual dietary choice for any insect, yet their ubiquitous distribution and chemical arsenal likely has allowed the lithosiines to dramatically radiate around the world. Based on the chemical analyses cited, in addition to behavioral studies, a diet of lichen likely provides lithosiines protection from predators throughout their life. Ants avoid eating the larvae, and as adults they advertise distastefulness to birds via aposematic coloration, and from bats via acoustic aposematism. The variety of lichen- derived metabolites found in the adult moths makes it likely that the lithosiines may also utilize the compounds for other functions, such as pheromone synthesis and perhaps as nuptial gifts, as seen in other arctiids (Boppre and Schneider 1985, Conner 1990, Eisner 2003). Lichen toxins may further serve a function in protecting the eggs, larvae and pupae of lithosiines from the ever-present threat posed by parasitic wasps and flies and even microbial pathogens, as anthraquinones recovered from some are known for their antimicrobial activity (Cudlin et al. 1976, Hesbacher et al. 1995). With around 3600 described species and perhaps as many still undescribed, there is so much to learn about these diminutive and colorful lichenivores. Most of those currently described have

58 unknown life-histories, which means we also have no idea how much fidelity some might have for particular lichens, their fortunes linked. Some lichens are said to be declining due to pollution and climate change, and thus are useful as environmental indicators (Lawrey 1993, Tesky 2001, Government of Canada 2008). How this might affect the moths that feed on them is unknown, but it seems clear that the biology of the lichens and the lichen moths are intricately intertwined, an elegant example of how complex multi- species interactions have shaped the amazing diversity of life on our planet.

Acknowledgments

The author would like to thank Bob Johnson and Christina Kwapich from Arizona State University, Tempe, AZ. for their help investigating the possibility of lithosiine larvae in ant colonies and Barbara Bartell for her hospitality and encouragement in searching for evidence of Crambida in Formica mounds on her property. The author is deeply grateful to David Wagner and Ray Nagle for their help obtaining and photographing the larvae needed for this work. Special thanks to Yves Carriere for his guidance while performing these studies, and to Wendy Moore for her support and many helpful comments to improve the manuscript.

REFERENCES

Acharya, L. and M.B. Fenton. 1992. Echolocation behavior of vespertilionoid bats (Lasiurus cinereus and Lasiurus borealis) attacking airborne targets including arctiid moths. Can. J. Zool. 70:1292-1298.

Anderson, T., D.L. Wagner, B.R. Cooper, M. E. McCarty and J. Zaspel. 2017. HPLC-MS Analysis of lichen-derived metabolites in the life stages of Crambidia cephalica (Grote and Robinson). J. Chem. Ecol. 43: 66-74.

Ayre, G.L. 1958. Notes on insects found in or near nests of Formica subnitens Creighton (Hymenoptera: Formicidae) in British Columbia. Insectes Sociaux 5(1): 1-7. Bendib, A. and J. Minet. 1999. Lithosiinae main lineages. Annales de Societe Entomologique de France 35(3-4): 241-263.

Bendib, A. and J. Minet. 1998. Female pheromone glands in Arctiidae (Lepidoptera): evolution and phylogenetic significance. Comptes rendus de l’Academie des Sciences Paris 321:1007-1014.

Boppre, M. and D. Schneider. 1985. Pyrrolizidine alkaloids quantitatively regulate both the scent organ morphogenesis and pheromone biosynthesis in male Creatonotos moth (Lepidoptera: Arctiidae). J. Comp. Physiol. 157: 569-577.

Common, I.B.F. 1990. Moths of Australia, Melbourne University Press, Carlton. Comstock, J.A. and C. Henne. 1967. Early stages of Lycomorpha regulus Grinnel with notes on the imago (Lepidoptera: Amatidae). J. Res. Lepid. 6(4): 275-280.

Conner, W.L. ed. 2009. Tiger Moths and Woolly Bears: Behavior, Ecology and Evolution of the Arctiidae. Oxford University Press, New York, NY.

Covell, C.V.Jr. 2005. A Field Guide to the Moths of Eastern North America, 2nd ed., Virginia Museum of Natural History, Martinsville, VA.

Consortium of North American Lichen Herbaria. 2020. https://lichenportal.org/cnalh/index.php

Cudlin, J. M. Blumauerova, N. Steinerova, L. Mateju and V. Zalabak. 1976. Biological activity of hydroxyanthraquinones and their glucosides toward microorganisms. Folia Microbiol. 21:54-57.

Dowdy, N.J. and W.E. Conner. 2016. Acoustic aposematism and evasive action in select chemically defended arctiine (Lepidoptera: Erebidae) species: nonchalant or not? PLoS One 11(4): e0152981.

Dyer, L. A. 1997. Effectiveness of caterpillar defense against three species of invertebrate predators. J Res. Lepid. 34:48-68.

Eisner, T. 2003. For the Love of Insects. Cambridge: Belknap Press of Harvard University Press. Gardiner, J.C.M. 1943. Immature stages of Indian Lepidoptera (5) Indian Journal of Entomol. 5:89-102. Gilmer, P.M. 1925. A comparative study of the poison apparatus of certain lepidopterous larvae. Ann. Entmol. Soc. Am., 18: 203-239. Government of Canada. 2008. Biobasics: the science and the issues. Bioindicators. http://biobasics.gc.ca/english/View.asp?x+740. Grube, M. and G. Berg. 2009. Microbial consortia of bacteria and fungi with focus on lichen symbiosis. Fungal Biol. Rev. 23:72-85.

Grube M., T. Cernava, J. Soh, S. Fuchs, I. Aschenbrenner, C. Lassek, U. Wegner, D. Becher, K. Reidel, C.W. Sensen and G. Berg. 2015. Exploring functional contexts of symbiotic sustain within lichen-associated bacteria bacteria by comparative omics. ISME Journal 9: 412-424.

Hall, J.P.W., D.J. Harvey and D. H. Janzen. 2004. Life history of Calydna sturnule with review of larval and pupal balloon setae in the Riodinidae (Lepidoptera). Ann. Entomol. Soc. Am. 97(2): 310-321.

Hesbacher, S., I Giez, G. Embacher, K Fiedler, W. Max, A. Trawoger, R. Turk, O. L. Lange and P. Proksch. 1995. Sequestration of lichen compounds by lichen-feeding members of the Arctiidae (Lepidoptera). J. Chem Ecol. 21(12): 2079-2089.

Hilker, M., U. Eschbach and K. Dettner. 1992. Occurence of anthraquinones in eggs and larvae of several Galerucinae (Coeloptera: Chrysomelidae). Naturwissenschafften 79: 271-274.

Holloway, J.D. 2002. The moths of Borneo: family Arctiidae, subfamily Lithosiinae. The Malayan Journal 55:279-469.

Jacobson, N. L. and S.J. Weller. 2001. A cladistics study of the Arctiidae (Lepidoptera) using characters of immature and adults. Thomas Say Monographs, Entomological Society of America.

Kageyama, A. and S. Sugiura. 2016. Caterpillar hairs as an anti-parasitoid defence. Sci. Nat. 103:86 (7 pages).

Knowlton, C.B., Jr. 1967. A revision of the species of Cisthene known to occur north of the Mexican Border (Lepidoptera: Arctiidae: Lithosiinae). Trans. Am. Entomol. 93:41- 100 (33 figures).

Komatsu, T. and T. Itino. 2014. Moth caterpillar solicits homopteran honeydew. Sci. Rep. 4: 1-3.

Lafontaine J.D., J.G. Franclemont and D.C. Ferguson. 1982. Classification and life history of Acsala anomala (Arctiidae: Lithosiinae). J. Lepid. Soc. 36: 218-226.

Lawrey, J.D. 1993. Lichens as monitors of pollution elements at permanent sites in Maryland and Virginia. The Bryologist 96: 339-341.

Lewis, J.A. 2014. A new species of Crambidia (Lepidoptera: Noctuidae: Arctiinae: Lithosiini) from the Eastern United States. Proc. Entomol. Soc. Wash. 116(4):394-401.

McCabe, T.L. 1981. Clemensia albata, an algal feeding arctiid. J. Lepid. Soc. 35(1): 34- 40.

Menzel, F., N. Bluthgen, T. Tolasch, J. Conrad, U. Beifub, T. Beuerle and T. Schmitt. 2013. Crematoenones—a novel substance class exhibited by ants functions as appeasement signal. Front. Zool.10:32.

Mitchell, M.E. 2105. Autonomy’s long shadow: a report on issues concerning lichen classification, 1870-1981. Huntia 15:87-104.

Moskowitz, D. and L. Haramaty. 2012. Fungi feeding by the Agreeable Tiger moth (Spilosoma congrua Walker) (Erebidae: Arctiinae). J. Lepid. Soc. 66(4):230-232.

Nakano, R. T., A. Takanashi, N. Surlykke, N. Skals and Y. Ishikawa. 2003. Evolution of deceptive and true courtship songs in moths. Sci. Rep. (3): 1-8.

Pankoke H., M.D. Bowers and S. Dobler. 2012. The interplay between toxin-releasing B- glucosidase and plant iridoid glycosides impairs larval development in a generalist caterpillar, Grammia incorrupta (Arctiidae). Insect Biochem. Molec. 42: 426-434.

Powell, J. A. and P.A. Opler. 2009. Moths of Western North America. University of California Press.

Rankovic, B. and M. Kosanic. 2015. Lichens as a potential source of bioactive secondary metabolites. In Rankovic, B. (ed) Lichen secondary metabolites: bioactive properties and pharmaceutical potential. Springer, Switzerland, pp. 1-26.

Rawlins, J.E. 1984. Mycophagy in Lepidoptera. In Fungus Insect Relationships, ed. Q .Wheelerand M. Blackwell, Columbia University Press, NY: pp. 382-483.

Sanchez-Pena, S.R., D.R.Davis and U.G. Mueller. 2003. A gregarious, mycophagous, myrmecophilous moth, Amydria anceps Walshingham (Lepidoptera: Acrolophidae),

62 living in Atta mexicana F.Smith (Hymenoptera: Formicidae) spent fungal culture accumulations. Proc. Entomol. Soc. Wash. 105(1): 186-194.

Schmidt, B.C. and P.A. Opler. 2008. Revised checklist of the tiger moths of the Continental United States and Canada. Zootaxa 1677: 1-23.

Scott, C. H., J.M. Zaspel, P. Chialvo and S. J. Weller. 2014. A preliminary molecular phylogenetic assessment of the lichen moths (Lepidoptera: Erebidae: Arctiinae: Lithosiini) with comments on palatability and chemical sequestration. Syst. Entomol. 39: 286-303.

Scott Chialvo, C.H., P. Chailvo, J.D. Holland, T. J. Anderson. J. W. Breinholt, A.Y. Kawahara, X. Zhou, S. Liu and J.M. Zaspel. 2018. A phylogenomic analysis of lichen- feeding tiger moths uncovers evolutionary origins of host chemical sequestration. Mol. Phylogenet. Evol. 121:23-34.

Simonson, S. 1996. Lichens and Lichen-feeding moths as bioindicators of air pollution in the Rocky Mountain front range. www.interlibrary.narod.ru/GenCat/GenCat.Scient/400000011.htm.

Singer, M.S., Y. Carriere, C. Theuring and T. Hartmann. 2004. Disentangling food quality from resistance against parasitioids: diet choice by a generalist caterpillar. Am. Nat. 164(3): 423-429.

Singer, M.S. and J.O. Stireman III. 2005. The tri-trophic niche concept and adaptive radiation of phytophagous insects. Ecol. Lett. 8:1247-1255.

Singer M. S., K.C. Mace and E.A. Bernays. 2009. Self-medication as adaptive plasticity: Increase ingestion of plant toxins by parasitized caterpillars. PLoS One 4, e4796.

Singer, M. S., I.H. Lichter-Marck, T. E. Farkas, E. Aaron, K.D. Whitney and K.A. Mooney. 2014. Herbivore diet breadth mediates the cascading effect of carnivores in food webs. Proc. Natl. Acad. Sci. USA 111: 9521-9526.

Snodgrass, R.E. 1935. Principles of insect morphology, McGraw-Hill Book Company, 515-523.

Spribille, T., V. Touvinen, P. Resl, D. Vanderpool, H. Wolinski, M.C. Aime, K. Schneider, E. Stabentheiner, M. Toome-Heller, G. Thor, H. Mayrhofer, H. Johannesson and J. P. McCutcheon. 2016. Basidomycete yeasts in the cortex of ascomycete macrolichens. Science 353 (6298): 488-492.

Stehr, F.W. 1987. Immature Insects, Kendall/Hunt Publishing, 538 and 293. Tesky, D. 2001. Biological Indicators. http://www.suite101.com/article.cfm/ecology/57858. 63

Thomas, J.A. and J. Settele. 2004. Evolutionary Biology: butterfly mimics of ants. Nature 432 (7015): 283-4.

Thomas, J., K. Schönrogge, S. Bonelli, F. Barbero and E. Balletto. 2010. Corruption of ant acoustical signals by mimetic social parasites. Commun. Integr. Biol. 3 (2): 169–171.

Wagner, D.L. 2005. Caterpillars of eastern North America: a guide to identification and natural history. Princeton University Press, Princeton, N.J.

Wagner, D. L., J. Rota and T. L. McCabe. 2008. Larva of Abablemma (Noctuidae) with notes on algivory and lichenivory in macrolepidoptera. Ann. Entomol Soc. Am. 101(1): 1-13.

Wilkinson, D. 2018. What is a lichen? Changing ideas on the lichen symbiosis. British Wildlife June: 351-357.

Associated Table and Figure Legends:

Table 1: Ant interaction table listing species, localities and results where interaction studies were conducted.

Figure 1: Screen grabs from a video of foraging Myrmecocystus sp. ants attacking and carrying away a larva of Trichoplusia ni (control) while ignoring a larva of the lithosiine Hypoprepia inculta placed next to it. Video was shot in Peppersauce Canyon, Santa Catalina, Mts., Pinal Co., Arizona.

Figure 2: Larvae of some of the lithosiines and controls used in this study, along with the predatory ant species with which interactions were observed: A) Arachnis zuni B) Estigmene albida C) Trichoplusia ni D) Hypercompe permaculata E) Cisthene tenuifascia F) Eudesmia arida G) Crambidia myrlosea H) Hypoprepia inculta.

Figure 3: Histological sections through the verrucae or pores of control and lithosiine larvae, stained with hematoxylin and eosin (H&E). Red arrows indicate presumed glandular cells, while the white arrows point to an eosinophilic pink substance present in the lithosiines but not in the control. Figure A shows a section through the verrucae of E. albida at 20X and B at 40X; trichogen cells are the largest cells, while the glandular cells are smaller and more simple than those of lithosiines with no pink substance present. Figures C and D show the complex dorsal gland cells of Lycomorpha regulus containing the pink substance (both 40X). Figure E shows the bulbous-based setae of C. cephalica containing a granular pink substance inside (40X). Figure F shows the dorsal gland cells of Eudesmia arida containing a pink substance (40X).

Figure 4: Histological sections through the verrucae or pores of control and lithosiine larvae, stained with Periodic Acid Schiff’s (PAS) stain (top row) and Mucicarmine stain (bottom row). Figures A and B show the control E. albida exhibiting a lack of reactivity with both PAS and Mucicarmine. Figures C and D show the dorsal gland of Bruceia hubbardi exhibiting strong reactivity to PAS (C), but negative to mucicarmine (D). Figures E and F show the glands and hollow setae of C. cephalica showing reactivity with PAS (E) but not with mucicarmine (F). All images 40X except E, which was taken at 20X to show the glandular cell staining.

Figure 5: Crambidia cephalica larva (5A) and histological sections, showing the distinct hollow setae. Figures 5B-D orient the viewer to the location of the first dorsal glands relative to the head of the larva. Bulbous-based, hollow setae containing a pink granular substance seem to be unique to this genus: 5E shows C. cephalica setae at 40X , 5F shows the setae of C. myrlosea at 40X. Red arrows highlight the pink substance and the glandular cells that presumably produce it.

APPENDIX C.

A new species of Hypoprepia from the mountains of central Arizona (Lepidoptera, Erebidae, Arctiinae, Lithosiini)

John D. Palting 1, Douglas C. Ferguson2 and Wendy Moore3

Graduate Interdisciplinary Program in Entomology and Insect Science, University of Arizona, Forbes 410, Tucson, AZ, 85721-0036, USA

1 Graduate Interdisciplinary Program in Entomology and Insect Science, University of Arizona, Tucson, Arizona, 85721-0036, USA,

2 Systematic Entomology Laboratory, PSI, Agricultural Research Service, U.S. Department of Agriculture, c/o Smithsonian Institution, Washington, D.C., 20013- 7012, USA,

3 Department of Entomology, University of Arizona, Forbes 410, Tucson, Arizona, 85721-0036, USA, Corresponding author

Email: [email protected]

Zookeys. 2018; (788): 19–38. Published online 2018 Oct 8. doi: 10.3897/zookeys.788.26885

A new species of Hypoprepia from the mountains of central Arizona (Lepidoptera, Erebidae, Arctiinae, Lithosiini)

John Douglas Palting,1,3 Douglas C. Ferguson,2 and Wendy Moore3

1 Graduate Interdisciplinary Program in Entomology and Insect Science, University of Arizona, Tucson, Arizona, 85721-0036, USA, 2 Systematic Entomology Laboratory, PSI, Agricultural Research Service, U.S. Department of Agriculture, c/o Smithsonian Institution, Washington, D.C., 20013-7012, USA, 3 Department of Entomology, University of Arizona, Forbes 410, Tucson, Arizona, 85721-0036, USA, Corresponding author. John Douglas Palting: [email protected] Corresponding author: John Douglas Palting ([email protected]) Academic editor: C. Schmidt

Received 2018 May 22; Accepted 2018 Jul 9.

This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Abstract

A new firefly-mimicking lichen moth of the genus Hypoprepia, H.lampyroides Palting & Ferguson, sp. n., is described from the mountains of east-central Arizona and the Sierra Madre Occidental of Mexico. Hypoprepia Hübner, 1831 is a North American genus of lithosiine tiger moths, previously consisting of five species: H.fucosa Hübner, 1831 and H.miniata (Kirby, 1837), both of eastern and central North America; H.cadaverosa Strecker, 1878 from the Rocky Mountains into New Mexico and west Texas; H.inculta H. Edwards, 1882, a widespread western USA species and H.muelleri Dyar, 1907 from the vicinity of Mexico City. The latter is herein synonymized under H.inculta (= H.muellerisyn. n.), resulting in the total number of taxa in the genus unchanged at five.

Keywords: Lithosiini , Madrean fauna, mimicry, Sky Islands

Introduction

The mountains of southeastern Arizona and northeastern Sonora are well known as a biological blending zone between the fauna of the Rocky Mountains to the north and Mexico’s Sierra Madre Occidental to the south. Positioned between these two great mountain ranges, the Sky Island Region contains a series of smaller mountain ranges that have oak and pine at higher elevations, each range being separated from one another by expanses of drier grasslands and desert. Sky Island ranges often harbor relict populations of plants and animals that suggest that in the distant past, both geology and climate allowed connections between the flora and fauna of the Rockies and the Sierra Madre (Warshall 1995). Examples among Lepidoptera of this connection include Chiricahuamultidentata (Guedet, 1941) and Chiricahualichenaria Ferris, 2010 (Geometridae, Ennominae), known in the US only from the highest elevations of the Chiricahua Mountains in SE Arizona, with the next nearest recorded population being in El Salto, Durango, nearly 900 miles to the south. A similarly striking disjunct population occurs with Nemoriasplendidaria (Grossbeck, 1910) (Geometridae, Geometrinae) known only from the top of the Huachuca Mountains, Arizona in the US with the nearest Mexico records also being from Durango. Alexiclesaspersa Grote, 1883 (Erebidae, Arctiinae) occurs sporadically from Colorado to several places in the White Mountains of central Arizona, adjacent parts of New Mexico, and not again until the top of the Sierra Madre in the vicinity of Yecora, Sonora, Mexico, skipping the Sky Island ranges entirely. Other rare US Lepidoptera that exhibit similar but less dramatically disjunct distributions include the lasiocampids Caloeciaentima Franclemont, 1973 and C.juvenalis (Barnes & McDunnough, 1911) (Lasiocampidae, Lasiocampinae), C.entima known in the US only from the high elevations of the Chiricahuas, and C.juvenalis only from the Chiricahuas and Huachucas, with spotty distributions in the Mexican state of Sonora (Sierra Mariquita, Sierra del Tigre and 72

Yecora). Agyllaseptentrionalis Barnes & McDunnough, 1911 (Erebidae, Arctiinae, Lithosiini) is also known from isolated high-elevation populations in the Chiricahua and Huachuca Mountains, separated from the nearest Sierra Madre populations in Yecora, Sonora by 400 miles. These are just a few of many examples among Lepidoptera species with relict disjunct distributions indicating an historical Rocky Mountain-Madrean connection in this region. We can now add another rare species of Lepidoptera from Arizona to the body of evidence supporting this past faunal connectivity. The moth was first noticed by the late Ron Leuschner, who collected a specimen on the door of a rental cabin in the hamlet of Greer, White Mountains, Arizona in 1988. Leuschner sent this specimen to Ferguson, who, prior to his death in 2002, recognized it as new and started to describe it based upon this specimen and two additional specimens he located in collections. Ferguson had dissected and made some comments on the male genitalia, but had not examined the internal structures of the female. In June 2017, JDP and Ray Nagle had the good fortune of collecting more than 30 specimens of this new species along Highway 191 in the vicinity of Rose Peak, Blue Ridge Primitive Area, Greenlee County, Arizona. Flying sympatrically with Hypoprepiainculta Edwards, 1882 was the similar-looking, but much larger bodied and more boldly colored, H.inculta look-alike (Figs 1–5). Finally, here was the almost mythical moth that Leuschner had found nearly 30 years prior in Greer. Its similarity to H.inculta (Figure 6), combined with narrow endemicity and an early flight period just prior to or at the onset of the summer rains, may account for the paucity of records of this new species. It appears to fly throughout the night, with new individuals showing up on the sheet with regularity until dawn, outnumbered by H.inculta by approximately 4 : 1. Most of the specimens collected were males, but two females of the new species were collected and kept live for ova, allowing for the larvae to be reared and photographed for the first time.

F gures 1–.2

Two views of living male Hypoprepia lampyroides.

Fi gures 5–6.

Adult male 5 Hypoprepia lampyroides and 6 Hypoprepia. inculta.

Other noteworthy species flying alongside the Hypoprepia were Nadatagibbosa (JE Smith, 1797) (Notodontidae, Phalerinae) and Spilosomavirginica (Fabricius, 1798) (Erebidae, Arctiinae, Arctiini), both common northern and eastern species, but at the extreme southern limit of their ranges here, as well as Apantesisf-pallida (Strecker, 1878) (Erebidae, Arctiinae, Arctiini), a primarily Rocky Mountain species, very rare this far southwest. Also present was the strikingly beautiful Erastriaviridiruferia (Neumoegen, 1881) (Geometridae, Ennominae), another Madrean species that occurs in central Arizona, with sporadic records from the Sky Islands Region through the Sierra Madre proper, where it occurs regularly at mid to high elevations.

Methods and materials

Phylogenetic analysis Total genomic DNA was extracted from the right middle leg of each voucher specimen using the Qiagen DNeasy Blood and Tissue Kit (Qiagen, Valencia, CA), according to manufacturer suggested protocol. The barcoding region of the mitochondrial gene cytochrome oxidase subunit 1 (COI) was PCR amplified with primers LCO1490 and HCO2198 (Hebert et al. 2003). PCR products were cleaned, quantified, normalized, and sequenced in both directions at the University of Arizona’s Genomic and Technology Core Facility using a 3730 or 3730XL Applied Biosystems automatic sequencer. Chromatograms were assembled and initial base calls were made for each gene with Phred (Green and Ewing 2002) and Phrap (Green 1999) as orchestrated by Mesquite Ver. 3.4 (Maddison and Maddison

2018) and Chromaseq vers. 1.3 (Maddison and Maddison 2017). Final base calls were made in Mesquite and ambiguous bases were designated by a standard ambiguity code. Resulting sequences were deposited in GenBank (Table 1). Previously published COI sequences of Hypoprepia and all other members of the tribe Lithosiini were downloaded from GenBank and the Barcode of Life Database (Table 1). All 500 sequences were assembled into a single matrix and were aligned using MAFFT vers. 7 (Katoh and Standley 2013). The aligned matrix was partitioned by codon position, with each codon position allowed to have independent parameter values for the model of evolution. Maximum likelihood (ML) heuristic searches were conducted using RAxML 8.0.9 (Stamatakis 2014) under the GTR+gamma model of evolution on CIPRES Science Gateway portal (Miller et al. 2010). 500 search replicates were conducted to find the maximum likelihood tree.

Table 1. GenBank/BOLD accession number of the species.

Species name GenBank/BOLD Accession Number Outgroup Abrochocis esperanza KC571047.1 Balbura dorsisigna KC571053.1 Balbura intervenata KC571052.1 Chrysochlorosia magnifica KC571057.1 Cisthene angelus BBLOE1648-12 Cisthene barnesii ABLCW009-10 Cisthene barnesii LMEM919-09 Cisthene barnesii RDNMF900-08 Cisthene deserta ABLCW126-10 Cisthene dorsimacula RDNMF903-08 Cisthene faustinula LOCBC003-06 Cisthene juanita IAWL658-09 Cisthene kentuckiensis HKONS224-08 Cisthene liberomacula LOCBC697-06 Cisthene martini LMEM065-09 Cisthene packardii LSUSA097-06 Cisthene perrosea ABLCW128-10 Cisthene picta LPOKA060-08 Cisthene plumbea KC571059.1 Cisthene polyzona BLPDD935-09 Cisthene sp. LPYPC028-08 Cisthene sp. LPYPC119-08 Cisthene subjecta HKONS229-08 Cisthene subrufa LPYPB681-08 Cisthene subrufa LPYPC078-08 Cisthene tenuifascia BBLSW086-09

Cisthene unifascia ABLCW140-10 Dolichesia falsimonia KC571062.1 Gardinia anopla KC571075.1 Lycomorphodes correbioides KC571088.1 Lycomorphodes sordida KC571089.1 Talara cara KC571098.1 Talara lepida KC571099.1 Talara nr. mona KC571100.1 Ingroup Hypoprepia cadaverosa KC571080.1 Hypoprepia cadaverosa MF922743.1 Hypoprepia cadaverosa MF923063.1 Hypoprepia cadaverosa MF923535.1 Hypoprepia cadaverosa MF923758.1 Species name GenBank/BOLD Accession Number Hypoprepia cadaverosa MF923893.1 Hypoprepia cadaverosa MF924076.1 Hypoprepia fucosa MF923771.1 Hypoprepia fucosa MF924037.1 Hypoprepia fucosa KC571078.1 Hypoprepia fucosa tricolor KC571079.1 Hypoprepia inculta ABLCW242-10 Hypoprepia inculta CMAZA783-10 Hypoprepia inculta 4170 MH337839 Hypoprepia inculta RDNMG037-08 Hypoprepia inculta 3259 MH337840 Hypoprepia inculta ABLCW240-10 Hypoprepia inculta ABLCW241-10 Hypoprepia inculta ABLCW244-10 Hypoprepia inculta ABLCW245-10 Hypoprepia inculta RDNME352-07 Hypoprepia inculta MF923496.1 Hypoprepia inculta 3573 MH337833 Hypoprepia inculta 3574 MH337841 Hypoprepia inculta ABLCW071-10 Hypoprepia inculta ABLCW056-10 Hypoprepia inculta ABLCW055-10 Hypoprepia lampyroides sp. n. 3566 MH337834 Hypoprepia lampyroides sp. n. 3567 MH337835 Hypoprepia lampyroides sp. n. 3568 MH337836

Hypoprepia lampyroides sp. n. 3569 MH337837 Hypoprepia lampyroides sp. n. 3570 MH337838 Hypoprepia miniata BBLOB1474-11 Hypoprepia miniata LBCC462-05 Hypoprepia miniata LBCC769-05 Hypoprepia miniata LGSMB301-05 Hypoprepia miniata LGSMB302-05 Hypoprepia miniata LOFLB682-06 Hypoprepia miniata LOFLC311-06 Hypoprepia sp. KT706007.1

We identified the closest relatives of Hypoprepia in the resulting maximum likelihood tree, selected these as our outgroup taxa, and re-ran the ML heuristic searches (as described above) on the smaller matrix of 73 taxa. Clade support was conducted using rapid bootstrapping with a subsequent ML search and letting RAxML halt bootstrapping automatically (using MRE- based bootstopping criterion).

Taxonomic treatment Genitalic preparations were made following the methods of Jaeger (2017) by staff at the CNC. Genitalia were slide-mounted using Euparal and photographed with a Leica DFC450 camera, Leica Application Suite 4.8 with a Leica M205C stereo microscope, and processed in Adobe Photoshop. Photographs of the pinned adult male and female paratypes were made using Visionary Digital Imaging System with a Canon EOS 7D digital camera and Canon MP-E65mm f/2.8 1–5× lens. Multiple images were combined using Zerene Stacker version 1.04. Repository abbreviations are as follows: CNC Canadian National Collection of Insects, Arachnids and Nematodes, Ottawa, ON USNM National Museum of Natural History (formerly United States National Museum), Washington, DC UAIC University of Arizona Insect Collection, Tucson, AZ UNAM Universidad Nacional Autonoma de Mexico, Mexico, DF DEBC Don E. Bowman Collection, Golden, Colorado JDPC John D. Palting Collection, Tucson, AZ RBNC Ray B. Nagle Collection, Tucson, AZ

Results and discussion

Phylogenetic analysis

Our molecular phylogenetic analyses reveal strong support for the monophyly of Hypoprepia and a close relationship between H.inculta and H.lampyroides (Figure 15). It is noteworthy that H.lampyroides is recovered as a single well-supported clade. However, recognizing this clade as a new species renders H.inculta paraphyletic in the COI gene tree. Focusing on gene tree topology alone, one might decide not to recognize H.lampyroides as a new species, but rather view it as a unique population of H.inculta. However, we contend that these are two valid species since specimens of both occur in strict sympatry at the Rose Peak locality and they are easy to distinguish morphologically by size, wing color, antennal structure, as well as the form of both male and female genitalia. We predict that the 657 base pair fragment of COI does not contain enough phylogenetic information to infer the Hypoprepia species tree with accuracy. This is a common result of phylogenetic analyses of the COI barcoding region within some Lepidoptera (Beltran et al. 2002, Wiemers and Fiedler 2007) and within noctuoids in particular (Schmidt and Sperling 2008, Zahiri et al. 2017). The lack of reciprocal monophyly among species in the tree could also result from ongoing hybridization events resulting in mtDNA introgression, and/or incomplete lineage sorting (Funk and Omland 2003).

Fi gure 15.

Maximum-likelihood tree of Hypoprepia species based on COI. Bootstrap values are reported on the branches subtending nodes with a support value greater than 50.

The phylogeny also suggests that Hypoprepia is in need of further revisionary work, particularly with respect to species boundaries between H.miniata and H.cadaverosa. These fully allopatric species (H.miniata common in the eastern US and H.cadaverosa common in the western US) look quite different from one another. Even so, several authors have suggested that they should be synonymized (Zahiri et al. 2017, Powell and Opler 2009). Given this and that both nominate forms are polyphyletic in our tree, it seems likely that these forms represent regional variation in the same species. Future investigations comparing their anatomy and phylogenetic analysis of additional genes, particularly nuclear genes, will help resolve this taxonomic question.

Taxonomic treatment

Hypoprepia lampyroides Keywords: Animalia, Lepidoptera, Arctiidae Palting & Ferguson sp. n. http://zoobank.org/746F6BFE-47B9-4E47-832B-F75A954A75C2

Figures 3–4. Adults of Hypoprepia lampyroides. 3 male and 4 female.

Fi gures 7–8.

Lateral view of male antennae: 7 Hypoprepiainculta 8Hypoprepia lampyroides

Fi gures 9–12 .

9 –10 Male genitalia of Hypoprepia lampyroides 11 – 12 Male genitalia of Hypoprepia inculta .

Fi gures 13–14.

Female genitalia of 13 Hypoprepiainculta and 14 Hypoprepialampyroides .

Fi gures 17–18.

Larvae of Hypoprepialampyroides . 17 Living last instar larva and 18 Penultimate instar larvae, preserved.

Fi gures 19–20.

Last instar larvae of 19 Hypoprepialampyroides and 20 Hypoprepiacadave rosa, preserved.

Type material. Holotype ♂. Arizona:, [Apache Co.], White Mountains, Greer, 8,200 ft., 4–5 July 1988, R.H. Leuschner [USNM]. Paratypes 32♂ 3♀. Arizona: Santa Cruz Co., 8.5 mi. SE of Patagonia, Harshaw Canyon, 4,850 ft., 24 July 1998, D.E. Bowman, 1♀ [DEBC]; 29♂ 2♀, Greenlee Co., Blue Ridge Primitive Wilderness, US Hwy 191, vicinity of Rose Peak, 33°26'N 109°22'W, 8084 ft., 19 June 2017 [specimens distributed between JDPC (8♂), UAIC (6♂), CNC (5♂ 1♀), USNM (8♂ 1♀), UNAM (2♂), and RBNC (1♀)]. Mexico: 10 mi. W. of El Salto, Durango, 9,000 ft, 13 June 1964, J.E.H. Martin, 1♂ [CNC]; 2♂, Sonora, Mesa del Campanero, Barranca El Salto, elevation 6561’, Municipio de Yecora, , 2 July 2013, J. Palting [JDPC, UNAM].

Etymology. The specific epithet lampyroides means “like Lampyra” referring to this species’ remarkable mimicry of a sympatric lampyrid beetle species, as discussed below.

Diagnosis. Hypoprepialampyroides (Figs 1–5) occurs sympatrically with H.inculta (Figure 6) and is easily distinguishable externally by its larger size; unmarked blackish forewings; brighter more extensively pink hindwings; somewhat different palpi; and different male antennae that more nearly resemble those of H.cadaverosa. The antenna differs structurally from that of H.inculta (Figure 7), which exhibit squarish, closely set segments (flagellomeres) with little space between them. The laminae of the antennal segments of H.lampyroides (Figure 8) are conspicuously raised, tapered, and appear farther apart when viewed laterally. The antenna of H.lampyroides is more like that of H.cadaverosa, a species that it does not otherwise resemble.

Internally, the male H.lampyroides (Figs 9–10) differs from H.inculta (Figs 11–12) in the form of the spinose cornutus on the dorsal vesica chamber, which is apically elongated in H.lampyroides versus sawblade-like in H.inculta. Hypoprepialampyroides males always have three well-developed spinose cornuti (Figure 10), whereas the left ventrolateral cornutus (adjacent to the ductus) is often missing or reduced in H.inculta (Figure 12). The shape of the valve and tegumen is stouter and less elongate than in H.inculta. In females, the corpus bursae is globose (Figure 13) versus irregularly elongate in H.inculta (Figure 14), with four instead of three signa, the right-ventral signa possessing smaller spines than the corresponding right-ventral signa in H.inculta.

Description. Sexes similar externally (Figs 3–4), but females with pink area on dorsal hindwing not quite as extensive, and with boundary between pink part and dark outer border more diffuse. Head. Vestiture of frons and vertex dark grey; labial palpus dark grey, upturned, slightly larger and longer d d than that of H.inculta, terminal (3rd) segment 1.25 × longer than 2nd; eye large, protuberant, more clearly exceeding a half sphere than those of the other Hypoprepia species; male antenna blackish, laminate, densely clothed with short setae beneath and with a few longer setae protruding sublaterally along the sides; female antenna simple, flagelliform. Thorax. Dark brown or dark gray except for the tegula, which is mostly bright pink, matching basal spot of forewing; patagium blackish; legs entirely blackish or dark gray. Abdomen. Vestiture gray, flushed with pink basally and terminally, ventrum entirely blackish or dark gray, except for some pink scales at distal end (H.inculta also may have a pink-tipped abdomen); ventral sternite A8 of males with reinforced, sclerotized rim-like anterior margin, but no pockets, coremata or androconial setae are visible on segments A7–A8. In females, pleurite of A7 with membranous but thick pockets, appearing somewhat rugose and more heavily sclerotized than surrounding integument. Forewing. Uniformly dark brown to charcoal gray, appearing blackish, unmarked except for a pink spot at base next to thorax, and lacking the pale streak on basal half of cubital vein seen in many H.inculta; male forewing length 17–20 mm, mean 17.5 mm (n = 6); female average forewing length 15.8 mm (n = 2) (usually 12–15 mm for both sexes of H.inculta). Hindwing. Hindwing pink, with a uniform, dark-gray costal and outer margin, ending just before anal angle; fringes gray to dark brown; ventrum of both wings similar to dorsum but slightly paler, and with more diffuse boundaries between pink and gray areas. Male genitalia (Figs 9–10) Generally similar to those of H.inculta; uncus cylindrical, flattened slightly laterally, oval in cross section, 8.8 × longer than wide; apex formed by slightly ventrally-curved, fine spine; basal two thirds with sparse, latero-basally directed setae; tegumen well-defined, rounded quadrate and dorsoventrally flattened with a slight constriction at juncture with vinculum; dorsal surface convex and bubble-like on either side of midline, densely covered in setal sockets distally; valve without clasper or process, slightly constricted basally, distal half rounded triangular, apex a rounded point, with short, broad somewhat spine-like setae along distal third of costal margin; sacculus not differentiated from remainder of valve, with a slight subbasal, setose bulge; juxta indistinct, forming a dorsally emarginate rounded-rectangular transverse plate, approximately 4 × wider than long; phallus a straight, simple cylinder, 2.5 × longer than wide, coecum

88 lacking; vesica consisting of three adjoining, globose chambers, the phallus appearing more or less as a tripartite club when vesica expanded; ventral chamber adjacent to ductus ejaculatorius, with additional lobe-like diverticulum, and with a spinose crest-like patch apically; laterodorsal chambers also with spinose crests. Female genitalia. (Figure 13) Papillae anales broadly diamond-shaped, sparsely setose; anterior and posterior apophysis relatively short, approximately equal in length to width of papillae; postvaginal aree with triangular scerlotization; ductus bursae short and broad, 1.5 × wider than long, highly flattened dorsoventrally and recurved ventrally; corpus bursae relatively small and globose, diameter 1.5–2 × width of ductus; signa consisting of two pairs of spinose straps, situated laterally near junction of ductus; cervix bursae situated right caudo-laterally and recurving left across ventral side of ductus.

Biology and distribution. The brown eggs of H.lampyroides (Figure 16) were laid in small clusters inside a vial containing a piece of paper, and under magnification exhibit the “hammered copper” surface texture typical of lithosiine ova. These hatched after 14 days, the larvae being light yellowish initially then darkening as they fed. The larval stages are basically dark brown and unmarked throughout their development. Like other Hypoprepia (and other members of the subtribe Cisthenini) the larvae lack true verrucae (Bendib and Minet 1999) and instead have structures technically known as panniculae (Stehr 1987) with just one or two, stiff, black setae emerging from each (Figs 17–19). The larva is similar to H.inculta, which is also predominantly brown with black setae, while H.cadaverosa, reared by JDP at the same time as H.lampyroides, are marked with bright yellow bands (Figure 20). The larval mandible, dissected (Figure 21), shows the enlarged molar region found in other lithosiines. This feature has been suggested as a synapomorphy for the Lithosiini (Bendib and Minet 1999) and is believed to be related to their lichen diet. The larvae fed successfully on a mixed population of lichens obtained by shaving bark off oak trees, and developed through six instars into a caterpillar large enough to pupate. Unfortunately, lab conditions failed to yield a successful pupation, and the larvae eventually died. It is likely that H.lampyroides over-winter as a fully mature larva, pupating in the spring and emerging in early summer.

Fi gure 16.

Eggs of Hypoprepia lampyroides , approximately 20×.

Fi gure 21.

Mandible of last instar Hypoprepia lampyroides , approximately 20×.

The striking resemblance of this moth at rest (Figs 1–2) to a common southwest species of firefly, Ellychniacorrusca Linnaeus, 1767 (Coleoptera: Lampyridae) (Figure 23), points to them being part of a mimicry ring, which also includes another common montane beetle, Discodonbipunctatum Schaeffer, 1908 (Coleoptera: Cantharidae) (Figure 22). Ellychniacorrusca was common during the day in the Rose Peak area, and the bright pink markings on its pronotal region closely match the pink markings at the base of the forewing in H.lampyroides, likely affording the resting moths protection should a bird or other predator come upon them. Lampyrids are known to be chemically protected and distasteful to birds, but unlike most familiar nocturnal fireflies, Ellychnia lacks an abdominal light and is primarily diurnal. Research on sequestration of lichen polyphenolic compounds by other lithosiine arctiids (Hesbacher et al. 1995, Conner 2009, Scott et al. 2014) suggests that H.lampyroides itself has some chemical protection, thus the mimicry between these organisms is likely Mullerian. H.inculta is also likely part of this mimicry ring, although with its smaller size, dull pink markings, and grey wing color, it is a much less dramatic match to Ellychnia than H.lampyroides.

Fi gures 22–23.

Mullerian mimicry with Coleoptera . 22 Discodon bipunctatum ( Cantharidae) 23 Ellychniacorrusca ( Lampyridae).

Hypoprepialampyroides is known from over 30 specimens collected in Arizona, two specimens from Yecora, Sonora, Mexico and one from Durango, Mexico (Figure 24).

Fi g ure 24 . Range map of Hypoprepia lampyroides.

Remarks. When examining the nearest relatives of H.lampyroides, Ferguson found that H.inculta from the southwestern United States is indistinguishable from the type material of H.muelleri Dyar, described from the vicinity of Mexico City, H.muelleri tends to have darker, more grayish hindwings, although in some H.inculta from Arizona they are equally grayish. Such a difference by itself is hardly significant. Unfortunately, fresh collected material of H.muelleri was not available for molecular analysis, but Ferguson’s conclusion based on his examination of the type material results in the following taxonomic change: Hypoprepiamuelleri Dyar, = Hypoprepiainculta Henry Edwards, syn. n. This extends the known range of H.inculta from as far north as Utah to the vicinity of Mexico City. H.muelleri had previously been the only member of the genus found exclusively in Mexico. Ferguson found the Durango, Mexico specimen of H.lampyroides among unidentified arctiids from the Canadian National Collection. The region of El Salto, Durango, where it was collected, is mesic, conifer-dominated forest similar to that around Greer, Rose Peak, and Yecora, Sonora. The Harshaw specimen, a female, was collected by Don Bowman of Golden, Colorado and sent to Ferguson for identification. The Harshaw region is rather dry mid- elevation oak woodland/mesquite grassland, very unlike where all the other specimens of this moth have been collected.

Acknowledgments

JDP is greatly indebted to Chris Schmidt for sending him Ferguson’s unfinished manuscript on this moth, for extensive help with the genitalic descriptions and for many helpful suggestions to improve the paper; to Christi Jaeger for the genitalic dissections and photos; to Ray Nagle for many years of mentoring and friendship, reviewing the manuscript and for photographing the adult and larvae of H.lampyroides; to Charles “Chip” Hedgcock for the exquisite portraits of the new taxa and to Margarethe Brummermann for allowing the use of her combined image of Ellychniacorrusca and Discodonbipunctatum to illustrate the similarity in color pattern between these taxa and H.lampyroides. This work is in partial fulfillment of JDP’s Doctorate of Philosophy degree in the Graduate Interdisciplinary Program in Entomology and Insect Science at the University of Arizona and is the product of the Arizona Sky Island Arthropod Project (ASAP) based in WM’s laboratory. JDP would like to thank extend special thanks to WM for her guidance in writing this manuscript and for her support and mentoring in the molecular systematics of lithosiines.

Citation Palting JD, Ferguson DC, Moore W (2018) A new species of Hypoprepia from the mountains of central Arizona (Lepidoptera, Erebidae, Arctiinae, Lithosiini). In: Schmidt BC, Lafontaine JD (Eds) Contributions to the systematics of New World macro-moths VII. ZooKeys 788: 19– 38. https://doi.org/10.3897/zookeys.788.26885

References

Beltran M, Jiggins CD, Bull V, Linares M, Mallet J, Owen McMillan W, Bermingham E. (2002) Phylogenetic discordance at the species boundary: comparative gene genealogies among rapidly radiating Heliconius butterflies. Molecular Biology and Evolution 19(12): 2176–2190. 10.1093/oxfordjournals.molbev.a004042 [PubMed] [CrossRef] [Google Scholar] Bendib A, Minet J. (1999) Lithosiinae main lineages. Annales de Societe Entomologique de France 35(3/4): 241–263. Conner WL[Ed.] (2009) Tiger Moths and Woolly Bears: Behavior, Ecology and Evolution of the Arctiidae. Oxford University Press. Funk DJ, Omland KE. (2003) Species-level paraphyly and polyphyly: frequency, causes, and consequences, with insights from animal mitochondrial DNA. Annual Review of Ecology and Evolutionary Systematics.34: 397–423. 10.1146/annurev.ecolsys.34.011802.132421 [CrossRef] Green P. (1999) Phrap. Version 0.990329 http://phrap.org Green P, Ewing B. (2002) Phred. Version 0.020425c. http://phrap.org Hebert PDN, Cywinska A, Ball SL, deWaard JR. (2003) Biological identifications through DNA barcodes. Proceedings of the Royal Society B 270(1512). 10.1098/rspb.2002.2218 [PMC free article] [PubMed] [CrossRef] Hesbacher S, Giez I, Embacher G, Fiedler K, Max W, Trawoger A, Turk R, Lange OL, Proksch P. (1995) Sequestration of lichen compounds by lichen-feeding members of the Arctiidae (Lepidoptera). Journal of Chemical Ecology 21(12): 2079–2089. 10.1007/BF02033864 [PubMed] [CrossRef] [Google Scholar] Jaeger CM. (2017) Phylogeny of Tortricidae (Lepidoptera): A morphological approach with enhanced whole mount staining techniques. MSc Thesis, Mississippi State University, Strakville, MS, 117 pp. [Google Scholar] Katoh K, Standley DM. (2013) MAFFT Multiple Sequence Alignment Software Version 7: Improvement in Performance and Usability. Molecular Biolology and Evolution 30(4): 772–780. 10.1093/molbev/mst010 [PMC free article] [PubMed] [CrossRef] [Google Scholar] Maddison DR, Maddison WP. (2017) Chromaseq: a Mesquite package for analyzing sequence chromatograms. Version 1.3. http://mesquiteproject.org/packages/chromaseq Maddison WP, Maddison DR. (2018) Mesquite: a modular system for evolutionary analysis. Version 3.40. http://mesquiteproject.org Miller MA, Pfeiffer W, Schwartz T. (2010) Creating the CIPRES Science Gateway for inference of large phylogenetic trees. Proceedings of the Gateway Computing Environments Workshop (GCE). New Orleans, Louisiana, 1–8. 10.1109/GCE.2010.5676129 [CrossRef]

Powell JA, Opler PA. (2009) Moths of Western North America. University of California Press. Scott CH, Zaspel JM, Chialvo P, Weller SJ. (2014) A preliminary molecular phylogenetic assessment of the lichen moths (Lepidoptera: Erebidae: Arctiinae: Lithosiini) with comments on palatability and chemical sequestration. Systematic Entomology 39: 286– 303. 10.1111/syen.12047 [CrossRef] [Google Scholar] Schmidt BC, Sperling FAH. (2008) Widespread decoupling of mtDNA variation and species integrity in Grammia tiger moths (Lepidoptera: Noctuidae). Systematic Entomology 33(4): 613–634. 10.1111/j.1365-3113.2008.00433.x [CrossRef] [Google Scholar] Stamatakis A. (2014) RAxML version 8: a tool for phylogenetic analysis and post analysis of large phylogenies. Bioinformatics 30: 1312–1313. 10.1093/bioinformatics/btu033 [PMC free article] [PubMed] [CrossRef] [Google Scholar] Stehr FW. (1987) Immature Insects, vol. 1. Kendal V Hunt Publishing, Dubuque, Iowa, 754 pp. [Google Scholar] Warshall P. (1995) The Madrean Sky Island Archipelago: A planetary overview. In: DeBano, LH, Ffolliot PF, Ortega-Rubio A, Gottfried G, Hamre RH, Edmister CB (Eds) Biodiversity and management of the Madrean Archipelago: The Sky Islands of Southwestern United States and Northwestern Mexico. 1994 Sept. 19–23; Tucson, AZ. Tech. Rep. RM-GTR- 264. Fort Collins, CO: US Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experimental Station. Wiemers M, Fiedler K. (2007) Does the DNA barcoding gap exist? – a case study in blue butterflies (Lepidoptera: Lycaenidae). Frontiers in Zoology 4(8). 10.1186/1742-9994-4-8 [PMC free article] [PubMed] [CrossRef] Zahiri R, Lafontaine JD, Schmidt BC, deWaard JR. Zakharov EV, Hebert PDN. (2017) Probing planetary biodiversity with DNA barcodes: The Noctuoidea of North America. PLoS ONE 12(6): e0178548. 10.1371/journal.pone.0178548 [PMC free article] [PubMed] [CrossRef]

Articles from ZooKeys are provided here courtesy of Pensoft Publishers